UNITED STATES DEPARTMENT OF THE INTERIOR FISH AND WILDLIFE SERVICE ^-y.,.--- yy^ FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE VOLUME 61 1. > ^ BULLETINS 178 TO 197 ISSUED BY THE FISH AND WILDLIFE SERVICE ^ 1960-1962 UNITED STATES DEPARTMENT OF THE INTERIOR FISH AND WILDLIFE SERVICE Bureau of Commercial Fisheries TITLE PAGE AND INDEX VOLUME 61 ISSUED 1963 4 U.S. GOVERNMENT PRINTING OFFICE • WASHINGTON, D.C. • 1963 39- 5:3-60 CONTENTS OF VOLUME 61 Bullotiii No. Pae« ITS. Xatuhal History of the Sandkau Shakk, Eulamiu mUlxrti. By Stewiiit Sprinuer. (Issued November 1960) 1-3S 179. (Irowth of Bliefiv Tuxa of the Western' Xorth Atlantic. By Fiaiik J. Mather III iind Howard A. Schuck. (Issued November 1960) ISO. Feiin i)itv of Red Salmon' at Brooks axd Karlt'k Lakes. Alaska. By Wilbur L. Hartman and Charles Y. Conkle. (Issued Noveml)er 1960) LSI. Filefishes (M(inacanth)(la() of the Western North Atlantic. By Frederick H. Berry and Louis E. Voegle. (Issued April 1961 ),_ 61-109 1S2. Embryological Stages in the Sea Lamprey and Effects of Temperature on Development. By George W. Piavis. (Issued March 1961) 111-143 lt;3. Blood Properties of Prespawning and Postspawning Anadro- Mors Alewives {Alosa pseiHloharengun). By Carl J. Sindermann and Donald F. Mairs. (Issued November 1961) 145-151 1S4. Effects of Copper Ore on the Ecology of a Lagoon. By Kenneth T. Marvin, Larence M. Lansford and Ray S. Wheeler. (Issued November 1961) 153-160 185. Validity of Age Determination From Scales of Marked Aaieri- CAN Shad. By Mayo H. Judy. (Issued January 1962) 161-170 ls(i. Calanoid Copepods From Equatorial Waters of the Pacific Ocean. By George D. Grice. (Issued February 1962) 171-246 1S7. Abundance and Distribution of Eggs and Larvae and Survival OF Larvae of Jack Mackerel {Trachurus syminetricus). By Da\ad A. Farris. (Issued May 1962) 247-279 188. DiSTRIBU'TION AND ABUNDANCE OF SkIPJACK IN THE HaWAII FISH- ERY, 1952-53. By Herbert H. Shippen. (Issued June 1962) 281-300 189. Abundance and Age of Kvichak River Red Salmon Smolts. By Orra E. Kerns, Jr. (Issued February 1962) 301-320 190. Early Developmental Stages of Pink Shrimp, Penaeus duorarum, From Florida Waters. By Sheldon Dobkin. (Issued January 1962) / 321-349 I'.il. Serological Studies of Atlantic Redfish. By Carl J. Sindermann. (Issued January 1962) 351-354 192. E.stim.\ting Red Salmon Escapements by Sample Counts From Observation Towers. By Clarence Dale Becker. (Issued March 1962) 355-369 193. Atlas of the Oceanographic Clim.\te of the Hawaiian Islands REGiON'j,,*'^«7RiLtttsr R. Seckel. (Issued Aui^ust 19()2). 371-427 ''^'li^^-'^ 81892 IV CONTENTS L^ Bulletin No. Page - 194. IXFLUEXCE OF E.\RLY MATURING FeM.\LES OX REPRODUCTIVE PoTEX- Ti.-^L OF Columbia River Blueback Salmon {Oncorhynchus nerka). By Ricliard L. Major and Donovan R. Craddock. (Issued June 1962) 429-437 195. Determixixg Age of Youxg Haddock From Their Scales. By Albert C. Jensen and John P. Wise. (Issued June 1962) _ 439-450 196. Development, Distribution, and Comparison of Rudder Fishes Kxjphosus sectatiix (Linnaeus) and K. incisor (Cuvier) in the West- ern North Atlaxtic. By Donald Moore. (Issued June 1962)_. 451-480 "V 197. Raft Culture OF Oysters in Massachusetts. By William N. Shaw. ^ (Issued June 1962) 481-495 /' UNITED STATES DEPARTMENT OF THE INTERIOR • Fred A. Sea ton, Secretary FISH AND WILDLIFE SERVICE • Arnie J. Suomela, Commissioner NATURAL HISTORY OF THE SANDBAR SHARK EULAMIA MILBERTI By Stewart Springer FISHERY BULLETIN 178 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 PUBLISHED BY U.S. FISH AND WILDLIFE SERVICE • WASHINGTON • 1950 PRINTED BY U.S. GOVERNMENT PRINTING OFFICE, WASHINGTON. D.C. For sale by the Superintendent of Documents, U.S. Government Printing Office, WashinHIon IS, D.C. Price 35 cents The series, Fishery Bulletin of the Fish and Wildlife Service, is cataloged as follows: U. S. Fish and Wildlife Service. Fishei'y bulletin, v. 1- Washington, U. S. Govt, Print. OtT., 1881-19 V. in illus.. maps (part fold.) 23-28 cm. Some vols. Issued in the congressional series as Senate or House documents. Bulletins composing v 47- also numbered 1- Title varies : v. 1^9, Bulletin. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, V. 1-23) 1. Fisheries— U. S. 2. Fish-cultur«^-U. S. SH11.A25 639.206173 Library of Congress coOrSoblj I. Title. 9—35239* CONTENTS Fae« I ntroduction 1 Nomenclature 2 Populations of Eulamia milberli 3 Ecological and systematic relationships of the genus Eulamia 4 Specimens examined 5 Field recognition 6 Distribution of Eulamia milberli 10 General nature of distribution 10 Factors affecting distribution 12 Nursery grounds and distribution of young 13 Distribution off Atlantic coast of the United States and in eastern Gulf of Mexico 14 Distribution in Bahamas and West Indies 16 Distribution in western Gulf of Mexico and western Caribbean 17 Migration 19 Reproduction 20 Courtship and mating 20 Time of mating 21 Development of the embryo 22 Number of young in litter 23 Length of young at birth 23 Abnormal embryos 25 Sex ratios 25 Growth and size at maturity 26 Food and feeding habits 29 Abundance 32 Enemies 32 Summary 34 Literature cited 36 ni ABSTRACT Populations of sandbar sharks of the eastern and western parts of the Atlantic Ocean are defined and general problems of nomenclature, the ecology of large carcharhinid sharks, and field recognition of sandbar sharks are discussed. A more-detailed account of observations on Eulamia milberii, restricted to the population of the western North Atlantic, is given, outlining distribution of adults and young, migrations, development, and behavior, based on observations from the commercial shark fishery which operated from centers in the Southeastern States from 1935 to 1950 and supplemented by data from research vessels operating after 1950. Comparisons with other species in the area, lists of large species of sharks taken at certain times off Salerno, Florida; Bimini, Bahamas; the mouth of the Missis- sippi River, Louisiana; and the Caribbean coast of Nicaragua-Costa Rica, as well as discussions of interspecies competition, are included. NATURAL HISTORY OF THE SANDBAR SHARK, EULAMIA MILBERTI By Stewart Springer, Fishery Methods and Equipment Specialist Bureau of Commercial Fisheries This account of the sandbar shark, Eulamia milberti ( Miiller and Henle) , is an attempt to bring together all the significant information on one kind of common and moderately large shark. Sharks have been studied because they are occa- sionally dangerous to man, often a nuisance to fishermen and, in the past at least, have been valuable as a source for food, leather, vitamin A, fish meal, and some specialty products. A rather comprehensive body of knowledge exists about some of the smaller species, such as the compara- tively valuable soupfin shark of the coast of North America (Ripley, 1946) and the school shark of Australia (Olsen, 1954), botli si)ecies of Galeorh'mus^ and the common spiny dogfish, Squalus (Ford, 1921; Hickling, 1930; Temple- man, 1944). Information on the natural history of the larger species is fragmentary. This is to be expected, because large species not only are difficult to catch and handle, but also are far- ranging and require observation over a wide geo- graphical area. The sharks, together with their relatives, the skates, rays, and chimaeroids, form a class of vertebrates that is sharply set off from the classes which contain the fishes, amphibians, reptiles, birds, and mammals. The sharks and other mem- bers of the class Chondrichthyes have cartilagi- nous skeletons, and wliile elements of the shark skeleton may become calcified, no true bone is formed. This is the basis for the definition that is generally used to distinguish the Chondrich- thyes from the higher vertebrates. But there are other differences in the chemistry and physiology that are very likely of great importance but are little understood. The evolutionary connections of the modern sharks and their allies with other modern vertebrates are obscure and of great an- tiquity. Note. — Approved for publication, October 27, 1958. Fishery Bulletin 178. Sharks occupy a place in nature at the top of tlie food chain. As predators they compete with man, but it is by no means established that their predatory activities ai'e always harmful. They are a nuisance or are harmful to fishermen chiefly because of the damage they do to nets or to fish that have been caught on setlines. In some locali- ties, in England and Australia, for example, sharks are utilized and are consequently of some value. In the United States, landings at present are of no great importance. Sharks may be dangerous to man through at- tacks on swimmers and survivors of marine dis- asters, but EuJamia milberti is not a species implicated in well-documented records of attacks. There may be several reasons for this. E. milberti ordinarily stays away from beaches and does not often feed at the surface. It usually seeks small prey. During the summer, when the female sand- bar sharks come inshore near the heavily popu- lated centers from New York southward along the Atlantic coast to give birth to their young, it is not their habit to seek food. The large males do not come inshore. So the sandbar shark, while large enough to be dangerous and perhaps the most common of the larger sharks in shore waters .southward from New York, is isolated by its habits from encounters with man. Nevertheless, the sandbar shark is potentially dangerous to man and might become a more serious danger with minor shifts in the environmental situation. The most annoying aspect of my work with sharks, prior to the publication in 1948 of the first volume of Fishes of the Western North At- lantic on sharks by Dr. Henry B. Bigelow and "William C. Schroeder. was that many we.'^tern Atlantic sharks could not be identified with con- fidence because of a scattered literature of vary- ing (]iiality. It is ai)]iroi)riate (liat I acknowledge tlie importance of this excellent general work to me, because without it ;uul witlioiit the encourage- 1 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE ment of its authors. I would not have attempted preparation of this report. Dr. Richard H. Backus of the Woods Hole Oceanographic Institu- tion, Dr. Cxiles AV. Mead of the U.S. Fish and Wildlife Service, and Dr. Leonard P. Schultz of the U.S. National Museum made many helpful suggestions during the preparation of this report. Captain B. W. Winkler was especially helpful in keeping fishing logs and measurements of about 1,300 large sharks he took off the Bahama Banks and off Nicaragua. Records obtained while I was employed by the Shark Industries Division of the Borden Company and while I was aboard the exploratory fishing vessel Oregon of the Fish and Wildlife Service comprise the basic data used here. Special assistance was given me also by the Lerner Marine Laboratory of the American Mu- seum of Natural History, by permitting 2 months of field study at Bimini, Bahamas, in the summer of 1948. In all my work with sharks, I have been given the most generous help by my associates in commercial shark fishing and aboard exploratory fishing vessels. NOMENCLATURE This report is not intended to settle problems of nomenclature and taxonomy, but to be useful it is necessary to name the sharks under discus- sion and to define the names used. My choice of a name for the sandbar shark is Eulamia miJhertl (Miiller and Henle) 1841. Use of Eulamia fol- lows my partial revision of the carcharhinids (Springer, 19.50). For the specific name milherti I follow Bigelow and Schroeder (1948) who note that, if it is finally proved that the ^Mediterranean form is identical with the American, the name pJumbeus Nardo 1827, must be used for the com- bined species in place of milherti. I disagree, however, with Fowler (1936), with the preceding statement l)y Bigelow and Sclirne- der, and witli Toi-tonese (19.51, 19.5(i) that Nardo's description is valid. The description by Nardo would api)ly to almost any carcharhinid and the specific mention of the rounded snout ' would apply better to some other carcharhinids than to the sandbar shark. Because there is no type and because Nardo's description would apply to al- 1 The total description .Tnd dinpnosis of ^(jtinltin i>lunihriis liy Nardo, 1827. p. 35, is as follows : "Speeiei seciindae eonvenit exacte Squal. Glaucim. Bloc, si color exciperetiir I't fiiini:i Toatri quae in exemplar! nostro rotunda est." most any carcharhinid if applied liberally but to none if applied strictly, I regard Squahis plumbeus as a nomen nudem. I am also unable to acce^rt Nardo's description as .specifically applicable to the sandbar shark based on the argument that the sandbar shark is the mo.st common large carcharhinid of the Adriatic. A most extraordinary snarl has developed over the years in tlie determination of the scientific name to be applied to the sandbar shark. The origin of this complication probably lies in the peculiarities of the distribution of species of carcharhinid sharks along the Atlantic coast of the Ignited States. Mistakes in identification of specimens have been frequent, probably because the descriptive accounts of the early authors were very brief and did not select truly diagnostic features for emphasis. Systematists had too few specimens and too little data on distribution to note that segregation of the sexes and segregation of the adults and young characterized these sharks at some seasons. In the latitudes from New York to Chesapeake Bay at depths wnthin easy reach of collectors or fishermen, two common large carcharhinid sharks occur, the sandbar shark, E. milherti (Miiller and Henle), and the bull shark, Oarcharhimts leucas (Miiller and Henle). The sandbar shark is rep- resented in this area by adult females and by the young of both sexes, but rarely by adult males in the observable elements of the population. The hull shark is represented usualJy by adult males, but females and young are also present sporadi- cally. - The ranges of the sandbar shark and the bull shark will be discussed later as well as the ap- parent competition between these species. An effect of the occurrence of the two species to- ^LarKP male Carrhnrhinux leucas were reported from the Chesapeake Bay area by Schwartz (19.58) : one was taken from the Patnxent River in 1957 and another at Flag Pond in the summer of 1958. This is apparently the first published report of the species from Chesapeake Bay. Specimens of large sharks came to the attention of Edgar H. Hollis. of the Maryland De- partment of Tidewater Fisheries in 1957, because Chesapeake flshermeii regarded them as rarities. Photographs of the speci- mens sent me by Mr. Hollis were sufficient to permit Identifica- tion as C. leiicns. Nichols (1918) and Nichols and Breder (1927) reported C. leucas from the south shore of Long Island, noting that these specimens were large males. Attention Is called to this parenthetically because the appearance of adult males at the periphery or in the cooler parts of ranges of carcharhinid sharks is frequent. NATURAL HISTORY OF THE SANDBAR SHARK gether. liowever. has been to foster confusion in the nomenclature. There apjiears to have been a tacit assumption by some naturalists that sexual dimori)hism accounted for differences in sandbar sharks and bull sharks despite recognition of tlie existence of both species. Superficially, the sandbar shark and the bull shark i-esemble one another but, as will be shown later, sandbar sharks can easily and positively be separated from bull sharks on the basis of several anatomical characteristics. Identification of specimens from the Middle Atlantic States, par- ticularly the Atlantic coast from Cape Cod to Chesapeake Bay, presents an added difficulty be- cause of the several very similar offshore species which may be caught occasionally, but probably rarely in inshore waters: Eulnmia obsciira (Le Sueur), E. faJcifo-rmis (Miiller and Henle), E. fo-ridana (Bigelow, Schroeder, and Springer), and E. ultima Springer. Recent unpublished records of the occurrence of E. inUherti young and of the occurrence of Carcharhinus leucas adults are rather numerous, and following publi- cation of the first volume of Fishes of the West- ern North Atlantic (Bigelow and Schroeder, 194S) there appears to be little confusion of the two species. Belaboring the point tliat descriptive accounts of carcharhinids must be detailed and selective to have any meaning seems necessary to affect the intrenched misconceptions about E. jnUherti that can be derived from the literature. Fowler's (lO-'iG) description of E. plumbeus (plumbeu-s = mUherti), which was based on American Middle States exam])les, although in a report on West African marine fishes, is not unique in confusing E. mUherti with another species,^ but it is de- tailed enough to be especially vulnerable to criti- cism. More elements of his description fit the bull sliark, Carrharhhwis leucoK. than E. m/lberfi. but additional confusion is inti-oduced by the probability that juvenile and adult characteristics of both species are mixed. There is no selection of diagnostically useful characteristics for em- phasis. The result is a plausible literary syn- 3 Garman'8 illustration of CnrcJitirhinus phityoilon (1913. pi. 3. flgs. 4 and fi) appears to bo very weli drawn from a sppcimen of Eulnmia tnilberti. Bigelow and Scliroeder (1948) note tlmt Carman's illustration is mislabeled. Tbe acoompanyinR illustra- tion of the teeth In Carman's plate .3. figure .5. appears to have been drawn from the teeth of platyodon. CnrchnrliinUH iilalii- odon (Poey) Is a synonym of f. Irurnii (MUUer and Henle). thesis that is a hazard to one attempting to fit a real shark to a position in zoological classification. POPULATIONS OF EULAMIA MILBERTI Sandbar sharks. Eulmnia milberti, occur in portions of the temperate and tropical Atlantic, the Caribbean, the Gulf of Mexico, and the Medi- terranean. Our data in this study primarily cover the population inhabiting the western North Atlantic, the Gulf of Mexico, and the eastern Caribbean. Later, the normal movements and distribution of this population will be discussed. We need fii-st to consider the relationships of the various populations. In addition to the popula- tion of the western North Atlantic two others may be roughly defined. One occurs along the coast of South America from Trinidad eastward and southward. The other is found along the west coast of Africa and is presumed to be con- tinuous with the stock entering the Mediter- ranean. The population occurring on the coast of South America appears to be a minor one. The species has been reported and figured by Ribiero (1923) from the coast of Brazil. While engaged in commercial shark fishing in April and May 1949, I made shipboard examina- tions (Springer. 1949) of a series of sandbar sharks from the north and east coasts of Trinidad and identified them in error as E. phmbem. I now believe that differences between the Trinidad specimens and typical milberti from the Atlantic coast of the United States are insufficient to war- rant recognition of sejiarate species, and that the conservative course, jiending accumulation of new data, is to look upon the various Atlantic sandbar sharks as representing a single species. At the Trinidad locations, adult males and females as well as young of all sizes from 4 feet upward were taken on single setlines. Although com- mercial shark fishing was carried on throughout 1949 from the coast of French Guiana westward to the Gulf of Venezuela, sandbar sharks were reported only from the north and east coasts of Trinidad and chiefly from depth-s of 5 to 20 fathoms near Galera Point. Tliere are no records of sandbar sharks from tlie West Indies north of Trinidad except from the north coast of Cuba and from the western inii't of the Bahama Banks. This is not, of 4 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE course, conclusive evidence of their absence from a region so poorly known ichthyologically as the West Indies. Nevertheless, all of the evidence points to a discontinuous distribution with no regular contact between the population known from Cape Cod to Costa Rica and the South American population known from Trinidad and the east coast of South America south of the Amazon. In connection with possible future work with the sandbar sharks, it should be noted that the E. milherti from Trinidad were taken in eddies of the very strong, westerly current flowing be- tween Trinidad and Tobago; and t^ii^ recruit- ment for this stock could take place in part by tran.sport by tlie Equatorial Current of the grow- ing young to Trinidad from shore waters of the African coast. The stock of Eulam/a mllberti in the eastern Atlantic is known from scattered records from the Mediterranean and the west coast of North Africa as summarized by Tortonese (1956). These records cover a long period of time and although critically reviewed by Tortonese and unquestionably accurate, they give little basis for an estimate of the abundance of the sandbar shark in relation to the abundance of other large species of the area. For our purposes hei-e, that is, to estimate the relative importance and abun- dance in comparison with other large sharks of the area, reports by Cadenat (1950, 1957) on E. milherti and other species from the coast of Senegal are quite informative. Cadenat has been able to make observations on fresh material from a fishery taking relatively large numbers of the larger species of shai'ks. His reports suggest that the stock of E. miThei-ti off northwest Africa is a strong one. The list of species of large sharks reported by Cadenat is quite similar to lists of large sharks from the southeastern coast of the United States. The endemic species of both areas are the smaller sharks. Precisely the same factors of prevailing wind and surface currents that make the southern crossing from the North African coast to Trini- dad easier for man when it is from east to west may be expected to ojierate for sharks. Simi- larly, for a more northerly crossing, the one from west to east is more easily followed. The postu- late that such contacts as exist between the stocks of the western Atlantic and the stock of the eastern Atlantic result from exchanges following this general clockwise circulation is a reasonable one. No actual evidence of regular contacts be- tween the three stocks exists, however, and there is substantial reason for the belief that move- ments of individual sharks from one stock to an- other are relatively infrequent occurrences. Knowledge of the distribution of large sharks in oceanic situations at considerable distance from land was extremely meager until very re- cently when data from oceanographic vessels and exploratory fishing vessels became available. The niost comprehensive study covers sharks of the Central Pacific (Strasburg, 1958) in which data .showing patterns of distribution of some of the larger species is given. Before the appearance of that study and of a less comprehensive account of Atlantic pelagic sharks (Backus et al., 1956), questions of shark distribution seaward of the continental shelves were unanswerable. Now, while it is known that neritic species of large sharks are capable of moving over great distances of open ocean, there is increasing evi- dence that they rarely do. , \ ECOLOGICAL AND SYSTEMATIC RELA- TIONSHIPS OF THE GENUS EULAMIA The genus EiiJ-amia may be divided into two groups on the basis of the structure and arrange- ment of the dermal denticles. The group to which E. milherti belongs is characterized by noninibricate denticles as contrasted with the other group which has denticles with overlapping edges or points. The milherti group includes comparatively few species. Probably Eulamia dussumieri. (Miiller and Henle) and E. japonicm (Schlegel) of the western Pacific belong here. In the Atlantic, the group is represented by a deep- water species, Enlamia ultima Springer, which is quite different from milberti, not only in its mor- phology but in its habitat. Aside from E. altima, the only Atlantic representative of the genus Eulamia (or any carcharhinid genus) with widely spaced, noninibricate denticles is Eulamia milherti. the sandbar shark. E. milhfrt) lias the shoalest range and occupies the most inshore habitat of any of the 5 or 6 species of Evlam/n of the Atlantic coast of Nortli NATURAL HISTORY OF THE SANDBAR SHARK Anicrii'u. Altlioufih milherti may be in compi'li- tioii with other species of Euhimia for food in some parts of its range, it does not compete with otlier si)ecies of Enlam'ia for nursery grounds. Of the other carcharhinids of the northwestern Atlantic, the genera Prionace and Pterulamiops are pelagic surface dwellers; Hypoprion is con- fined to waters generally deeper than 100 fathoms near shelves or banks; Negapnon, Apriondon, ScoUodon, and Carchcirhinus are shallow water sharks that spend at least some part of their lives in shallow lagoons, river mouths, or estuaries, and venture into deeper water rarely except for transitory movements; the species of Eulamia are sharks of the continental shelves, oceanic banks, and island terraces, although some species extend their langes well insliore and also for consider- able distances beyond the limits of the Conti- nental Sjielf. The only other western North At- lantic carcharliinid genus, Galeocerdo^ is repre- sented by a single species in subtropical and ti'opical waters out to depths of at least 200 fathoms. It does not exhibit the specialized schooling habits of the other carcharhinids. shows no strong migratory tendencies, and is less re- stricted in habitat choice than the others. There seems to be a tendency to greater variation in the number of young produced as well as a greater number per litter in Galeocerdo and Prionance and possibly also in Pterolamiops than in other northwestern Atlantic carcharhinids. Insofar as is known, there are no very important differences in the general outlines of the life history patterns of Necjapncm^ Apnonodon. Scoliodon, Carchar- himif!, and EitJamia, although there appear to be many differences in detail. Barriers which may restrict the movements of the larger sharks including Eulamia mUherti are not i-eadily apparent. Occasional captures of sharks outside areas of normal concentration of the species prove that they can and do wander. The remarkable thing is that large sharks tend to remain within definable iial)itats and geographi- cal ranges. Since species of EuJnnna are, in general, less dependent on land masses than Carcharhimis and extend their activities regularly to surface waters of the open ocean beyond the Continental Shelf. it would not be surprising to find that some species have a very wide distribution in temperate 55250S 0—60 -2 and tropical seas. Euhimia fiaridana (Bigelow, Scliroeder, and Springer) ma}' be an example of such a distribution (see Strasburg, 1958). Those species of Eulamia, such as milherti, which are tied to shallow-water habitats are presumably subject to a greater degree of isolation. SPECIMENS EXAMINED Specimens, records, and field observations for this report have been assembled over a period of about 25 years during which time I have exam- ined several thousand sandbar sharks. Available records of the commercial shark fishery cover more than 100.000 adult Eulamia milberti. About half of these sharks were measured at the point of landing. Earlier records of the stations in- cluded specimens of Eulamia altima and Eulamia foridana under the heading sandbar sharks. Since I visited most of the stations frequently, and during part of the period between 1935 and 1950 supervised recording procedures, I saw rela- tively large numbers of sandbar sharks. Speci- mens which appeared unusual to station employ- ees were retained when practicable for my inspec- tion. ]\Iost of my observations were made along the coasts of southern Florida. Adequate num- bers of specimens for some purposes have been examined from the eastern and northern parts of the Gulf of Mexico, the Atlantic coast of the United States south of Cape Cod, and from the Caribbean coast of Nicaragua and Costa Rica. The available material in several museum col- lections in the United States was studied, but this consisted chiefly of preserved embryos or very young sharks and dried jaws. The collection of data in the shark fisheiT suf- fered from interruptions and was assembled to aid an industrial operation rather than for a biological study. The difficulty in handling speci- mens, averaging nearly 7 feet in length as adults with an average weight of about 135 pounds, has made it necessary to select different series or samples for diffei'ent objectives: one sample for length-weight relations; another for tooth counts, and so on. 1 was unable to find spirit-jireserved specimens of eastern Atlantic or Mediterranean origin re- ferable to either E. plumbeus or E. milberti dur- ing a hasty examination of catalogs and specimens at the Museum d'llistorie Naturelle in Paris or in 6 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE the British Museum of Natural Histoiy, aUhough specimens from the western Atlantic were pres- ent. Tortonese noted (1938) that there are two specimens in the Musee di Trieste collected in 1869 and 1871 and, after examination of speci- mens labeled inilberti from the western Atlantic in the Museum at Paris and the British Museum, he indicated (in 1951) that he regards milhertl as a synonym of phi?nieus. FIELD RECOGNITION Eulamia mWbei'ti is commonplace in appear- ance. It has neither unusual color markings nor spectacular structural features. The length of the shark at maturity, 7 feet, makes the species too large for the biological collector and too small to interest the journalist. It is necessary to search for distinguishing features (fig. 1). The opportunity for comparison of series under ordi- nary circumstances is negligible, and almost all identifications of the larger sharks are necessarily made in tlie field. The suggestion that many of tlae presently recognized species of carcharhinid sharks are not in fact separable from one another but .should be regarded as unidentifiable parts of a species complex has been advanced in specula- tive conversation by some of my friends who are ichthyologists. This view may easily develop from unsatisfactory attempts to make identifica- tions with methods which are quite adequate and successful in application to teleosts but fall short when applied to sharks, and particularly to car- charhinid sharks. E. milberti, in waters off the United States, is readily defined and problems concerning it are not complicated by the existence of geographic or environmental races or sub- species, insofar as the available evidence shows. This is apparently not true of some of the other carcharhinids where separate populations may be defined on the basis of morphological diflferences shown in the analysis of adequate series from different areas. The keys and descriptions given by Bigelow and Schroeder (19-18) are adequate for the identifica- tion of the carcharhinid sharks of the western North Atlantic excepting Eulamia altima, which was described (Springer, 1950) after publication of this work. Nevertheless, identifications need to be made carefully because of the general struc- tural similarity of the species which look alike on superficial examination. Sharks of the genus Eulainia in the falciforjnh-springeri group are Figure 1. — Eulamia milhcHi in .an exhibition tank. Tho high, triangular first dorsal fin, nonfalcate pectoral fins, and relatively high second dorsal and anal fins, nearly equal to one another in area, are characteristic of the species. (Phor tograph courtesy of Marine Studios, Marineland, St. Augustine, Fla.) NATURAL HISTORY OF THE SANDBAR SHARK not well known and possibly are incompletely de- fined. Minor differences between Atlantic and Gulf of Mexico populations of E. ohscura need further study. There is no difficulty, however, in distinguishing E. milherti from these species or from other species of sharks ordinarily found within its geographical range. The importance of determining the presence or absence of a middorsal ridge (a low ridge in the skin extending for all or a part of the distance between the first and second dorsal fins) for the identification of carcharhinid sliarks cannot be overemphasized. This minor structural feature is certainly nonadaptive and its usefulness as an indicator of probable relationships should be great (see Springer, 1950: p. 1, and Backus, Springer, and Arnold, 1956: p. 180, for discus- sion). The first mention of this characteristic in publislied work was by Nichols and Breder (1927), but correct identifications of tlie common large ground .sharks of the east coast of the ITnited States were made by Nichols and by Rad- cliffe independently before lOlfi. In one of the more valuable papers on sharks, Radcliffe (1916) made the first general use of the structure of the dermal denticles to show differ- ences in western North Atlantic carcharhinid species; and his illustrations show clearly the distinctive denticle type and arrangement which sets E. m/'lberti off from other carcharhinids within its range, except for the newly described E. alt I ma. Both E. miJberti and E. aJtima differ from all other North American carcharhinid sharks in having nonimbricate denticles without strongly projecting points; however, the denticles of E. ultima are mucli smallei- tlnin tliose of E. 7nUherti. Commercial shark fishermen at Salerno and Kej* West, Fla., recognized altlma as distinct from the sandbar .shark and called it the bignose .shark or Knopp's shark before it received a scientific name. The diagnosis given with the original de- scription of E. althna (Springer. 1950) should be adequate for the determination of specimens of all sizes. All of the known examples of E. nltima have been taken at depths of 50 to 150 athonis off Salerno, Florida, in the Straits of Florida, in the northeastern Gulf of Mexico, and n the Dragon's Mouth between Trinidad and Venezuela. Its vertical range overlaps that of tiie shallower water E. milberti in the Straits of Florida aiea and extends well into the nighttime range of tlie night shark. Ilypapnon signatus Poej'. The geographical range of E. altlma may be quite extensive, but it is unknown because comparatively little fishing has been carried out at the depths whei-e this species might be expected to occur. Such fisliing as has been done in mid- water and just beyond the edges of the Conti- neTital Shelf by connnercial shark fishermen indi- cates that the species is relatively common. Probably many more E. altima would have been taken by the connnercial fishery were it not for the fact that in the Florida-Caribbean region the liver oil of althna is characteristically lower in vitamin-A content than is that of any of the other species of Eulamio or of Hypo prion in that area. In a large measure, the confusion in the nomen- clature of the larger American carcharhinids that existed before tlie publication of the 1948 work by Bigelow and Schroeder would undoubtedly have been avoided if descriptive literature had included information on tlie presence or absence of the middorsal ridge. A fine replica of a shark, wliicli in the light of the better descriptions now available can easily be identified as Carcharhinvs leucas (Midler and Henle), a species without a middorsal ridge, is shown in an illustration in an informative article (Rockwell, 1916: p. 161) under the caption Carcharhinus obscurus {Eula- inia obseura), a species with a middorsal ridge. Determination of the presence or absence of the ridge is sometimes difficult, particularly for mu- .seum specimens or for specimens that have been exposed to the sun for a long period. Although identifications can be made without reference to the ridge, they are likely to be difficult and use of all of the available ditferentiating character- istics is desirable. The confusion of the sandbar shark with the bull shark extends to the Pacific. References liave fi-equently been made to tlie sandbar shark. Eulamiii milherti, as occui'ring on the Pacific roast of Panama. There is no evidence of this and tlie species probably is not found there. Gar- man's (1913) synonymy of milherti included Eulamia nicarafjuensifi (till and Bransford, the fresh-water bull shark. The two bull sharks. nicaraguensis and leucas are so similar to one I 8 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE another that their separation is doubtful. There is a Pacific species so similar to nicaragitensis and leucas that conunercial shark fisliermen who fished on both coasts claimed they were unable to dis- tinguish one from the other except by area of capture. I do not know the scientific name for the form if it has a name, but whatever the species, it is like nicamgucnsis or leucas and not like mUherti. Meek and Hildebrand seemingly had difficulties with this one in Fishes of Panama (1923), wherein they discuss a Pacific species Carcharh'mus azureus (Gilbert and St arks) as a synonym of rrdlberti. But Meek and Hildebrand did not see a specimen from the area which they themselves could identify as inUberti and the significant sentence in their treatment states — We certainly must regard the present arrangement as tentative only, for more specimens must be compared before the true affinities of the specimens from the op- posite coasts can be established. A paper by Rosenblatt and Baldwin (1958) on some of the carcharhinids of the eastern Pacific presents for the first time information on the presence or absence of the middorsal ridge in Pacific species. This is an important contribution and includes more comprehensive descriptions than have hitherto been available for sharks of the eastern tropical Pacific. These authors find the separation of Eulaonia from Carcharhinus unacceptable for Pacific species. In support of this an vuifortunate choice of illustrative argu- ment is used. They say C. altima, for example, has a definite dermal ridge but teeth which are as narrow as those of any member of the smooth- backed group (Springer, 1950). This is an error. The teeth of altma in the upper jaw are similar in general shape to the teeth of the other species of Eulamia. These authors logically call atten- tion to the ill-assorted group left in the genus Carcharhinus by my 1950 revision, mentioning leucds and velox as examples. I am in complete agreement with this but find no cogent argument for the elimination of Eulamia^ since the species of Eulamia as resti'icted are remarkably similar to one another in all of their morphological fea- tures. The sharks allied to the genus Carchar- hinus are far too widespread and numerous and there is far too little known about them for an adequate study of the entire group. Additional revisions of the group are needed. Differences between adults of E. altima and inilhertl are quite apparent in field examination wlien the two are seen side by side. The snout of altima is longer and notably thicker dorsoven- trally. Furthermore, the first dorsal fin in E. altima looks quite different because it is not quite so far forward as in E. milberti and is neither so erect nor so high. The high and erect dorsal fin of E. milierti in a forward position (fig. 2) is a reliable and adequate character for field recogni- tion of adults in the water, if the size of the shark is taken into consideration. Gill (1862) based his classification of the car- charhinid sharks almost entirely on the structure of the teeth. His arrangement of genera was not satisfactory and it is apparent that short descrip- tions of shark teeth are inadequate and lead to confusion even though the number and form of the teeth show comparatively little variation within species and are of considerable diagnostic value. The persistence of essentially similar shape and structure in the successively larger teeth appearing in some carcharhinid sharks as they grow has been fairly well established by ob- servation. In E. milierti, at least, this appears to be true, although this is neither universal among sharks nor adequately demonstrated for many species. To obtain some verification of the extent of variation in the number of tooth rows in car- charhinid sharks, I took advantage of a situation requiring the preparation of several hundred clean, dry shark jaws for a commercial order. I carefully identified the sharks and tagged the ^^jaws of a series of 110 E. milierti together with all other sharks appearing at the same time on the dock at Salerno, Fla. All of the milberti and most of the other sharks were adults ; sex was not noted. After the jaws were cleaned I counted and recorded the number of tooth rows (table 1). To the extent thut this sample represents the population of milberti, the counts of rows of teeth indicate that variation is small in that species. The shape and the relative position of the fins in carcharhinid sharks are reasonably useful characteristics for identification. Small differ- ences in the size of fins or even in their positions, however, are of comparatively little value because of differential growth and the diverse trends this ^^^r NATURAL HISTORY OF THE SANDBAR SHARK *i: Figure 2. — Eulamia milberli turning in front of the camera at Marineland, Fla. Note that the pectoral fins are quite broad at their bases, relatively pointed and not strongly concave on their trailing edges. With the exception of the caudal fin, all fins function as rudders or stabilizers and cannot be used independently i for locomotion The pectorals provide lift to offset the lift of the asymmetrical caudal since without a forward lift-the shark would tend to somersault. The large stiff fins in forward positions reduce the ability of this species to roll and twist but may be expected to increase the precision of its forward swoops at creatures on the sea bottom. (Photograph courtesy of Marine Studios, Marineland, Fla.) growth may take in different species. Data to show adequately the differential growth in car- charhinids are lacking. But one example will suffice to show how unreliable proportional meas- urements can be for comparisons between species in which specimens of different sizes and ages are involved, and in which the growth patterns of the species l)cing compared are mikncnvn. In three examples of young ^- milberti, 685. 680, iiiid r)?);") mm. long, from the vicinity of Woods Hole, Mass.. the lengths of pectoral fins (meas- ured on their outer margins or leading edges) are 16.2, 15.0. and 15.0 percent of the total length of the sharks. In three adult milberti from oflt Englewood. Fla., 2.210. 2,070, and 2,240 mm. long, tlte pectoral fin lengths are 21.3, 21.5. and 21.0 jiercent of the total length. Let us compare these ])roportions with measurements of pectoral fin lengths of the whitetip shark. Ptcrolamiops longiminiiis (Poey). A late-embryo whitetip 530 mm. long, taken l.''>5 miles off Xew Smyrna. Fla., has a pectoral fin length 22. (i percent of the total length; a young wliiteti]) 1.020 mm. long, taken olf Tumpico, Mexico, lias a pectoral fin 25.5 per- cent of the total length; and an adult whitetip, 2.310 mm. long, from the central Caribbean, has a pectoral fin 22.0 percent of the total length. The figures indicate a proportionately longer 10 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 1. — Tooth-row counts in carcharhinid sharks taken off Salerno, Fla., summer of 19^7 Number of specimens having tootlirow counts ' of— 21 22 23 24 25 26 27 28 29 30 31 32 33 34 36 36 37 EvXamia milberti (110 specimens): 3 20 37 44 55 30 11 3 1 6 12 8 29 13 12 E. floridana (48 specimens): 11 23 2 1 3 4 1 1 3 1 15 10 E. obscura (39 specimens): Upper jaw 3 15 1 1 15 3 3 1 CarchaThinus leucas (24 specimens): 2 18 5 3 C. Hmbatits (31 specimens) : 1 1 25 4 5 14 11 1 C. maculipinnis (13 specimens): 1 9 2 1 4 3 ? 2 Gakocerdo cuvier (21 specimens): 4 1 3 2 12 13 2 2 3 Negaprion brevirostris (8 specimens): 1 1 1 2 6 5 * All counts made from cleaned jaws from whicii all membranous sheatliing had been removed to permit accurate counts whether or not teeth of the functional row were missing. pectoral fin in adults than in young for E. ndl- ierti, but an entirely diiferent condition in F. longimanus. The sandbar sharks available to me were re- markably uniform in general appearance and in those features that I could measure, count, or compare. In an attempt to learn something from morphometries, a considerable number of milberti and other species were measured carefully and in detail. However, the principal value that I de- rived from this excessively laborious task was in the deliberate examination of specimens enforced by measurement of detail and in the notes made to accompany the measurements. The exercise served also to impress upon me the difficulties attending attempts to get adequate series to show growth patterns among some of the species of large sharks which are not only migratory but probably short lived. A characteristic of great importance for field recognition of specimens of carcharhinid sharks is the total length of the specimen considered in connection witli its .sex and maturity (fig. 3). The mammalogists and ornithologists have long considered total length important in identifica- tion because mammals and birds have determi- nate growth patterns. As will be shown later, E. milherti has growth characteristics which re- sult in adults of predictable size. Furthermore, the size range of adults within the known seg- ments of the population falls within limits which are narrow enough to facilitate field identifica- tion by process of elimination. Thus, an adult Eulamia more than 92 inches in total length is probably not milberti, and adult males less than 70 inches or adult females less than 72 inches in total length are unknown. DISTRIBUTION OF EULAMIA MILBERTI General nature of distribution The distribution of Eulamia milberti is diffi- cult to treat adequately because, even though further discussions will be limited to the popula- tion of the western North Atlantic, the distribu- tion patterns are extremely complex. The adults segregate by sex and to some degree have differ- ent vertical ranges. The nurseiy areas occupied by the very young sharks are free of adults ex- cept when the females come inshore to give birth to their young. The migratory patterns of young and adults differ greatly. Finally there is a well-defined principal range occupied by at least nine-tenths of the western Nortli Atlantic populaticm and an accessory range of uncertain importance. It is quite possible that the population occupying the accessory range is not self-sustaining and exists only because there is continuous but quite acci- NATURAL HISTORY OF THE SANDBAR SHARK 11 Corcharias lourus Caleocerdo cuvier Negopr/on brevirostris Aprionodon isodon Carcharhinus limbatus* Carcharhinus macutipinnis Carcharhinus leucas Eulamia milberti Eulamia altima Eulamia floridarya Eulamia obscura Sphyrna sp.** 115 ' '" y-i';""— 1 ?.? INCHES I L I _L _L J_ J_ _L J_ J_ J_ _L J_ _1_ _L _L J_ J_ _L 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 size range of embryos d- ^^ = size of adults Figure 3. — Comparative sizes of adults and joung of common large sharks found within the geographical range of E. milberti. The figures at the right indicate the number of specimens in the sample used to determine size range. The size ranges for embr_vos are estimates based on maximum observed lengths of embryos and minimun lengths of free- swimming young observed in Florida collections. (* Size range in Florida-Antillean specimens. Western Gulf of Mexico and Central American coast specimens are smaller and produce smaller young. The western stock may prove to be distinct and if so should take the name Carcharhinus natator Meek and Hildebrand. ** The great hammerhead of the West Indian region, following Bigelow and Schroeder (1948). The nomenclature is now unsettled. The name Sphyrna tudes is not available for the great hammerhead and probably should be replaced by Sphyrna mokarran (Ruppell).) dental recruitment from the principal population. Within the expected vertical and geographical range of the species are some areas which appear to be avoided. IL ia^^iell to^'uientioii a^*iit that the sandbar shark, like other large sharks, is not prevented by well-defined barriers from wander- ing out of its normal range. The limits of distribution are therefore not sharph- defined. Following the traditional pat- tern in descriptions of distribution it may be said that the sandbar shark, as rejiresented by the western Xorth Atlantic population, is common in summer off the Atlantic coast from Cape Cod to West Palm Beach, Fla., and in winter from the coast of the Carolinas around the tip of Florida to the gulf coast of Florida as far north as Tarpon Springs. It occurs uncommonly in the western part of the Gulf of Mexico and along the continental shores southward to Costa Rica. It is a casual visitor on the northern coast of Cuba and the western edges of the Bahama Banks. Its vertical range is from the shoreline out to li'?5 fathoms. It enters bay mouths but is not found in fresli waters. The ])rincipal source of information on the distribution of tlie sandbar shark comes from the commercial shark fishery. Atlantic 'coast shark fishernxen iised bottoiu setlines -Aiore often than 12 FISHERY BULLETIN OF THE FISH AND "WILDLIFE SERVICE any other gear. Each unit of gear consisted of a main line of chain or wire rope a half-mile or more in length. This was set on the bottom, anchored at both ends with the anchors rigged with buoys so that the lines could be retrieved. Short, branch lines made of chain, each with a baited hook, were spaced at intervals of 20 to 40 feet along the central part of the main line. The typical unit, had 100 hooks. Flof^ting lines and anchored gill net's wer^'alao, -tt^d-eScasibnally in the fishery. Positive information from the shark fishery on the distribution of E. milherti is quite volumi- nous and detailed. Systems of payments to fish- ermen required detailed records involving iden- tification and measurement of all sharks landed by ves.sels of the principal fishing company. Al- together, records of landings of more than 100,000 milberti during a period of 15 years have been examined. Information on the absence of E. milberti from si>ecific areas has been difficult to assemble, but here also the records of the shark fishery supply most of the data. The species was firet repoi'ted in the Florida area from correctly identified specimens after the commercial shark fishery began (Springer, 1938), so the earlier scientific literature has been useless in the establish- ment of the range of the species in the Florida area southward. Offsliore records, from areas where water depths are more than 500 fathoms, are exclusively from catches made on tuna longlines used by the exploratory fishing vessel Oregon (for descriptions of this gear see Bullis, 1955, and Captiva, 1955). Some infonna- tion on the distribution of the young was obtained from otter-trawl catches made by the exploratory fishing vessel Delaware off the coast of North Carolina. Additional scattered records were picked up from accidental catches made by com- mercial and sport fishermen who used various types of gear, from catches made by collectors fishing for aquarium specimens, and from biolo- gists who captured specimens incidental to other collecting activities. The area of greatest uncertainty is in the off- shore and midwater range. Recent marine ex- ploration has shown that substantial populations of large sharks, fishes, and invertebrates live in subsurface waters beyond the Continental Shelf where they have escaped the attention of natu- ralists. On June 11, 1954, the first E. milberti known from waters beyond the Continental Shelf was taken on a tuna longline hook at USFWS Oregon station 1099, 85 miles off the coast of Texas where the depth was approximately 600 fathoms. Since the hook was set to fish at about 30 fathoms, this shark, an adult male, was cruis- ing in midwater. Throughout the second half of 1954, all of 1955, and the first part of 1956, long- line fishing was carried on in the offshore waters of the Gulf of Mexico by the M/V Oregon and a few commercial vessels. Large numbers of sharks were taken, chiefly species known to be partly or entirely pelagic. No additional milberti were taken until early February 1955, when a com- mercial vessel caught two adult females about 50 miles off the northern edge of the Campeche Bank where depths were estimated to be more than 1,000 fathoms. These sharks were caught on longlines with hooks fishing not more than 50 fathoms deep. These three captures, outside the principal range, appear to have little signifi- cance in the general picture of the distribution of Eula7nia milherti. Factors affecting distribution It may be assumed that water temperature and salinity are important in limiting the distribution of the sandbar shark and that there are other factors clearly influencing the movements and distribution of the species. The reaction of sand- bar sharks to ocean currents, the availability of food, the relation between the growth rate or the reproductive pattern and the migratory move- ments, all appear to be important in forcing the species into a particular range. No data are available, however, to show the relative strengths of these conditions as determinants of the range of the sandbar shark. The facts, from superficial examination at least, do not support the thesis that competition with other species is a powerful influence in the selec- tion of a particular range. Young Eulamia mil- berti, for example, apparently caiuiot long sur- vive where large Carcharhimis lencas in propor- tion to milberti are relatively abimdant. The presence of large numbers of large C. levcas in the vicinity of the nioiitli of the Mississippi River seemingly does not deter gravid female milberti from moving into the area to give birth to young. NATURAL HISTORY OF THE SANDBAR SHARK 13 Whatever the particular reason or reasons may be, the general absence of young milberti in the Gulf of Mexico shows that conditions are un- favorable for them. Circumstantial evidence suggests that interspecies competition is respon- sible, because large C. leucas eat young milberti. No explanation is apparent for the common occurrence of E. milberti along the continental shores and its absence from most of the West Indian shallow waters except that the species seems to have preferences for certain types of bottom. E. ??ulberti is ordinarily not common in areas of coral reefs or where the bottom is rough. Since it is chiefly a bottom-dwelling species, it is not surprising that it would exhibit preference for one type of bottom over another. In its migratory passages around the southern tip of Florida and the Florida Keys, however, there appears to be active avoidance of the fringing reef. Here the migrating adults leave the rela- tively shallow areas they inhabit on both the east and west coasts of Florida and temporarily enter , and feed in much deeper water. J ■©» ^ ' Nursery grounds and distribution of young '•^* I The principal nursery grounds of the western ' North Atlantic population of Eulami.a milberti lie in relatively shallow water along the Atlantic coast of the United States from Long Island to Cape Canaveral, Fla. This range may be ex- tended slightly at its northern end to the south side of Cape Cod in favorable years but the southern limit is more definitely fixed. Not one young milberti has been taken south of Cape Canaveral, around the tip of Florida, or in the eastern Gulf of Mexico. On the east coast of Florida, south of Cape Canaveral, a few sexually immature milberti of almost adult size have been taken; but in this area adult milberti are com- mon. A great quantity and variety of fishing effort lias been concentrated south of Cape Canaveral on the Florida coast. The total ab- sence of young milberti here is remarkable in view of the somewhat indefinite range limits of the adults. A secondary nurseiy range apparently lies in the northwestern part of the Gulf of Mexico. It is indicated only by the capture of a few females with near full-term embryos near the mouth of the Mississippi River, the capture of a large 552508 — 60 3 milberti with nearly full-term embryos off the Texas coast (Henry Hildebrand, 1954), and a specimen 747 mm. (nearly 30 inches) from the Texas coast (Bigelow and Schroeder, 1948). It is probable that gravid females wandering away from the principal range of the species give birth to young along the Mexican and Central American coast, but no records of the capture of young in this area have been found. Shark fish- ing on a small scale has been carried out over most of this area and catches have been examined at various points from the mouth of the Rio Grande River to Costa Rica; but excepting the Gulf of Campeche, no young milberti appeared. The female Eulumia milberti^ which move into the principal nur.sery areas to give birth to their young, do not remain there long and do not feed actively while there. This may explain the scarcity of records of captures of adult E. m,il- bertl along the Atlantic coast. Great South Bay, Long Island, is one of the nursery areas of E. milbei-tiand accounts of the appearance of fe- males in the bay and birth of the young are given by Thome (1916). Additional mention of the appearance of E. milberti in the Great South Bay area is made by Nichols (1918), who notes that the interesting fact about them is that the adults of the two sexes of the same species are almost never taken together near Long Island. Here the adult females are E. milberti and the adult males are Carcharhinus leu<:as. Records of young sharks from Chesapeake Bay show that E. milberti gives birth to young in the bay in summer. William Massmann, of the Vir- ginia Fisheries Laborator}^ has kindly given me (in correspendenc*-) records of E. milberti from the lower Chesapeake Bay. He sa5's — Although young of this species are probably the most abundant shark In the Bay In summer, I would not say that It is numerous. • • • it is commonly caught by anglers and probably rather generally distributed in the lower Bay. I have not seen an adult in the Bay or any individual more than three and a half feet long. After a comparatively brief period in shallow water or in the mouths of bays, perhaps at the beginning of cool weather, young vrllbertl appear to move offshore. Tlie only area from which young milberti are known in the winter .season lies off the coasts of tiie Carolinas at depths out to 75 fathoms. ->( 14 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Distribution off Atlantic coast of the United States and in eastern Gulf of Mexico The northern limit of the range of the sandbar shark is easily established. There are no reliable records of its capture from the Gulf of Maine (Bigelow and Schroeder, 1953), but south of Cape Cod it has been taken frequently but irregu- larly at Woods Hole, Mass. Numerous records of the species along the Atlantic coast of the United States are summarized by Bigelow and Schroeder (1948). Sandbar sharks may be said to be common in summer along the Atlantic coast of the Ignited States from Long Island to the tip of Florida and in winter along the waters of the Continental Shelf off the Carolinas southward to the southern tip of Florida, in water of moderate depths in the Florida Straits and along the west coast of Florida northward to Tarpon Springs or the Middle Grounds (the rough bottom area south of Cape San Bias, Fla.). This area is the principal known range, but the species has also been taken in small numbers from the northern and western Gulf of Mexico, the western borders of the Bahama Bank, the northern coast of Cuba, and the Caribbean coast of Nicaragua and north- ern Costa Rica (fig. 4). , i" The hypothesis is advanced hejie that the sand- / bar sharks of the eastern and southern sides of the Gulf Stream in the Straits of Florida are casual visitors to those areas and that the stock of the northern and western parts of the Gulf of Mexico is a breeding stock which is not self- sustaining, but is recruited in part from migra- tory adults moving to the northern and western parts of the Gulf by mistake or through error in orientation and navigation during the regular winter migration of the principal stock. The sandbar shark has been taken from the shallows along beaches out to a depth of about 135 fathoms. The young have been taken most often in shallow waters to depths of about 5 to 25 fathoms in summer, but in winter they move offshore to warmer water and depths as great as 75 fathoms off the Carolina coast. The sandbar shark is known only from the shallower part of the Continental Shelf in the warmer months in the extreme northern part of its range. Probably the adults are more common off beaches tlian in major bays or inlets. Hilde- brand and Schroeder (1928) found the species rather rare in Chesapeake Bay although more common than any other shark except the spiny dogfish. Radcliffe (1916) states that the species appears to be rare in the Beaufort, North Caro- lina, region. However, it was regarded as com- mon in bays on the ocean side of Long Island from mid-June to mid-September, by Nichols and Breder (1927). The apparent scarcity of adult E. milberti noted by Hildebrand and Schroeder and by Rad- cliffe is easily explained. It is possible that E. milberti enters the mouths of bays to give birth to young more frequently than records suggest. Female Carch/irhinus Jeucas and Negaprion hrevi- rostris move inshore and stop feeding for a short period at the time of the birth of their young, and immediately after the young are born the females move into comparatively deeper water. This may be a common habit among carchar- hinids and certainly a very useful one to provide for survival of the species. The Long Island records are to a large extent based on harpooned specimens, and adult females should probably not be expected to be easily available to capture on baited hooks in areas where the young are born. The best fishing depth for adult E. mWberti from the Carolinas south to Miami was found by the commercial shark fishermen to be 15 to 30 fathoms. On this stretch of coast it was rarely if ever taken beyond 50 fathoms on bottom set- lines and made up less than 5 percent of the catch on floatlines set beyond the 100-fathom curve. Southward from Miami, mUberti was rare among the keys, in Hawk Channel, or along the shallower portions of the reefs south of the Florida Keys. In the winter, a few appeared in catches made in the Northwest Ship Channel but, in general, these waters were left to other species. The sandbar sharks, however, were the common- est sharks on the bottom beyond the fringing reef out to depths of 50 fathoms and made up sub- stantial portions of catches out to 100 fathoms. They were appreciably more numerous in catches off the lower keys where currents were not so strong. Northward from the keys along the west Florida coast as far as Tampa, E. milberti was found to be most abundant in depths of less than 30 fathoms. Shark-fishing vessels operated out of Salerno, Fla., almost every day that weather permitted NATURAL HISTORY OF THE SANDBAR SHARK 15 NORTH AMERICA / / (I ^ / ^ \\) f f IV I * \ M * \ M ^ z::^^'y\^ ■A ^ PRIMARY NURSERY RANGE ^ ON MUD BOTTOM AREAS IN LESS THAN 10 FATHOMS ^\ * ~^^^PRIMARY WINTER RANGE »■ ^^-'^^^^ OF ADULTS , \ SECONDARY NURSERY \^ ^ GROUNDS -^ ^ 1 a II t F Of MEXICO \\ ^ ^^^^ ' s 'a V ^ ">— _^ ^^ ^ — ^ v^ ' / -r ^SECONDARY WINTER ^ / / VrANGE OF ADULTS ^r>> ^ - • • « W C 1 II 1 ( B E A N S E 1 > t • e Figure 4. — Geographical distribution of the western North Atlantic population of the sandbar shark, Eulamia milberti. from 1936 to 1050, except for parts of 1939 and 19-10 when activities were suspended due to over- production. These vessels landed .their catches daily at Salerno where the sharks were identified, measured, and recorded. Catches included in tlie Salerno landings were made from Bethel Shoal, north of Fort Pierce, Fla., to the offings of Jupiter Light. Some details of these Tandings, showing catch per unit of ^ffort, were reported in an earlier publication (Springer, 1951). These data -sliow that at Salenio the higliest average rate of catch of E. jriilberti occurred annually in the month of February and the lowest average rate of catch in SepteJiiber. These data concern 16 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE only adult milbej'ti and the comparatively few sexually immature mUherti of adult size occur- ring with the adults. On the west coast of Florida, shark-fishing vessels caught no E. m.nbertl at all from May through November and reported their largest catches from January through March in each of several years for which data are available. At Key West, Fla., high catch rates were ob- tained for milberti from deepwater sets made in the winter and early spring but catches at other seasons were poor. High catch rates of adult E. milberti were ob- tained by shark-fishing vessels off the Carolinas in September, and from 1946 to 1949 one or two of the more able shark-fishing vessels followed the fisliing for adult milberti southward, arriving off Salerno in January or February. Scouting by shai"k-fishing vessels showed that some adult 7nilberti were present in each month of the year along the Atlantic coast between the latitudes of Charleston, South Carolina, and Miami, Florida. Although E. milberti was the principal species sought in this area, other species contributed variously to the value of the total catch. Table 2 shows the comparative availa- bility of large sharks to the kinds of fishing gear employed in the Salerno area in the late spring. Distribution in Bahamas and West Indies E. m,ilberti is common only on the western side of the Gulf Stream. On the eastern side of the sti'eam, it is replaced by E. falciformis as the common inshore Eulamia. The wandering of milberti into the Antillean area may be quite limited. There are no records of milberti east- ward through the West Indies nor from the southern shores of Cuba. A shark fishing opera- tion on the eastern part of the Bahama Bank in the period from 1947 to 1949 did not take mil- berti. E. falciformis was reported by Evermann and Marsh (1902) from Puerto Rico; by Beebe and Tee Van (1928) from Port-au-Prince Bay, Haiti ; by Nichols (1929) from Puerto Rico; and by Backus (19.57) from open sea situations east of the Gulf Stream. None of these authors noted the presence of E. milberti. Frank Mather III has told^-me in correspond- ence.of the capture of E. fwhiformis and E. flori- d^nu from the vicinity of St. Croix and St. Thomas in the Virgin Islands. Mather's fisliing operations covered the depth range in which E. milberti would be expected if its geographical range ex- fends through the West Indies and if the absence of milberti follows the usual pattern in the West Indies. \ Table 2. — Sharks taken by commercial fishing vessels in the Salerno-Fort Pierce area {Bethel Shoal to Jupiter Light) and landed at Salerno, Fla., in May and June 1945-46 |1 to 3 vessels; only sharks with hide length of 55 in. or more included; fishing depths from 18 to 40 fathoms] Number of sharks Species Eulamia milberti (Miiller and Henle), sandbar shark., Spht/rna sp.,' hammerheads Eulamia obscura (LeSueur), dusky shark Eulamia Horidana (Bigelow, Schroeder, and Springer), silky shark .,, Galeocerdo curier (LeSueur), tiger shark Carctiarhinus leucas (Miiller and Henle). bull shark.. . Negaprion breviToatris (Poey), lemon shark Carcharhinus sp.,^ blacktips Qinglymostoma cirratum (Gmelin), nurse shark.. Carcharodon carctiarins (Linnaeus), great white shark. Isurus oiyrinchus Raflnesque, mako Unidentified • 3 species; station records do not distinguish kind. 3 2 species; station records do not distinguish kind. That the Gulf Stream is not itself a barrier to E. milberti is apparent from the occasional cap- tures along the Bahama Banks and off the north- ern coast of Cuba. It is possible that large num- bers of migratory sharks may wander away from normal migratory routes at times when unusual conditions prevail. Certainly a few Tnilberti JA .-t/^ cross the stream. From May 18, 1948, to July 8, 1948, I under- took a program of experimental shark fishing along the western edge of the Bahama Bank from Riding Rock northward. Fishing operations were carried out from a base at the Lerner Ma- rine Laboratory of the American Museum of Natural History at Bimini, using a fishing vessel and gear provided by the Shark Industries Divi- sion of the Borden Co. Fishing was carried on chiefly by bottom setlines at depths from 10 to 200 fathoms, but some floating lines were used to assure collection of as wide a variety of sharks as possible. Sets included some made at various levels along the extremely pr.ecipitous slope of the bank, which drops off abruptly from about 20 fathoms down to the floor of the Gulf Stream channel where depths are more than 150 fathoms. NATURAL HISTORY OF THE SANDBAR SHARK 17 Durino; the entire fishing period along the Ba- hama Banks only 14 E. miJherti were caught in the lot of 447 sharks. Of the total, 197 were reef sharks, Eulamia spj'ingeri (Bigelow and Schroeder). Table 3. — Sharks taken in exploratory fishing from the Duskv along edges of northwestern Bahama Banks, May 'l8 to July 8, 1H8 Number of sharks taken in— Species Bimini area (20-200 fathoms) Walker Key (10-100 fathoms) Eulamia springeri (Bipelow and Schroeder), reef shark. 10 60 46 2 17 15 10 12 11 2 S 3 3' 1 1 1 1 1 187 13 Eulamia obacura TLcSueur.. dusky shark 5 18 Eulamia altima Springer, bignose shark Fulamia noridaua (Bigelow, Schroeder, and Springer), 2 Eulamia milherti (Mullerand Henle). sandbar sharks. 4 2 Scoliodon terra-noiae (Richardson), sharpnose shark Oi-nglymostoma cirratum (Gmelin), nurse shark,. Splii/rna leitini OritTith, southern hammerhead Heianchus sp. (not H. oriseus). little con-shark Carcharhinus limbatu^ (Miiller and Henle),' little blacktip 3 3 2 1 2 Carcharfiinus leucas (Miiller and Henle). bull shark Muslelus canis (Mitchlll), common smooth dogshark... Pterolamiops longimaiius (Poey). whitetip 3 CaTCharhinus maculipinnis (Poey), big blacktip Carcltarltinus acronotus (Poey), blacknose shark Ftilmnin sp , nnHptprminpd 1 ' Antillean form. ' Nomenclatorial status of this species not determined. During the same season of the year in which the exploratory fishing was done, a great number of sharks were landed across the Gulf Stream at Salerno, Fla.. about 80 miles from the northern end of our Bahama fi.shing area. But more than half of them were E. in.iJberti, and no E. falci- formis was landed. Results of the Bahama fish- ing are summarized in table 3, which shows catches made in two areas off the Bahama Banks. For purposes of camparison, catches made in the Salerno-Fort Pierce area in May and June 1945 and 1946 are shown in table 2. The two fishing operations are not exactly comparable, of course, not only because they were carried on in differ- ent years but because the Bahama fishing was essentially exploratory while the Salerno-Fort Pierce fishing was a part of a continuing com- mercial operation concentrated in limited depths and locations. Exploratory fishing with sets .scat- tered in different depths and locations off Salerno would presumably produce a few E. springeri because there are normally a few to be found along inshore reefs adjacent to the St. Lucie Inlet at Salerno. This is a poor fishing spot, however, so the commercial vessels rarely caught springeri. Distribution in western Gulf of Mexico and western Caribbean Positive knowledge of the distribution of Eulamia milherti in the western part of the Gulf of Mexico and along the Caribbean coast of Cen- tral America is based on records of 8 gravid fe- males off the Mississippi River delta, 1 gravid female and 1 young male from the coast of Texas, 1 adult male and 2 adult females from the south- central part of the Gulf of Mexico over deep water, and 51 adults from the Caribbean coast of Nicaragua and Costa Rica. From the mouth of the Mississippi River east- ward and southward around the Gulf to the vicinity of Tarpon Springs on the Florida west coast, milherti appears to be absent or at least rare from inshore waters out to 30 fathoms. No catches were reported by shark fishermen and no specimens were seen. I should note that while employed in the shark fisherj', seasonal shark- fishing stations were maintained at various times at all of the fishing ports of any consequence from the mouth of the Mississippi River around the tip of Florida to the Carolinas. I visited all of these stations frequently and sometimes par- ticipated in fishing operations. The presence of milherti in appreciable quantity would almost certainly have been noted in catches from the sta- tions at Panama City and Carrabelle, Fla., had specimens been taken. It is not possible to present a meaningful ac- count of the sandbar shark, Eulamia milherti, without frequent reference to the bull shark, Cureharhinus leu^'os, and its fresh-water repre- sentative, Carcharhin'us nirm^aguensis (Gill and Bransford). The sandbar shark has been con- fused with the bull shark because of the peculiar manner in which their ranges overlap. ItT-TTddi- tion, bull sharks ajipear to be the most important of the predators on young .sandbar sharks and the primary factor that prevents E. milherti from extending its nursery range into otherwise suita- ble areas in tropical seas. As has already been noted, bull .sharks are found along the Atlantic coast as far north as Long Island but increase in numbers somewhat in the latitude of Salerno, Fla.. wliere, as noted in table 2, they were sixth in number of large sharks .^M. 18 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE landed by the shark fishery. The abundance of bull sharks at various locations in the Gulf of Mexico varies seasonally, but from the vicinity of Apalachicola, Fla., westward along the north- ern coast of the Gulf of Mexico and southward along the coasts of Central and South America as far at least as French Guiana, bull sharks fonn the major part of catches of large sharks made by shallow-water setlines in some seasons. Table 4 shows the comparative frequence of capture of large sharks during a test-fishing period off the mouth of the Mississippi River. Table 4. — Large sharks taken from the Joe Leckich on hMom lines set in 5 to 35 fathoms off the mouth of the Mississippi River, June 25 to July '29, 19^7 Species ' Carctiarhinus leucas TMiiller and Henle), bull shark... Fulamia obscura (LeSueur), dusky shark. Oaleocerdo cutier fLeSueur), tiger shark. Carcliarliinus limbatus (Miiller and Henle), little blacktip ' Spttyrna sp., great hammerhead CaTcharhinus macrilipinnif:, big blacktip . Fulamia milherti ( Miiller and Henle), sandbar shark. Carcharias tauruf: R;ifincsque, sand shark.. Negaprwn brerirofilris d'oey), lemon shark Sphyrna tewini GrilTith, southern hammerhead Eulamia floridana (Bigelow, Schroeder, and Springer), silky shark :^.: Eulamia springeri (Bigelow and Schroeder) .U. Average weight of liver (lb.) 38 53 129 3 49 13 U 26 20 12 16 4 ' Also taken but not recorded because of small size: several Scoliodon terra-novae (Richardson), Carcharhinus porosus Ranzani, and Mustelus canis (Mitchill), and one young Qiftgtymostoma cirratum (Gmelin). 2 Continental form— typical of western Gulf of Mexico. The eight E. milherti were adult females and five of them were gravid. Since milberti has liver oil of comparatively higher potency than the oil from other species in the area, a special effort was made to catch them in subsequent, larger scale fishing efforts. Nevertheless, catches of milherti were not made later during 1947 and commercial fishermen operating in the area re- ported sandbar sharks absent in 1948 and 1949. JThe few records of E. milherti from offshore Gulf waters would not be important except that they serve to show the species can move into and across deep areas of the ocean. Off the Atlantic States there has been comparatively little long- line fishing beyond the Continental Shelf and there are no records of milherti from deep water. There are, however, records of catches of Eulamia faUifonnis (Miiller and Henle) and E. obscura from longline sets made by the exploratory fish- ing vessel Delaware beyond the limits of the Continental Shelf off the Middle Atlantic States. These two species are also reported by Backus ( 1957) from the Atlantic beyond Continental Shelf limits. A lot of 51 adult sandbar sharks was taken by Captain B. W. Winkler from off the Caribbean coast of Nicaragua and Costa Rica. This group included all of the sandbar sharks in a collection of 854 sharks of all species which Captain Wink- ler measured and recorded for me from Septem- ber through December, 1948. The most inter- esting feature of this collection of E. milherti was that it was made up of approximately equal numbers of adult males and females (26 males and 25 females). The .shark fauna of the area as represented by Captain Winkler's collections included a large number of species that are predatory on sharks of the size of young E. milherti. The following sharks were taken : Species: Number Carcharhinus, ' chottos or bull sharks 421 Galeocerdo cuvier (LeSueur), tiger shark Eulamia obscura (LeSueur), dusky shark Eulamia floridana (Bigelow, Schroeder, and Springer), silky shark Carcharhinus limbatus (Muller and Henle),* little blacktip Eulamia milberti (Muller and Henle), sandbar shark Sphyrna sp. (not determined), hammerhead shark Eulamia sp.,' reef sharks Scoliodon sp., sharpnose shark Hexanchus sp.,* cow sharks Ginglymostoma cirratum (Gmelin), nurse shark.. Eulamia altima Springer, bignose shark Negaprion brevirostris (Poey), lemon shark Undetermined ' 85 76 70 54 51 27 25 15 13 3 2 1 4 ' Either C. leucas (IVIiiller and Henle) or C. nicaraguensis (Gill and Brans- ford) or both. 2 Continental form. s Probably F. springeri (Bigelow and Schroeder) or F. falciformis (Muller and Henle). * Includes two species. 5 Possibly a small species of Oaleorhinus. n^. few days' fishing with bottom longlines along the outer edge of the Continental Shelf off northern Nicaragua and on Serrana and Seranilla Banks in February 1949 failed to produce E. 7nilherfi or any adult sharks, but moderate num- bers of young E. faridana were taken. As a grader and buyer, I examined several lots of dried sliark fins said to have been taken off the coasts of Colombia and Venezuela. E. milherti fins were not noticed althougli it is possible that NATURAL HISTORY OF THE SANDBAR SHARK 19 a few might have been overlooked. The fins of E. milberti adults (pectorals, first dorsal, and lower caudal lobe) are more desirable for com- mercial purposes than the fins of some other species because they are thicker and have a rela- tively lar>::e proportion of the material used for shark-fin soup. Shark fishermen withi.' wliom I tallced and who would recognize imlberti also re- ported the absence of the species from their catches made oif Colombia and Venezuela. Thus, evidence for the occurrence of E. milberti in the southwestern Caribbean, while not very satisfac- tory, is negative. ~a ,n)v f MIGRATION It is probable that the migrations of E. mil- berti are of two kinds. One is simply the gradual withdrawal of the sharks from waters that be- come too cold or too warm — a movement that is accompanied by normal feeding acti\nties and is characteristic of immature sharks. The other is a movement, generally, over a greater distance that may or may not be induced by temperature changes. The general patterns of the major movements of adult sandbar sharks suggest that ocean currents greatly influence the direction and extent of the movements. It is necessary to consider the migratory move- ments of the adult male, the adult female, and the immature Eid-amia milberti separately, "^'e may look upon the Atlantic coast from the vi- cinity of Cliarleston, South Carolina, to the northern part of Florida as the core of distribu- tion of the principal stock of the western North Atlantic population because it is only in this area that all three groups are known to be found. This may mean merely that in this area there is overlapping distribution. "We know tliat the adult females go as far north as Long Island to give birth to young in summer and in some years even farther, to tlie vicinity of Cape Cod. There are no data to show whether the adult males or nongravid females mo\e nortliward into the por- tion of the species' range lying nortii of the Caro- linas. All that is known of the distribution of the young is tliat the voting are born in water of moderate salinity from Cape Canaveral to (\ape Cod and tliat some of them move in winter into the comi)aratively warmer offsliore water found at depths of 50 to 75 fathoms on the Carolina coast. Until the young sharks reach adult size they do not take part in the long southern migra- tion characteristic of adults or move south of Cape Canaveral. One capture of a lot of nearly 200 young off North Carolina indicates that the young occur in schools of both sexes and of mixed sizes. Migratory movements of adult E. milberti south of Cape Canaveral are more clearly out- lined from the data available from the shark fishery. The annual southward movement ap- pears to be coincidental with the beginning of cooler weather and to be accelerated by cold snaps. A very much larger number of adult female sandbar sharks were taken than adult males by the Florida shark fishermen. This and the tend- ency to segregation by sex, will be discussed later in connection with reproduction. But it should be pointed out here that no adult males are recorded from inshore nursery grounds and probably occur there but rarely. In the Florida shark fishery, adult males of most species brought more money than adult females and were particularly sought by fishermen. The fishermen were convinced by observation that adult males of most species in- cluding m,ilberti were usually in deeper and cooler water than the females, that they usually pre- ceded the females in migration, and that they usually were to be found in more compact aggre- gations so that fishing was best where males could be found. Such figures as are available bear this out. A rationale for this condition is suggested by the thermal sensitivity (decreased fertility with application of heat) of the male germ plasm among vertebrates in general (see Cowles, 1945). The sandbar shark is properly considered a "ground shark" and is rarely to be seen from the surface except when it comes into shallow water. An exception to this occurs off Salerno immedi- ately following ]>eriods of especially cold winter weather nortli of that area. At such times, when weather and water surface conditions permitted observations from a boat, large schools of mil- berti were to be seen headed south, swimming at about .S to 5 knots 5 to 10 feet below the surface, but where water deptiis were about 20 fathoms or more. Shark fishermen have tTt>kt:jHe, a«d-=fiay< o^vn expedience bears this out, that it is useless to trv to follow these sharks or to trv to divert I 20 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE them by chumming or by setting lines ahead of them. The appearance of southbound schools at Salerno was generally accepted as a harbinger of better fishing a few days later. The southbound migratoiy movements of E. milherti at Salerno were inshore and within the southbound eddy of the Gulf Stream; north- bound movements were not observed. It is sug- gested that these movements were eitlier offshore movements or slower movements of more diffuse aggregations. Northbound movements offshore would be aided by the Gulf Stream. An hypothe- sis whicli may be more convenient than signifi- cant is that E. milherti tends to follow currents in migration and if the currents are strong does not go against them. Of course, sharks would not make appreciable headway against the cur- rent at the surface near the axis of the Gulf Stream without vigorous and persistent effort. The sandbar shai'k does appear to take advantage of eddies or countercurrents and the fishing plans of some of the more successful shark fisher- men were based on an assumption that the shark's seasonal movements would follow tlie currents available at the time. The distance traveled by various segments of the population probably does not extend from one end of the geographical range of the species to the other. From the southern end of the nursery range of the principal stock at Cape Canaveral, a seasonal gradient of availability was shown by catch per unit-of-effort data. This availability decreased in summer southward and around the tip of Florida to the west coast of Florida where the species was completely absent from summer catches. Thus, the minimum migratory travel of the part of the stock reaching the vicinity of Tampa would be approximately 600 miles. Catches of E. milherti throughout the area be- yond the southern end of the nursery range reach their highest peak in midwinter. Catch per unit- of-effort data previously published (Springer, 1951) show the catch of E. milberti at Salerno, Florida, as decreasing from 4.8 fish per 100 hooks for Februaiy to a low of 1.1 per 100 hooks for September. A cold upwelling over the narrow Continental Shelf immediately north of Jupiter Light usually occurs in June or July. It is prob- ably of brief duration but annually stuns great quantities of fish, although the sharks are not affected. This phenomenon coincides with spec- tacularly good shark fishing and possibly also with considerable mating activity on the part of E. milberti. This may give some bias to Salerno catch per unit-of-effort figures for early summer. REPRODUCTION Courtship and mating I have seen neither the courtship nor mating of Eulamia milherti. The general pattern may be constructed, however, from fragments of infor- mation and from inferences based on the few facts known about related large sharks. The comparative morphology of the secondary sexual apparatus of male sharks has been given compre- hensive discussion by Leigh-Sharpe in 11 papers. The functions ascribed by Leigh-Sharpe in three of these papers (1920, 1921, and 1924) to the vari- ous parts of the ajjparatus in carcharhinid sharks are in general accord with my observations on GaUocerdo, the tiger shark. The courtship pat- erns in Galeocerdo, Eula/mia milherti, and other large carcharhinids probably do not differ greatly. A brief outline of the mechanics of fertilization in tlie carcharhinid sharks is included here to oriept the reader in, following some of the infer- ences made in later discussion of differential death rates in the sexes. "^Carcharhinid sharks are born alive and fertilization is internal. Paired intromittent organs of the male known as clasp- ers are supported by cartilages. Immediately following the rapid enlargement of the testes, which occurs at maturity, layers of calcification appear at the surface on the principal clasper cartilages. At this time the claspers become semirigid except at the basal area of attachment of the claspers to the base of the pelvic fin adja- cent to the cloaca. The tip of each clasper, how- ever, is expandible. When expanded, the carti- lages of the tip are transverse to the main axis of the clasper and open as the ribs of a fan. The expanded tips are thought to serve both to liokl the oviducts of the female open and to prevent with- drawal of the claspers because of the rigid carti- lages in a transverse position. The very large clasper siplions are a distinguishing and peculiar feature of the apparatus in male carcharhinids. These siphons are a pair of separate sacs lying just under the skin of the belly on either side of NATURAL HISTORY OF THE SANDBAR SHARK 21 the midline and extendinjr from the pelvic to the pectoral ureas. They function as reservoirs for the sea water used to flush the male sex cells from the bases of the claspers into the oviducts of the female during mating. The siphons may hold a large amount of sea water, as much as 2 gallons in Guleottrdo. The siphons do not ordinarily contain the sea water which is presumed to enter the siphons during the period immediately pre- ceding mating. During the mating season the area at the bases of the claspers of the larger carcharhinids exhibit extraordinary vascular congestion. Characteris- tically, in male Galeocerdo, a mass of very soft spongy tissue appears around the cloaca. This is present to a lesser degree in the smaller carchar- hinids such as t!^ milberti. Unusual congestion, edema, and suboermal hemorrhage at the base of the claspers are evidences of courtship activity on the part of the male. Large sharks are not highly maneuverable and cannot swim backward, so it is necessary for the claspers to rotate and point forward during mat- ing. Since the muscle system in the typical carcharhinid clasper seemed functionally inade- quate or feeblf, I carried out an experiment which incidentally revealed the probable method by which the clasper siphons are filled with sea water. I oTitained an adult male Carcharhinus lirnhatus about 5 feet long and evidently in mat- ing condition. The choice of species was dic- tated by circumstances, one of which was the fact that a 5-foot shark was as large as I could man- aged By injecting a considerable quantity of an isotonic solution into the caudal vein, ^'^as able to induce the claspers to assume the normal mat- ing position. This action caused the claspers to revolve inward and forwai-d. As the claspers moved into a forward pointing position, a funnel, formed by a membrane supported by rods of cartilage, opened at the base of each clasper. The mouth of the funnel was also directed forward and the constricted end led into the siphon. The caudal vein was plugged experimentally to hold the claspers and funnel in position and the shark was moved forward as rapidly as possible through tlie water. This caused the clasper siphons to fill with water. Application of addi- tional pressure to the caudal vein resulted in com- plete expansion of the fanlike tip of each clasper. The course of courtship and mating in all of the larger carcharhinids including Ey milberti probably follows the pattern in which the male persistently follows and occasionally bites the female on the back until she swims upside down. Both claspers probably function at the same time, one entering each oviduct of the female by way of the lateral opening from the cloaca. The con- tact of the two sharks may be presumed to force the sperm-ladened sea water from the siphons into the oviducts. The mating pattern has been given in .some de- tail to emphasize the point that mating is very complicated in carcharhinids, and that the me- chanical difficulties are compounded among the larger species by their greater weight and lesser maneuverability. Time of mating The approximate period of the mating season is established by the appearance of males with enlarged testes and also with some evidence of vascular congestion of the pelvic-fin area and by the appearance of females with eggs of full size in the ovaries (about 1 to li^ inches in diameter in l^mllberti) . In the vicinity of Salerno, Fla., mating of ^milberti evidently takes place in the spring or early summer. Males appear commonly in inshore catches after the firet of Februaiy but remain segregated for some time. Catches of both sexes indicating mixed schools are more fre- quently made in April and May than in other months. After Ma}\ male i^. milberti are rela- tivel}' scarce in Salerno catches. Among car- charhinids, the males may stop eating during the courtship period. This is an inference drawn partly from the general reduction of catches of males on baited hooks during mating seasons, partly from the observed smaller size of the livers of males immediately following the mating season, and partly fiom observations on the mat- ing activity of Galeocerdo. Fertilized eggs and the smallest detectable em- bryos were observed first in Salerno catches from the first part of July to the first ]>art of August. Tlu' time of mating can be established with more l)recision, however, by observation of the time of appearance of the fresh courtship scars on the females. Scars, tooth marks, and ismall open wounds produced by shark teeth are conunonly 22 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE found on adult female carcharhinids and are gen- erally restricted to the dorsal surfaces between the two dorsal fins. These are never present on males or immature females and are obviously produced during courtship. Scars or wounds are not always present on gravid females or at least are not always detectable but were found on about half the gra'vid E. milberti taken at Salerno. The coexistence of old and completely healed scars with fresh scars on some females is one bit of evidence that female E. mnhei'ti pro- duce more than one litter of pups in a lifetime. All available evidence points to the month of June as the time of maximum mating activity of E. milberti in southeastern Florida waters. It has- already been pointed out that 4nales were rarely taken during the month of June when mating activity is assumed to be at a peak. There is some evidence from catches that the males were present in substantial numbers at that time. From 3 to 5 percent of the catches of the better fishermen at Salerno, who kept their hooks very sharp, were snagged sharks; that is, the sharks were caught by hooks in the fins or tails or occa- sionally in other parts of the body but not in the ^noiuth. More males were (jauglit in thik wa'y dur- ing June than were iTOoked by mourh. Development of the embryo ''^■ In E. milbei'fi. as in other carcharhinids, it is presumed that fertilization occurs after the large egg leaves the single functional ovary. It is also presumed that fertilization occurs before the egg has been moved through a shell gland. Shell glands are located near the anterior end of each of the two functional oviducts. In passing through the shell-gland area of an oviduct, a single egg is enveloped by a diaphanous tubelike shell capable of great expansion to accommodate the growth of the embryo to a very large size. The nutrient material from the egg yolk is suffi- cient only to provide for early growth of the embryo and to supplement nutrient materials necessary for intermediate growth. The means by which nourishment is supplied to the growing embryos probably varies in different species of carcharhinids, but in species of Eulamia, three principal methods appear to be involved. In addition to that sui)plied by the yolk some ab- sorption of nutrient material from fluids in the oviducts may be assumed to take place. This would appear to be necessary to provide sufficient material to carry the embryo to a length of about 12 inches at which length the pseudoplacenta is formed from the yolk sac. M^ observationKtHi the^mbryology of the sand- bar shark are limited to general notes on the ex- ternal appearance of the eggs and embryos at several stages during development. The spherical, unfertilized eggs in the single functional ovary reach a diameter of 1 to li/i inches. In winter and early spring, large num- liers of adult females not carrying embryos were fomul to have developing eggs 1/2 to 3^ inch in diameter in the ovaries. In a few instances, fe- males taken in July and August were found with eggs of maximum size in the ovaries as well a& fertilized eggs in the oviducts. In the greatest disparity of development noted, there were two large yellow eggs remaining in the functional ovary while embryos in the oviducts ranged from less than 6 mm. to 10 mm. in length. A female 7niJherfi. 6 feet 7 inches long collected off Salerno on July 2, 1948, in 25 fathoms, was typical of a series taken in early July of that., year. This female contained 10 egg cases, 5 in each oviduct ; no large eggs remained in the fimc- tional ovaiy. Each stringlike egg case was about 120 centimeters long, with thin membranous, amber-colored, transparent walls. A single yolk was contained in one expanded oval section of each egg case. The expanded section, approxi- mately 6 cm. long, was located about 10 to 12 cm. from one end of the egg case. This section also contained a clear fluid in each of 8 of the egg cases that had developing embryos 9 to 13 mm. long. The remaining 2 egg cases, one anteriorly ' in each oviduct, contained milky fluid and there was no evidence of fertilization nor development of the single egg yolk contained in each. The section of the egg case occupied by the embryo, spherical yolk, and clear fluid was held in shape by two longitudinal folds and by folded constric- tions of the egg-case membrane at either end. Xlie eo-fT case could, however, be unfolded and ex- panded with relatively light internal pressure. The egg case surrounds the embryo until birth and unfolds or stretches to accommodate the de- veloping embryo. "When the embryo reaches full term the pseudoplacental mass extends outside of NATURAL HISTORY OF THE SANDBAR SHARK 23 rlie e42 {^collected at Salerno. Fla., a series of ^uVirhia inilherti embryos of both sexes and near full term which were ap- parently perfect except for having the eyes on the lower side of the snout, almost in contact with one another and just posterior to the nos- trils, and having no trace of an opening in the skin for the mouth, although the jaw cartilages were apparently normal. 'Pfie specimens were pregwivfitLi n form alHF%Trt dried put during the following years and were discjlfded. Again in I J9*46 similar embryos were collected, and a^out / half fi ibafrel were preserved; but all were lost in * a -Wfrtlcane whidi destroyed^ a dock building. ^ No ab« ormal yoonfr-Were found m the rela- tively Irffge series of litt«irs.£iamined^in 104$ and J had iio later ojjportunity to see substantial mimbers of E. milherti embryos. All litter mates exlillmed the same abnormal condition and were remarkably uniform structurallj\ A very rough estimate of the frequency of occurrence is one set of abnormal young in .500 to 1.000 sets of ap- parentl)' normal pups. SEX RATIOS ^^ It was the general observation that landings of adult Eulamia miJherti at Salerno were in the ratio of 5 females to 1 male. A similar sex ratio was estimated for Salerno landings of E. ohscwa and for Bahama landings of ^^v'K.^spviMj'Ti. A disproportionately large numter of female E. ilorid(Mut were landed at Sa>rno but the records do nqt furnish an adequate basis fop'an estiipiue. An insufficient number of E. alfiina or E. npring- eri were recorded for estimate. Murphy and Nichols (1916) say that the commonest large sharks in the waters about New York are the gromid sharks (Carcharhmm), and also that males of tliese fishes are rarely .seen but toward midsummer many of the females enter our bays where they give birth to their young. They further stii,te that the cojinnonest ground shark is Carcharhinus ntilbertii The only record sug- gesting etjuality in the number of adults of the sexes of E. milbet'tl is Captain Winkler's record of the captuie of 26 males and 2.5 females off the Caribbean coasts of Nicaragua and Costa Rica in the fall. ' / / Data on the sex ratio in young E. milherti is limited to a series of 203 young from 26 to 50 inches long collected by otter trawl in Februaiy 1958 off/N"oith (>rolin!v. Tli/re vy^re Ol^^tiales and It^ fejuliles'in thii? collection. Florida shark fishery records of carcharhinids otlier than those of the genus Eulamia show local segregation by .sex and size but in no other car- charhinid nor in the hammerheads is there any clear indication ft^onr'availa^e _r©eords from Florida of/greivt imbalance m the sex ratios of adults, y' Tilt unavailability of male milherti to baited hooks during the mating season maA' explain in ])art the smaller number of males in the landings. However, beca/use Florida commercial shark fish- ing after 1946 was carried on out to depths greater than the maximum known depth range of the fspecies, and because the males brought the fishermen a higher price than the females, it is certain that there was no intentional selection of females^^Tlie fishermen believed that the schools of males, if found, were easier to catch in large nunibers.- Females were far more abundant than males in the deeper water catches made off the Florida Keys in the late fall and early spring, and a much greater abundance of females char- acterized the winter catches on the west coast of Florida. It has already been shown that approxinuitely equal numbers of male and female milherti are born. The evidence that there are substantially more females than males in the adult ]in]iulation is very strong, if sui' information adecjuately covers the geographical range of the species. Al- though if is quite possible that segments of the adult population have been entirely overlooked in the offshore and midwater depths in the north- ern jiart of its range, the shortage of males in the population around .southeiii Florida is remark- able. There is some indirect evidence also of a short- age of males in the breeding population. If the females bear pups in alternate years. .50 percent of the adult females wduld be expected to be gravid in winter. I have previously reported (Springer, 194(» ^uit tmly, ;i^qut ,1J pe^ceut ufjhe 26 FISHERY BULLETIN OF THE FISH AND "WILDLIFE SERVICE adult females taken in winter off Englewood, on the west coast of Florida, were pravid. I saw large numbers of E. miJherti in 1942 and again iiovn 1946 through 1949 at Salerno, and my notes include several estimates of the proportion of gravid to nongravid adult females seen on the dock. For the late-winter and early spring periods, it was estimated that substantially less than a third of the females were carrying pups.' In n^ sani]ile of 399 adult females takeri' for length-frequency data, apjuoximately 18 percent were gravid. Three interesting, if theoretical, explanations are suggested to account for the apparent differ- ences in the number of males and females in milherti and in other Enlamia. Tlie mating pattern appears to be particularly dangerous to the males, since mating occurs when the females are in a feeding cycle while the males ai-e not. That is, during courtship males may nip or slash to some extent but do not take large bites. The females have no such inhibitions ex- cept at the time the young are born, and fatal accidents to males may be frequent during court- ship. However, if this explanation is to be ac- ceptable, some further speculation is needed to suggest why Galeocerdo with a similar mating pattern is not represented in the adult population by a preponderance of females. Geiser (1924) summarized a variety of reports on the higher death rate for males in some mam- mals, fishes, and invertebrates, and suggests that there is a genetic basis for this in certain cases wliere the possession of two sex chromosomes by the females ensures a greater lonerevitv of the female by "canceling out" possible mutations in the x-chromosome, especially associated lethals. while in the male there is no such "canceling out."' The third explanation is that the males occupy wider geographical and vertical ranges than the females, remain in the cooler parts of the ranges, and exhibit a greater tendency to wander than the females. Thus, greater numbers of males than females are lost to the breeding population by wandering and death in unfavorable environ- ments. Whatever the explanation for the unequal sex ratio, the smaller number of males is not a suffi- cient handicap to prevent E. milberti from being one of the conunoner sharks. GROWTH AND SIZE AT MATURITY In Florida catches, adult male Eulamia mil- herti average 4.2 inches shorter than the average 1 adult female and weigh 32 pounds less than the ,' average nongravid adult female. The smaller size of the adult male is characteristic of all of the western North Atlantic carcharhinids al- though the size of males and females at birth is approximately the same. For about 20 years I maintained a clo.se watch oji landings of one or more commercial vessels and saw \io mUherti females longer than 92 inches and no males longer than 89 inches among the thousands that were examined. ^ The smallest sexually mature male in the ma- terial examined was 71 inches and the smallest sexually mature female was 72 inches in total length. Sexual maturity in the male is easily and {lositively determined because enlargement of the testes to functional size is immediately fol- lowed by the appearance of a ring of calcium at the surface of the major clasper cartilage. This ring is easily seen in cross section but since its effect is to stiffen the segments of the clasper, sectioning is unnecessary for positive determina- tions. Determination of sexual maturity in fe- male specimens where the specimens were non- gravid or had no courtship scars was made by examination of the ovary and -the oviducts. The females were regarded as sexually immature if none of the eggs in the ovary had l)egun to increase in size and if the oviducts were smaller in diame- ter than is characteristic of the ftilly contracted ovi^ ducts in females following jiarturition. It may be noted in table 6 that, while at least 2 female milherti were mature at a length of 72 inches. 5 immature females of greater lenglh were collected. Obviously the length at which the fe- males may become sexually mature varies more tlian 4 inches. The left skew of the length-freiinency polygons shown in figure .5 may be the result of any of sev- eral variables including the length at wjiich maturity is reached. A total of 513 adult sandbar sharks was select- ed from southeastern Florida catches for meas- urement of total length and for comparison of NATURAL HISTORY OF THE SANDBAR SHARK 27 40 z 30- 20 r^- \ - J \ \ - /' \ \ -^ ^'^ — V ''' r I I I I I i""'-!— -I I ^ 72 74 76 78 80 82 84 86 88 90 TOTAL LENGTH IN INCHES Figure 5. — Length-frequency polygons for adult male and female Eulamia milberti from southern Florida. leno^h frequency with the length frequencies of 76 sharks from off Fort Myers on the west coast of Florida and of 51 sharks from the Carib- bean coast of soutliern Nicaragua and northern Costa Kica (table 6). To reduce bias, all sharks of all species in any catcli were measured and recorded or none were recorded. Selection was affected by the scarcity of males. The compara- tively large sample of fall females at Salerno had to be measured to get any catches that included males which rarely appeared there at that season. In the sample taken from the lower east coast of Florida, 10 sharks, o males and 5 females, were found to be immature. These immature shark.s were excluded from the calculations of mean lengths of adults, but the sizes together with the dates of capture are given in footnotes to table 6, which shows the mean length of tlie sample lots. By its migratoiy movements and its restriction to limited nursery areas, the North American population of E. milherti appears to be subject to constant mixing. It does not seem reasonable to expect a rigid segregation by area of origin of those milberti mating off southern Florida. This may be one factor in the apparent homogeneity of the population. In some of the other caicharhinids, environ- mental or racial factors appear to affect the size at which the species becomes mature. For ex- ample, the average size of the bull shark. Car- charhmus leiicas. from the A'icinity of Trinidad is appreciably less than the average size of adults of the same species from the Gulf of Mexico. Important differences in tlie size at maturity as well as the size at birth separate-tl*^ Texas and southei-n Floi-ida ]>opulations of the little black- Table 6. — Mean length, number measured, and length range of adult Eulamia milberti, by sex, area, and season of collection (Lengths In inches; length range in samples in parentheses] 1 January-March .\pril-June July-September October-December Combined data Area Number in sample Mean length Number in sample Mean length Number in sample Mean length Number in sample Mean length Number in sample Mean length Males: 69 29 78.1 (71-89) 79.0 (73-84) 36 78.4 (72-86) 5 78.8 (76-82) 5 78.8 (73-82) 114 78.7 / Southwestern Florida (from 2 schools) (71-89) Nicaragua-Costa Rica 16 64 79.9 (72-89) 83.0 (74-91) 10 161 75.0 (71-84) 82.9 (73-90) 26 399 71 78 4 Females: Southeastern Florida ' III 82.8 (72-91) 63 82.9 (76-92) (71-89) 82 9 Gravid > (72-92) 83 2 Gravid (from 1 school) 59 48 82.8 (75-88) 83.8 (75-87) (73-88) N ongrav id (from 1 school) Southwestern Florida: Nongravid 47 83.3 (72-90) NMcaragua-Costa Rica: Nongravid 7 82.6 (78-88) 18 83.9 (76-90) 25 83.5 (76-90) ' .5 immature males were collected with thus sample but excluded from tabulation and from calculations of mean length: I specimen 64 inches long collected In .March. 2 specimens each 66 inches long collected in .\ugust. 1 specimen 5'j inches lone collected in Octoher. and 1 specimen 6S inches long collected \n November. - .5 immature females were collected with this sample but e\cludet\ Figure 1. — Projected impressions of scales of bluefin tuna. Arrows indicate annuli. A. Scale from fish 34 cm. long taken off Martha's Vineyard, September 1952; no annulus. B. Scale from fish 60 cm. long taken oflf Martha's Vineyard, July 1950; 1 annulus. C. Scale from fish 80 cm. long taken off Martha's Vineyard, August 1952; 2 annuli. D. Scale from fish 104 cm. long taken off Cape Cod, October 1950; 3 annuli. GROWTH OF BLUEFIN TUNA 41 >4.^:ir D Figure 2. — -Centra of vertebrae of bluefin tuna. Arrows indicate annuli. A. Vertebra from fish 77.5 em. long taken off Martha's Vineyard, August 1951 ; 2 annuli. B. Drawing of stained vertebra from a fish 104 cm. long, taken off Cape Cod, October 1950; 3 anmili. Scale in figure 1 D is from the same fish. (Reproduced from fialtsoff (1952) by permis- sion of the author and publishers.) C. Stained vertebra of fish 110 centimeters long taken off Cape Cod, October 1950; 4 annuli. D. Vertebra of giant tuna taken off Cape Cod; about 11 annuli. Annuli beyond the 9th or 10th are usually small and indistinct and disappear soon after dissection. (A, C, and D, unrelouched.) 42 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 1. — Fork lengths (in cm.) of blue fin tuna taken in the vicinity of Cape Cod, for each number of annuli INumbers In parentheses estimated from regression of length on caudal spread or on weight] 1 2 3 4 6 6 7 8 9 10 11 12 13 14 34.2 51.5 71.0 84.0 107.2 (120.3) (144.4) 155.8 162.9 179.0 182.0 223.4 207,6 236.6 247.0 35.3 52.1 71.0 84.7 107.7 (124.5) (146.0) 156.7 170.8 179.4 184.5 225,6 229.0 237,0 248.0 (37. 5) 52.2 73,0 85.3 108.0 (126.0) (151.6) (167.6) 174.8 183.0 190.4 234,0 240.0 (249. 0) 37.8 53.3 74.0 85.7 108.6 126.2 153.0 (161.7) 176.5 185.2 218.0 241.5 240.6 41.2 53.8 74.0 85.7 1 10.0 (127. I) (153.2) (164.4) 176.6 193.6 221.6 244.3 244.0 42.2 55.4 74.0 85.9 110.0 (129.0) (163.7) (165.8) 177.8 (196.6) 224.0 245.7 248.0 56.0 74.5 85.9 112.1 130.0 154.0 (169.0) 182.0 257.0 66.1 74.7 86.1 (113.7) (130.4) (156.0) 58.2 74.8 87.1 114.1 130.8 168.2 58.8 74.9 87.4 114.5 (131.3) (168.7) 59.0 75.0 87.8 114.6 132.0 (166,0) 59.7 75.0 88.2 115.0 (132.0) (170.9) 60.0 76.6 76.7 75.7 76.3 76.7 76.9 77.1 77.6 77.6 77.6 (77.9) 78.2 78.5 79.1 79.5 80.0 (80. 8) (81.8) 81.9 88.4 88.7 89.0 89.0 89.7 90.2 90.8 91.1 91.6 92.1 92.9 95.0 (98. 0) 103.3 104.4 104.8 115.0 116.1 116.6 117.5 (118.3) 118.6 (118.6) (119.0) 119.6 120.0 122.0 (122.8) (123,0) 124.4 124.5 125.6 129.0 (130.2) (130.4) (130.7) (132. 1) (132,0) (133,0) (133.5) (134.0) 136.0 136.0 (135.6) (137.2) (137.4) 140.0 140.0 (140. 0) 142.2 146.0 (150.2) (157.0) (160.3) by depressions in the surface and also by variations in color, which were accentuated when the verte- brae were soaked in water, or when they were stained (Galtsoff, 1952). We examined the verte- brae either witli the naked eye or with the aid of a wide-field binocular microscope. We believe that these annuli on scales and vertebrae are probably formed during winter or early spring. Scales were legible for most fish weigliiiig 50 pounds or less, but rarely for larger ones. As scales could be more readily collected than vertebrae, we used scales for most of our age determinations of small fish, resorting to vertebrae for larger speci- mens. The material from each specimen was usually examined independently, and often also by our colleague Donald Allen. When readings Table 2. — Average foil, length of htuefin tuna taken in the vicinity of Cape Cod, for each year of age from readings of annuli on scales and vertebrae Age in years Length in cm. Number of specimens --- 38.0 56.9 76.5 90.5 118.8 136.0 156.4 161.6 174.4 188.1 203.4 224.6 233.7 243.3 248.0 1 13 2 - - 3 - 28 4... - _.. 34 6 -.- _ 29 6 12 7 - 7 8 7 9 6 10 6 11 2 12 _ 6 13 7 14 3 differed, material was reexamined. If the differ- ence between extremes remained greater than 2, the specimen was discarded. For differences of 2 or less, the average value or the unit closest to it was used. Actually, there were few disagree- ments in readings for fisli up to 50 or 60 pounds. Legible scales and vertebrae were found for 28 fish and counts of annuli on scales agreed with those on the vertebrae. Readings for fish of 70 to 270 pounds often differed by 1 year; those for larger fish sometimes differed by 2 years or more. Lengths (table 1) are from annuli counted from scales or vertebrae, or both; length-fre- quency distributions (fig. 3) are for each year of age ; average length (table 2) is for each year. ANALYSIS OF LENGTH FREQUENCIES Another method of estimating age and growth is by following the seasonal progression of dom- inant size groups. This is especially useful for species that spawn over a fairly short season and grow rapidly. Evidently tlie bluefin tuna meets these conditions, as even casual observers notice the regularity with which catches of small tuna can be ranked in size categories by eye. Moore (1952) and Postel (1954) analyzed size frequencies to determine the ages of yellowfin tuna in the Pacific and the tropical eastern Atlantic, re- spectively. Aikawa and Kato (1938) used the GROWTH OF BLUEFIN TUNA 43 saiiic iiH'thoil ill coiijunctiuii with counts of vertebral rings in studying tlic growth of western Pacific tunas, as did Partio (1955) for northeastern Pacific albacore. Westnian and Gilbert (1941) and Westnian and Neville (1942) used the method in conjunction with scale studies for bluefin tuna taken off Long Island, X.Y. We have based our size-frequency study on lengths rather than on weights, as we believe tluit, for fish of a given age, lengths are subject to smaller and more regular variations. Because we lack sufficient data on the large sizes, we have Table 3. — Length frequencies of bluejin tuna from iO.H to 56.5 inches long taken off Long Island ' in 1941 ii^d off Kew England - in 1950-57 from late June to mid- October Number of tuna Length in inches July August September Octo- ber Ace in years J 1-15 16-31 1-16 17-31 1-15 16-30 1-18 21 7 9 5 31 33 25 4 1 4 13 17 12 1 1 9 31 10 1 4 1 42 4 58 7 26 14 3 1 2 2 22 23 24. .. 25 I 26.. 27. 1 23 60 !)7 25 10 2 5 24 68 114 64 28 1 27 50 120 100 5fi 20 3 10 88 122 65 22 6 7 44 119 178 141 96 31 9 28... 29 3 5 41 97 72 17 4 3 1 1 1 3 7 3 15 2 3 8 15 9 4 30 31. 32 33 II 34. 9 L 30 69 87 53 14 1 35 6 5 35... a 13 46 153 253 203 112 33 11 4 5 36... 6 8 22 26 18 10 5 3 38. 24 6 40 26 94 29 76 28 52 23 16 14 4 2 39... 40.. 41 1 19 28 37 58 29 24 5 41.. 8 2 III 42 42 10 15 17 14 11 8 3 3 4 11 18 11 8 4 ti 43 23 57 23 6 8 9 4 8 2 15 6 9 4 5 1 1 6 11 13 21 9 7 7 1 2 45 46. 47 48 49 IV 49. 5 5 2 5 1 2 3 1 2 2 3 2 2 4 1 2 4 5 50... 51 52 2 2 1 2 1 6 6 1 54 55 56 ' Westman and Xeville's (1942) sample consisting of 1,129 fish, was meas- ured at Frecport, L.I., (N.Y.). ' A tew fish caught off Nova Scotia, Long Island (N.Y.), and .Vew Jersev arc included. Most of this sample was taken in the vicinity of Cape Cod', but many of the fish were from the offshore waters, mostiv in the vicinity of Georges Dank. Frank Kilcv measured 1,8'Jl fish at I'rovincelown (\I:iss.") in 1953-54. We are indchted to Lewis R. Day of the Fisheries Hoard of Canada for measuring 5 fisti in Nova Scotia in ly.TO, Jean .McClean Wight of nest Hartford (Conn.) for measuring 5 fish in Nova .-^cotia in l'J.54. Capt. Charles A. Mayo. .Ir.. of I'rovlncetown (Mass.) for measuring ,57 fish there in 19.W, and Dr. Kobert II. Oibbs, Jr., recently of the Wooiis Hole Occano- ?r,''l'"''' '"'•"""""• '">■ mciisuring 69 fish In the Cape Cod area in 19.W and > . ' •. ., '' remainder of the sample was measured by the authors and Donald ' Includes 102 fish measured June 28-30, 1953. 551440 O — 60 2 not attempted to analyze lengths greater than 56.5 inches. From various sources, we have compiled length measurements of 4,990 bluefin tuna less than 56.5 inches long. With the exception of the 1941 sample which was measured on JjOng Island by Westman and Neville (1942), and several fish less tiian 20 inches long which were from more south- erly waters, most of these were taken in the New England area from 1950 to 1957. With the excep- tion of a few specimens less than 20 inches long, the fish were caught from late June to mid- October. We have followed the metiiod of Westman and Neville in measuring the fish to the nearest inch with a tape from the snout to the fork of the tail, following the curvature of the body. Where we used calipers for determining length, we estimated the "curved" measurement from a conversion factor.-' The measurements for fish more than 20 inches long, for all localities and years combined, are listed by half monthly periods in table 3. The data for smaller fish were gathered from more diverse localities and extend over a greater portion of the year, hence are listed in more detail in table 4. Table 4. — Lengths of bluefin tuna less than 20 inches long, with dates and localities of capture and so^trces Length Num- Date m inches ber of fish Locality Source June 9, 1953 1.8 Miami area, Fla.. Rivas (1954). Julv 14. 1953 11.7 do Do. July 17, 1953 12.3 do Do. July 19, 1953 10.3 do Do. July 16-23, 1954.. 12.0 Atlantic City, N.J. William Upper- man. i July 19-26. 1954.. 9.0 do. Do. Julv 23, 1953 11.4 Miami area, Fla.. Rivas (1954). July 23. 1957 7.7 do Al Pflueger.s July 24, 1953 11.8 do Rivas (1954). Do 12.9 12.7 do do Do. July 25. 1954 Al Pflueger.' Jldy26, 1953 11.9 do Rivas (1954). Julv 29, 1953 12.2 do Do. Do 13.5 do Do. Julv 30, 19,53 11.2 do Do. July 31. 1953 11.7 do Do. Julv 31, 1954 12.0 do Al Pflueger.' Aug. 10, 1953 12.4 do Rivas (1954). Aug. 12, 1954 13.0 Brielle, N.J Mrs. K. R. Mayer." Aug. 23. 1940 14.2 do Manasquan Mar- lln and Tuna Club. Aug. 25. 19,53 9.0 do Mrs. K. R. Mavcr.J Aug. 29, 1938 12.7 do C. W. Hoffman. 1 Aug. 29, 1952 15.5 Off Martha's Vineyard. Mass. R/V Htar. R. Wolf.! Sept. 2, 1939 12.5 1 Brielle. N.J Westman :md OillH-rt (1941). Sept. 1-7, 1953.... 12.7 4 Oulf of Mexico, r.S. National (29°03' N, Museum.' 88°54' W). Do 13.8 1 do Do.< See footnotes at end of table > A straight line fitted by inspection to a i)lot of straight (cali|)er) length against curved (tai)e) length, based on measurements tor 185 individuals 37 to 257 cm. long, indicated that straight length was 0.958 ot curved length. 44 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 4. — Lengths of bluefin tuna less than 20 inches long, with dates and localities of capture and sources — Continued Date Sept. 6, 1963 16.0 Do --. 16.5 Sept. 5, 1957 Sept. 6, 1952 12.0 15.6 Sept. 7, 1957 Do 12.0 13.0 Sept. 12, 1952 16.0 Sept. 12, 1963 9.8 Do 11.0 Do 14.0 Sept. 14, 1957 Sept. 16, 1953 13.2 18.0 Sept. 17, 1952 14.1 Sept. 18, 1952 Sept. 20, 1967 Sept. 21, 1964 16.9 14.0 13.9 Sept. 21, 1957 Do 31.0 14.0 Sept. 22, 1957 Do Sept. 23-24, 1953.. 14.0 16.0 17.4 Sept. 24, 1967 Do 13.0 14.0 Do Oct. 11, 1953 15.0 15.0 Oct. 13, 1962 14.6 Oct. 18, 1963 17.0 Nov. 10, 1953 Nov. 12, 1952 Nov. 16, 1952 Nov. 27, 1952 Jan. 5, 1951 Jan. 27, 1959 15.6 17.7 16.6 17.6 18.3 18.8 Length in inches Num- ber of flsh Locality Brielle, N.J do Ocean City, Md. 0£E Martha's Vineyard, Mass. Ocean City, Md. do Off Long Island, N.Y. Brielle, N.J. do ....do Ocean City, Md. Brielle, N.J Off Martha's Vineyard, Mass. ...do.... Ocean City, Md.. Off the Carolinas, (33°10' N, 77°25' W). Ocean City, Md. . ....do ....do ....do.. Off Martha's Vineyard, Mass. Ocean Citv, Md.. ...-do ....do Brielle, N.J Cape Hatteras, N.C. Brielle, N.J Miami area, ....do ....do. ....do ....do.... Off Cape Hat- teras, N.C. ria. Source Mrs. K. R. Mayer.3 Do.a F.J. Mather. Do. M. L. Dennis. Do. Rivas (1964). Mrs. K. R. Mayer.3 Do.3 Do.s M. L. Dennis. Mrs. K. R. Mayer.3 R/V Carpn. F. J. Mather. Do. M. L. Dennis. U.S. Fish and Wildlife, Bruns- wick, Ga. M. L. Dermis. Do. Do. Do. MIVAlbalwss III, J. Taylor.' M. L. Dennis. Do. Do. Mrs. K. R. Mayer.3 F. J. Mather. Mrs. K. R. Mayer. 3 Rivas (1964), Do. Do. Do. Do. UIV Albatross 1 11. R. Brigham and L. Lawday.2 ' Measurements were checked with ruler on photographic prints. ' Specimens were made availaljle to us by kindness of the individuals listed. 3 Measurements taken by charter boat captains who tagged the tuna, an d collected for us by Mrs. Mayer. * Measurements were made by Isaac Ginsburg and transmitted to us by Dr. L. P. Schultz. The length frequencies for all localities combined between late June and mid-October are shown graphically for each half monthly period by years in figures 4-10, and for all years combined, by half monthly periods in figure 11. The number of fish in any given size group and period varies consider- ably from year to year, due to nonuniform sam- pling and availability, and to variations in the numerical strength of year classes. The general pattern of size groupings is consistent, however, with maximum and minimum numbers occurring around the same lengths year after year. It seems obvious that these groupings represent dif- 50 100 150 200 250 FORK LENGTH-CENTIMETERS 300 Figure 3. — Frequency distribution by 5-centimeter groups of lengths of bluefin tuna for counts of annuli. ferent ages. We have arbitrarily designated with vertical lines the points which we believe best separate the various age groups. Usually there seems to be little question as to where these lines should be drawn. If doubtful we based our judg- ment on a comparative study of data for the en- tire series of years. In a few such cases, we as- signed some of the fish at a low point to one age and the rest to the other. Corresponding broken lines separate the data in table 3. AGE DETERMINATION The question arises as to whether or not the fish forming the first modal group appearing in our length frequency study are young of the year. In figure 12, we have compared the lengths of fish in each age group as determined by counts of an- nuli with the lengths of those in corresponding age groups as determined by length frequencies. On examining this figure, we find the assumption that fish in the first modal group are young of the FORK LENGTH-INCHES Figure 4. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken July 1-15, by years. The 1953 sample includes 102 fish measured June 28-30. GROWTH OF BLXJEFIN TUNA d 60 40 20 X iZ 100 o 80 S 60 CD § 40 ^ 20 40 20 45 NONE MEASURED |9 5 5 in_^_g ,1956- NONE MEASURED '9^7 o4" I 1 1 J 1 1 1 1 1 1 1 1 1 1 1 1 . 1 [ 1 1 1 1 20 30 40 FORK LENGTH-INCHES 50 Or--- if 2_' ^ 1954 J n 1955 1956 1957 I M M I I I I I I I ' 10 I I I M I I I I 20 30 40 FORK LENGTH-INCHES 50 FifiiTRE 5. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken July 16-31, by years. Figure 6. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken August 1-16, by years. year is supported by tlie fact that no annuli are found on scales and vertebrae of fisli of this size. Moreover, it is in accord with the conclusions of Sella (1929) and others studying the European bluefin tuna, which were officially accepted by the International Council for Exploration of tlie Sea in 1932 at Malaga (Conseil International pour I'Ex- ploration de la Mer, 1933), and witli tliose reached by Westman and Gilbert (1941), Westman and Neville (1942), and by Rivas (1954), for bluefin taken ofi' New York and Miami, Fla. It is sup- ported by our failure in all our observations, in- quiries, and searching of literature and records, to find any evidence that a smaller size group exists. We conclude therefor that tuna in the second size group (corresponding to fish with 1 animlus) are 1 year old; in the third size group (2 annuli) are 2 j'ears old; and so on through 4 \ 46 FISHERY BULLETIN OF THE FISH AND "WILDLIFE SERVICE 60 40 20 40 20 X 220 en u- 200 o 180 ^ 160 § 140 ^ 120 100 80 60 40- 20 NONE MEASURED 1950 i H- -A. n nr 1951 - n m NONE MEASURED 1952- 1953 1954 m J955 W~ 1957 Vf H- • ' 20 30 40 FORK LENGTH-INCHES 50 Figure 7. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken August 17-31, by years. years. Although length data on older ages are not sufficient for analysis, we believe that counts of annuli are useful for estimating the age of older fish despite the decreasing reliability of readings with increasing age. We find also that the analysis of length-fre- quency curves is consistent with age determina- tions h\ counts of annuli on scales and vertebrae. Such discrepancies as exist probably result from the fact that the samples for age readings were smaller and less uniformly distributed through the seasons than were those taken for length measure- ments. For example, most of our samples for counts of annuli of 3-year-olds were collected in early summer or fall rather than in midsummer. Even so, correspondence in conclusions reached from the two kinds of data is close. 20 30 40 FORK LENGTH-INCHES 50 Figure 8. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken September 1-15, by years. 20 30 40 FORK LENGTH-INCHES Figure 9. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken September 16-30, by years. Weight frequencies of landings of medium- sized bluefin tuna in Cape Cod Bay and Nova Scotia in the years 1948-51 (fig. 13) show a ten- dency for modal weights to coincide with sizes determined by counts of annuli for ages 5-7. Most clear cut cases are the 5-year-olds in Cape Cod Bay and 6-year-olds in Nova Scotia in 1948, GROWTH OF BLUEFIN TUNA 47 20 30 40 FORK LENGTH-INCHES I Figure 10. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken October 1-18, by years. and the 6-year-olds in Nova Scotia in 1949. Split modes show a preponderance of 7-year-olds in both areas in 1950, suggesting the progression of the year class of 1943 through the Cape Cod Bay fishery 1948-50 and the Nova Scotia fishery 1949-50. Two successful tagging experiments with bluefin tuna yielded approximate data on their actual growth. One fish, tagged off Cape Cod, Mass., July 27, 1954, was recaptured by French fisher- men in the Bay of Biscay August 16, 1959. When tagged, the fish measured 72.5 cm., and its weight when recaptured was reported as approximately 65-70 kilograms (143-154 pounds equivalent to about 150-154 cm.). These sizes are near the lower limits for ages 2 and 7, respectively, from table 1. The other was tagged August 11, 1957, off Chatham, Mass., and recaptured August 30, 1959, off Gloucester, Mass. Its weight when tagged was estimated as 65 pounds (equivalent to about 114 cm.) by an experienced fisherman, and it weighed 130 pounds (equivalent to about 150 cm.) when recaptured. These lengths are in good agreement with those listed in table 1 for ages 4 and 6, respectively. Hence the results of these experiments are in reasonable agreement with our age determinations by counts of annuli. GROWTH OF YOUNG BLUEFIN TUNA As length measurements are several times as numerous as counts of annuli, and permit us to trace growth during each summer as well as from year to year, we shall base our discussion of growtli of young tuna on length frequency analvsis. 10 20 30 40 FORK LENGTH-INCHES Figure 11. — Frequency distribution by 1-inch groups of lengths of bluefin tuna taken in 1941 off Long Island and in 1950-57, mostly off Xew England, by periods. Three very small tuna taken off New Jersey in 1938-40 are also included. Table 5 lists the average lengths of the fish in each age group of fish more than 20 inches long, as demarcated by vertical lines in figures 4-11. It identifies year classes and also shows average lengths for all years combined. We plotted these lengths by periods in figure 14 and fitted curves to them empirically, taking into account the number of measurements represented by each point, ex- cept in 2 or 3 where tlie preponderant samples were not, in our opinion, composed of average- sized fish. 48 FISHERY BtlLLETIN OF THE FISH AND WILDLIFE SERVICE 40 20 40 (- 40 UJ20 £ a. to 40 20 1 1 M I '''' M 1 1 ' M H I I I I I 1 1 1 I I 1 1 1 1 I I I I I I I I I I I I I I I I I I I ■' I I I I I I I I I I I I I IZ ■y^i::^\ _!ZL. 'I I I "I I II I ll I I 'I" I ll "" I" " I" "I " " ll ' "I" " I' Mlh.l I ll M 20 30 40 50 FORK LENGTH-INCHES 60 Figure 12. — Frequency distribution by 1-inch groups of lengtiis of bluefin tuna for age groups 0-IV as deter- mined by length frequencies (solid lines) and by counts of annuli (dotted lines). The fish less than 20 inches long listed in table 4 obviously form a distinct age group. Figure 15 shows the averages of these measurements with a curve drawn by inspection to fit the points and also to fit in with the curve for 1-year-olds from figure 14. Although our unpublished studies of the distribution of the bluefin tuna indicate that all these fish belong to one population, samples from different areas have been designated by different symbols. Figure 16 shows a curve of estimated growth of bluefin tuna for the first 4}i years of life, and table 6 lists the average sizes at the middle of each month, as indicated by this curve. Figure 16 indicates extremely rapid initial growth and distinct seasonal variations in growth rate. Bluefin tuna spawn during an undefined period in spring (Rivas, 1954; Bullis and Mather, 1956). Assuming, as we did in drawing figure 16, that hatching occurs in mid-May, the young may grow at a rate of nearly 6 inches per month to reach a size of 8)^ inches by July 1. In the ensuing discussion, however, we shall consider July 1 as the date of birth and shall refer to the period July 1-October 16 as "summer" and the remainder of the year as "winter." The growth rate diminishes rapidly during the first summer, but the average rate is estimated at 2 inches per month. The rate continues to decrease during most of the first winter, averaging about four- 70 100 200 270 WEIGHT-POUNDS Figure 13. — Weight frequencies by 10-pound groups of bluefin tuna from 70 to 270 pounds, taken in Cape Cod Bay and ofT Nova Scotia in 1948-51, by years. fifths of an inch per month. For the remainder of the period studied, ages 1-4, the growth rate does not change greatly with age, averaging about iy2 inches per month in summer and about one- third of an inch per month in winter. ESTIMATED ANNUAL GROWTH OF BLUE- FIN TUNA THROUGH 10 YEARS The average sizes at each age, determined by length frequency analysis for ages 0-4, and by counts of annuli for older ages, are plotted in figure 17. The curve shown was derived graphically, starting from the point for 3-year-olds, from the GROWTH OF BLUEFIN TUNA 49 Wiilford (1946) transformation of our data. As we found it possible to read only a small percent- age of the material available for ages beyond 10 years, we luive drawn the curve to that point only. Average sizes at mid-summer for bluefin tuna of ages 0-1 as indicated by this curve are listed in table 7. 3 40- FiGURE 14. — Average length.s of bluefin tuna by age groups as indicated in figures 4 to 11. Curves were fitted empirically. The numbers of fish in the samples for all years combined are indicated. Samples consisting of less than 5 fish were not shown. CONCLUSIONS Our results through the fifth summer of life for bluefin tuna taken off New England are substan- tially in agreement with those of Westman and Neville (1942) for fish taken off Long Island. Our study of the growth of these young tunas, however, was based on a nmch larger sample, NEW JERSEY- C»«*E COO MARVLAND-CAROLINAS MIAUI AflEA GULF OF UEIICO I AUGUST [sEPTEweml oc'obew | tovcwetHf ptcExatw I .lainjaB* \ 30 £ -25 I Z Ul - 20 o I o 5 Figure 15. — Lengths of bluefin tuna less than 20 inches long (young of the year), from table 4. The curve of estimated growth was fitted by inspection. AGES AND SEASONS „, 4 „,. iO - 1 li i2 ; li i 3; Ij ; 4:1! MONTHS ANO YEARS Figure 16. — Estimated growth of young bluefin tuna. Broken lines indicate estimated lengths in periods for which data are lacking. The upper scale shows ages and seasons as assumed in the text, and the lower scale shows ages assuming that hatching occurs at mid- May. and the sampling was spread over several years rather than 1 year. Differences in growth be- tween year classes were found to be slight and the results of the analysis of the composite sample are believed to approximate the average encount- ered in nature. Although it was not possible to fully verify the readings of annuli for older fish by the analysis of size frequencies, and difficulties in reading the annuli increased with their number, we have exlendcd our (let(>rminations to consider- ably older ages than lias previously been done for western Atlantic bluefin tuna. 50 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 5. — Average lengths of biuefin tuna, in each age group as indicated in figures 4 to 11, listed by ages, years of measurement, and year classes Year of measurement Average lengths of fish in inches (numbers ot fish in parentheses) Age In years July August Septemljer October Year class 1-15 16-31 1-16 17-31 1-16 16-30 1-18 n941 21.8(7) 22.9(40) 23. 1 (39) 23.4(9) 23. 6(9) 26.0(1) 24.0(28) 24.5(4) 24.0(1) 25.2(29) 24. 6(59) 25.2(36) 23.7(7) 27.0(5) 1940 1950 1951_- 21.9(7) 23.0(2) 21.0(1) 22.0(4) 24.0(48) 25.1(12) 25.6(14) J 1952 23. 3(3) 26.0(1) 1951 1953 1962 1954 22.6(8) 24.5(4) 23.1(94) 31.3(126) 29.2(10) 31.2(29) 30.7(25) 30.7(75) 30.6(39) 31.5(2) 30.9(306) 38.9(78) 25.0(1) 1963 1955 26.0(1) 24. 1 (53) 33.8(9) 1964 21.9(21) 31.1(71) 23.9(47) 32. 1 (9) 24.8(131) 33. 6(8) 26.6(31) 33.3(14) 26.0(2) All years. n941 1950 32.5(7) 34.6(7) 1948 1951 30.5(33) 30.2(57) 31.2(9) 30.8(7) 31.6(16) 31.4(6) 30.9(214) 31.2(71) 33.2(6) 32.3(9) 33.4(21) IE 1952 1960 1953 31.1(6) 1951 1954 32.2(216) 33.5(2) 34.2(12) 33.6(49) 41.1(72) 1962 1955 34.4(13) 33.3(26) 40.8(202) 34.2(29) 33.9(41) All years 30.6(176) 37.5(37) 31.0(316) 39.3(208) 32.2(23?) 40.4(163) All years. 1941 1950 43.0(1) 1947 1951 36.6(72) 36.0(1) 36.9(147) 37.2(5) 36.5(30) 36.0(2) 37.6(1) 40.5(9) 38.8(17) 1948 1952 III 1953 37.2(259) 37. 6(8) 37.8(285) 38. 4(126) 195C 1954 39.0(648) 38.0(1) 40.0(5) 39.3(827) 46.7(6) 39.8(34) 39. 8(66) 40.4(27) 1961 1955 39.6(113) 40.0(1) 40.4(31S) 47.9(4) 1956 39.0(7) 37.5(382) 45,0(19) 39.8(6) 38.4(625) 44.4(30) 40.3(10) 40.6(134) 19.53. 36.7(263) 43.5(4) 40. 1 (98) All years. 1937 (1941 1950 60.0(15) 1946. 1951 45.8(16) 45.8069) 43.3(25) 1947. 1953 45.5(24) 47.0(1) 46.7(27) 48.0(1) 46.3(6) 47.4(65) 1949 IV 1954 .. 45. 1 (99) 44.3(4) 46. 6(48) 47.6(8) 1950. 1955 1961 1956 43.5(45) 43.4(10) 47.7(11) 1952 1957 44.6(4) 46.1(109) 44.5(2) 48.0(76) 1953. All years 45.5(214) 44. 4(89) 45.1(69) 46.5(55) 46.8(18) All years. Table 6. — Estimated sizes of young biuefin tuna at the middle of each month derived from figure 16 Month May June July August September, October November, December. January... February.. March April Fork length in inches curved measurement Age in years I II III 21.0 28.7 35.8 6.6 21.7 29.5 36.2 10.3 22.7 30.6 37.3 12.fi 23.9 32.0 38.6 14.3 25.0 33.2 40.0 16.3 26.1 34.5 41.4 16.7 27.2 35.3 42.6 17.6 27.8 35.4 43.0 18.4 28.2 36.6 43.1 19.1 28.3 36.6 43.2 19.7 28.4 35.7 43.3 20.2 28.6 35.8 43.4 IV 43.5 43.7 44.5 46.7 46.9 48.0 Fork length in centimeters straight measurement Age in years 15.8 25. 1 30.6 34.8 37.2 40.6 42.9 44.8 46.5 48.0 49.1 61.1 52.8 55.3 58.2 60.9 63.5 66.2 67.7 68.6 68.9 69.1 69.4 71.8 74.5 77.6 80.9 84.0 85.9 86.2 86.4 86.6 86.9 87.1 III 87.1 88.1 90.9 94.0 97.4 100.8 103.9 104.8 105.0 105.1 105.3 105.7 IV 106.0 106.7 108.5 111 3 114.1 117.0 Weight in pounds Age In years 0.2 0.6 1.1 1.9 2.1 2.9 3.5 4.1 4.6 6.0 5.4 6.2 6.8 7.8 9.4 10.0 11.5 13.6 14.2 14.6 14.9 15.1 15.3 15.5 17.0 19.0 22.0 24.4 26.5 29.0 30.0 30.4 30.6 30.8 31.0 III 31.2 32,0 34.0 36.6 42.0 46.0 49.0 49.6 50.0 50.2 50.4 50.7 IV 51.0 52.0 55.0 60.5 65.0 68.0 GROWTH OF BLUEFIN TUNA 51 I ^ eo 1*1 K I 60 -1 I I I 1 I I 1 I I I I T" '.4 [/A '#' 379 ,1, 99 'I LEGEND NO OF FISH IN SAMPLE 33 AVERACE 700 600 500 450 400 350 300 250 100 0- AGE-YEARS Fir IRE 17. — Estimated growth curve for bluetin tuna (heavy broken line), with points derived from length frequencies for ages 0-IV years and from counts of annuli for older ages. Lighter broken lines, fitted by inspection, show estimated limits of variation. Table 7. — Estimated sizes of bluefin tuna during the summer for ages 0-10 years From figure 17 Length Weight i Age in years Centimeters (straight) Inches (curved) n pounds Average Range Average Range Average Range 32 57 77 95 IH 133 149 163 177 190 201 22-43 50-66 66-88 83-109 102-127 122-148 134-165 HS-lS'i 161-197 172-210 182-221 13.0 23.5 31.5 39.0 47.0 55.0 61.5 67.0 73.0 78.0 82.5 9-18 20-27 27-36 34-45 42-52 50-61 55-68 61-75 66-81 71-86 75-91 1.5 8.5 22.0 40 69 100 140 185 240 290 340 0. 6-3. 5 I 5-13 11 Ill 14-30 25-50 IV 45-90 V 80-140 VI 105-190 VII 140-250 VIII IX X . . 180-320 220-395 255-455 Our growth data for fisli up to 10 years of age are in good agreement with tliose of -Selhi (1929) for Mediterranean bhiefiu tuna. Our few reaihiigs for older ages indicated somewhat larger sizes for western Atlantic fish for the respective ages. Bella's work was based on a larger sample than ours, and he used special instruments which we did not have. Therefore tliis apparent difference may result from less complete sampling and less precise reading of annuli for the western Atlantic bhiefin, rather than to an actual differ- ence in the growth of the larger fish from the respective areas. LITERATURE CITED AiKAWA, HiROAKi, and M. Kato. 1938. Age determination of fish (Preliminary Rept. 1). Bull. Japanese Society Scientific Fisheries, vol. 7, No. 1, pp. 79-88. Translation from the Japanese by W. G. Van Campen in U.S. Depart- ment of the Interior, Fish and Wildlife Service, Special Scientific Report — -Fisheries No. 21. Arnold, Edgar L., Jr. '1951. An impression method for preparing fish scales for age and growth analysis. U.S. Depart- ment of the Interior, Fish and Wildlife Service, Progressive Fish Culturist, vol. 13, pp. 11-16. BuLLis, Harvey R., Jr., and F. J. Mather, III. 1956. Tunas of the genus Thunnus of the northern Caribbean. American Museum Novitates, No. 1765, pp. 1-12. Conseil International pour l'Exploration ue la Mer. 1933. Conference d'experts pour I'examen des m^thodes scientifiques et techniques k appliquer k r^tude des poissons de la famille des thonid^s. Conseil International pour l'Exploration de la Mer, Rapports et Procfes-Verbaux, vol. 84, pp . 91-103. Galtsoff, Paul S. 1952. Staining of growth rings in the vertebrae of tuna {Thunnus thynnus). Copeia, No. 2, pp. 103-105. Moore, Harvey L. 1951. Estimation of age and growth of yellowfin tuna {Ne.othunnus macroplerus) in Hawaiian waters by size frequencies. U.S. Department of the Interior, Fish and Wildlife Service, Fishery Bulletin, No. 65, vol. 52, pp. 133-149. Partlo, J. M. 1955. Distribution, age and growth of eastern Pacific albacore {Thunnus alalunga Gmelin). Journal of the Fisheries Research Board of Canada, vol. 12, pp. 35-60. Postel, E. 1954. La croissance du thon d nageoires jaunes Neothunnus albacoia (Lowe) dans I'Atlantique tropical. Bulletin de la Soci6t6 Zoologique de France, vol. 79, No. 2-3, pp. 85-90. Rivas. Luis Rene. 1954. .\ preliminary report on the spawning of the western North Atlantic bluefin tuna {Thunnus thynnus) in the Straits of Florida. Bulletin of Marine Science of the Gulf and Caribbean, vol. 4, pp. 302-322. Sella, Massimo. 1929. Migrazioni c habitat del tonno {Thunnus thynnus L.) studiati col metodo degli ami, con osservazioni su I'accrescimento, sul regime delle tonnare ecc. R. Coinitato Talassografico Italiano. 52 FISHERY BtTLLETIN OF THE FISH AND WILDLIFE SERVICE Sella, Massimo. Mem. 156, pp. 3-24. Translation in U.S. Department of the Interior, Fish and Wildhfe Service, Special Scientific Report — Fisheries No. 76. Walford, Lionel A. 1946. A new graphic method of describing the growth of animals. Biological Bulletin, vol. 90, pp. 141-147. Westman, J. R., and P. W. Gilbert. 1941. Notes on age determination and growth of the Atlantic bluefin tuna, Thunnus thynnus (Linnaeus). Copeia, No. 2, pp. 70-72. Westman, James R., and W. C. Neville. 1942. The tuna fishery of Long Island, New York. Board of Supervisors, Nassau County, Long Island, N.Y., pp. 1-31. O UNITED STATES DEPARTMENT OF THE INTERIOR, Fred A. Sea ton, Secretary FISH AND WILDLIFE SERVICE, Arnie J. Suomela, Commissioner Bureau of Commercial Fisheries, Donald L. McKernan, Director FECUNDITY OF RED SALMON AT BROOKS AND KARLUK LAKES, ALASKA FISHERY BULLETIN 180 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 Published by U.S. Fish and Wildlife Service • Washington • 1960 Printed at V.S. Governminl Prinlina Office. WashlniSfon For sale by the Superintendent of Documents, U.S. Government Prlntlnft Office, Washlneton 25, D.C. - Price 15 cents \ Library of Congress catalog card for the scries, Fishery Bulletin of the Fish and Wildlife Service: U. S. Fish and Wildlife Service. Fishery bulletin, a'. 1- Washington, U. S. Govt. Print. Off., 1881-19 V. in illus., maps (part fold.) 23-28 cm. Some vols, issued in the congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies : v. 1^9, Bulletin. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, V. 1-23) 1. Fisheries— U. S. 2. Fish-culture— U. S. i. Title. SH11.A25 639.20C173 9—35239* Library of Congress i59rlj5blj CONTENTS Page I lit roduct ion 53 Collection and treatment of materials 54 Relation between size of fish and fecundity 54 Variation between paired ovaries 55 Reproductive potential of spawning populations 56 Summary 58 Literature cited 60 in ABSTRACT The relation between mideye-fork length and number of eggs for red salmon at Brooks Lake (1957-58) and Karluk Lake (1958) is established. A review of available literature on fecundity at Karluk Lake indicates that there may have been a long-term decrease in the size of females and correspondingly in the average number of eggs per female. Annual variations in age composition by life-history categories and in sex ratios affect the number of eggs available for deixtsition. Analyses of Karluk Lake red salmon stocks show a relation between the ocean age and size of fish and their fecundity, those fish of a greater length of ocean residence and size having the largest number of eggs. Since a distinctive seasonal pattern in the occurrence of life-history categories and related sex ratios exists, it is theorized that the commercial fishery could be so concentrated as to deplete that i>ortion of the run of highest egg production potential. FECUNDITY OF RED SALMON AT BROOKS AND KARLUK LAKES, ALASKA BY WILBUR L. HARTNIAN AND CHARLES Y. CONKLE, Fishery Research Biologists Bureau of Commercial Fisheries Fecundity of red salmon, Onrarhynchus n^rka (Walbiuiin), was studied at Brooks and Karluk Lakes, Alaska, for use in estimating the repro- ductive potential of spawning stocks. Eepro- ductive potential is defined here as the total number of eggs available for seeding in a partic- ular spa^Tiing population. Such information Lake by way of Shelikof Strait, and Karluk Kiver (fig. 1). The Bureau of Commercial Fisheries maintains research stations at both lakes to investi- gate factors responsible for fluctuations in the abundance of salmon runs that have occurred in these areas (U.S. Fish and "Wildlife Service, 1958). Certain specific problems ditfer between p I I I I I ^ii^ 50 100 Mi FiGUUE 1. — Brooks and Karluk Lakes in western Alaska. forms the basis for determining sui-vival rates of red salmon during various life-liistory stages in fresh water. Adult red salmon enter Brooks Lake by way of Bristol Bay and Xaknek I\iver, and enter Karluk N' il:. Ai.prciv.'il fur puWlOiition, .Ian. l;!, IDCIl. Itiilli-tin ISO. 552701—60 the lakes, but all research is integrated info a broad study of the physical, chemical, and bio- logical facfore aft'ecting the fresh-water survival of red salmon. I\ed salmon were sampled from spawning mi- grations into Brooks Lake in I!).")" and 1958, and into Karluk Lake in in5s. Karlier data are avail- 53 54 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE able for Karluk Lake and are used for comparison in the discussion that follows. We wish to express our thanks to Allen IMc- Cready, William Pogue, and Patrick Tomlinson for aid in developing and conducting field opera- tions, and to Theodore R. Merrell and John B. Owen of the Bureau of Connnercial Fislieries, for reviewing the manuscript. COLLECTION AND TREATMENT OF MATERIALS Ovaries used in these studies were obtained from females trapped at adult innnigration sites at both lakes. They were taken throughout the season and over the size range of females in the stocks. A few females killed during beach-sein- ing and gill-netting operations in Brooks Lake were also used. Only females not fully ripe were examined. This reduced the possibility of includ- ing partially spawned females. Both ovaries were removed intact and placed in 20-percent formalin. An identification tag was attached to the right ovai-y to distinguish it from the left. After hardening for at least 48 hours, ovaries were removed from the formahn and tlior- oughly washed in water. The eggs were stripped from the ovarian tissue by hand and also thor- oughly washed in water. Total numbers of eggs were counted in each ovary of each female sampled at both lakes. A mechanical hand tally was used. Sampling niethods involving volume or weight were experimented with at Brooks Lake in 1958. The most reliable method was to extract 3 ran- dom 100-egg samples from each gonad and esti- mate the total count from the average count- weiglit relation of the selected samples. Fecundity was usually estimated within 2 percent of the actual count. This method is sufficiently accu- rate and should be considered where extensive fecundity studies are scheduled. RELATION BETWEEN SIZE OF FISH AND FECUNDITY A relation exists between the size of fish and the number of eggs in the body cavity. Bicker (1932) shows that the relation between fish length and egg count in brook trout (S'ah'efhn/.s: font'm- 5000' (9 o 4000" cc liJ m =3 3000- 2000- 1957 BROOKS LAKE • n = 38 958 BROOKS LAKE • n=22 1958 KARLUK LAKE + n = 39 I ' ' ' I ' I 45 50 55 60 MIDEYE-FORK LENGTH IN CENTIMETERS 65 FioiiRE 2. — Relation of egg coniils lo niideye-fork length for rod salmon at Krouks Lake, 1».'>T and l!)r,,s, and at Karluk Lake, 10.">S. RED SALMON OK BliOOKS AND KARLUK LAKES 55 a: 64-* u t- UJ < 2 t- Z 60- UJ o z < X H 56- e) z UJ _J • ^ cr o b2- ti. UJ > ' UJ Q 2 48- 44 T • • • • r 50 60 SNOUT- FORK LENGTH IN CENTIMETERS Figure 3. — ReLition of snout-fork length to mideye-foi-k length for 193 female red salmon sampled in the Karluk River, 1952. ali.s) is curvilineiir. Rounsefell (1957) states that over the narrower raiifres of lenortli at maturity found in Oneorhynchus sp., the straio;ht-line equa- tion adequately describes tlie relationship. This view was held earlier for red sahnon Ijy Foerster and Prit chard (1041). Tiiey believed that the overall relation between lish length and nunilicr of eirjrs was prol)ably loii'arithinic, but since adult s])awnin<); red salmon "renerally fall witltin a lim- ited size rano'e, the st rai^ilit-line equation was adequate. Total egg counts for females examined at Brecame increasingly greater than in the right. REPRODUCTIVE POTENTIAL OF SPAWNING POPULATIONS The number of salmon spawning in a locality is often regarded as indicative of the reproductive potential. Escapement figures have been used since the inception of management of our salmon resources to predict the strength of futuie runs. However, considerable variability in the repro- 1 studies on the spawning bioIoRy of the red salmon, Oncor- huiichiix tierkn (Walhaum). in Bristol Bay, Alaslia, with special reference to the effect of altered sex ratios. Ole A. Mathisen, doctoral dissertation. University of Washington. ductive potential may actually exist indepentlent of the actual number of spawners. Annual differences in sex ratios alone can cause substantial differences in the number of eggs available for deposition. Average fecundity for female red salmon was relatively stable during three years at Babine Lake (Withler, 1950), but a high preponderance of males in one of these years reduced by one-half the reproductive poten- tial of that spawning population, even though the RED SALMON OF BROOKS AND KARLUK LAKES 57 3000-1 T ^ 2000 NUMBER EGGS RIGHT OVARY Figure 4. — Egg counts for right and left civaries of female red salmon sampled at Brooks Lake, lO'iT and 195S, and Karluk Lake, 1958. number of spawners was approximately the same (table 1). Mathisen - in a study of e,) ranged from 100 to 75 per- cent, the 2-ocean-year categories (53 and fij ranged from 62 lo 32 percent, and the 3-ocean- year categories ranged from 38 to 35 percent. The reproductive potential and the egg contri- liution from each of the six major life-history categories comprising the 1958 run at Karluk Lake is shown in figure 5. Examination of these data indicates that 89 jwrcent of the eggs for lX)tential deposition came from the two 2-ocean- year categories, 5^ and 64. Barnaby (1944) shows that a seasonal pattern of appearance of the different life-history cate- gories exists at Karluk Lake. As the daily com- position of the run changes throughout the season, obviously so does the reproductive potential. It is possible that the commercial fishery could have been concentrated during that seasonal time of migration when important groups such as the 53 and 64 categories were moving through the fish- ing grounds. As a result, tlie reproductive potential of the spawning populations could be seriously reduced because of a shift in age com- position to younger ocean-age groups. These fish on a one-for-one basis are less fecund. It is interesting that the 3-ocean-year categories (63 and 74) accounted for only about 6 percent of the 1958 escapement, which is the lowest contribution by these groups ever recorded (Rounsefell, 1958: 156). At the same time the 1-ocean-year cate- gories (43 and 54) accounted for 17 percent of the 1954 escapement, which is by several times the highest contribution from those categories ever recorded. Analysis of future Karluk escape- ments will determine if this substantial shift in ocean-age composition is the result of a real trend or merely variability in year-class strength. It is concluded that a detailed study of the re- productive potential of the spawning populations is necessary to establisli a basis for fresh-water survival studies of red salmon. It may also serve to help explain, at least in part, the causes for declines in red salmon runs over the jjast decades. SUMMARY The relation between egg count and mideye- fork length was derived for Brooks and Karluk Lakes red salmon. Comparing recent Karluk Lake data with ptist fecundity studies indicates that a long-term decrease in mean size and fecun- dity of females has occurred. Red salmon at Brooks and Karluk Lakes con- sistently had more eggs in the left ovaiy than in the right. RED SALMON UK BROOKS AND KARLUK LAKES 59 I I MALES FEMALES NUMBER OF FISH IN THOUSANDS Figure o.— Length-frequency distributions and ri'prdductive i]otential by life-history categories of the escapement at Karluk Lal;e, 11)58. I 60 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Variability in the reproductive potential of spawning stocks was shown to be attributable to at least 4 factoi-s: (1) biological differences in fecundity between fish of the same size, (2) change in the life-history composition of spawning stocks, (3) differences in sex ratios between life-histoi-y categories, and (4) seasonal differences in repro- ductive ix>tential evidenced from the well defined pattern of occurrence by life-history categories at Karluk Lake. The possibility of the commercial fishery being concentrated on the most fecund life-history cate- gories is theorized as a contributing factor to the declines in red salmon runs over the past several decades. LITERATURE CITED Barnaby, Joseph T. 19-14. Fluctuations in abundance of red .salmon, Oiicor- hynchux »erka (Wall)a\im), of tlie Karluk River, Alaska. Fishery Bullrtiu, U.S. Fish and Wildlife Service, vol. r,0. No. 3!), i.ii. 2.37-29.">. Brown, C. J. D., and (Jertri'de C. Kamp. 1912. Gonad measurements and egg counts of brown trout (Salmo trutta) from the Madison River, Mon- tana. Transactions of the American Fisheries Soci- ety, vol, 71, iip. 19.V200. Chamberlain, F. M. 1907. Some observations on salnmu and trout in Ala.ska. Report Conuuissioner of Fisheries for 1900. Bureau of Fisheries Document 027, 112 pp. FoERSTER, R. Earle, and Andrew L. Pritchard. 1941. Observations on the relation of egg content to total lengtli and weight in the smkeye salmon (0»- rorhyncli IIS iicrka) and the pink salmon (0. gor- biischa). Transaction Royal Society of Canada 3~>. sec. V:51-60. Gilbert. Charles II., and Willis H. Rich. 1927. Investigations concerning the red-salmon runsi to the Karluk River, Alaska. Bulletin U. S. Bureau of Fisheries, vol. 4.3, pt. 2, i)p. l-(>9. Xelson, Philip R. 19.19. Effects of fertilizing Bare Lake, Alaska, on growth and production of red salmon ( O. iicrka). Fishery Bulletin, U.S. Fish and Wildlife Senice, vol, 60, No. 1.59, pp. .59-86. Ricker, William E. 1932. Studies of speckled trout { SaJreliniis fonthinliK) in Ontario. Publications Ontario Fisheries Research Laboratory, No. 44, pp. G7-110. In University of To- ronto Studies, Biol. Ser. No. .36. Rounsefell, George A. 1957. Fecundity of North American Salmonidae. Fishery Bulletin, U.S. Fish and Wildlife Service, vol. .57, No. 122, pp. 451-468. 19.5S. Factors causing decline in sockeye salmon of Karluk River, Alaska. Fishery Bulletin, U.S. Fish and Wildlife Service, vol. .58, No. 130, pp. 83-169. U.S. Fish and Wildlife Service. 1958. Progress report on Ala.ska fisheries management and research for 1957. U.S. Fish and Wildlife Serv- ice, Si>ecial Scientific Report-Fisheries No. 25S, 22 pp. WlTHLER, F. C. 19.50. Egg content of Babine sockeye. Fisheries Re- search Board of Canada, Progress Reports, Pacific Coast Station, No. 82, pp. 16-17. o UNITED STATES DEPARTMENT OF THE INTERIOR, Stewart L. Udall, Secretary FISH AND WILDLIFE SERVICE Bureau of Commercial Fisheries, Donald L. McKernan, Director FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC By Frederick H. Berry and Louis E. Vogele fKS^\CAT\^ FISHERY BULLETIN 181 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 PUBLISHED BY UNITED STATES FISH AND WILDLIFE SERVICE • WASHINGTON . 1961 PRINTED BY UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON, D.C. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington 25, D.C. Price 40 cents Library of Congress catalog card for the series, Fishery Bulletin of the Fish and Wildlife Service: U. S. Fish and Wildlife Service. Fishery bulletin, v. 1- Washington, U. S. Govt. Print. Off., 1881-19 V. in illus., maps (part fold.) 23-28 cm. Some vols, issued in the congressional series as Senate or House documents. Bultetins composing v. 47- also numbered 1- Title varies : v. l-i9, Bulletin. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, V. 1-23) 1. Fisheries— U.S. 2. Fish-culture— U.S. r. Title. SH11.A25 639.206173 9—35239* Library of Congress [59r55bl] CONTENTS Page Introduction 61 Methods 62 Identification 62 Key to genera of Monacanthidae from the western North Atlantic 63 Keys to species of Monacanthidae from the western North Atlantic 63 Description of genera and species 64 Alulera Cloquet 1816 65 Alutera monoceros (Linnaeus) 1758 65 Alutera scripta (Osbeck) 1765 66 Alutera schoepfii (Walbaum) 1792 66 Alutera heudelotii HoUard 1855 67 Monacanthus Oken 1817 68 Monacanihus tuckeri Bean 1906 69 Monacanthus ciltatus (Mitchill) 1818 69 Stephanolepis Gill 1861 70 Stephanolepis hispidus (Linnaeus) 1758 71 Stephanolepis setifer (Bennett) 1830 72 Amanses Gray 1833 73 Amanses pullus (Ranzani) 1842 73 Literature cited 74 Appendix 76 A. Figures 76 B. Tables 96 C. Specimens examined 100 Addendum 109 in ABSTRACT Filefishes of the western North Atlantic, important forage fish because of their abundance, have been inadequately described and are difficult to identify to species. In determining the species that occur in the western North Atlantic, several thousands of young and adult specimens were examined, and four genera and nine species were found to be valid and separable by external characters: Alulera monoceros, A. scripta, A. schoepfii, A.heudelotii, Monacanthus ciliatus, M. tuckeri, Stephanolepis hispidus, S. setifer, and Amanses pullus. Considerable intraspecific variation in profile was found, resulting in the synonymizing of several species, among them Alulera punctata and Stephanolepis spilonotus. Evidence is presented for the use of Monacanthidae as the family name, rather than Aluteridae. Keys to juvenile and adult specimens are presented, and proportional meas- urements and fin-ray counts are given in tabular form. Graphs, charts, photo- graphs, and drawings supplement the data to facilitate identification. Information on ranges and behavior is also presented. IV FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC By Frederick H. Berry and Louis E. Voegle, Fishery Research Biologists Bureau of Commercial Fisheries Filefishcs of the family Monacanthidae that occur in the western North Atlantic Ocean have been studied as a part of the biological research program of the United States Bureau of Com- mercial Fisheries Biological Laboratory at Bruns- wick, Georgia. This program is concerned with an evaluation of the fauna off the southeastern Atlantic coast of the United States. However, the study of the filefishes, because of their wide distribution, was not limited to this area. Several of the species concerned may occur north to Newfoundland and south to Brazd, and two of them probably have a worldwide distribution. Examination of collections made in recent years — during cruises of the M/V Theodore N. Oill, Oregon, Combat, and Silver Bay off the south Atlantic coast of the United States — by dip net, plankton tows, meter larvae net, and from stomach contents of larger fish taken by trolling, has indi- cated that the filefishes are a numerically abundant group and comprise an important part of the planktonic and forage-fish fauna. Recent catches at trawling stations in this area have furnished additional specimens for taxonomic, morphological, and environmental evaluation. Many of these specimens are in the collection of the Brunswick Biological Laboratory, but several museum and university collections of filefishes were examined to augment this material. Nine species of filefishes from the western North Atlantic were identified — primarily from speci- mens taken off the United States. These species are Alutera monoceros (Linnaeus), Alutera scripta (Osbeck), Alutera schoepfii (Walbaum), Alutera heudelotii Hollard, Monacanthus tuckeri Bean, Monacanthus ciliatus (Mitchill), Stephanolepis hispidus (Linnaeus) , Stephanolepis setijer (Bennett), and Amanses pullus (Ranzani). It is our purpose to give reasons for use of these names, to briefly diagnose and distinguish the Approved for publication, December 22, 1959. Fishery Bulletin 181. genera and species, to furnish illustrations of the species, and to list the specimens examined that they shall be readily available for more detailed future studies. In accomplishing these objectives we have determined several aspects of the life histories of these fishes, added to the knowledge of their distribution, discovered and confirmed certain anatomical features, described morphological and meristic variation, and through our own studies and from published accounts we have summarized their ta.xonomic relationships. We have been dogmatic in our taxonomic pronouncements con- cerning the several taxonomic problems that remain in order to stabilize the nomenclature until adequate numbers of specimens from the entire geographical ranges of these groups can be studied. We have explained the problems involved. Previously, the two most useful references for identifjang specimens of Monacanthidae from the western North Atlantic were publications by Fraser-Brunner (1940, 1941). We have found that the early hfe-history stages of the Atlantic coast species occur pelagi- cally in offshore waters, and beheve these waters probably are the principal habitat for those stages. The late juvenile and adult stages tend to adopt inshore or benthic-offshore habitats. Preliminary inspection of the extensive plankton collections made off the coasts of North and South Carolina, Georgia, and east Florida by the Gill in 1953-54 (see Anderson, Gehringer, and Cohen, 1956) has indicated that larval filefish are rela- tively abundant in waters of the Gulf Stream in this area. Samples taken by dip net on the Gill and other vessels have indicated the abun- dance of juvenile specimens in offsiiore waters, particularly in association with floating seaweed. Although juveniles are taken in inshore waters and are seined on the beaches, the specimens from offshore waters appear to be more abundant and of a smaller average size. Available data 61 i 62 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE on larger juveniles and adults show that they were usually taken from or near the bottom in shallow waters out to depths of about 25 fathoms. We have heard reports of large specimens of Alutera floating at the surface far out at sea, and we have also been told that these and large speci- mens of other filefish genera have been seen by skin divers on or near the bottom. All of the large specimens on which we have adequate collection data were taken by bottom-collection methods. We are indebted to the following persons for making specimens available that were instru- mental in this study: James E. Bohlke, Academy of Natural Sciences of Philadelphia; Eugenie Clark, Cape Haze Marine Laboratory; Earl E. Deubler, Jr., University of North Carolina; W. I. Follett, California Academy of Sciences; John D. Kilby, University of Florida; George S. Myers, Stanford University; Leonard P. Schultz, U.S. National Museum; Victor G. Springer, Florida State Board of Conservation; and Royal D. Suttkus, Tulane University. We are grateful for the assistance of the entire staff of the U.S. Bureau of Commercial Fisheries Biological Lab- oratory at Brunswick. METHODS Body measurements greater than about 6 millimeters were taken with dial calipers or with dividers and a metric scale ; smaller measurements were taken with a microscope and calibrated micrometer eyepiece. Measurements of less than 100 mm. were generally recorded to the nearest 0.1 mm.; larger measurements, to the nearest millimeter. Counts of rays of the soft dorsal, anal, and pectoral fins were made with a microscope and transmitted light on all specimens of less than about 250 mm. standard length. Each discernible ray was counted, including the small or rudimen- tary ray that occasionally is present at the posterior end of the fins. Only the pectoral and caudal fins have branched rays. All species examined have two dorsal spines and one pectoral spine; the second dorsal spine and the pectoral spine become minute or vestigial with growth of the fish. The pectoral spine was not included in the count of that fin. By definition, a ray in a fin may be of two types: a spine (which usually has a pointed tip, is never segmented, and is never branched) and a soft ray (which usually has blunt or fimbriated tip, is segmented, and may or ^ may not be branched) . Obvious deformities were neither counted nor measured. The following measurements are illustrated in figures 1 and 2. Standard length (S.L.). — Distance from tip of snout (upper lip) to middle of caudal-fin base. The caudal-fin base is distinguished externally as the curved ridge formed by the proximal ends of the caudal -fin rays. This ridge is not to be con- fused with the line formed by the extension of body skin and scales onto the bases of the caudal rays. Percent of standard length is recorded as "% S.L." Body depth. — Distance between origins of second dorsal fin and anal fin. Head length. — Distance from tip of snout (upper lip) to upper end of gill slit. Snout length. — Distance from tip of snout (upper lip) to anterior margin of orbit. Eye diameter {orbit diameter) . — Horizontal diam- eter of orbit. Eye to dorsal spine. — Straight-line distance from top of orbit to front center of base of first dorsal spine. Dorsal-spine length. — Distance from front center of base of dorsal spine to its tip. Caudal-Jin length. — Distance from middle of caudal base to tip of longest caudal ray. Peduncle depth. — Least depth of caudal pe- duncle, a vertical measurement from posterior end of anal-fin base. Peduncle length. — Shortest distance, from poste- rior end of anal-fin base to caudal-fin base along ventral surface of peduncle. IDENTIFICATION Our dichotomous keys to filefishes of the western North Atlantic have been constructed to allow for intraspecific and interspecific variation, and for ontogenetic changes in form and morphometries. When our series of specimens was small or in- complete in size range, we attempted to anticipate variation and ontogenetic changes, particularly by not using or qualifying the use of characters that we suspect might not be valid at specimen sizes we did not have. filefishes ( monacanthidae ) of the western north atlantic 63 Key to Genera of Monacanthidae from the Western North Atlantic A. Pelvic bone without an external spine or with only a very small rudimentary barbed spine present in three species (fig. 5). Gill slit usually very oblique (at an angle of about 45° from longitudinal body axis on specimens larger than 40 mm. S.L.) (fig. 4). First dorsal spine located over middle or back of eye (fig. 4). Anal-fin rays, 35 to 52 (fig. 3) Alutera. A A. Pelvic bone with a prominent external spine (fig. 5). Gill slit nearly vertical or only slightly oblique (fig. 4) B. B. A deep groove behind the dorsal spines (fig. 27B). Pelvic spine not movable (fusion may be broken on damaged specimens) (fig. 5). First dorsal spine inserted over anterior part of eye (on specimens 30 mm. S.L. and larger) (fig. 4). Anal-fin rays, 29 to 32 (fig. 3) Amanses. BB. No deep groove behind the dorsal spines. Pelvic spine movable in anterioposterior direction (fig. 5). First dorsal spine inserted over posterior part of eye (fig. 4) . Anal-fin rays, 26 to 36 (fig. 3) C. C. Scales with 1 to 8 or more spines, each spine arising individually from the scale base, and none of the spines branched; the spines usually separate but joined basally by a thin bony connection on larger specimens (95 mm. S.L. and larger; fig. 8). Body relatively shallower (table 12). Caudal peduncle of specimens 20 mm. S.L. and larger with 2 to 4 pairs of enlarged spines on each side (spines recurved in males). No elongated dorsal rays. Ventral flap relatively large (fig. 30) Monacanthus. CC. Scales usually with 1 spine, but with about 3 to 8 closely joined spines in larger specimens (100 mm. S.L. and larger). On specimens larger than about 40 mm. S.L. the spines branched one to many times above their bases; on specimens between about 19 and 40 mm. S.L. the spines of only a part of the scales are branched; and on speci- mens smaller than about 19 mm. S.L. spines are not branched (fig. 8). Body relatively deeper (table 12). No enlarged paired and recurved spines on caudal peduncle. Second dorsal ray elongated in mature males. Ventral flap relatively small (fig. 31) Stephanolepis. Keys to Species of Monacanthidae from the Western North Atlantic Genus Alutera Cloquet A. Dorsal rays 43 to 50. Anal rays 46 to 52 (fig. 3). Pectoral rays modally 14 B- B. Caudal peduncle longer than deep; peduncle length into peduncle depth 0.65 to 0.95 times. Caudal fin relatively short, about 18 to 26% S.L. Eye to dorsal spine distance relatively large, 7.0 to 8.6% S.L. Depth relatively great on specimens smaller than 175 mm. S.L., 36.8 to 43.8% S.L. (fig. 35) Alulera monoceros. BB. Caudal peduncle deeper than long on specimens larger than 30 mm. S.L.; peduncle length into peduncle depth 1.24 to 1.60 times on specimens larger than 50 mm. S.L., 1.03 to 1.05 times on specimens of 31 to 46 mm. S.L., 0.86 on a 27-mm. specimen. Caudal fin relatively long, about 33 to 61% S.L. Eye to dorsal spine distance relatively small, 5.0 to 6.7% S.L. Depth relatively shallow on specimens smaller than 175 mm. S.L., 21.5 to 33.1% S.L. (fig. 35) Alutera scripta. AA. Dorsal rays 32 to 41. Anal rays 35 to 44 (fig. 3). Pectoral rays modally 12 and 13 C. C. No pelvic spine. Eye to dorsal spine distance variable and relatively large on specimens larger than 100 mm. S.L. (fig. 34), 7.3 to 13.5% S.L. Eye relatively small on specimens larger than 175 mm. S.L., 4.8 to 6.8% S.L. Body depth relatively small in specimens smaller than 35 mm. S.L., 17.3 to 23.2%, S.L. Snout relatively short on specimens smaller than 45 mm. S.L., 12.0 to 23.9 %> S.L. Body scales relatively large and sparse; spines on scales relatively long and not close set (fig. 6), producing a comparatively rough feeling to the touch. Dorsal spine relatively long, thin, and with small barbs (fig. 7). Ventral profile of specimens smaller than about 45 mm. S.L. flatly curved, not produced into an angle (fig. 11). Pigment pattern of preserved specimens of about 70 to 200 mm. S.L., usually consisting of relatively fewer rounded spots or stripes mainly present on the ventral portion of the body; however, this pigmentation may be entirely absent (fig. 23) . Coloration of live specimens with few to many orange spots Alutera schoepfii. CC. Rudimentary pelvic spine present (on specimens 30 to 135 mm. S.L.) (figs. 4 and 5). Eye to dorsal spine distance relatively small on specimens larger than 100 mm. S.L. (fig. 34), 4.6 to 6.6% S.L. Eye relatively large on specimens larger than 175 mm. S.L., 6.2 to 7.7% S.L. Body depth relatively great in specimens smaller than 35 mm. S.L., 27.6 to 30.6% S.L. Snout relatively long on specimens smaller than 45 mm. S.L., 23.8 to 26.7% S.L. Body scales relatively small and numerous; spines on scales relatively short and close set (fig. 6), producing a "velvety" feeling to the touch, especially on specimens larger than 70 mm. S.L. Dorsal spine relatively short with large barbs (fig. 7); this condition pronounced on specimens between 40 and 140 mm. S.L. Ventral profile on specimens smaller than about 45 mm. S.L. produced in a convex angle by the extended pelvic bone (fig. 12). Pigment pattern of preserved specimens larger than about 70 mm. S.L. consisting of rounded or elongated and rounded spots, these more numerous on the dorsal half of the body (fig. 25). Color markings on live specimens bluish purple Alutera heudelotii. g4 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Genus Monacanthus Oken A. Body depth relatively shallow, 31.3 to 38.6% S.L. (fig. 36). Head relatively long on specimens larger than 40 mm. S.L., 33.1 to 36.1% S.L. Snout relatively long on specimens larger than 30 mm. S.L., 25.4 to 28.1% S.L Monacanthus tucker i . AA. Body depth relatively great, 39.1 to 54.5% S.L. (fig. 36). Head relatively short on specimens larger than 40 mm. S.L., 29.0 to 33.3% S.L. Snout relatively short on specimens larger than 30 mm. S.L., 21.9 to 25.7% S.L Monacanthus ciliatus. Genus Stephanolepis Gill A. Dorsal rays usually 31 to 34, rarely 29, 30, or 35. Anal rays usually 31 to 34, rarely 30 or 35 (table 10). Pigment pattern of preserved specimens between about 27 and 65 mm. S.L. consisting of a longitudinal arrangement of relatively few, small, dark dashes in several rows and several relatively large, dark, oblique or vertical blotches on the sides; the breast and snout without small flecks of pigment; and two moderately distinct dusky bars of about equal intensity on the caudal fin (fig. 31). On larger specimens the body dashes and caudal bars tend to become indistinct, the blotches on the sides become larger and more irregular, and the breast and snout become generally darker, but still lack a spotted effect ' _ Stephanolepis hispidus. AA. Dorsal rays usually 27 to 29, rarely 30. Anal rays usually 27 to 29, rarely 26 or 30 (table 10) . Pigment pattern of preserved specimens between about 27 and 65 mm. S.L. consisting of a relatively greater number of rows of dark dashes (which are more sharply defined and which give a broken-lines effect to the sides) and relatively small vertical or oblique blotches present on the sides; the breast and snout with few to many small flecks or spots of pigment; and two very distinct bars on the caudal fin, the anterior bar the darker (fig. 31). On larger specimens the body dashes and blotches and the caudal bars are less distinct, but the broken-lines effect on the sides and the spots on the breast and snout remain apparent ' Stephanolepis seiifer. Genus Amanses Gray A. A single species Amanses pullus. DESCRIPTION OF GENERA AND SPECIES the more common acceptance of this name. In addition, Gill (1884: p. 417) gives "Monacanthini, Monacanthidae is the correct name for this Nardo . . . 1844" as the name used earhest and group of fishes, although the name Aluteridae ^^^^ gj^^g ^y^ Q^j^er uses of Monacanthus as a was used by Fraser-Brunner (1941: p. 176) and family group name that predate the first use of others. Whitley (1929: p. 138) stated that ^/tt^era (in 1873) as a family group name (p. 416). "Aluterus Cloquet is an earUer name than ^^j^g separation of Monacanthidae (under the Monacanthus Shinz, the first Latinization of name of Aluteridae) as a family distinct from 'Les Monacanthes' Cuvier, so the family hitherto Balistidae by Fraser-Brunner (1941) provides known as Monacanthidae should be named adequate justification for this subjective distinc- Aluteridae." Two factors govern the propriety ^^^^^ although some recent authors have not of family names— priority of the generic names acknowledged the separation and have treated as they have been used as family names, and, Monacanthidae as a subfamily of Balistidae. especially in the case of well-known groups, the j^^ addition to the trenchant characters given by generally used and accepted name that has become Fraser-Brunner, comparison of larval forms — attached to a family (International Trust for larval Monacanthidae are very laterally com- Zoological Nomenclature, 1953: p. 33, art. 45(1)). pressed, contrasted with the laterally expanded Both of these factors apply to the acceptability and rotund larval Balistidae— provides additional of the name Monacanthidae. A review of the reason for the famihal separation. Fraser- literature and abstracting journals clearly supports Brunner's (1941: p. 176) separation follows: The division Balistiformes of the suborder Balistoidea consists of two families, which are separable as follows: I. Palatine T-shaped, the foot of the T movablv articulated with the ectopterygoid. 8 outer teeth m each jaw; 6 inner ones in upper jaw. 2 or no caudal vertebrae with epipleurals. 5 precaudal interneurals; ^ 4 formmg trough for spmous dorsal fin, the first movably articulated between exoccipitals, the others free from skull; the fifth formmg a prop between trough and vertebral column. Distal ends of caudal interneurals and interhaemals not expanded. 3 dorsal spines. Scales moderate or small, in regular series, imbricate. All soft fins with branched rays Balistidae. 1 The described pigmentations develop between 22 and 27 mm. S.L. We record specimens that have 30 soft rays In the dorsal or anal fins and lack the definitive pigment pattern as specifically unidentifiable. > "Excepting one, which may be thickened, in series with the caudal interneurals, in front of soft dorsal fin" (Fraser-Brumier, 1941; p. 176). FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 65 II. Palatine a simple bar, not directly connected with ectopterygoid. 6 outer teeth in each jaw; 4 inner ones in upper jaw. 4 or 5 caudal vertebrae with epipleurals. 3 precaudal interneurals ^ fused to form trough for spinous dorsal fin, immovably attached to exoccipitals, unconnected with vertebral column. Distal ends of caudal interneurals with prominent lateral expansions. Normally 2 dorsal spines, the second very small and sometimes absent. Scales small or minute, not in regular series, rarely in contact. Soft dorsal, anal and pectoral rays simple -_ _ _ __ Aluteridae [Monacantbidae]. Alutera Cloouet 1816 Aluterus as first proposed by Cloquet (1816: p. 135) is correctly emended to Alutera because the stem, aluta, is a feminine noun (Andrews, 1851 : p. 89) and thus is in accord with the Copen- hagen decisions on zoological nomenclature (In- ternational Trust for Zoological Nomenclature 1953: p. 49, art. 84(1)). Cloquet's proposal of the name appeared in a French dictionary and was based on a manuscript of Cuvier. Cuvier's (1817: p. 153) first apphcation of the name, however, was in the vernacular, "Les Aluteres." Oken (1817: p. 1173) furnished Alutera in its nomenclatorially acceptable form. This genus includes the following nominal genera as syn- onyms: Ceratacanthus Gill 1861, Osbeckia Jordan and Evermann 1898, and Davidia Miranda Ribeiro 1915. Fraser-Brunner (1941: p. 187) separated the first two of these names as sub- genera of Alutera on the basis of fin-ray counts and shape of the snout and caudal peduncle. We recognize four species of Alutera from the western North Atlantic: Alutera monoceros (Lin- naeus) 1758, Alutera scripta (Osbeck) 1765, Alutera schoepjii (Walbaum) 1792, and Alutera heudelotii Hollard 1855. A fifth species, Alutera punctata (Cuvier) in Spix 1831, has been reported from this area, but we regard it as a synonym of Alutera schoepjii, and it is discussed under the account of that species. Alutera heudelotii, A. scripta, and A. monoceros possess an external, rudimentary pelvic spine near the distal end of the pelvic bone. This spine usually has several short, thick, and irregular barbs, that appear to wear off in large specimens. On some of the largest specimens the pelvic spine could not be located, presumabl}' because of its degeneration and the corresponding increase in number and thickness of the spines on the body scales in this area. At its maximum development on smaller fish, the barbs of this rudimentary spine are much thicker and extend farther from the body surface than the spines of the associated body scales (fig. 5). This type of spine does not 566129 0—61 2 occur in Alutera schoepjii, and therefore its presence or absence is useful in distinguishing this species from A. heudelotii, particularly so in specimens 50 to 90 mm. S.L. where other charac- ters used to separate these two species are relative or overlapping. This spine was found in all specimens of Alutera heudelotii from 30.5 to 135 mm. S.L., but could not be located in specimens of 136 mm. S.L. and larger. It was noted in specimens of A. monoceros from 53 to 137 mm. S.L. In A. scripta, it was noted in specimens from 27 to 200 mm. S.L. Smith (1935: p. 359, pi. XLII D) recorded pelvic spines in A. monoceros and A. scripta from South Africa. Longley (1935: p. 86) noted the pelvic spine in Alutera ventralis and referred to it as "a micro- scopic vestige of the reduced ventral girdle of Monacanthus." Hildebrand (in Longley and Hildebrand, 1940: p. 279) corroborated its pres- ence in this species, but described it as freely movable in the skin. The skin surrounding the spine can readily be lifted away from the bone, and we have found the spine to be directly fused to the pelvic bone. However, the spine can be broken away from the bone, and if retained in position in the surrounding skin it is then movable. We have not determined a homologous relation- ship between this rudimentary spine and the pelvic spine of Monacanthus, Stephanolepis, or Amanses, and accept Longley's interpretation only on circumstantial evidence. Alutera monoceros (Linnaeus) 1758 (Figures 19, 20, and 21) This species was described in a pre-Lmnaean publication by Osbeck (1757) from a specimen taken off the coast of China. The name was docu- mented nomenclatorially by Linnaeus ui 1758. Some authors have regarded Alutera monoceros as a Pacific species and have distinguished the At- lantic form under the name of Alutera guntheriana Poey 1863; but comparisons of our Atlantic ma- terial with specimens from the China Sea and the Philippines show them to be identical in all 66 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE respects, and we regard Alutera monoceros as a species of worldwide occurrence. Comparisons of specimens from the western North Atlantic with specimens from Brazil, South Africa, and the Pacific coast of Panama show slight differences in contour and depth, but we attribute this to indi- vidual variation and perhaps varj'^ing rates of ontogeny in different geographical areas. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 46 to 50 ; anal soft rays, 47 to 52 (table 1) . Pectoral spine, 1 rudimentary. Pectoral soft rays, 14 (table 11). Pelvic spine, rudimentary and not movable (as Alutera heudelotii in fig. 5), absent in large specimens. Gill slit, oblique at an angle of about 45° to horizontal to body axis (as in fig. 4). First dorsal spine, inserted over middle or posterior part of eye (as in fig. 4) . No deep groove behind dorsal spines. Body depth, 34.4 to 43.8% S.L. (table 12; fig. 35). Head length, 26.6 to 34.7% S.L. (table 13). Snout length, 23.4 to 27.5% S.L. (table 14). Eye diameter, 4.2 to 8.3% S.L. (table 15). Eye to dorsal spine distance, 7.0 to 8.6% S.L. (table 16). Caudal peduncle longer than deep; peduncle length into peduncle depth 0.65 to 0.95 times. Caudal fin relatively short, about 18 to 26% S.L. Specimens examined. — From the western North Atlantic: 10 of 53 to 545 mm. S.L., from southern Massachusetts, the Carolinas, eastern Florida, and the Florida Keys (fig. 38). Alutera scripta (Osbeck) 1765 (Figures 9, 19, 20, and 22) This species has usually been regarded as of worldwide distribution. We have examined spec- imens from Hawaii, Okinawa, and the Pacific coast of Panama that appear to be identical with our western Atlantic material. Whitley (1952: p. 30) attempted to limit the Atlantic population under the name of Osheckia picturata (Poey) 1863. The brief description of Batistes scriptus that has been assigned to this species was m a publication by Osbeck (1757: p. Ill) that predates nomen- clatorial acceptabilit3^ Linnaeus did not record this name in the tenth edition of his Systema Naturae, although he did include (1758: p. 327) the listing of Osbeck's Balistes monoceros from the preceding page of Osbeck's book (1757: p. 110). The fu-st nomenclatorially acceptable publication of the name scripta is in a translation of Osbeck's 1757 book from Swedish to German hy J. G. Georgi in 1765 (p. 145). Since Georgi apparently made a direct translation without any emenda- tions, we do not consider him as the author of Osbeck's names and descriptions. Diagnostic characters. — ^Dorsal spines, 2. Dor- sal soft rays, 43 to 49; anal soft rays, 46 to 52 (table 2). Pectoral spine, 1 rudimentary. Pec- toral soft rays, 13-15 (table 11). Pelvic spine, rudimentary and not movable (as Alutera heudelotii in fig. 5), absent in large specunens. Gill slit, oblique at an angle of about 45° to horizontal body axis on specimens larger than 40 mm. S. L. (as in fig. 4). First dorsal spine, inserted over middle or posterior part of eye (as in fig. 4). No deep groove behind dorsal spines. Body depth, 21.5 to 35.0% S.L. (table 12; fig. 35). Head length, 29.3 to 33.9% S. L. (table 13). Snout length, 21.9 to 28.8% S. L. (table 14). Eye diameter, 5.3 to 9.1% S. L. (table 15). Eye to dorsal spine distance, 5.0 to 6.7% S. L. (table 16). Caudal peduncle deeper than long on specunens larger than 30 mm. S. L. ; peduncle length into peduncle depth 1.24 to 1.60 tunes on specimens larger than 50 mm. S. L., 1.03 to 1.05 times on specinaens of 31 to 46 mm. S. L., 0.86 on a 27-mm. S. L. speci- men. Caudal fin relatively long, about 33 to 61% S. L. Specimens examined. — From the western North Atlantic: 48 of 27 to 377 mm. S. L. (skin and skull examined of a specunen about 410 mm. S. L.) from Bermuda, off the North Carolina coast, southward around Florida, in the Gulf of Mexico, and the Caribbean (fig. 38). Color. — In live specunens taken off North Caro- lina in September 1959, the scrawled markings and spots were dark green and the background color was mottled olive-brown. This color fades and may disappear upon preservation, but the pigmentation in the markings and spots remains dark on most specimens even after prolonged preservation. Alutera schoepfii (Walbaum) 1792 (Figures 10, 11, 23, and 24) This species is extremely variable in certain morphological characters. Early in the study when we had only a few specimens, it appeared that two forms existed, one of which we would have called Alutera punctata (Cuvier) in Spix, FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 67 1831. Specimpiis were examined that were ex- tremely diverse in eye diameter, distance from eye to dorsal spine, shape of the snout to dorsal spine profile, body depth, and pigmentation. However, when our complete size series of speci- mens had been acquired and examined, we found that tlie specimens intermediate in these morpho- logical characters were more abundant than the extremely diverse specimens. We were convinced that these specimens represented a single highly variable species. If ^4. punctata exists, we have no specimens of it, and there is no available pub- lication to differentiate it from A. schoepjii. The inadequacy of the original description of ^4. punc- tata was pointed out by Longley (/;) Longlej- and Hildebrand, 1941 : p. 292). Longley also examined the spechnen Jordan and Rutter (1897: p. 127) used for the first redescription of A. punctata, and considered it to be A. schoepjii. We assume ar- tist's license in the excessively low numbers of dorsal fin rays and high numbers of body spots on the drawing of the type specimen of A. punctata {in Spix, 1831, pi. LXXVI). "Cuv. in litt." is given by Agassiz as the author of this species {in SpLx, 1831: p. 137), inferring that Cuvier should be recorded as the author of this name. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 32 to 39; anal soft rays, 35 to 41 (table 3). Pectoral spine, 1 rudimentary. Pectoral soft rays, 11 to 14 (table 11). Pelvic spine, absent at all sizes. Gill slit, oblique at an angle of about 45° to horizontal body axis on specimens larger than 40 mm. S.L. (as in fig. 4). First dorsal spine, inserted over mid or posterior part of eye (as in fig. 4). No deep groove behind the dorsal spines. Body depth, 17.3 to 47.4% S.L. (table 12; fig. 35). Head length, 23.3 to 34.2% S.L. (table 13). Snout length, 12.0 to 28.6% S.L. (table 14). Eye diameter, 4.8 to 8.8% S.L. (table 15). Eye to dorsal spine distance, 3.9 to 13.5% S.L. (table 16; fig. 34). Specimens examined. — 258 of 15.0 to 410 mm. S.L., from Bermuda, from Nova Scotia southward along the eastern and Gulf coasts of the United States, and from along the coasts of Cuba, Jamaica, Haiti, Atlantic Panama, and Brazil (fig. 38). Color. — In live specimens taken off North Carohna in September 1959, the coloration was variable with background shades of white, orange, or metallic gray. When white was present, it was usually most prevalent over the anterior regions of the fish. Orange was nearly always present, at least in the form of spots along the midventral region of the body. The dark metallic graj' color was often present on the dorsal half of the body as well tis on the peduncle. In a few specimens the body was entirely dark, but even in these orange spots were present, and in several specimens the orange spots were extremely numer- ous. Usually when the anterior regions of the fish were white, some orange blotches extended onto the white background. Often the dark gray occiured as large blotches over the orange. The entire coloration fades rapidly when specimens are placed in a preservative — the orange spots are extremely ephemeral. Alutera heudelotii Hollard 1855 (Figures 12 and 25) Alutera heudelotii Hollard (1855: p. 13, de- scribed from Senegal, West Africa) occurs in both the eastern and western Atlantic, and its syn- onymy has only recently been determined.' It includes the following nominal species: Alutera Juscus (Fischer, 1885: p. 75, pi. II, fig. 6, from Cameroon, West Africa) ; Alutera blankerti (Met- zelaar, 1919: p. 295, fig. 64, from Cape Blanco, West Africa) ; and Alutera ventralis (Longley, 1935: p. 08, from Tortugas, Florida; redescribed by Hildebrand in Longley and Hildebrand, 1940: p. 278). This species has largely been overlooked or con- fused, and we have re-identified specimens in several museums that were incorrectly identified as A. scripta, which species it superficially resem- bles, and as A. schoepjii and its synonym A. punctata. A number of early and recent refer- ences to A. punctata were undoubtedly based on specimens of A. heudelotii. The 44-mm. specimen from off West Africa described and illustrated as A. blankerti by Poll (1959: p. 247, fig. 83) repre- sents this species, as does the 291-mm. West African specimen Poll illustrated as A. punctatu-s (1959: fig. 82). Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 36 to 41 ; anal soft rays, 39 to 44 (table 4). Pectoral spine, 1 rudimentary. Pectoral soft rays, 12 to 14 (table 11). Pelvic spine. ' Berry, Frederick H., and Max Poll. Manuscript, Synonymy of the Atlantic Ocean fllefish Alutera heudelotii HoUard. 68 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE rudimentary and not movable (fig. 5), absent in specimens larger tlian 135 mm. S.L. Gill slit, oblique at an angle of about 45° to horizontal body axis in specimens larger than 40 mm. S.L. (fig. 4). First dorsal spine, inserted over mid or posterior part of eye (fig. 4). No deep groove behind dorsal spines. Body depth, 27.6 to 46.5% S.L. (table 12; fig. 35). Head length, 29.1 to 35.2% S.L. (table 13). Snout length, 23.8 to 28.7% S.L. (table 14). Eye diameter, 6.2 to 10.0% S.L. (table 15). Eye to dorsal spine distance, 4.0 to 7.3% S.L. (table 16; fig. 34). Specimens examined. — 68 of 30.5 to 240 mm. S.L., from Bermuda, from southern Massachusetts, off the coast of the Carolinas, around the Florida coast, in the Gulf of Mexico, and off Brazil (fig. 38). Color. — In live specimens taken off North Carolina in September 1959, the scrawled mark- ings and spots were bluish purple ; the background color was a mottled olive brown that faded upon preservation. The pigmentation of the markings and spots remains dark on most specimens even after prolonged preservation. Monacanthus Oken 1817 We have examined two valid species of this genus from the western North Atlantic, M. tuckeri Bean 1906 and M. ciliatus (Mitchill) 1818. The two species of Monacanthus in the western North Atlantic were recorded in a new subgenus, Leprogaster, by Fraser-Brunner (1941: p. 184). He distinguished it as an Atlantic subgenus characterized by a shorter pelvic spine and a smaller ventral flap than are present in the Pacific subgenus Monacanthus. We found no elongation of the upper caudal ray in the Atlantic species as was depicted by Fraser-Brunner for his new Pacific species, Monacanthus macrolepis (1941: p. 190, fig. 4). Monacanthus tuckeri apparently is a smaller species than M. ciliatus, both in not growing to so large a size and in maturing at a smaller size. Based on the specimens we examined it appears to be the less abundant of the two along the United States coast, but more equally common with M. ciliatus in the Bahamas and Bermuda. In his revision of the Aluteridae Fraser-Brunner (1941) recorded both Monacanthus and Stephano- lepis as valid and distinct genera. Since then several workers have disagreed with this pro- nouncement and have regarded Stephanolepis as a s_\monym of Monacanthus. The probable reason for this disagreement is the interpretation of scale structure of the two nominal genera. We have found the scale structure is subject to ontogenetic change — not adequately accounted for by Fraser- Brunner. Scales of various sizes of specimens of Stephanolepis hispidus and Monacanthus ciliatus are diagrammatically illustrated in figure 8. In the structure and ontogeny of its scales, Stephano- lepis setifer is essentially similar to S. hispidus, as is Monacanthus tuckeri to M. ciliatus, except that M. tuckeri is smaller at maturity than is AI. ciliatus and exhibits changes in its scale structure at smaller sizes. The scales of all four genera of filefish examined during this study have one or more spines arising perpendicularly from the scale base, the number of spines increasing with growth or size of the fish. Above the scale base the spines are usually curved posteriad, and they may undergo certain modifica- tions as secondary sexual characteristics, partic- ularly in the region of the caudal peduncle. The scales of Alutera and Amanses are similar to those of Monacanthus. The scales of Monacanthus and Stephanolepis are similar up to a size of about 20 mm. S.L., the scales of each having a single spine (fig. 8). At sizes larger than 20 mm. S.L., the spines of some of the scales of Stephanolepis have become branched — this branching occurs well above the scale base usually on the distal one- fourth of the spine. Between 30 and 40 mm. S.L. the spines of essentially all of the scales of Stepha- nolepis have become branched. Two or more closely joined spines are present on scales of Stephanolepis of more than 100 mm. S.L., and eight were present on the scales of a 150-mm. S.L. speci- men — aU of these spines are branched. Con- versely, the scale spines of Monacanthus never branch — each spine arises individually from the scale base. Two spines were found on a few scales of a 41-mm. S.L. specimen of Monacanthus ciliatus, three at 46.5 mm. S.L., and seven at 95 mm. S.L. (fig. 8) . Some of the spines on larger specimens of Monacanthus are jomed basally by a thin bony partition. After analyzing these concrete differences in scale structure in the two groups, as well as distinct differences in secondary sexual characters, we FILEFISHES (MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 69 recognize the value of Fraser-Bruniier's generic distinction of Siephanolepw from Monacanthus. The pelvic spine in Monacanthus is very similar to that in Stepkanolepis (fig. 5). Monacanthus tuckeri Bean 1906 (Figm-es 13 and 29) Although this species was described more than 50 years ago, it Has never been adequately dis- tinguished from Monacanthus ciliatus, and many museum collections we have examined contained both species, usually cataloged as M. ciliatus-. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 32 to 37; anal soft rays, 31 to 36 (table 5). Pectoral spine, 1 rudimentar}-. Pectoral soft rays, 10 to 12 (table 11). Pelvic spine, large and movable (as in Stephanolepis; fig. 5). GUI slit, nearh' vertical with respect to horizontal body axis (as in Stephanolepis; fig. 4). First dorsal spine, inserted over posterior part of eye (as in Steph- anolepis; fig. 4). No deep groove behind dorsal spines. Body depth, 31.3 to 38.6 % S.L. (table 12; fig. 36). Head length, 33.1 to 41.5 %S.L. (table 13). Snout length, 20.7 to 28.2 % S.L. (table 14). Eye diameter, 8.7 to 14.4 % S.L. (table 15). Eye to dorsal spine distance, 6.3 to 10.6 mm. S.L. (table 16). Specimens examined. — 60 of 15.3 to 56.5 mm. S.L., from Bermuda, off the Carolmas, off eastern Florida, in the Bahamas and the Lesser Antilles (fig. 39). Sexual characters. — The seven largest specimens available had gonads large enough to permit deter- mination of sex (2 males, 56.5 and 50.5 mm. S.L.; 5 females, 53, 51.5, 50.5, 49, and 48 mm. S.L.). The next largest specimens, 44 and 36 mm. S.L., had visible gonads, but they were too smaU for the sex to be interpreted. The males have a dorsal and a ventral pair of enlarged recurved spines on each side of the caudal peduncle, and the spmes on other scales on the sides of the peduncle are elongated, forming a bristlelike patch. The females have similar pairs of spines on the pedmicle but they are smaller and are directed posteriorly, and the spines of scales on the peduncle are not much, if any, larger than other bodj' scale spines. These dorsal and ventral pairs of spines are dis- cernible on specimens as small as 19 mm. S.L., since at this size and larger the scale bases from which they arise are larger (of greater diameter) than the bases of the other peduncle scales. The larger or more expandible ventral flap of the male with the dark stripe near its margin was described and illustrated by Clark (1950: p. 162). Clark listed males of 39, 59, and 60 mm. S.L., a female of 45 mm. S.L., and immatm-e specimens of 17 to 30 mm. S.L. Monacanthus ciliatus (Mitchill) 1818 (Figures 14, 15, 29, and 30) As noted before, this species has frequently been confused with Monacanthus tuckeri. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 29 to 37 ; anal soft rays, 28 to 36 (table 6). Pectoral spine, 1 rudimentary. Pectoral soft rays, 9 to 13 (table 11). Pelvic spine, large and movable (as in Stephanolepis; fig. 5). Gill slit, nearly vertical with respect to horizontal body axis (as Stephanolepis; fig. 4). Fu-st dorsal spine, inserted over posterior part of eye (as in Stephanolepis; fig. 4). No deep groove behind dorsal spines. Body depth, 39.1 to 54.5 % S.L. (table 12 ; fig. 36). Head length, 29.0 to 38.7 % S.L. (table 13). Snout length, 16.4 to 25.7 % S.L. (table 14). Ej-e diameter, 7.4 to 14.5 % S.L. (table 15). Eye to dorsal spine distance, 6.7 to 10.1 % S.L. (table 16). Specimens examined.- — 347 of 11.0 to HI mm. S.L., from Bermuda, Massachusetts, the coast of the Carolinas, around Florida, in the Gulf of Mexico, the Bahamas, and throughout the Carib- bean (fig. 39). Sexual characters.— Clark (1950: p. 159) de- scribed the sexual characters of this species. Im- matm-e specimens have a dorsal and a ventral pair of posteriorly directed spines on each side of the caudal peduncle — discernible on specimens as small as 20 mm. S.L. In the three largest females we examined (92.5, 101, and 109 mm. S.L.) the anterior spine of each pair was slightly recurved. In male fish larger than about 60 mm. S.L., these spines enlarge and become strongly recurved. Although the original pairs of spines remain distinct, additional and similar spines form with growth — the largest male examined (107 mm. S.L.) had 5 dorsal and 4 ventral spines on each side. On males 90 nmi. S.L. and larger, the spines on the other scales on the sides of the peduncle are elongated, forming a bristlelike patch. 70 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Stephanolepis Gill 1861 After examination of thousands of specimens of this genus from the western North Atlantic, we were able to distinguish only two species' — S. hispidus (Linnaeus) and S. setifer (Bennett) . The name setifer was applied to specimens from Cuba and Atlantic Colombia with a relatively low number of fui rays (D. 27-28, A. 26-27) by Fraser- Brunner (1940: p. 519). His reasons for applying and restricting this name certainly appear to be justified. Stephanolepis setifer is identical with Monacanthus oppositus Poey described by Meek and Hildebrand (1928: p. 798) from Panama, but we can not confu-m their species range of from Massachusetts to BrazU. Five species of Stephanolepis were identified from the western Atlantic by Fraser-Brunner (1940). S. setifer has low niunbers of fin rays, whereas the other four species were reported to have 30 or more dorsal and anal rays. S. insignis Fraser-Brunner 1940 and S. varius (Ranzani) 1842 were recorded from Brazil. Our specimens of Stephanolepis do not represent either of these forms; their distinguishing characteristics are not too convmcing; none has been recorded from the western Atlantic with the exception of the type material. The remaining two species reported by Fraser-Brunner were S. hispidus (Linnaeus) 1758 and S. spilonotus (Cope) 1871. We record S. spUonotus as a sjmonym of S. hispidus, because we judge our specimens to represent a single species with moderate variation m morphological characters, and because we found a complete overlap in every character that Fraser-Brunner (1940: p. 523, 535) used to separate the two nominal species. Certainly there is no difference between populations of S. hispidus from the At- lantic and from the Gulf of Mexico, as his obser- vations suggest. The five specimens Fraser- Brunner designated as S. spilonotus from Florida, Mississippi, and Cuba in the Museum of Com- parative Zoology cannot be located. In Stephanolepis the first dorsal spine has two rows of large, ventrally dkected barbs on its posterior margin. The number of barbs present is difficult to count, because those near the base of the spine abruptly decrease in size, particularly in larger specimens; but the number of these barbs has been used previously as a taxonomic character (Fraser-Brunner, 1940: p. 523) to separate S. hispidus with 6 or 7 strong barbs from S. spilonotus with 12 or more small barbs. We have determined two features that invalidate this character: (1) the barbs become relatively smaller as the fish increases in size, and (2) the number of barbs increases with growth of the fish. The following counts of barbs from one side of the spine of S. hispidus illustrate this sec- ond invalidating feature (standard length in milli- meters and number of barbs in parentheses) : 8.4 (3), 8.9 (2), 9.5 (3), 16 (3), 16.5 (3), 16.5 (2), 17 (3), 17.5 (2), 18 (2), 20 (3), 24.5 (4), 26.5 (4), 29.5 (2), 42.5 (5), 44 (5), 48 (5), 52.5 (6), 59 (6), 62 (5), 66 (5), 67 (6), 70 (6), 72 (6), 72 (7), 73 (8), 81 (10), 83 (8), 97 (8), 114 (8), 122 (10), 136 (8), 139 (11), 142 (10), 143 (12), 145 (9), 151 (11), 158 (13), 167 (11). Frequently the number of barbs on each side of the back of the spines varies by 1 or 2, and in these cases the count from the side having the greater number of barbs was recorded. It is characteristic that a small percentage of specimens of most of the species of Monacan- thidae examined had a background pigmentation much darker than average. This was observed in specimens preserved in both alcohol and for- malin. Conversely, some few of the preserved specimens were almost unpigmented. This caused some difficulty in confirming pigmentation char- acteristics for S. hispidus and S. setifer, but usually when moist specimens were examined the correct determination could be made (drying or partly drying of specimens makes the pigmenta- tion more difficult to see) . This feature produces excessive difficulties in utilizing the key to the species of Stephanolepis by Fraser-Brunner (1940: p. 521), in which the primary couplet concerns pigment (longitudinal pattern of bars, patches, or bands vs. transverse and mottled pattern). The profile from snout to dorsal spine of S. hispidus is concave in most specimens, but in some it is nearly straight, and in others slightly convex. The distance from the upper edge of the orbit to the base of the first dorsal spine in S. hispidus is also variable and may be either greater or less tlian the diameter of the eye. We have found no specimens of Stephanolepis setifer from the coast of the United States, where S. hispidus is relatively common. Analysis of FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 71 the specimens examined indicates that S. setifer is more common around Cuba, Jamaica, other islands of the West Indies, and in open waters of the Gulf Stream or Florida Current. Apparently it is a smaller species than S. hispidus, maturing at a smaller body size. The relationship of this genus to Monacanthus is discussed under the generic account of Mona- canthus. Photomicrographs of the scale struc- tures of Stephnnolepis (under the name of Stefanolepis hispidus) were published by Sanzo (1930, pi. Ill, figs. 32-35). The pelvic spine of Stephanolepis hispidus and of S. setifer possesses barbs and is articulated with the end of the barbed portion of the pelvic bone that protrudes through the skin (fig. 5). It is freely movable for a short distance (about 45°) in an anteroposterior direction. The color patterns on sides, breast, and caudal fin are of about equal value in separating the two species; that is, the prominence of one of the characters is usually accompanied by an equal prominence of the other two. At sizes less than about 22 mm. S.L. the species cannot be sepa- rated on this basis as the patterns described below are nearly always absent. From about 22 mm. to about 27 mm. these patterns are often present. Specimens between about 27 to 65 mm. normally have good and distinguishable color patterns; the pattern tends to become less distinguishing at the larger sizes. Sides: S. setifer normally has more rows of dashes arranged longitudinally, giving a broken-lines effect, the dashes being narrower and more sharply defined than the cor- responding small bars and spots of S. hispidus; and both species have smiilar broad, dusky bands of varying intensity, that may be vertical or oblique. Breast: Both species have the broad dusky bands continuing onto the breasts, but in addition, S. setifer has few to many small flecks or spots, especially in the region ventral to and anterior to the bases of the pectoral fins; these flecks are entirely absent in S. hispidus. Caudal fin: Botli species have two dark vertical bands on the caudal fin, however, these bands are nar- rower and usually much darker in S. setifer; the first band in S. setifer is usually nmch darker tlian the second, while in .S. hispidus both bands are of about equal intensity and not very prominent. Stephanolepis hispidus (Linnaeus) 1758 (Figures 16, 17, 31, 32, and 33) The close relationship of this species to Stephanolepis setifer has been discussed under the account of the genus. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 29 to 35; anal soft raj's, 30 to 35 (tables 7 and 10). Pectoral spine, 1; rudimentary at larger sizes, pronounced in larvae (see fig. 16). Pectoral soft ra3's, 12 to 14 (table 11). Pelvic spine, large and movable (fig. 5). Gill sUt, nearly vertical with respect to horizontal body axis (fig. 4). First dorsal spine, inserted over posterior part of eye (fig. 4). No deep groove behind dorsal spines. Body depth, 43.3 to 65.8% S.L. (table 12; fig. 37). Head length, 29.5 to 41.4% S.L. (table 13). Snout length, 14.4 to 27.5% S.L. (table 14). Eye diameter, 6.9 to 17.1% S.L. (table 15). Eye to dorsal spine distance, 7.3 to 17.1% S.L. (table 16). Specimens examined. — 3,539 of 5.6 to 211 mm. S.L., from Georges Bank southward aU along the Atlantic and Gulf coasts of the United States, off Mexico and Brazil (fig. 40). It has been suggested (Fraser-Brunner, 1940: p. 535) that the number of dorsal and anal fin rays is greater in specimens from more northern local- ities than from more southern localities. The following values tend to indicate such a trend: Eighty-seven specimens from the Gulf of Mexico ranged from D 29-A 30 to D 34-A 34 with a 26.2-percent mode at D 32-A 32 ; 267 specimens from Georgia ranged from D 30-A 31 to D 34-A 33 with a 27.3-percent mode at D 32-A 32; 199 speci- mens from North Carohna ranged from D 31-A 31 to D 35-A 35 with a 29.6-percent mode at D 33-A 33: but a smaller sample of 20 specimens from Massachusetts ranged from D 32-A 32 to D 34-A 34 with a 40-percent mode of D 33-A 32, inter- mediate between that of Georgia and North Carolina. Sexual characters. — Two secondary sexual char- acters develop on maturing males: the second soft ray of the dorsal fin becomes very elongated, and the spines of the scales on the sides of the caudal peduncle become prolonged and form a patch of bristles. The elongation of the second dorsal soft 72 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE ray begins between about 104 and 128 mm. S.L. The patch of bristles on the peduncle forms be- tween about 104 and 134 mm. On specimens larger than 140 mm. the elongated second soft ray of the dorsal fin was 21 to 95 mm. longer than the third soft ray. No secondarj^ se.xual characters were found in females. Although females aver- age a slightly greater body depth than males, appreciable variation occurs in this character and the values for the sexes overlap. The following observations were made on a sample of 140 specimens of 73.5 to 211 mm. S.L. taken by bottom trawling on the M/V Silver Bay off the coast of North Carohna during September 1959: Seven immature specimens or specimens with gonads too small to be evaluated, 73.5 to 95 mm. (mean, 85.3 mm.); 62 males, 78 to 211 mm. (mean, 131.8 mm.); 71 females, 77 to 180 mm. (mean, 120.1 mm.) of which five (146 to 180 mm.) had large macroscopic eggs in the ovaries. Such large eggs were found in other specimens ranging from 81 to 139 mm. S.L., taken at other times and areas. Occurrence. — The locations of specimens of Stephanolepis hispidus and S. setifer taken at the surface off the southeastern Atlantic coast of the United States on cruises of the Gill, Combat, and Silver Bay are shown in figure 41. These speci- mens were less than 70 mm. S.L., and represent developing young, the majority of which were being carried northward by the Gulf Stream. The total number of records and of specimens of S. hispidus was much greater than for S. setifer. The records of S. hispidus are distributed from inshore out to beyond the axis of the Gulf Stream, whereas the records of S. setifer are generally confined to the boundaries of the Stream. On cruise 18 of the M/V Silver Bay off the North Carolina coast in September 1959, records were made of all of the bottom-trawling stations at which Stephanolepis hispidus was taken. Figure 42 shows that the species was broadly distributed over the area at that time. Most of these speci- mens were matiu-e, and some of the females had macroscopic eggs and apparently were near spawning condition. Stephanolepis setifer (Bennett) 1830 (Figures 31 and 32) The resemblance of this species to Stephanolepis hispidus has been discussed under the account of the genus. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 27 to 30; anal soft rays, 26 to 30 (tables 8 and 10). Pectoral spine, 1 rudimentary. Pectoral soft rays, 11 to 13 (table 11). Pelvic spine, large and movable (fig. 5). Gill slit, nearly vertical with respect to horizontal body axis (fig. 4). First dorsal spine, inserted over posterior part of eye (fig. 4). No deep groove behind dorsal spines. Body depth, 46.8 to 59.6 % S.L. (table 12; fig. 37). Head length, 31.3 to 40.2 % S.L. (table 13). Snout length, 18.4 to 26.8 % S.L. (table 14). Eye diameter, 7.6 to 15.9 % S.L. (table 15). Eye to dorsal spine distance, 7.7 to 13.3 % S.L. (table 16). Specimens examined. — 139 of 11.0 to 136 mm. S.L., from Bermuda, the Carolinas, southward around Florida, into the Gulf of Mexico, and throughout the Caribbean (fig. 40). Sexual characters. — Sex was determined on 37 specimens, 15 males of 56.5 to 136 mm. S.L., and 22 females of 46.5 to 98 mm. S.L. ; 18 other speci- mens of 36 to 53.5 mm. S.L. were either immature or had gonads too small to be interpreted. Second- ary sexual characters apparently are similar to those of Stephanolepis hispidus, except that S. setifer matures and secondary sexual characters develop at smaller sizes. The females showed no secondary sexual development. Females 62.5 to 98 mm. S.L. had large macroscopic ovarian eggs, but females of 76.5 mm. and of 61.5 mm. S.L. and smaller had microscopic eggs. Males 82.5 mm. S.L. and larger had a patch of bristles on each side of the caudal peduncle; smaller specimens lacked this bristle patch. All males examined had the second soft ray of the dorsal fin elongated : 5.5 mm. longer than the other rays in the 56.5-mm. S.L. male and more than 33 mm. longer in the 136-mm. S.L. male. A 98.5-mm male had the third ray elongated also, about one-half the extent of elongation of the second ray. FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 73 Amanses Gray 1833 Fraser-Brunner (1941) reduced Cantherines Swainson 1839 to subgeneric rank within the genus Amanses Gray 1833. The subgenus Aman- ses reportedly possesses "A patch of long spines on side between soft dorsal and anal fin, at least in male." Since this feature has never been reported for western North Atlantic monacan- thids, A. pullus should be of the subgenus Cantherines. The pelvic spine of Amanses pvllus is fused to the end of the barbed portion of the pelvic bone that protrudes through the skin (fig. 5). It is similar to the pelvic spine of Monacanthus and Stephanolepis, but unlike the spine in those genera, it is not movable, unless damaged. (With exces- sive pressure the plane of fusion may part, and the spine may be abnormally movable.) Amanses pullus (Ranzani) 1842 (Figures 18, 26, 27, and 28) Ranzani (1842) described Monacanthus pullus from a large, blackish specimen without spines on the caudal peduncle, from the coast of Brazil. Cope (1871) described Monacanthus amphioxys from a smaller, hghtly colored specimen, also without caudal spines, from St. Martins Island in the West Indies. The relationship of these two nominal forms is still uncertain, but we believe the forms are identical. The variation in color pat- tern of specimens 38 to 148 mm. S.L. was described by Clark (1950: p. 163) under the name of Cantherines pullus. In addition to her observa- tions, we have examined a large freshly preserved female (158 mm. S.L., University of Florida 7266) that has a black body and caudal fin and tlie other fins pale or colorless. Larger specimens preserved for a long time have brownish bodies and clear fins. Diagnostic characters. — Dorsal spines, 2. Dorsal soft rays, 33 to 37; anal soft rays 29 to 32 (table 9). Pectoral spine, 1 rudimentary. Pectoral soft rays, 12 to 14 (table 11). Pelvic spine, large and not movable, fused to pelvic bone (fig. 5). Gill slit, nearly vertical with respect to horizontal body axis (fig. 4). First dorsal spine, inserted over anterior part of eye on specimens 30 mm. S.Ij. and larger (fig. 4). A deep groove present 566129 0—61 3 behind the dorsal spines into which they can be depressed (fig. 27, B). Body depth, 38.6 to 49.3 % S.L. (table 12; fig. 37). Head length, 29.0 to 42.9 % S.L. (table 13). Snout length, 25.6 to 33.3 % S.L. (table 14). Eye diameter, 5.2 to 14.9 % S.L. (table 15). Eye to dorsal spine dis- tance, 5.9 to 9.0 % S.L. (table 16). Specimens examined. — 99 of 17.5 to 325 mm. S.L., from Massachusetts, southward along the Atlantic coast, around the coast of Florida into the Gulf of Mexico, the Bahamas, and throughout the West Indies (fig. 39). The specialized scalation and spination on the caudal peduncle (usually a sex-associated character) is not clearly understood. Two large fresh specimens with orange-colored curved spines on the peduncle have been reported to us (per- sonal communications, Eugenie Clark, Cape Haze Marine Laboratory, and Craig PhiUips, U.S. Fish and Wildlife Service). The three largest speci- mens examined have recurved spines on the peduncle — a 325-mm. S.L. male has 3 dorsal and 2 ventral strongly recurved spines on each side of the peduncle; a 322-mm. male has 2 dorsal and 2 ventral spines similarly located (both of these specimens have a patch of bristles extending from the recurved spines onto the body); a 288- mm. female has 2 dorsal and 2 anal spines on each side of the peduncle, that are smaller and only slightly recurved in comparison to the spines of the males, and the patch of bristles on the pe- duncle of this female is relatively smaller. A 182-mm. specimen (sex unknown) has 2 pairs of large recurved spines and sparse patches of bristles on each side of the peduncle. A 115-mm. speci- men (sex unknown) has 2 pairs of small and only slightly recurved spines on each side of the peduncle. No other specimens of this species examined had paired peduncle spines. Specimens with patches of bristles on each side of the pe- duncle included females 136 and 158 mm. S.L., males 124 and 136 mm. S.L., and sex unknown 105, 123, 127, 131, and 138 mm. S.L. Several specimens between 100 and 142 mm. lacked these bristle patciies, and no specimens less than 100 mm. had them. A 288-mm. female had large ovaries but no macroscopic eggs, although females of 158, 142, 136, and 136 mm. luul large macro- scopic eggs. 74 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE LITERATURE CITED Anderson, William W., Jack W. Geheinger, and Edward Cohen. 1956. Physical oceanographic, biological, and chemi- cal data, south Atlantic coast of the United States, M/V Theodore N. Gill Cruise 1. U.S. Fish and Wildlife Service, Special Scientific Report — Fish- eries No. 178, 160 p., 15 figs. Andrews, Ethan A. 1851. A copious and critical Latin-English lexi- con . . . Harper & Bros., New York, 1663 p. Bean, Tarleton H. 1906. Descriptions of new Bermudian fishes. Bio- logical Society of Washington, Proceedings, vol. 19, p. 29-33. Bennett, Edward Turner. 1830. Observations on a collection of fishes, formed during the voyage of H.M.S. "Chanticleer," with the characters of two new species. Zoological Society of London, Proceedings, vol. 1, p. 112. Clark, Eugenie. 1950. Notes on the behavior and morphology of some West Indian plectognath fishes. Zoologica, vol. 35, no. 13, p. 159-168, figs. 1-7, pis. 1-2. Cloquet, Hippolyte. 1816. Dictionnaire des Sciences Naturelles. Vol. 1, supplement, p. 135. F. G. Levrault, Strasbourg. Cope, Edward D. 1871. Contribution to the ichthyology of the Lesser Antilles. American Philosophical Society, Trans- actions, vol. 14, p. 445-483. Cuvier, Georges. 1817. Le Regne Animal .... Vol. 2, Deterville, Paris, 528 p. Fischer, Johann Gustav. 1885. Ichthyologische und herpetologische Bemer- kungan. II. Uber einige afrikanische Fische des Naturhistorischen Museums in Hamburg. Jahr- buch der Hamburgischen. Wissenschaftliehen Anstalten, vol. 2, p. 66-77, pi. II. Fbasbr-Brunner, Anton. 1940. Notes on the plectognath fishes. III. On Monacanthus setifer Bennett and related species, with a key to the genus Slephannlepis and descrip- tions of four new species. Annals and Magazine of Natural History, ser. 11, vol. 5, no. 30 (art. 51), p. 518-535, figs. 1-7. 1941. Notes on the plectognath fishes. VI. A synopsis of the genera of the family Aluteridae, and descriptions of seven new species. Annals and Magazine of Natural History, ser. 11, vol. 8, no. 45 (art. 16), p. 176-199, figs. 1-9. Gill, Theodore Nichols. 1861. [Communication on several new generic types of fishes.] Academy of Natural Sciences of Phila- delphia, Proceedings, vol. 13, p. 77-78. 1884. Synopsis of the plectognath fishes. U.S. National Museum, Proceedings, vol. 7, no. 448, p. 411-428. Grat, John Edward. 1833. Illustrations of Indian zoology of new and hitherto unfigured Indian animals from the collec- tion of General Hardwicke. Vol. 2. London. Hollard H. 1855. Monographie de la Famille des Balistides. Annales des Sciences Naturelles, Zoologique, ser. 4, vol. 4, p. 5-27, pi. 1. International Trust for Zoological Nomenclature. 1953. Copenhagen decisions on zoological nomencla- ture. 14th International Congress of Zoology. London. 135 p. Jordan, David Starr, and Barton Warren Evermann. 1898. The fishes of North and Middle America: A descriptive catalogue of the species of fish-like vertebrates found in the waters of North America, north of Isthmus of Panama. U.S. National Museum, Bulletin No. 47 (pt. 3), p. 2183a-3136. Jordan, David Starr, and Cloudslet Rutter. 1897. A collection of fishes made by Joseph Seed Roberts in Kingston, Jamaica. Academy of Nat- ural Sciences of Philadelphia, Proceedings (3d. series), p. 91-133. Linnaeus, Carl. 1758. Systema Naturae. . . . Laurentii Salvii, Holmiae. Ed. 10, vol. 1, 823 p. Longlet, William H. 1935. Osteological notes and descriptions of new species of fishes. Carnegie Institution of Wash- ington, Yearbook, vol. 34, p. 86-89. Longlet, William H., and Samuel P. Hildebrand. 1940. New genera and species of fishes. Carnegie Institution of Washington, Publication No. 517 (Papers of the Tortugas Laboratory, vol. 32), p. 223-285, 28 figs., pi. 1. 1941. Systematic catalogue of the fishes of Tortugas, Florida, with observations on color, habits, and local distribution. Carnegie Institution of Washingtion, Publication 535 (Papers of the Tortugas Labora- tory, vol. 34), 331 p., 34 pis. [Edited and completed by Samuel F. Hildebrand.] Meek, Seth E., and Samuel F. Hildebrand. 1928. The marine fishes of Panama. Field Museum of Natural History Publication No. 249, Zoological Series, vol. 15, pt. 3, p. 709-1045, pis. 72-102. Metzelaar, Jan. 1919. Report on the fishes, collected by Dr. J. Boeke, in the Dutch West Indies, 1904-1905, with com- parative notes on marine fishes of tropical West Africa. In Boeke, J. (ed.) ; Rapport . . . in de Kolonie Curacao [etc.]. Firma F. J. Belinfanti, The Hague. 315 p., 64 figs. Miranda Ribeiro, Alpio de. 1915. Fauna Brasiliense. Peixes. V. [Eleuthero- branchios Aspirophores] Physoclisti. ."Vrchivo do Museu Nacional do Rio d6 Janeiro, vol. 17, 777 p. FILEFISHES (MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 75 MiTCHiLL, Samuel Latham. 1818. The fishes of New- York described and ar- ranged. In a supplement to the Memoir on the same subject, printed in the New- York Literary and Philosophical Transactions, vol. 1, p. 355-492. American Monthly Magazine and Critical Review, February, vol. 2, no. 4, p. 242-248, 1 fig.; March, vol. 2, no. 5, p. 321-328. Oken, L. 1817. Isis, p. 1182-1183 (misprinted 1782-1783). London. [Not seen.] OSBECK, PeHR. 1757. Dagbok ofver en Ostindisk resa aren 1750-52, med anmarkningar uti naturkundigheten, from- mande folkslags sprak, ets. (En Ostindisk resa til Suratte, China, etc. Fran 1750. . . ). Ludv. Grefing, Stockholm, 376 p., 13 pis. 1765. Herrn Peter Osbeck. . . . Reise nach Ostin- dien und China. . . . Aus dem Sehwedischen tibersetzt von J. G. Georgi. J. C. Koppe, Rostock, 552 (28) p., 13 pis. Poet, Felipe. 1863. Descriptions des poissons nouvelles ou peu connues. Academy of Natural Sciences of Phila- delphia, Proceedings, vol. 15, p. 180-188. [In Ichthyological Papers, American Authors, vol. 8.] Poll, Max. 1959. Poissons. IV. Teleosteens Acanthoptery- giens (deuxieme partie). Expedition Oceanogra- phique Beige dans les Eau C6tieres Africaines de I'Atlantique Sud (1948-1949), vol. 4, no. 3B, 417 p., 127 figs. Ranzani, C. 1842. De novis speciebus piscium. Dissertatio secunda. Novi Comment. Acad. Sci. Inst., Bonon., vol. 5, p. 1-21, pis. 1-7. Sanzo, LUIGI. 1930. Ricerche biologiche su materiali raccolti dal Prof. L. Sanzo nella Cainpagna idrografica nel Mar Ros.so della R. N. Ammiraglio Magnaghi 1923-1924. (VII). "Plectognati." R. Comitato Talasso- grafieo Italiano, Memoria 167, 111 p., pis. 1-7. Venezia. Smith, James L. B. 1935. The South African species of the family Aluteridae. Albany Museum, Records, vol. 4 (pt. 2), p. 358-364, pis. 40-42. Spix, Johann Baptist von. 1831. Selecta genera et species piscium quos in itinere per Brasiliam .... p. 83-138, pis. 49-101. C. Wolf, Monachii. Swainson, William. 1839. The natural history and classification of fishes, amphibians, and reptiles, or monocardian animals. Vol. 2. London. Walbaum, Johann Julius. 1792. Petri Artedi Sueci Genera piscium .... Emendata et aucta a Johanne Julio Walbaum. Grypeswaldiae, Ant. Ferdin. Rose, pt. 3, 723 p., 3 pis. Whitley, Gilbert P. 1929. Studies in ichthyology. Xo. 3. Australian Museum, Records, vol. 17, no. 3, p. 101-143, figs. 1-5, pis. 30-34. 1952. Figures of some Australian fish types. Royal Zoological Society of New South Wales, Proceed- ings, 1951-1952, p. 23-31, figs. 1-8. APPENDIX FIGURES Figure 1. — Diagram of a filefish showing measurements Figure 2. — Enlarged front view of upper part of head used in this study. showing measurement from eye to insertion of dorsal spine. 2e 27 28 29 30 27 28 29 30 31 32 33 DC 34 35 36 L 37 SOFTRAYS 38 39 40 41 42 43 44 45 46 47 48 49 50 1 '• i 1 i : : : ^5 f....i etifei' .- *: !-- ■: / .^.^pullus 31 1 n 32 IJ - "H ! ... 33 '-^ ^ciliatus 34 ni 1. ^ — tiirkf^ri CD >- 35 nispiQ ■•■■1 1 i L 1 36 AY' 37 1 38 o en 39 J \ 40 /^^ 1 • -J 41 ^ 'choepfii^ ^heudelotii < < 42 43 44 45 monoceros 46 y 47 \ \ ~ 48 i 49 y : 50 scripta--^ 1 .1..., 51 52 - 1 1 FifiURE 3.— Dorsal and anal soft riiy correlation in western North Atlantic Monacanthidae. The various outlines (dots, dashes, solid lines, and dots and dashes) encompass the dorsal and anal ray combinations found in the nine species. 76 FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 77 ANQLt Of SILL OPENlNC Sfephanolepis Stephanolepis hispidus "LINE OF ARTICULATION Amanses pullus LINE OF FUSION ^Sa Alutera <;#\ heudelotii A heudelotit 37inm 5 L. Figure 4. — Outlines of Stephanolepis, Amanses, and Alutera, illustrating location of the pelvic spine, posi- tional relation of the first dorsal spine to the eye, and angle of the gill opening to the horizontal body axis. Figure 5. — Pelvic spines: Stephanolepis hispidus, 38 mm. S.L.; Amanses pullus, 36.5 mm. S.L.; and Alutera heudelotii, 37 mm. S.L. The ratio of magnification of the drawings is 1.0, 1.33, and 4.0. 78 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Alutera schoepfii 68 mm. S.L. Alutera heudelotii 69.5 mm. S.L. Figure 6. — Diagrammatic section of head region between eye and nostrils, illustrating the relative number and position of scale spines in this area: Alutera schoepfii, 68 mm. S.L.; Alutera heudelotii, 69.5 mm. S.L. Figure 7. — Dorsal spines: Alutera schoepfii, 68 m Alutera heudelotii, 69.5 mm. S.L. m. S.L.; riLEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 79 Sfephanolepis hispidus Monacanthus ciliatus 10.5 mm 21 mm 33 mm 12.7 mm. 41 mm. 46.5 mm. 95mm. 105mm. 150 mm. J^ Figure 8. — Scales of Stephanolepis hispidus and Monacanthus ciliatus, illustrating development of the scale spines with increase in body size. The drawings are semidiagrammatic and are not drawn to the same relative proportion. Figure 9. — Alutera scripta, 31.0 mm. S.L. {Combat station 438). 80 FISHERY BULLETIN OF THE FISH AND "WILDLIFE SERVICE Figure 10. — Alutera schoepfii, 15.0 mm. S.L. (Gill cruise 2, regular station 6). Figure U. — Alutera schoepfii, 32.5 mm. S.L. (Combat, Port Canaveral, Fla.). Figure 12. — Alutera heudelotii, 30.5 mm. S.L. (Oregon station 1074, University of Florida 3829). Figure 13.—Monacanthus tuckeri, 15.3 mm. S.L. (Acad- emy of Natural Sciences of Philadelphia 84471). Figure 14. — Monacanlhus ciliatus, 11.0 mm. S.L. (Silver Bay station 476) . FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 81 FiGiiRE 15. — Monacanthus ciliatus, 15.3 mm. S.L. {Gill cruise 7, regular station 54) . ^^^^^^g^ Figure 18. — Amanses pullus, 17.5 mm. S.L. (Gill cruise 9, from stomach contents of Katsuwonus pelamis, Nov. 15, 1954; 1600). FiGUBB 16. — Stephanolepis hispidus, 6.5 mm. S.L. (Gill cruise 7, regular station 38) . Figure 17. — Stephanolepis hispid^is, 15.2 mm. S.L. (Gill cruise 8, regular station 48) . Figure 19. — Top: Alutera monoceros, 53 mm. S.L. (U.S. National Museum 117022). Bottom: Alulera scripta, 53 mm. S.L. (Gill cruise 8, regular station 52). 566129 0—61- 82 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE ■"*»'*♦?, ,^ ^ Figure 20. — Top: Alutera scripta, 119 mm. S.L. Bottom: Alutera monoceros, 122 mm. S.L. (Both from Combat station 459.) Note differences in body depth, caudal peduncle, and caudal-fin length. FiGUBE 23.— Alutera schoepfiL Top: 126 mm. S.L. (Uni- versity of Florida 2542). Bottom: 176 mm. S.L. (Uni- versity of Florida C-9-2053-3). Body pigment spots that are frequently present on A. schoepfii of this size are not present on these two specimens. Figure 21. — Alutera monoceros, 545 mm. S.L. {Silver Bay station 1550). Figure 22. — Alutera scripta, 377 mm. S.L. (U.S. National Museum 170118). Figure 24. — Alutera schoepfii, 317 to 343 mm. S.L. (Tulane University 17106). Note variations in head profile and in size and position of the eye in this species. FILEFISHES (MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 83 Figure 25. — Alulera heudelotii, 230 mm. S.L. (Tulane Universitj' 16316). Note the small distance from eye to dorsal spine. Figure 26. — Amanses puUus, 45.5 to 58 mm. S.L. {Combat station 474). Note variation in pigmentation. 84 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 27. — Amanses pullus, 158 mm. S.L. female (University of Florida 7266). A. Lateral view. B. Oblique view showing deep groove behind dorsal spines. Figure 28. — Amanses pullus, 322 mm. S.L. male (U.S. National Museum 32096). Xote prominent patch of bristles and pairs of large recurved spines on peduncle. : ^-.. -«*«■ t^ mr ^■■^^i'" -^ -u*^. ' Figure 29. — Left, Monacanthus tuckeri. Top: 56.5 mm. S.L. male. Bottom: 49 mm. S.L. female. (Both Academy of Natural Sciences of Philadelphia 84478.) Right, Monacanthiis ciliatus. Top: Immature specimen, 54 mm. S.L. Bottom: Immature specimen, 47.5 mm. S.L. (Both from Sanibel Island, Fla., August 19, 1959.) Note black line on ventral flap of the male M. tuckeri and compare body profiles of the two species at similar sizes. FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 85 Figure 30. — Monacanlhus ciliaius. Top: 92.5 mm. S.L. female. Bottom: 103 mm. S.L. male. (Both Uni- versity of Florida 3611.) Note larger, recurved spines on caudal peduncle and black line on margin of ventral flap of male. Figure 31. — Top: Stephanolepis hispidus, 52.5 mm. S.L. (Gill cruise 4, regular station 46). Bottom: Stephano- lepis selifer, 55 mm. S.L. (Combat station 459). Note small spots and dashes of pigment and bars on caudal fin of S. selifer. 86 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 32. — Left: Stephanolepis setifer (Stanford University Natural History Museum 4772). Standard length and sf\ from top to bottom, 101-mm. male, 95-mm. male, 98-mm. female, 96-mm. female. Right: Stephanolepis hispidiif:. immature specimens (top two. Silver Bay station 1315, bottom two. University of Florida No. C-7-1253-4). Standard lengths from top to bottom, 103, 83.5, 88.5, and 88 mm. Note intraspecific and interspecific variation in size of eye. The elongated second ray of the two males of S. setifer is not clearly shown. FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 87 Figure 33. — Stephanolepis hispidus. Top: 167 mm. S.L. female {Silver Bay station 1297). Bottom: 169 mm. S.L. male (Silver Bay station 1210). Note the elongated second dorsal ray and patch of caudal peduncle bristles on the male, absent on the female. 88 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE ' 1 ' • Alutera schoepfii \ ' 1 1 1 45 X Alutera heudelotii 40 - • - 35 • • . 1 e LU O 30 ' — y^ z • y^ < y H , U) Q ^ yv- i^J 25 y^'° z Q. to _l • ^ ' • < w • ^/"^ IT 20 — /^^ — o ^^ Q ^^ O y^ H y^ UJ y^ >- 15 y< - LlI s * X X X X 'f 10 5 1 1 1 1 1 50 100 150 200 250 300 350 400 STANDARD LENGTH (mm.) FiotTRE 34. — Relation of eye to dorsal-spine distance and standard length for Alutera schoepfii and Alutera heudelotii. Specimens larger than 100 mm. can be distinguished by this character. FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 89 > Q O CQ 75 T J I I t Alufero I r ■■■^ 1 ' 1 I 1 . • schoepfii « heudelotii « - K -. - , * - •• /So 1 1 . 1 . 1 1 1 ^ /to "^-T" 10 20 30 40 STANDARD LENGTH " /A .•s* •1^* ..C^" Alufero • schoepfii X heudelotii o scripto A monoceros 60 100 150 200 250 300 350 400 STANDARD LENGTH (mm) Figure 35. — Relation of body depth to standard length for Alutera schoepfii, Alulera heudelotii, Alulera scripla, and Alutera monoceros. Insert graph illustrates difference in depth between Alulera schoepfii and Alulera heudelotii at sizes less than about 35 mm. 90 60 "I — ' — r FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE r~~' — I — ' — I — 50 40 £ E g 30 >- O o CD 20 10- J_ _L I r T X _L Monacanthus • cillafus o tucker i _L _L _L, n5 ilo Figure 36.— Relation of body depth to standard length for Monacanthus ciliatus and Monacanthus luckeri. 10 20 30 40 50 60 70 80 STANDARD LENGTH (mm.) 90 100 FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 91 E £ a. LlI Q >- Q O CD 110 I ' 1 ' I ' 1 ' 1 ' 1 ' 1 ' 1 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' I ' I ' I 100 - . 90 " , • ' 80 ~ . .:. 70 • •• O o • 60 - • . - • . ^ 0° 50 _ • . » • « o _ :" X " o . 40 - ' 8 o* - 30 20 - • Stephanolepis hispidus " Stephanolepis setifer ° Amanses pullus 10 1.1,1,1,1.1,1.1. I.I.I. 1 1 1 1 . 1 . 1 . 1 . 1 . 1 . 1 20 80 100 120 140 STANDARD LENGTH (mm.) 160 ISO 200 Figure 37. —Relation of body depth to standard length for Stephanolepis hispidus, Stephanolepis setifer, and Amanses pullus. 92 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 40' 30' 20= Alutera o monoceros • scripta ° heudelotii • schoepfii Figure 38. — General distribution of specimens examined of Ahdera monoceros, Alutera scripta, Alutera heudelotii, and Alutera schoepfii. FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 93 40' Monacanthus • ciliafus — o tuckeri Figure 39. — General distribution of specimens examined of Monacanthus ciliatus, Monacanthus tuckeri, and Aynanses pullus. 94 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 40' 30° 20° Stephanolepis hispidus a sefifer 100 Figure 40. — General distribution of specimens examined of Stephanolepis hispidus and Stephanolepis setifer. (No attempt has been made to indicate all records where collecting stations were closely spaced.) li FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 95 -1 1 n r V Stephonolepis • hispidus » selifer ■^^^^ SX. Figure 41. — Locations of specimens of Stephonolepis hispidus and Stephanolepis setifer taken at the surface off the southeastern Atlantic coast of the United States by dip net and meter larvae net on cruises of the Gill, Combat, and Silver Bay. The 20-fathom contour line is represented by the line of dots. The approximate axis of the Gulf Stream is represented by the line of dashes. .^^ 5 hispidus m SPECIMENS T4KEN o SPECIMENS NOT TAKEN Figure 42. — Chart of the waters off North Carolina showing the concentrated bottom-trawling stations made by the Silver Bay in September 1959 and indicating the stations at which Stephanolepis hispidus was taken (black squares) and was not taken (open squares). B. TABLES Table 1. — Dorsal ray-anal ray relation for 10 specimens of Alutera monoceros Table 5. — Dorsal ray-anal ray relation for 62 specimens of Monacauthus tuckeri Dorsal soft rays 47 48 49 50 Dorsal soft rays 34 35 36 37 47 1 &^« 2 42 49 •3 60 a ■< 61 1 2 1 1 1 52 1 31 32 33 34 35 36 1 I 1 2 3 2 1 1 11 1 9 9 2 I 1 4 2 Table 2. — Dorsal ray-anal ray relation for 1^7 specimens of Alutera scripta Table 6. — Dorsal ray-anal ray relation for Z39 specimens of Monacanthus ciliatus 43 44 Dorsal soft rays 45 46 47 48 49 Dorsal soft rays 32 33 34 46 1 2 47 1 1 I 48 1 6 3 49 2 7 8 50 3 2 4 2 51 1 2 52 1 Table 3. — Dorsal ray-anal ray relation for 126 specimens of Alutera schoepfii 28 29 30 31 32 33 34 35 36 6 23 10 2 14 30 12 2 13 22 5 2 1 10 8 17 6 2 4 16 9 3 1 1 3 5 6 4 1 2 Dorsal soft rays 34 35 36 37 38 39 Table 7. — Dorsal ray-anal ray relation for 976 specimens of Stephanolepis hispidus 35 w 36 S 37 ° 38 "3 2 39 40 41 1 2 1 1 2 1 1 2 2 3 6 1 3 9 15 2 7 17 11 1 2 9 9 5 1 2 7 2 30 Dorsal soft rays 31 32 33 34 30 31 32 33 34 35 1 3 18 3 1 4 67 97 3 21 216 164 6 1 37 209 17 2 1 13 38 6 3 6 Table 4. — Dorsal ray-anal ray relation for 68 specimens of Alutera heudelotii Dorsal soft rays 38 39 40 Table 8. — Dorsal ray-anal ray relation for 133 specimens of Stephanolepis setifer 41 39 40 41 42 43 44 1 2 3 8 4 3 10 7 1 2 3 10 1 1 1 I Dorsal soft rays 28 29 30 26 27 28 29 30 2 18 16 1 6 50 8 6 20 2 4 90 FILEFISHES (MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 97 FiouRE 9. — Dorsal ray-anal ray relation for 81 specimens of Amanses puUus 33 Dorsal soft rays 34 35 36 £ 29 "o - 31 a ^ 32 2 1 2 13 4 12 32 4 1 5 4 1 Table 10. — Dorsal ray-anal ray relation for 133 speci- mens of Stephanolepis setifer and 976 specimens of Stephanolepis hispidus [s= S. setifer; h = S. hispidus; figures are the percentages of the total for each ray CODibination] 28 29 Dorsal soft rays 30 31 32 33 34 35 26 s 1.6 27 s 13.6 S12.0 s 0.8 28 s 4.6 S37.7 s 6.0 R 29 s46 S15.0 s 1.6 i ^° hO. 1 s 3.0 ho. 3 hl.8 hO.3 hO.l 1 31 ho. 4 h6.9 h9.9 hO.3 32 h2.2 h22.3 hl5.8 ho. 6 33 hO. 1 h3.8 h21.4 h6.9 ho. 2 34 hO.l hi. 3 h39 hO.5 1 35 hO.3 ho. 5 Table 11. — Numbers of pectoral soft rays in western North Atlantic Monacanthidae (Counts of rays from both sides; not recorded with respect to right and left sides; rudimentary pectoral spine not included in counts) Species 9-10 10-10 10-11 11-U 11-12 12-12 12-13 13-13 13-14 14-14 14-16 15-15 10 41 2 8 Alutera scripta 2 61 40 2 3 9 2 2 6 63 23 3 3 18 34 148 2 11 6 88 1 4 10 1 10 4 3 236 42 63 19 16 64 8 13 Table 12. — Relation of body depth to standard length in Monacanthidae of the western north Atlantic, by species and millimeter intervals |In percent of standard length for grouped millimeter intervals of standard length] Body depth in percent of standard length Standard length (mm.) Alutera Monacanthus Stephanolepis Arrtanses monoceros scripfa schoepfii heudelotii tuckeri ciliatus hispidus setifer puUus 0-4.9 5-9.9 43. 3-48. 6 46. 2-51. 7 48 7-57. 9 51. 2-58. 7 51. 2-59. 6 62. 8-55. 7 52. 9-58. 61.3-58.0 53.9-69.0 53.4-60.0 53. 9-62. 6 48. 3-62. 2 54.S-63.6 64. 4-65. 2 51.3-61.0 49. 3-60. 1 49. 4-65. 8 60. 4-68. 2 60.0-50.2 10-14.9_ 39. 1-48. 6 39, 9-48. 5 43. 2-48. 8 42. 1-49. 6 47. 7-49. 6 46. 1-48. 7 45. 7^6. 8 45. 6-51. 9 46. 4-49. 9 48. 4-52. 6 47. 7-54. 6 46.5-54.0 49. 5-62. 9 45. 2-54. 6 47. 2-52. 2 48.4-57.8 60.2-54.6 61. 8-66. 6 61. 7-59. 6 62.0-67.3 54.3-56.3 51.9-54.9 60. 4-68. 6 52. 7-55, 7 56. 6-68. 4 49.1-55.9 48. 4-56. 8 46. 8-60. 9 16-19.9 17.3 18. 2-18. 6 17. 4-22. 2 19. 1-23. 2 22. 7-23. 9 25. 9-30. 9 25. 1-33. 1 27. 8-39. 25. 2-36. 4 30.0-10.4 35. 6-39. 8 29. 6^1. 1 32.4-14.0 38 5-40. 1 38 8-42. 6 40. 4-44. 7 41.2.-15.3 35. 4-43. 1 38.2-47.4 41.4 38 6-46 37.4-46.3 38.3-42.9 33. 1-38. 5 31. 3-37. 9 38.9 20-24.9 _ 25-29.9 21.6 22.6-24.5 23.1-25.9 26.9 25.2 27.4-28.0 27. 7-29. 2 28. 8-30. 2 27.8 27. 6-30. 6 31.1 30. 1-32. 6 30-34.9 . . . 34. 5-36. 4 33.0-33.9 36.4 36.3-36 6 35. 8-38. 6 40.3 36-39.9 38 6-39. 1 40-44.9 38. 8-44. 2 46-19.9 40. 8-43. 3 60-69.9 36.8 42.0 41.4 37.0 36.7 36. 2^2. 9 36.4-41.0 43.0 39. 8-44. 6 40. 0-43. 4 38. 7-46. 6 39. 5-39. 6 36.8-41.2 38. 8-42. 1 41 8-46.9 60-69.9 _ 40 8-43.2 70-79.9 43, 3-45. 1 80-89.9... 45 1-46.6 90-99.9 30.0 29. 5-31. 4 31.0-31.6 30. 2-33. 1 46 4-46. 2 100-124.9 37. 7-38. 6 43.8 43.1 42. 1-47. 2 125-149.9 41.6-46.7 160-174.9 46.8 175-199.9 47.3 200-224.9 36.0 31.6-31.7 225-249.9 250-274.9 275-299.9 49.3 300-349.9 46.9-48.3 350-399.9 36.0 400-1-.... 34.4-36.7 98 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 13. — Relation of head length to standard length in Monacanthidae of the western North Atlantic, by species and millimeter intervals [In percent of standard length for grouped millimeter Intervals of standard length] Head length in percent of standard length Standard length (mm.) Alutera Monacanthus SUphanoUpis Amanses monocerot scripta schoepfti heudelotii tuckeri ciliatus hispidus settler pullits 5-9 9 34. 4-41. 4 34.3-38.3 34.0-38.2 33. 3-37. 1 34.4-37.3 34. 8-35. 7 33. 5-36. 5 33. 0-36. 1 33. 5-34. 9 32. 6-34. 8 32. 6-36. 9 33. 0-34. 7 31.9-35.1 31. 5-34. 4 31. 7-34. 6 31. 1-34. 2 29. 5-36. 4 30.4-33.0 29. 8-31. 6 10-14.9 .- 34.6-38.2 32. 7-38. 7 32. 6-38. 5 34.4 33. 6-35. 37. 6-40. 2 35. 8-37. 7 35. 7-37. 35. 7-36. 3 35. 0-37. 8 34. 5-36. 7 34. 1-36. 3 35.4 33. 6-35. 3 35.7 37.9 31.3 33. 5-37. 1 31. 6-34. 2 15-19 9 23.3 25.2-26.9 26. 3-25. 5 27. 4-27. 9 29. 5-29. 7 31.6-31.8 28. 5-32. 2 29. 4-32. 2 30. 3-32. 7 30. 8-32. 3 30. 1-32. 4 29. 3-34. 2 28.0-31.8 29. 8-31. 7 27. 4-30. 8 28. 8-31. 1 27. 8-29. 7 27. 9-30. 3 27.8-32.2 29.8 27. 4-31. 1 27.0-31.0 27. 9-29. 6 37.8-41.6 35. &-39. 1 42 9 20-24.9 25-29.9 31.5 31. 9-32. 5 31. 0-32. 7 29.3 32.6 32. 6-32. 9 32.0-33.1 31. 4-33. 9 30-34.9 31.8-35.2 32.4 33. 7-34. 1 36.3-37.0 35. 2-35. 8 34.8 34. 8-34. 9 33. 1-36. 1 38 8 35-39.9 38 6-40 1 40-44 9 32. 4-33. 3 32 8-33. 3 31.0-33.3 36 6-39 3 45-49.9 60-59.9 34.7 31.7 33.7 32.2 33.4 32.9-34.6 30. 1-33. 7 34 4-39.0 60-69 9 33 3-35 70-79.9 31.0-31.8 29.4-31.7 29. 8-32. 6 29.0-30.1 32 9-37. 8 80-89 9 32 9-33 8 90-99 9 -- 32.2 30. 9-32. 8 31.4-33.0 31.7-32.0 29 0-32 9 100-124.9. -- 31.4-31.6 31.6 30.6 31. 3-31. 8 30. 2-32. 7 30. 1-31. 6 30. 8-31. 4 29. 1-30. 9 29.8-31.3 32 35. 4 125-149.9 31. 9-32. 4 160-174.9 32 175-199.9.. 200-224 9 32 3 32.2-32.4 225-249.9 260-274.9 275-299.9 33 9 300-349.9 - - 31 4-33.8 350-399.9 31.3 26. 6-26. 9 Table 14. — Relation of snout length to standard length in Monacanthidae of the western North Atlantic, by species and millimeter intervals |In percent of standard length for grouped millimeter Intervals of standard length] Snout length In percent of standard length Standard length (mm.) Alutera Monacanthus Stephanolepis Amanses monoceros scripta schoepfii heudelotii tuckeri ciliatm hispidus seiifer puUus 6 -9.9 . . 14.4-20.0 17.5-19.3 18. 2-20. 6 19. 4-20. 4 19. 6-22. 3 22. 0-24. 3 21. 5-24. 8 21.4-24.2 22. 5-24. 9 22. 7-25. 23.2-26.0 22. 5-25. 1 22. 8-25. 2 21.2-24.4 22. 5-25. 1 23.3-26 21.3-27.6 23. 3-26. 9 22. 4-24. 4 10-14.9... 15 4-21.0 16 6-21. 9 19. 3-25. 2 21. 4-25. 1 22.2-23.9 23. 7-25. 1 22. 2-23. 23. 3-24. 9 22. 8-24. 9 23. 8-24. 6 23. 3-25. 2 21. 9-24. 5 23. 4-25. 7 22. 4-24. 3 18.4-21.0 19. 0-21. 6 20.4-21.3 21. 2-23. 2 21. 7-25. 7 23. 5-25. 1 23. 5-24. 3 26.1 23. 4-24. 7 24.3 26.8 20. 6-24. 9 24.2-25.6 22.1-23.9 16-19.9 12.0 14. 1-16. 7 16. 8-16. 6 18. 8-18. 9 20. 8-21. 8 23. 8-23. 9 23.0-24.7 23. 1-25. 5 23. 6-27. 25. 5-27. 3 24. 7-27. 6 24. 2-28. 6 22. 6-27. 5 24. 8-25. 7 26. 1-27. 1 24.8-27.8 23. 6-27. 24. 2-26. 6 24. 2-25. 9 27.1 23. 6-27. 6 23. 9-26. 7 23.3-26.1 21.8-28.2 20.7-25.6 26.7 20-24.9 26-29.9 21.9 22. 5-23. 8 23. 7-25. 1 22.4 25.0 26. 7-25. 9 26. 8-26. 7 26.0-28.0 30-34.9 23. 8-25. 2 25.4 25. 9-26. 7 251-27.0 25.4-25.9 26.4 26.7 25. 1-28. 1 27.3 36-39.9 27. 8-28. 8 40^4.9-.-- 27. 4-28. a 45-49.9 27.1-28.0 50-69.9 27.2 26.5 27.0 26.0 28.1 27. 7-27. 8 25.0-28.7 26. 5-28. 4 60-69.9 26. 9-27. 4 70-79.9 - 25 9-33. 3 80-89.9 26. 9-27. 8 90-99.9- -- 27.2 26. 3-28. 8 26. 8-28. 27. 4-28. 3 25. 5-28. 9 100-124.9- --.- 26. 2-27. 5 26.7 26.5 27.2-27.9 26. 2-28. 7 26. 7-27. 9 27. 1-27. 7 26. 6-28. 26.9-28.2 26. 0-27. 8 126-149.9 25. 9-27. 2 150-174.9 -.- 26.9 175-199.9 200-224 9 28,3 28.0-28.1 225-249.9 250-274.9 275-299.9 27.4 300-349.9 25. 6-28. 8 350-399.9 26.8 400H- 23.4^24.2 FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 99 Table 15.- -Relation of eye diameter to standard length in Monacanthidae of the western North Atlantic, by species and millimeter intervals [In percent of standard length for grouped millimeter intervals of standard length] Eye diameter in percent of standard length Standard length (mm.) Alultra Monacanthus Stephanolepii A manses monoceroa scripta achoepfii heudelotii tuckeri ciliatm hispidu> tetifer pullus 5-9.9 .. 13. 3-17. 1 13.3-15.7 13.2-15.6 11.8-14.0 11. 3-15. 8 13. 3-14. 8 12.0-13.9 11.3-13.5 10. 3-12. 4 10.4-11.5 9.5-11.7 9. 7-10. 5 9.1-11.6 8. 7-9. 6 7.4-9.9 7. 5-8. 9 7. 0-9. 1 7. 1-7. 8 6.9-7.1 10-14.9 - 11.8-14.6 10.8-13.1 11.0-12.6 10. 9-12. 8 11.2-12.1 12.6 10.2-11.1 9. 9-10. 3 8.6-9.8 9.2 9.1-10.2 7.4-9.1 7.7-8.9 7. 4-8. 3 14.0-15.9 13 8-15.0 13.1-14.4 12-1-13. 9 12.9-13.3 11.7-12.6 11.1-11.6 10.9-11.4 11.0-11.8 9.9-10.1 9.9-12.0 9.4-11.3 8.7-11.6 7.6-9.9 15-19 9 8.0 7.8-8.4 7. 2-7. 8 7. 1-8. 4 7. 0-7. 6 8.0-8.4 7. 4-8. 7 7.8-8.3 7.0-8.6 5.8-8.8 6. 7-7. 9 4. 8-8. 1 6. 0-7. 3 6. 4-7. 1 6. 1-7. 2 5. 1-6. 8 5. 7-6. 1 5.3-6.1 5. 1-6. 7 5.9 5.0-6.4 4.8-6.0 4. 9-5. 1 .. 11.4-14.4 11.0-12.7 14.9 20-24.9 25-29 9 8.5 8. 4-9. 1 7.8-9.0 7.3 7.4 7. 4-7. 9 6. 6-7. 2 6.6-8.0 ' 30-34 9 . . . 8. 6-9. 7 10. 8. 8-9. 1 11.0-11.2 10. 0-10. 1 9.8 9.4 8. 7-10. 1 10.6 35-39.9 9.7-11.0 40-44.9... 10.3-11.3 45-49.9 9. 5-10. 50-59.9 8.3 7.8 7.4 8.0 8.3 8. 2-9. 6 7.3-8.4 9. 0-10. 4 60-69.9... 8. 5-9. 4 70-79 9 8. 0-9. 80-89.9 8. 0-8. 2 90-99 9 6.7 6.9-6.8 6. 7-6. 6 6.6-6.4 7.9-8.1 100-124.9 6.4 6.7 5.4 6.4-8.1 6. 3-7. 7 6.4-8.5 7. 0-7. 6 6. 2-7. 7 6.6-7.5 7.5-8.8 125-149 9 7. 0-7. 6 150-174 9 7.0 175-199.9 6.8 200-224 9 6.4 6.6 225 ''49 9 250-274 9 275-299 9 5.6 300-349 9 5. 2-5. 3 350-399 9 6.3 4.2-4.4 Table 16. — Relation of eye to dorsal-spine distance to standard length in Monacanthidae of the western North Atlantic, by species and millimeter intervals (In percent of standard length for grouped millimeter Intervals of standard length] Eye to dorsal spine distance In percent of standard length Standard length (mm.) Alutera Monacanthus Stephamlepis Amanaes monoceros scripta sckoefii tteudelotii tuckeri ciliaiws hispidus setifer pullus 5-9.9 11. 7-17. 1 11.5-13.9 11. 6-12. 7 10.8-11.4 10.4-11.9 9.6-11.1 9. 7-11. 8.7-11.0 9. 5-10. 5 8. 8-10. 1 8. 9-10. 6 8. 7-10. 7 8. 8-10. 7 8. 3-10. 2 8. 4-10. 6 7. 4-10. 7. 4- 9. 8 8.0- 8.6 7. 3- 8. 5 10-14 9 8. 6- 9. 5 11.8-12.8 12.7 11.1 10.6-11.2 9. 9-13. 3 10.3-11.5 9.1 9.3- 9.9 8. 7-10. 9 10.9 9.8 9.4 8. 4-10. 4 7. 7- 8. 6 15-19.9... 4.7 3.9-4.8 4. 3- 5. 1 4.0 5. 1- 5. 6 4.0 5. 1- 6. 7 5. 4- 7. 1 6.4- 6.8 6. 4- 6. 9 6. 7- 8. 1 6.3- 8.3 7.3- 8.9 7. 8- 8. 6 7. 7- 9. 2 8. 4-11. 2 8. 8-12. 7 7.5-11.0 7.6-11.8 8.1 7. 9-13. 6 7. 9-12. 9 12.0-13.0 9. 5-10. 6 7. 7- 8. 8 7.4 20-24.9 8. 8- 9. 8 10.1 25-29 9 6.7 5.8-6.6 5. 1-6. 4 6.1 5.7 6.0-6.4 6.4-6.1 5.7-6.4 30-34.9 . . . 4. 3-7. 3 4.9 4. 0-5. 4 7. 5- 7. 7 7.0- 7.2 7.3 6. 5- 6. 7 6. 3- 7. 4 7.6 35-39.9 7.5-7.9 40-44.9 9.6 8. 0- 8. 4 8.0- 9.6 7.8-9.0 45-49.9 6.9-7.9 50-59.9 8.1 7.8 7.3 7.2 6.6 5. 7-6. 7 6. 7-6. 7 6.4 4.6-6.6 6. 2-6. 1 5. 1-6. 1 4. 9-5. 6 5. 1-6. 5 4. 8-6. 3 6.6-8.6 60-69.9 7. 1-7. 7 70-79 9 8.0- 9.1 6. 7- 7. 9 7. 8- 8. 5 7.0- 7.9 7 3-8 5 80-89.9 7. 2-7. 9 90-99.9 6.1 6.8-6.0 6.9 5. 1-6. 8. 2-8. 3 100-124.9 7.8 8.3 8.6 6 6-8.3 125-149.9 5.9-6.5 150-174.9 6.6 175-199.9 200-224.9 6.1 5.0-5.6 225-249.9 250-274.9 275-299.9 8.9 300-349.9. 7.4-8.3 350-399.9 6.3 400-1- 7.0-7.2 100 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE C. SPECIMENS EXAMINED The specimens examined are listed by species and arranged generally from north to south along the Atlantic coast of the United States, including Bermuda and the Bahamas; then north, west, and south around the Gulf of Mexico; next throughout Cuba and the rest of the West Indies; and finally southward through the Caribbean and to Brazil. The few records from outside the western North Atlantic are listed terminally on the individual species lists. The following abbreviations are used for col- lections : ANSP, Academy of Natural Sciences of Phila- delphia; BLBG, U.S. Fish and Wildlife Service Biological Laboratory, Brunswick, Ga. ; CAS, California Academy of Sciences; CHML, Cape Haze Marine Laboratory; FSBC, Florida State Board of Conservation Marine Laboratory, St. Petersburg, Fla.; SU, Stanford University Natiu-al History Museum; TU, Tulane University; UF, University of Florida; UNC, University of North Carolina Institute of Fisheries Research; USNM, U.S. National Museum. The following abbreviations are used for collect- ing methods, in those instances in which the methods are definitely known : D.N., dip net; M.L.N. , meter larval net; P.T., plankton tow; S.C., stomach contents of a larger fish; Sn., seine; Tr., bottom trawl. The following additional abbreviations are used : • Cr., cruise; Reg., regular station; Spc, special station; Std., standard station; Sta., station; S.L., standard length. Alutera monoceros Woods Hole, Ma.ss., 22 Aug. 1898, (1 specimen) 114 mm. S.L., USNM 85771 25 miles southwest of Cuttyhunk Island, Mass., 17 Sept. 1935, (1) 144 mm. S.L., caught in lobster pot, USNM 107273. ___34°05' N., 76°21' W., to 34°01' N., 76°18' W., Alhalross Cr. 31-A, Sta. 2, tow 2, 19 Jan. 1950, (1) 520 mm. S.L., Tr. 25-75 fathoms, USNM 152089.___34°05' N., 76°21' W., to 34°01' N., 76°18' W., Alhalross Cr. 31-A, Sta. 2, tow 2, 19 Jan. 1950, (1) 401 mm. S.L., Tr. 25-75 fathoms, USNM 152090... .Kitty Hawk, N.C., (1) 167 mm. S.L., USNM 163881. .-.30°34' N., 80°17' W., Silver Bay Sta. 1550, 17 Jan. 1960, (1) 545 mm. S.L., Tr. 22-21 fathoms, BLBG....26°47' N., 79°53' W., Combat Sta. 459, 28 July 1957, (1) 122 mm. S.L., D.N.y BLBG....Tortugas, Fla., (3) 53.0-73.0 mm. S.L., collected by W. H. Longley, USNM 11 7022.... Port of Fortaleza, Mucuripe, Brazil, Mar. 1945, (1) 91.0 mm. S.L., SU 52309, No. 558 Port of Fortaleza, Mucuripe, Brazil, June 1945, (1) 78.0 mm. S.L., SU 52309, No. 841. ...Durban Harbor, Durban, Natal, South Africa, 8 Aug. 1933, (1) 96.5 mm. S.L., collected by H. W. B. Marley, Herre 1934 Exped., SU31365 China Sea, coast of Pahang, Malay Peninsula, Nov. 1926, (2) 115 and 137 mm. S.L., collection of Fisheries Dept. F. M. S., Herre 1934 E.xped., SU 30786.. ..Manado, Indonesia, (1) 115 mm. S.L., USNM 126630.... Chame Point, Pacific Panama, (1) 49.5 mm. S.L., collected by Tweedlie, USNM 82059. Alutera scripta 34°38' N., 74°46' W., Gill Cr. 2, Reg. 80, 12 May 1953, (1 specimen) 62.5 mm. S.L., D.N., BLBG 34°14' N., 76°03' W., Silver Bay, 15 Sept. 1959, (3) 78.5-103 mm. S.L., D.N., BLBG....32°34' N., 77°48' W., Gill Cr. 8, Reg. 52, 26 Sept. 1954, (1) 53.0 mm. S.L., D.N., BLBG..-. Bermuda, (1) about 410 mm. S.L. (skin and skull only), collected by G. B. Goode, USNM 21889. ..-31°57' N., 78°09' W., Gill Cr. 3, Reg. 50, 6 Aug. 1953, (1) 39.5 mm. S.L., D.N., BLBG.-..30°00' N., 80°10' W., Silver Bay Qia.. 476, 18 June 1958, (1) 33.0 mm. S.L., M.L.N., BLBG.... 29°40' N., 80°00' W., Gill Cr. 8, Reg. 18, 13 Sept. 1954, (1) 35.0 mm. S.L., D.N., BLBG..-.29°38' N., 80°12' W., Combat Sta. 474, 14 Aug. 1957, (5) 64.0-73.0 mm. S.L., D.N., BLBG....29°38' N., 80°09' W., Silver Bay Sta. 471, 17 June 1958, (1) 46.0 mm. S.L., M.L.N., BLBG...- 29°29' N., 80°09' W., Combat Sta. 485, 18 Aug. 1957, (1) 66.0 mm. S.L., D.N., BLBG....29°29' N., 80°10' W., Combat Sta. 490, 19 Aug. 1957, (1) 99.0 mm. S.L., D.N., BLBG...-29°10' N., 80°19' W., to 29°19' N., 80°15' W., Combat Sta. 336 to Sta. 337, 1 June 1957, (1) 41.0 mm. S.L., D.N., BLBG....28°18' N., 79°28' W., Gill Cr. 8, Reg. 8, 12 Sept. 1954, (1) 39.0 mm. S.L., D.N., BLBG.... 27°14' N., 79°50' W., Combat Sta. 462, 29 July 1957, (1) 53.5 mm. S.L., D.N., BLBG... .Jupiter Inlet, Fla., July 1958, (1) 27.0 mm. S.L., BLBG....26"'47' N., 79°53' W., Combat Sta. 459, 28 July 1957, (2) 119 and 128 mm. S.L., D.N., BLBG....25°11' N., 79°56' W., Combat Sta. 443, 22 July 1957, (1) 127 mm. S.L., D.N., BLBG....25°10' N., 80°02' W., Combat Sta. 438, 22 July 1957, (3) 31.0-121 mm. S.L., D.N., BLBG....Tortugas, Fla., (1) 160 mm. S.L., collected by W. H. Longley, USNM 117024.... Tortugas, Fla., (2) 70.5 and 121 mm. S.L., collected by W. H. Longley, USNM 117023. ...Gulf of Mexico, 30 miles southwest of Boca Grande, Fla., 21 Oct. 1956, (1) 200 mm. S.L., CHML Destin, Fla., about June 1958, (1) 163 mm. S.L., BLBG Choctawatchee Bay, Fla., June 1958, (1) 152 mm. S.L., BLBG..-.28°47' N., 87°56' W., Oregon Sta. 1589, 23-24 July 1956, (2) 141 and 164 mm. S.L., USNM 158763... .29°10' N., 88°08' W., Oregon Sta. 1525, 17 May 1956, (1) 67.0 mm. S.L., TU 11639.... 28°45' N., 88°03' W., Oregon Sta. 1590, 24 July 1956, (2) 164 mm. S.L., UF.-..26°40' N., 92°00' W., Oregon Sta. 1035, 8 May 1954, (1) 145 mm. S.L., TU 10937.. ..24°50' N., 92°35' W., Oregon Sta. 2198, 23-24 June 1958, (1) 107 mm. S.L., D.N., BLBG....20°50' N., 86°10' W., Oregon Sta. 1297, 28 Apr. 1955, (1) 35.0 mm. S.L., D.N., USNM 159168 Cuba, 1914, (1) 244 mm. S.L., collected by Henderson and Bartsch, USNM 82569.... Cuba, (1) 57.6 mm. S.L., collected by Poey, USNM 37466 Jamaica, (1) 230 mm. S.L., collection of Institute of Jamaica, USNM 32041 Haiti, (1) 79.0 mm. S.L., collected by Beebe, FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 101 USNM 178807.. -.Barbados, Antigua, (1) 377 mm. S.L., collection U. of Iowa Barbados-Antigua Exped. 1918, USNM 170118 No data, presumably from western North Atlantic, (1) 35.5 mm. S.L., BLBG Chame Point, Pacific Panama, (2) 50.0-67.5 mm. S.L., USNM 82059 Ecuador, 16 Nov. 1919, (1) 52.0 mm. S.L., USNM 84042..-. No data, (1) 98.0 mm. S.L., USNM 83814. Alutera schoepfii Halifax, Nova Scotia, (1) 95.5 mm. S.L., collected by Honeyman, USNM 22490 Woods Hole, Mass., Fish Hawk, Aug. 1899, (1) 170 mm. S.L., USNM 120557.. __ Woods Hole, Mass., (1) 142 mm. S.L., USNM 34397 Woods Hole, Ma.ss., (2) 63.0 and 137 mm. S.L., USNM 1 19226.. _ .Vineyard Sound, Mass., 4 Oct. 1927, (2) 118 and 132 mm. S.L., USNM 107272 West Falmouth, Mass., 22 Aug. 1899, (1) 126 mm. S.L., USNM 120556.... Buzzards Bay, Mass., (1) 117 mm. S.L., USNM 107263 New Bedford, Mass., (2) 96.0 and 102 mm. S.L., USFC 16542, TU 8911 Katama Bay, Mass., 19 Aug. 1938, (1) 101 mm. S.L., USNM 107275.... Newport, R.I., (6) 131-160 mm. S.L., collected by S. Powell, USNM 20727 Newport, R.I., (1) 143 mm. S.L., collected by S. Powell, USNM 20198.. .Long Island, N.Y., (1) 163 mm. S.L., collected by Jordan, SU 1959 Barnegat Bay, Seaside Park, N.J., July-Aug. 1926, (1) 355 mm. S.L., coUected by B. A. Bean, USNM 120493 Carson's Inlet, N.J., 9 Sept. 1916, (1) 196 mm. S.L., ANSP 46736. ...Bombay Hook, Del., 10 June 1884, (1) 407 mm. S.L., Sn., USNM 34977 Mouth of Great Wicinoco River, Chesapeake Bay, Albatross, (1) 355 mm. S.L., USNM 33333 Mouth of Great Wicinoco River, Chesapeake Bay, Albatross, (1) 390 mm. S.L., USNM 33332 Ocean View, Norfolk, Va., 22 Oct. 1922, (2) 116 and 130 mm. S.L., Sn., USNM 91 102.... Ocean View, Va., 1-5 Oct. 1922, (4) 101-131 mm. S.L., Sn., USNM 91103.. ..34°58' N., 75°54' W., Silver Bay Sta. 1258, 8 Sept. 1959, (1) 367 mm. S.L., Tr. 13-14 fathoms, BLBG.-..34°53.5' N., 75°32' W., to 34°54' N., 75°31' W., Albatross III, Cr. 31-A, Sta. 17, 22 Jan. 1950, (1) 363 mm. S.L., 23 fathoms, USNM 152092.. ..34°50' N., 76°14' W., Silver Bay Sta. 1254, 8 Sept. 1959, (4) 194-355 mm. S.L., Tr. 12-11 fathoms, BLBG....34°47' N., 76°20' W., Silver Bay Sta. 1252, 8 Sept. 1959, (2) 320 and 340 mm. S.L., Tr. 9-11 fathoms, BLBG....34°44' N., 75°53' W., Silver Bay Sta. 1271, 12 Sept. 1959, (1) 402 mm. S.L., Tr. 17 fathoms, BLBG... . 34°39' N., 76°01' W., Silver Bay Sta. 1270, 12 Sept. 1959, (1) 335 mm. S.L., Tr. 20 fathoms, BLBG....34°38' N., 76°49' W., Silver Bay Sta. 1291, 22 Sept. 1959, (1) 263 mm. S.L., Tr. 8-10 fathoms, BLBG....34°37' N., 77°01' W., Silver Bay Sta. 1311, 24 Sept. 1959, (1) 271 mm. S.L., Tr. 7 fathoms, BLBG....34°35' N., 76°23' W., Silver Bay Sta. 1250, 7 Sept. 1959, (1) 328 mm. S.L., Tr. 11-10 fathoms, BLBG....34°35' N., 77°03' W., Silver Bay Sta. 1310, 24 Sept. 1959, (4) 217-260 mm. S.L., Tr. 8 fathoms, BLBG.-..34°32' N., 76°33' W., Silver Bay Sta. 1317, 27 Sept. 1959, (1) 303 mm. S.L., Tr. 8 fathoms, BLBG....34°31' N., 76°53' W., Silver Bay Sta. 1239, 6 Sept. 1959, (1) 360 mm. S.L., Tr. 14-12 fathoms, BLBG....34°31' N., 76°35' W., Silver Bay Sta. 1242, 6 Sept. 1959, (1) 223 mm. S.L., Tr. 9 fathoms, BLBG.... Cape Lookout, N.C., July 1912, (1) 345 mm. S.L., col- lected by R. J. Coles, USNM 74300 Cape Lookout, N.C., 22 Oct. 1927, (1) 72.0 mm. S.L., surface, USNM 111836.....34°29' N., 76°57' W., Silver Bay Sta. 1307, 24 Sept. 1959, (1) 360 mm. S.L., Tr. 12 fathoms, BLBG 34°27' N., 76°20' W., Silver Bay Sta. 1249, 7 Sept. 1959, (1) 294 mm. S.L., Tr. 7 fathoms, BLBG....34°26' N., 77°05' W., Silver Bay Sta. 1308, 24 Sept. 1959, (1) 348 mm. S.L., Tr. 9-10 fathoms, BLBG....34°25' N., 76°51' W., Silver Bay Sta. 1294, 23 Sept. 1959, (2) 308 and 334 mm. S.L., Tr. 12-13 fathoms, BLBG....34°23' N., 76°54' W., Silver Bay Sta. 1295, 23 Sept. 1959, (2) 309 and 352 mm. S.L., Tr. 13-15 fathoms, BLBG.. ..34°21' N., 77°34' W., Silver Bay Sta. 1227, 4 Sept. 1959, (2) 240 and 293 mm. S.L., Tr. 7-8 fathoms, BLBG....34°16' N., 77°34' W., Silver Bay Sta. 1226, 4 Sept. 1959, (1) 350 mm. S.L., Tr. 8-7 fathoms, BLBG... .34°15' N., 76°36' W., Silver Bay Sta. 1236, 6 Sept. 1959, (1) 358 mm. S.L., Tr. 18 fathoms, BLBG....34°13' N., 76°48' W., Silver Bay Sta. 1296, 23 Sept. 1959, (1) 366 mm. S.L., Tr. 17 fathoms, BLBG....34°09' N., 76°55' W., Silver Bay Sta. 1230, 5 Sept. 1959, (2) 342 and 379 mm. S.L., Tr. 17 fathoms, BLBG 34°09' N., 76°35' W., Silver Bay Sta. 1297, 23 Sept. 1959, (2) 243 and 251 mm. S.L., Tr. 20 fathoms, BLBG..--34°07' N., 77°19' W., Silver Bay Sta. 1224, 4 Sept. 1959, (1) 372 mm. S.L., Tr. 13 fathoms, BLBG....33°56' N., 77°20' W., Silver Bay Sta. 1215, 3 Sept. 1959, (4) 330-372 mm. S.L., Tr. 15 fathoms, BLBG 33°47' N., 77°50' W., Silver Bay Sta. 1210, 2 Sept. 1959, (2) 136 and 355 mm. S.L., Tr. 8 fathoms, BLBG 33°45' N., 76°50' W., Silver Bay Sta. 1218, 3 Sept. 1959, (6) 374-410 mm. S.L., Tr. 23-24 fathoms, BLBG 33°41' N., 77°40' W., Silver Bay Sta. 1209, 2 Sept. 1959, (2) 309-332 mm. S.L., Tr. 11-12 fathoms, BLBG....33°32' N., 77°30' W., Silver Bay Sta. 1208, 2 Sept. 1959, (3) 300-358 mm. S.L., Tr. 14 fathoms, BLBG...-32°54' N., 77°04' W., Gill Cr. 3, Reg. 61, 10 Aug. 1953, (1) 20.5 mm. S.L., D.N., BLBG....32°24' N., 78°45' W., Gill Cr. 8, Reg. 48, 25 Sept. 1954, (1) 23.0 mm. S.L., D.N., BLBG Charleston, S.C, (1) 124 mm. S.L., collected by J. C. Mitchell, USNM 30727 Bermuda, (1) 365 mm. S.L., USNM 23859... .31 °00' N., 80°23' W., Gill Cr. 4, Reg. 32, 16 Oct. 1953, (1) 26.5 mm. S.L., D.N., BLBG Commercial Trawling Area, Bruns- wick, Ga., 9-13 Apr. 1956, (1) 187 mm. S.L., Tr., BLBG Commercial Trawling Area, Brunswick, Ga., 23 Sept. 1956, (1) 96.0 mm. S.L., Tr., BLBG.-..30°20' N., 80°36' W., Gill Cr. 7, Reg. 25, 26 June 1959, (1) 22.5 mm. S.L., D.N.,BLBG....29°32' N., 80°25' W., Combat Sta. 348, 2 June 1957, (1) 355 mm. S.L., Tr. 22 fathoms, BLBG Port Canaveral Anchorage, Fla., Combat, 28-29 Apr. 1957, (1) 32.5 mm. S.L., D.N. , BLBG... .Jupiter Inlet, Fla., Sept.-Nov. 1958, (3) 22.7-33.0 mm. S.L., UF Jupiter Inlet, Fla., Aug. 1958, (1) 68.0 mm. S.L., UF Jupiter Inlet, Fla., June 1958, (1) 377 mm. S.L., UF....27°04' N., 80°04' W., Gill Cr. 2, Reg. 5, 23 Apr. 1953, (1) 15.0 mm. S.L., D.N., BLBG Biscayne Bay, Fla., Launch 58, 5 Sept. 1938, (1) 126 mm. S.L., Tr., USNM 155986.... Tortugas, Fla., (5) 27.5-62.5 mm. S.L., collected by W. H. Longley, USNM 11 7026.... Cape Haze, Fla., 2 July 1958, (1) 49.0 mm. S.L., CHML....Gasparilla Bay, 102 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Fla., (1) 337 mm. S.L., CHML GaspariUa Bay, Fla., (1) 233 mm. S.L., CHML GaspariUa Sound, Placida, Fla., 10 May 1955, (1) 197 mm. S.L., CHML._. .Gas- pariUa Sound, Placida, Fla., 9 Apr. 1956, (1) 187 mm. S.L., CHML GaspariUa Sound, Placida, Fla., 26 May 1958, (1) 250 mm. S.L., TU 18493. -._Lemon Bay, Fla., 28 Sept. 1955, (1) 98.0 mm. S.L., CHML Lemon Bay, Fla., summer 1956, (1) 95.0 mm. S.L., CHML Lemon Bay, Fla., May 1957, (1) 41.0 mm. S.L., CHML Englewood Beach, Fla., 26 Aug. 1956, (1) 30.0 mm. S.L., CHML Pass-a-Grille to Venice, Fla., 10-11 Feb. 1958, (1) 306 mm. S.L., FSBC VGS 58-41._..Egmont Key, Fla., 23 Mar. 1958, (12) 218-257 mm. S.L., TU 17941.__. Mullet Key, Boca Ciega Bay, Fla., 4 June 1958, (1) 64.0 mm. S.L., FSBC VGS 58-152 Bird Key, Boca Ciega Bay, Fla., 7 July 1958, (2) 89.0 and 95.0 mm. S.L., FSBC VGS 58-181 Bird Key, Boca Ciega Bay, Fla., 7 July 1958, (1) 273 mm. S.L., FSBC VGS 58-276. .._ City dock, Clearwater, Fla., 5 Aug. 1948, (1) 96.0 mm. S.L., pole- and-line, USNM Cedar Keys, Fla., 18-20 Aug. 1949, (2) 62.5 and 93.5 mm. S.L., UF Cedar Keys, Fla., 7 Nov. 1950, (1) 73.0 mm. S.L., UF.,..Piney Point, Cedar Keys, Fla., 7 June 1950, (2) 37.0-38.5 mm. S.L. UF.__-Cedar Keys, Fla., 3 Oct. 1953, (1) 190 mm. S.L., UF C-10-353-4 Cedar Keys, Fla., 20 Sept. 1953, (2) 158 and 176 mm. S.L., UF C-9-2053-3.... Cedar Keys, Fla., 20 Sept. 1953, (1) 159 mm. S.L., UF.__-Cedar Keys, Fla., 6 Sept. 1953, (1) 126 mm. S.L., UF 2542 Cedar Keys, Fla., 16 Aug. 1953, (1) 103 mm. S.L., UF C-8- 1653-5 Cedar Keys, Fla., 25 July 1953, (3) 60.0-96.0 mm. S.L., UF C-7-2553-4.... Cedar Keys, Fla., 12 July 1953, (1) 98.0 mm. S.L., UF C-7-1253-5.___Cedar Keys, Fla., 12 July 1953, (1) 80.5 mm. S.L., UF C-7-1253-1.... Cedar Keys, Fla., 12 July 1953, (1) 118 mm. S.L., UF C-7-1253-4 Cedar Keys, Fla., 12 July 1953, (2) 91.0 and 165 mm. S.L., UF C-7-1253-7 Cedar Keys, Fla., 12 July 1953, (1) 68.0 mm. S.L., UF 2475 Cedar Keys, Fla., 1 July 1953, (1) 64.0 mm. S.L., UF 2457. _._ Cedar Keys, Fla., 28 June 1953, (4) 47.0-81.5 mm. S.L., UF C-6-2853-3 Cedar Keys, Fla., 28 June 1953, (1) 59.0 mm. S.L., UF C-6-2853-l.__ .Cedar Keys, Fla., 30 June 1954, (1) 85.5 mm. S.L., UF C-6-3054-6. _ _ . Cedar Keys, Fla., 15-16 Aug. 1955, (3) 47.0-72.5 mm. S.L., TU 11953 Cedar Keys, Fla., 11-13 June 1957, (39) 46.5-73.0 mm. S.L., USNM 176239 Cedar Keys, Fla., 23 Nov. 1957, (1) 86.5 mm. S.L., UF..._Destin, Fla., Nov. 1956, (1) 213 mm. S.L., UF..._Fort Walton Beach, Fla., Nov. 1956, (1) 325 mm. S.L., UF._._Pensa- cola, Fla., (1) 185 mm. S.L., CAS C87 3519. __ .Florida, Orian, (1) 126 mm. S.L., collected by B. A. Bean and J. A. Pine, USNM 62555.. ._28°47' N., 87°56' W., Oregon Sta. 1589, 23-24 July 1956, (1) 68.5 mm. S.L., D.N., USNM 158763.... 28 °44' N., 88°08' W., Oregon Sta. 1583, 20-21 July 1956, (1) 59.0 mm. S.L., D.N., UF.... 29°22' N., 88°48' W., Oregon Sta. 1109, 15 June 1954, (5) 317-343 mm. S.L., TU 17106.... Mississippi coast, (1) 124 mm. S.L., USNM 147796.... Grand Isle, La., (n 119 mm. S.L., USNM 125803.... Oyster Bayou, Terrebonne County, La., June-July 1954, (1) 60.5 mm. S.L., TU 9039....Freeport, Tex.,' Jan.-May 1947, (1) 309 mm. S.L., USNM 147808.... Aransas Pass, Tex., 11 Oct. 1926, (1) 35.8 mm. S.L., USNM 156000.... Aransas Pass, Tex., 5-7 June 1954, (4) 32.3-51.3 mm. S.L., TU 11781 Harbor Island, Corpus Christi, Tex., 5 Oct. 1926, (2) 74.5 and 78.0 mm. S.L., USNM 156002 Corpus Christi, Tex., (1) 72.0 mm. S.L., USNM 94553 Corpus Christi, Tex., (1) 140 mm. S.L., USNM 94554 Hahia Honda, Cuba, 5 June 1914, (1) 272 mm. S.L., collected by Henderson and Bartsch, USNM 82568 Hahia Honda, Cuba, 5 June 1914, (1) 283 mm. S.L., collected by Henderson and Bartsch, USNM 82567 Kingston, Jamaica, (1) 290 mm. S.L., SU 11808 Jamaica, B.W.I., (3) 105-170 mm. S.L., collected by J. S. Roberts, SU 4880 Jamaica, (1) 134 mm. S.L., collected by Adams, USNM 4910 Bizoton Wharf, Haiti, (1) 65.0 mm. S.L., collected by Beebe, USNM 178063 Bizoton, Haiti, (10) 25.0-53.5 mm. S.L., col- lected by Beebe, USNM 178917 Haiti, (1) 92.5 mm. S.L., collected by Beebe, USNM 178055 Off Nicaragua, Oregon, (1) 237 mm. S.L., USNM 159204.... Chiriqui Lagoon, Atlantic Panama, 12 July 1933, (1) 37.5 mm. S.L., USNM 178916 Fox Bay, Colon, Atlantic Panama, 5 Jan. 1911, (2) 91.5 and 115 mm. S.L., USNM 81516 Colon, Panama, summer 1916, (1) 132 mm. S.L., ANSP 49071 Brazil, Albatross, (1) 67.5 mm. S.L., USNM 43291 No data, presumably from western North Atlantic, (1) 328 mm. S.L., BLBG No data, presumably from western North Atlantic, (1) 44.0 mm. S.L., BLBG No data, (1) 184 mm. S.L., USNM 91471 No data, (2) 120 and 121 mm. S.L., USNM 91472 No data, (1) 130 mm. S.L., USNM 91469 No data, (1) 151 mm. S.L., USNM 91470. Alutera heudelotii Off southern Massachusetts, Nov. 1949, (1) 158 mm. S.L., USNM 148340.— North Carolina coast. Albatross III Cr. 31-B, Jan.-Feb. 1950, (1) 240 mm. S.L., USNM 152043. ...84°45'20" N., 75°38'10" W., Albatross Sta. 2599, 18 Oct. 1885, (1) 29.2 mm. S.L., USNM 131492 34°45' N., 75°38' W., Combat Sta. 386, 17 June 1957, (1) 125 mm. S.L., Tr. 45 fathoms, BLBG....34°39' N., 76°01' W., Silver Bay Sta. 1270, 12 Sept. 1959, (3) 140-172 mm. S.L., Tr. 20 fathoms, BLBG...-34°38'30" N., 75°33'30" W., Albatross Sta. 2603, 18 Oct. 1885, (1) 27.0 mm. S.L., USNM 131596.... 34°05' N., 76°21' W., to 34°01' N., 76°18' W., Albatross III Cr. 31-A, tow 2, (1) 220 mm. S.L., Tr. 25-75 fathoms, USNM 152091. ...33°29' N., 77°22' W., Silver Bay Sta. 1205, 1 Sept. 1959, (1) 232 mm. S.L., Tr. 16-20 fathoms, BLBG....33°2]' N., 77°24' \V., Silver Bay Sta. 1204, 1 Sept. 1959, (1) 171 mm. S.L., Tr. 15-16 fathoms, BLBG... -Bermuda, (1) 69.5 mm. S.L., collected by Beebe, USXM Miami, Fla., 24 Oct. 1953, (1) 78.0 mm. S.L., CHML.... 25° 10' N., 80°02' W., Co77ibat Sta. 438, 22 July 1957, (6) 31.0-108 mm. S.L., D.N., BLBG.._.24°13' N., 81°42' W., Combat Sta. 436, 21 July 1957, (1) 69.5 mm. S.L., BLBG.... Tortugas, Fla., (2) 71.5 and 92.5 mm. S.L., collected by W. H. Longley, USNM 117025.. ..Tortugas, Fla., (4) 42.5-54.0 mm. S.L., collected by W. H. Longley, USNM 88106 Channel west of White Shoal, Tortugas, Fla., 22 June 1932, (1) 78.5 mm. S.L., USNM 109177 (lecto- type) Channel west of White Shoal, Tortugas, Fla., FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 103 22 June 1932, (5) 48.5-63.5 mm. S.L., USNM 109178, (paratypes) Tortugas, J^la., (7) 29.7-69.5 mm. S.L , collected by W. H. Longley, USXM 117027 Tortuga.s, Fla., mid- April 1956, (1) 165 mm. S.L., CHML Tortugas, Fla., June- Aug. 1926, (1) 60.0 mm. S.L., USXM 88105.. ..26°n' N., 82°48' W., Oregon Sta. 987, 7 Apr. 1954, (1) 121 mm. S.L., Tr. 20 fathoms, UF 3880 28°23' N., 84°49' W., Oregon Sta. 916, 10 Mar. 1954, (1) 131 mm. S.L., Tr. 37 fathoms, UF 3603.,__28°22' N., 84°53' W., Oregon Sta. 917, 10 Mar. 1954, (3) 131-136 mm. S.L., Tr. 48 fathoms, TU 13180.. __29°07' N., 84°54' W., Oregon Sta. 890, 7 Mar. 1954, (1) 206 mm. S.L., Tr. 19 fathoms, UF 3588 29°00' N., 85°02' VV., Orego7i Sta. 891, 7 Mar. 1954, (1) 173 mm. S.L., Tr. 21 fathoms, UF 3648....29°36' N., 86°01' W., Silver Bay Sta. 159, 23 Aug. 1957, (2) 136 and 198 mm. S.L., Tr., USNM.... 28°47' N., 87°56' W., Oregon Sta. 1589, 23-24 July 1956, (1) 40.0 mm. S.L., USNM 158763 27°55' N., 88°b5' W., Oregon Sta. 1139, 24 July 1954, (1) 50.0 mm. S.L., TU 13164._._29°20' N„ 88°26' W., Silver Bay Sta. 14, 1 July 1957, (1) 230 mm. S.L., TU 16316. ..-27°35' N., 89°35' W., Oregon Sta. 1133, 22 July 1954, (1) 33.0 mm. S.L., TU 13032.... 28° 10' N., 94°05' W., Silver Bay Sta. 9, 29 June 1957, (1) 211 mm. S.L., TU 16270.. ..28°12' N., 94°10' W., Silver Bay Sta. 8, 29 June 1957, (.1) 220 mm. S.L., TU 16256. ...28°28' N., 94°20' W., Silver Bay Sta. 6, 25 June 1957, (1) 204 mm. S.L., TU 16230 28°02' N., 94°39' W., Oregon Sta. 143, 21 Nov. 1950, (1) 102 mm. S.L., TU 2021....24°00' N., 96°50' W., Oregon Sta. 1074, 25 May 1954, (1) 30.5 mm. S.L., D.N., UF 3829..._18°43' N., 93°30' W., Oregon Sta. 1060, 16 May 1954, (1) 81.0 mm. S.L., UF._._24°50' N., 92°35' W., Oregon Sta. 2198, 23-24 June 1958, (5) 31.5-43.0 mm. S.L., D.N., BLBG.... Ambergris Cay, Yucatan, (1) 136 mm. S.L., USNM 79247.... Recife, Brazil, (2) 188 and 195 mm. S.L., SU 52306. Monacanthus ciliatus Off Georges Bank, Mass., Caryn, (1 specimen) 30.3 mm. S.L., BLBG....35°08' N., 75°22' \V., Gill Cr. 8, Reg. 78, 29 Sept. 1954, (1) 18.0 mm. S.L., D.N., BLBG....34°32' N., 75°53' \V., Silver Bay Sta. 1268, 11 Sept, 1959, (1) 77.5 mm. S.L., Tr. 31-30 fathoms, BLBG 34°10' N., 77°30' W., Gill Cr. 8, Reg. 67, 28 Sept. 1954, (1) 16.2 mm. S.L., D.N., BLBG..._34°04' N., 76°14' W., Gill Cr. 7, Reg. 71, 10 July 1954, (1) 23.8 mm. S.L., D.N., BLBG.... 33°45' N., 76°50' W., Silver Bay Sta. 1218, 3 Sept. 1959, (7) 43.0-85.5 mm. S.L., Tr. 23-24 fathoms, BLBG.... 33°32' N., 77°30' W., Silver Bay Sta. 1208, 2 Sept. 1959, (4) 41.6-51.1 mm. S.L., Tr. 14 fathoms, BLBG 33°29' N., 77°22' W., Silver Bay Sta. 1205, 1 Sept. 1959, (2) 72.5 and 77.0 mm. S.L., Tr. 16-20 fathoms, BLBG....33°29' N., 76° 40' W., Gill Cr. 3, Reg. 64, 11 Aug. 1953, (3) 21,0-21.5 mm. S.L., D.N., BLBG..-.33°17' N., 78°38' \V., Gill Cr. 8, Reg. 55, 26 Sept. 1954, (15) 15.5-21.5 mm. S.L., D.N., BLBG....33°03' N., 78°21' VV., Gill Cr, 7, Reg. 54, 4 July 1954, (1) 15.3 mm. S.L., D.N., BLBG 33°03' N., 78°21' W., Gill Cr. 8, Reg. 54, 26 Sept. 1954, (1) 24.6 mm. S.L., D.N., BLBG....32°56' N,, 78°06' W., Combat Sta. 283, 19 Apr. 1957, (1) 101 mm. S.L., Tr. 50 fathoms, BLBG... .32°54' N., 77°04' \V., Gill Cr. 3, Reg. 61, 10 Aug. 1953, (2) 21,7 and 21.8 mm. S.L., D.N., BLBG....32°60' N., 77°27' W., Combat Sta. 295, 21 Apr. 1957, (2) 23.9 and 25.8 mm. S.L., D.N., BLBG....32°24' N., 78°45' \V., Gill Cr. 8, Reg. 48, 25 Sept. 1954, (1) 18.5 mm. S.L., D.N,, BLBG Bermuda, (1) 51.0 mm. S.L., collected by J. M. Jones, USNM 21249 Bermuda, (1) 51.0 mm. S.L,, collected by Beebe, USNM 178766 Castle Roads, Bermuda, 11-12 Sept. 1931, (6) 23.7-43.0 mm. S.L., dredge 12-20 ft., USNM 178785 Nonsuch, Bermuda, (6) 39.5-52,0 mm. S,L., collected by Beebe, USNM 178862 Nonsuch, Bermuda, 18 Oct. 1930, (4) 35.6-43.5 mm. S.L., USNM 178859.. ..31°00' N., 80°23' \V., Gill Cr. 4, Reg. 32, 16 Oct. 1953, (2) 19.0-24.0 mm. S.L., D.N., BLBG....30°00' N., 80°10' \V., Silver Bay Sta. 476, 18 June 1958, (26) 11.0-21.9 mm. S.L., M.L.N., BLBG....29°48' N., 80°12' W., Silver Bay Sta. 470, 17 June 1958, (14) 15.7-19.5 mm. S.L., D.N., BLBG 29°38' N., 80°09' W., Silver Bay Sta. 471, 17 June 1958, (25) 12.4-20,1 mm. S.L., M.L,N., BLBG....29°29' N., 80°09' VV., Combat Sta. 485, 18 Aug. 1957, (4) 14.0-18.7 mm. S.L., D.N., BLBG._._29°26' N., 80°08' VV„ Combat Sta. 315, 27 Apr. 1957, (2) 16.5-16.8 mm. S.L., D.N., BLBG....29°19' N., 80°18' VV., Combat Sta. 339, 1 June 1957, (1) 109 mm. S.L., Tr. 25 fathoms, BLBG 29°10' N., 80°19' W., Combat Sta. 336, 1 June 1957, (1) 21.2 mm. S.L., D.N., BLBG..._29°00' N., 79°26' VV., Gill Cr. 4, Reg. 16, 14 Oct. 1953, (1) 24,3 mm. S.L., D.N., BLB<5...-28°58' N., 80°13' VV,, Combat Sta. 333, 1 June 1957, (1) 107 mm. S.L., Tr. 30 fathoms, BLBG 28°19' N., 79°26' W., Gill Cr. 3, Reg. 8, 26 July 1953, (2) 18.0 and 20.5 mm. S.L., D.N., BLBG..-.28n5' N., 77°01' W., Gill Cr. 9, Spc. 6-7, 5 Nov. 1954, (1) about 23 mm. S.L., S.C. of Thunnus albacares (Bonnaterre) BLBG 28°09' N., 79°21' W., Gill Cr. 7, Spc. 9 to Reg. 8, 24 June 1954, (1) S.C. of Thunnus allanlicus (Lesson), BLBG 27°36' N., 83°40' VV., Oregon Sta. 935, 18 Mar. 1954, (1) 111 mm. S.L., Tr. 27 fathoms, UF 3665....27°34' N., 80°04' W., Gill Cr. 7, Reg. 4-5, 23 June 1954, (1) 15.5 mm. S.L., S.C. oi Euthynnus alletleralus (Rafinesque) BLBG 27°14' N., 79°50' VV., Combat Sta. 462, 29 July 1957, (1) 22,0 mm. S.L., D.N., BLBG....27°01' N., 80°04' W., Gill Cr. 2, Reg. 3, 23 Apr. 1953, (3) 15.3-19.9 mm. S.L., D.N., BLBG Settlement Point, Grand Bahama Island, Gill Cr. 8, 29 Aug. 1954, (15) 13.6-21,5 mm. S.L,, D,N., BLBG._..26°47' N., 79°53' VV., Combat Sta. 459, 28 July 1957, (3) 23.0-25.0 mm. S.L., D,N., BLBG 26°45' N., 78°55' W., Gill Cr. 4, 3 Oct 1955, 1600, (1) about 15 mm. S.L., S.C. of Thunnus atlanticus, BLBG 26°45' N., 78°55' W., Gill Cr. 4, 3 Oct. 1955, 1555, (2) about 17 mm. S.L., S.C, of Thunnus atlanticus, BLBG Hawks Bill Creek, Grand Bahama Island, Gill Cr. 4, 3 Oct. 1953, 1900-2100, (6) 12,8-22.7 mm. S.L., D.N., BLBG Hawks Bill Creek, Grand Bahama Island, Gill Cr. 4, 3 Oct. 1953, 2130-2330, (1) 21.8 mm. S.L., D.N., BLBG.... 26°09' N., 78°12' VV., Gill Cr. 7, 22 June 1954, (1) 16.5 mm. S.L., S.C. of Thunnus atlanticus, BLBG 26°05' N., 78°12' W., Gill Cr. 4, Nassau to Reg. 1,11 Oct. 1953, (6) about 12-18 mm. S.L., S.C. of Sphyraena barracuda (VValbaum), BLBG Hatchet Bay, Eleuthera Island, Bahamas, (1) 93,5 mm. S.L., UF 3504,. ..25° 16' N,, 80°07' VV., Combat Sta. 457, 26 July 1957, (11) 14.8-20.7 104 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE mm. S.L., D.N., BLBG....25°13' N., 80°10' W., Combat Sta. 455, 26 July 1957, (5) 16.0-18.7 mm. S.L., D.N., DLBG..__25n3' N., 80°10' W., Combat Sta. 455, 26 July 1957, (1) 39.0 mm. S.L., D.N., BLBG._..25°10' N., 80°02' W., Combat Sta. 438, 22 July 1957, (7) 16.5-29.5 mm. S.L., D.N., BLBG.___24°04' N., 79°15' W., Combat Sta. 448, 24 July 1957, (1) 13.5 mm. S.L., D.N., BLBG._._ Key West, Fla., (11) 39.0-58.0 mm. S.L., collected by D. S. Jordan, SU 2358 Pulaski Light, near Tortugas, Fla., (1) 100 mm. S.L., UF Tortugas, Fla., Apr. 1956, (2) 95.0-102 mm. S.L., CHML West of Loggerhead Key, Fla., (4) 75.5-87.0 mm. S.L., Ur.___Sanibel Island, Fla., 14 Aug. 1959, (10) 27.6-51.5 mm. S.L., BLBG.__. Sanibel Island, Fla., 19 Aug. 1959, (4) 47.5-58.0 mm. S.L., BLBG._.. Cedar Keys, Fla., Sept.-Oct. 1948, (18) 17.3- 69.5 mm. S.L., UF 730 Cedar Keys, Fla., 8-9 Oct. 1948, (1) 34.4 mm. S.L., UF Cedar Keys, Fla., 16 Aug. 1953, (18) 25.4-54.7 mm. S.L., UF C-8-1653-4.___ Cedar Keys, Fla., 6 Sept. 1953, (1) 41.6 mm. S.L., UF C-9-653-4 Cedar Keys, Fla., 20 Sept. 1953, (1) 64.0 mm. S.L., UF C-9-2053-3..-. Cedar Keys, Fla., 1 Nov. 1953, (3) 21.1-57.0 mm. S.L., UF C-ll-153-2.__. Cedar Keys, Fla., 13 Nov. 1953, (1) 57.5 mm. S.L., UF C-U- 1353-6.___ Cedar Keys, Fla., 30 June 1954, (1) 40.6 mm. S.L., UF C-6-3054-5.... Cedar Keys, Fla., (17) 30.9- 51.3 mm. S.L., UF....27°30' N., 84°14' W., Oregon Sta. 937, 18 Mar. 1954, (4) 92.5-103 mm. S.L., Tr. 38 fathoms, UF 3611.___22°13' N., 89°43' W., Silver Bay Sta. 404, 12 May 1958, (1) 92.5 mm. S.L., Tr. 25 fathoms, USNM... _ 22°23' N., 89°44' W., Oregon Sta. 2174, 11-12 May 1958, (30) 12.7-21.8 mm. S.L., D.N., USNM.__-24°50' N., 92°35' W., Oregon Sta. 2198, 23-24 June 1958, (1) 23.3 mm. S.L., D.N., BLBG Jamaica, (3) 66.0-85.0 mm. S.L., collected by J. S. Roberts, SU 4857 Bizoton, Haiti, 3 Feb. 1927, (1) 42.0 mm. S.L., Sn., USNM 178538 St. Thomas, Virgin Islands, (1) 45.0 mm. S.L., collected by Beebe, USNM 178812 Pointe-a-Pitre, Guadaloupe, May 1946, (1) 33.5 mm. S.L., USNM 132624 English Harbor, Antigua, Leeward Islands, (1) 61.0 mm. S.L., USNM 170300 Martinique, West Indies, 17 Apr. 1937, (1) 50.0 mm. S.L., collection of Smithsonian Institute Hartford Exped., USNM 117429 Union Island, Grenadines, (1) 19.5 mm. S.L., collected by Beebe, USNM 169962.___15°57' N., 82°06' W., Oregon Sta. 1935, 15 Sept. 1957, (1) 72.0 mm. S.L., USNM 185267. .-.07°55' N., 57°27' W., Oregon Sta. 2247, 31 Aug. 1958, (4) 70.5-77.5 mm. S.L., Tr. 44-37 fathoms, BLBG. ..-(23) 36.3-71.6 mm. S.L., SU 1958 No data, presumably from western North Atlantic, (2) 61.0 and 80.0 mm. S.L., BLBG. Monacanthus tuckeri Castle Roads, Bermuda, 11-12 Sept. 1931, (7) 25.0-43.0 mm. S.L., dredge 12-20 ft., USNM 178785.... 33°29' N., 76°40' W., Gill Cr. 3, Reg. 64, 11 Aug. 1953, (1) 21.4 mm. S.L., D.N., BLBG Nonsuch, Bermuda, 18 Oct. 1930, (3) 30.5-51.0 mm. S.L., USNM 178859.. _ .Nonsuch, Bermuda, (1) 55.5 mm. S.L., collected by Beebe, USNM 178862 Bermuda, (2) 36.0 and 43.0 mm. S.L., collected by J. M. Jones, USNM 21249._-.29°38' N., 80°09' W., Silver Bay Sta. 471, 17 June 1958, (1) 17.Q mm. S.L., M.L.N., BLBG Settlement Point, Grand Bahama Island, Gill Cr. 8, 29 Aug. 1954, (5) 17.4-23.0 mm. S.L., D.N., BLBG....26°45' N., 78''55' W., Gill Cr. 4, 3 Oct. 1953, 1600, (1) about 20 mm. S.L., S.C. of Thunnus atlanticus (Lesson), BLBG Hawks Bill Creek, Grand Bahama Island, Gill Cr. 4, 3 Oct. 1953, 1900-2100, (2) 17.9 and 18.7 mm. S.L., D.N., BLBG Hawks Bill Creek, Grand Bahama Island, Gill Cr. 4, 3 Oct. 1953, 2130-2330, (4) 16.7-20.3 mm. S.L., D.N., BLBG 26°30' N., 78°40' W., Gill Cr. 4, 3 Oct. 1953, 1730, (1) about 18 mm. S.L., S.C. of Thunnus atlanticus, BLBG 25°56.5' N., 77°54' W., Gill Cr. 7, 22 June 1954, (1) about 20 mm. S.L., S.C. of Katsuwonus pelamis (Linnaeus), BLBG....25°20' N., 77°15' W., Gill Cr. 6, 19 Apr. 1954, (1) 21.0 mm. S.L., S.C. of Coryphaena hippurus Linnaeus, BLBG Nassau Harbor, Bahamas, 20 Apr. 1958, (3) 15.3-17.2 mm. S.L., ANSP 84471 New Providence, Hog Island, Bahamas, 22 Mar. 1952, (2) 51.5 and 53.0 mm. S.L., ANSP 72669. ...Hog Island, Bahamas, 25 Apr. to 3 May 1957, (3) 16.0-24.0 mm. S.L., ANSP 84482 Periwinkle Rock, Rose Island, Bahamas, 4 Aug. 1955, (3) 49.0-56.5 mm. S.L., ANSP 84478 Rose Island, Bahamas, 31 July 1955, (13) 20.8-50.5 mm. S.L., ANSP 84481 Andros Island, Bahamas, (1) 18.7 mm. S.L., collected by P. Cloud, USNM 174977.. ..Antigua, Leeward Islands, 20 Apr. 1958, (1) 22.5 mm. S.L., dredge, USNM 183569 St. Croix, Virgin Islands, West Indies, (4) 20.3-23.3 mm. S.L., CAS 12403. Stephanolepis hispidus Georges Bank, Caryn, (19) 13.8-46.5 mm. S.L., collected by B. B. Leavitt, BLBG Newport, R.I., (1) HI mm. S.L., collected by S. Powell, USNM 21631. -..36°30' N., 74°33' W., Albatross, (2) 22.5 and 33.0 mm. S.L., USNM 38330.... 34°58' N., 75°54' W., Silver Bay Sta. 1258, 8 Sept. 1959, (1) 86.3 mm. S.L., Tr. 13-14 fathoms, BLBG....34°56' N., 75°56' W., Silver Bay Sta. 1257, 8 Sept. 1959, (3) 82.5-173 mm. S.L., Tr. 13-14 fathoms, BLBG.._.34°45' N., 75°38' W., Combat Sta. 386, 17 June 1957, (1) 137 mm. S.L., Tr. 40 fathoms, BLBG....34°44' N., 75°53' W., Silver Bay Sta. 1271, 12 Sept. 1959, (1) 59.6 mm. S.L., Tr. 17 fathoms, BLBG Beaufort Inlet, N.C., 17 July 1954, (1) 67.5 mm. S.L., UNC 2237.... Beaufort Inlet to Cape Lookout, N.C., Sept. 1956, (5) 79.0-93.5 mm. S.L., UNC 889 Newport River Narrows, Carteret County, N.C., 8 July 1955, (11) 58.0-74.5 mm. S.L., UNC 210... -River's Island, N.C., 8 July 1959, (64) 24.0-44.0 mm. S.L., Sn., BLBG River's Island, N.C., 8 July 1959, (6) 28.3-38.2 mm. S.L., Sn., BLBG River's Island, N.C., 10 July 1959, (135) 13.5-48.0 mm. S.L., Sn., BLBG Fiver's Island, N.C., 10 July 1959, (2) 43.9 and 47.1 mm. S.L., Sn., BLBG River's Island, N.C., 4 Aug. 1954, (1) 100 mm. S.L., UNC 2545 More- head City, N.C., 13 July 1959, (4) 37.0-46.5 mm. S.L., Sn., BLBG.-..Morehead City, N.C., 10 July 1959, (3) 30.0-40.0 mm. S.L., BLBG Bogue Sound, Carteret County, N.C., 29 July 1958, (2) 38.3 and 94.5 mm. S.L., UNC 2363.--.34°41' N., 76°50' W., Silver Bay Sta. 1288, 20 Sept. 1959, (1) 90.8 mm. S.L., Tr. 7-5 fathoms, BLBG...-34°39' N., 76°01' W., Silver Bay Sta. 1270, 12 Sept. 1959, (4) 104-108 mm. S.L., Tr. 20 fathoms, FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 105 BLBG....34°39' N., 76°27' W., Silver Bay Sta. 1262, 10 Sept. 1959, (1) 92.3 mm. S.L., Tr. 6-7 fathoms, BLBG 34°38' N., 76°49' W., Silver Bay Sta. 1291, 22 Sept. 1959, (5) 83.1-98.3 mm. S.L., Tr. 8-10 fathoms, BLBG.___34°38' \., 76°40' W., Silver Bay Sta. 1284, 20 Sept. 1959, (1) 87.0 mm. S.L., Tr. 6-8 fathoms, BLBG 34°38' N., 76° 40' W., Silver Bay Sta. 1284, 20 Sept. 1959, (3) 79.3-113 mm. S.L., Tr. 6-8 fathoms, BLBG..._ 34°38' X., 76° 49' W., Silver Bay Sta. 1240, 6 Sept. 1959, (3) 87.6-97.6 mm. S.L., Tr. 7 fathoms, BLBG 34°37' X., 77°01' W., Silver Bay Sta. 1311, 24 Sept. 1959, (2) 110 and 124 mm. S.L., Tr. 7 fathoms, BLBG..-.34°35' X., 75°15' W., Gill Or. 7, 13 July 1954, (1) 33.0 mm. S.L., S.C. of Eulhynnus allelteralus (Rafinesque), BLBG 34°35' X., 77°03' VV., Silver Bay Sta. 1310, 24 Sept. 1959, (1) 113 mm. S.L., Tr. 8 fathoms, BLBG 34°35' X., 76°23' W., Silver Bay Sta. 1250, 7 Sept. 1959, (1) 111 mm. S.L., Tr. 11-10 fathoms, BLBG..._34°34' X., 76°36' VV., Silver Bay Sta. 1315, 27 Sept. 1959, (2) 83.4 mm. and 105 mm. S.L., Tr. 8 fathoms, BLBG..._34°33' X., 77°06' W., Silver Bay Sta. 1309, 24 Sept. 1959, (1) 104 mm. S.L., Tr. 9 fathoms, BLBG.___34°32' N., 76°33' W., Silver Bay Sta. 1317, 27 Sept. 1959, (2) 77.6 and 86.6 mm. S.L., Tr. 8 fathoms, BLBG.___34°32' X., 76°49' W., Silver Bay Sta. 1293, 23 Sept. 1959, (1) 181 mm. S.L., Tr. 10-11 fathoms, BLBG 34°32' X., 76°49' W., Silver Bay Sta. 1293, 23 Sept. 1959, (9) 79.9-105 mm. S.L , Tr. 10-11 fathoms, BLBG 34°32' X., 75° 57' W., Silver Bay Sta. 1269, 12 Sept. 1959, (1) 94.5 mm. S.L., Tr. 25 fathoms, BLBG.._-34°32' X., 75°53' W., Silver Bay Sta. 1268, 11 Sept. 1959, (4) 53.4-172 mm. S.L., Tr. 31-30 fathoms, BLBG._-.34°31' X., 76°51' W., Silver Bay Sta. 1259, 10 Sept. 1959, (2) 84.0 and 89.2 mm. S.L., Tr. 11 fathoms, BLBG.___34°31' X., 76°53' W., Silver Bay Sta. 1239, 6 Sept. 1959, (4) 74.0-102 mm. S.L., Tr. 14-12 fathoms, BLBG.__.34°29' X., 76°49' W., Silver Bay Sta. 1305, 24 Sept. 1959, (4) 97.1-163 mm. S.L., Tr. 12 fathoms, BLBG....34°29' N., 76°57' W., Silver Bay Sta. 1307, 24 Sept. 1959, (6) 76.1-126 mm. S.L., Tr. 12 fathoms, BLBG..._34°27' N., 76°53' W., Silver Bay Sta. 1306, 24 Sept. 1959, (5) 98.5-172 mm. S.L., Tr. 12 fathoms, BLBG.-..34°26' X., 77°05' W., Silver Bay Sta. 1308, 24 Sept. 1959, (11) 89.1-128 mm. S.L.. Tr. 9-10 fathoms, BLBG 34°25' X., 76°5r W., Silver Bay Sta. 1294, 23 Sept. 1959, (4) 96.3-171 mm. S.L., Tr. 12-13 fathoms, BLBG 34°24' X., 76°46' W., Silver Bay Sta. 1238, 6 Sept. 1959, (3) 84.0-191 mm. S.L., Tr. 14-12 fathoms, BLBG..__34°23' X., 76°54' W., Silver Bay Sta. 1295, 23 Sept. 1959, (2) 179 and 189 mm. S.L., Tr. 13-15 fathoms, BLBG..__34°23' X., 76°54' \V., Silver Bay Sta. 1295, 23 Sept. 1959, (3) 75.2-96.7 mm. S.L., Tr. 13-15 fathoms, BLBG.__^.34°22' X., 77°09' W., Gill Or. 3, Reg. 68, 11 Aug. 1953, (48) 7.5-32.5 mm. S.L., D.N., BLBG 34°22' X., 75°38' W., Gill Cr. 8, Reg. 74, 30 Sept. 1954, (6) 12.0-23.5 mm. S.L., D.N., BLBG....34°22' X., 76°13' \V., Silver Bay Sta. 1248, 7 Sept. 1959, (2) 86.3 and 96.4 mm. S.L., Tr. 13-15 fathoms, BLBG._ .^34°22' X., 76°41' W., Silver Bay Sta. 1237, 6 Sept. 1959, (9) 73.6-93.3 mm. S.L., Tr. 14-15 fathoms, BLBG....34°2r X., 76°34' W., Silver Bay Sta. 1299, 23 Sept. 1959, (4) 49.0-61.6 mm. S.L., Tr. 14 fathoms, BLBG....34°2r X., 77°34' W., Silver Bay Sta. 1227, 4 Sept. 1959, (5) 93.7-188 mm. S.L., Tr. 7-8 fathoms, BLBG..__34°19' X., 77°19' W., Silver Bay Sta. 1228, 5 Sept. 1959, (4) 87.1-79.4 mm. S.L., Tr. 10 fathoms, BLBG....34°18' X., 76°32' W., Gill Cr. 3, Reg. 70, 12 Aug. 1953, (16) 17.0-49.0 mm. S.L., D.X., BLBG....34°16' X., 77°34' W., Silver Bay Sta. 1226, 4 Sept. 1959, (5) 79.0-157 mm. S.L., Tr. 8-7 fathoms, BLBG 34° 15' X., 77°07' W., Silver Bay Sta. 1229, 5 Sept. 1959, (17) 11.0-17.8 mm. S.L., M.L.X., BLBG.... 34° 14' X., 76°01' W., Silver Bay Sta. 1247, 7 Sept. 1959, (3) 68.5-97.3 mm. S.L., Tr. 33-24 fathoms, BLBG 34°13' X., 76°48' W., Silver Bay Sta. 1296, 23 Sept. 1959, (6) 109-168 mm. S.L., Tr. 17 fathoms, BLBG.._ .34°10' N., 77°30' W., Gill Cr. 8, Reg. 67, 28 Sept. 1954, (11) 10.0-21.5 mm. S.L., D.N., BLBG....34°10' N., 76°15' W., Silver Bay Sta. 1245, 7 Sept. 1959, (1) 91.2 mm. S.L., Tr. 22 fathoms, BLBG.__-34°09' N., 76°35' W., Silver Bay Sta. 1297, 23 Sept. 1959, (5) 94.7-211 mm. S.L., Tr. 20 fathoms, BLBG.___34°09' N., 76°55' W., Silver Bay Sta. 1230, 5 Sept. 1959, (1) 85.0 mm. S.L., Tr. 17 fathoms, BLBG.__-34°07' X., 76°32' W., Silver Bay Sta. 1298, 23 Sept. 1959, (2) 94.5 and 106 mm. S.L., Tr. 19 fathoms, BLBG.._-34°06' N., 77°46' W., Silver Bay Sta. 1213, 2 Sept. 1959, (5) 76.0-186 mm. S.L., Tr. 7-8 fathoms, BLBG.___34°05' X., 76°45' W., Silver Bay Sta. 1231. 5 Sept. 1959, (1) 87.6 mm. S.L., Tr. 20 fathoms, BLBG 34°04' N., 76°14' W., Gill Cr. 7, Reg. 71, 10 July 1954, (13) 12.5-34.0 mm. S.L., D.N., BLBG.__-34°03' N., 77°50' W., Silver Bay Sta. 1212, 2 Sept. 1959, (3) 88.4-180 mm. S.L., Tr. 5-6 fathoms, BLBG....34°02' N., 76°16' W., Gill Cr. 8, Reg. 71, 29 Sept. 1954, (6) 13.0-21.0 mm. S.L., D.N., BLBG.__.34°02' N., 77°35' W., Silver Bay Sta. 1214, 3 Sept. 1959, (7) 141-175 mm. S.L., Tr. 11 fathoms, BLBG.._.34°01' N., 76°37' W., Silver Bay Sta. 1316, 27 Sept. 1959, (1) 119 mm. S.L., Tr. 9 fathoms, BLBG._._33°57' X., 77°11' W., Gill Cr. 8, Reg. 66, 28 Sept. 1954, (3) 11.0-14.5 mm. S.L., D.N., BLBG.... 33°57' N., 77°01' W., Silver Bay Sta. 1222, 4 Sept. 1959, (4) 166-182 mm. S.L., Tr. 17-16 fathoms, BLBG.... 33°57' X., 77°01' W., Silver Bay Sta. 1222, 4 Sept. 1959, (1) 11.5 mm. S.L., M.L.N., BLBG._..33°57' N., 77°13' W., Gill Cr. 3, Reg. 66, 11 Aug. 1953, (42) 8.4-52.5 mm. S.L., D.N., BLBG....33°56' X., 77°20' W., Silver Bay Sta. 1215, 3 Sept. 1959, (18) 73.5-174 mm. S.L., Tr. 15 fathoms, BLBG._._33°55' N., 77°52' W., Silver Bay Sta. 1211, 2 Sept. 1959, (3) 75.1-146 mm. S.L., Tr. 5-6 fath- oms, BLBG.__,33°50' N., 75°59' W., Gill Cr. 7, Reg. 72, 10 July 1954, (1) 55.5 mm. S.L., D.N., BLBG..--33°50' N., 76°55' W., Silver Bay Sta. 1221, 4 Sept. 1959, (5) 117-128 mm. S.L., Tr. 20 fathoms, BLBG.. __33°49' N., 75°59' W., Gill Cr. 2, Reg. 72, 10 May 1953, (1) 15.0 mm. S.L., D.X., BLBG..__33°47' X., 77°50' W., Silver Bay Sta. 1210, 2 Sept. 1959, (8) 81.5-207 mm. S.L., Tr. 8 fathoms, BLBG.. -.33°44' N., 77°00' W., Gill Cr. 3, Reg. 65, 11 Aug. 1953, (1) 11.5 mm. S.L., D.N., BLBG.... 33°44' N., 76°56' W., Gill Cr. 8, Reg. 65, 28 Sept. 1954, (21) 12.0-26.0 mm. S.L., D.N., BLBG... .33°44' N., 76°58' W., Silver Bay Sta. 1217, 3 Sept. 1959, (9) 121-174 mm. S.L., Tr. 22-23 fathoms, BLBG....33°43' N., 76°56' W., Gill Cr. 4, Reg. 65, 8 Nov. 1953, (1) 19.0 mm. S.L., D.N., BLBG....33°41' N., 77°40' W., Silver Bay Sta. 106 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 1209, 2 Sept. 1959, (11) 79.7-182 mm. S.L., Tr. 11-12 fathoms, BLBG.--_33°29' N., 76°40' W., Gill Cr. 3, Reg. 64, 11 Aug. 1953, (107) 12.0-48.5 mm. S.L., D.N., BLBG.___33°29' N., 77°22' W., Silver Bay Sta. 1205, 1 Sept. 1959, (7) 79.5-127 mm. S.L., Tr. 16-20 fathoms, BLBG..._33°24' N., 76°25' W., Gill Cr. 3, Reg. 63, 11 Aug. 1953, (1) 11.5 mm. S.L., D.N., BLBG....33°22' N., 77°38' W., Gill Cr. 4, Reg. 59, 7 Nov. 1953, (1) 40.0 mm. S.L., D.N., BLBG._._33°21' N., 77°24' W., Silver Bay Sta. 1204, 1 Sept. 1954, (4) 83.7-91.2 mm. S.L., Tr. 15-16 fathoms, BLBG._._33°21' N., 77°24' W., Silver Bay Sta. 1204, 1 Sept. 1959, ^21) 81.2- 176 mm. S.L., Tr. 15-16 fathoms, BLBG.._-33°19' N., 77°34' W., Gill Cr. 3, Reg. 59 to Reg. 60, 10 Aug. 1953, (11) about 32 to 49 mm. S.L., S.C. of Coryphaena hippurus Linnaeus, BLBG 33°17' N., 78°38' W., Gill Cr. 8, Reg. 55, 26 Sept. 1954, (41) 9.0-48.0 mm. S.L., D.N., BLBG....33°15' N., 76°23' W., Gill Cr. 2, Reg. 63, 8 May 1953, (3) 11.0-12.5 mm. S.L., BLBG.-..33°03' N., 78°21' W., Gill Cr. 8, Reg. 54, 26 Sept. 1954, (121) 9.5-36.0 mm. S.L., D.N., BLBG.___ 33°03' N., 78°21' W., Gill Cr. 7, Reg. 54, 4 July 1954, (46) 8.0-45.0 mm. S.L., D.N., BLBG._._33°03' N., 77°09' W., Combat Sta. 289, 20 Apr. 1957, (1) 7.0 mm. S.L., D.N., BLBG.___32°58' N., 78°15' W., Gill Cr. 8, Reg. 53 to Reg. 54, 26 Sept. 1954, (1) about 10 mm. S.L., S.C. of Exithynnus alleUeralus, BLBG Off South Caro- lina, Combat, 18 Apr. 1957, (1) 18.0 mm. S.L., D.N., BLBG.-_.32°54' N., 79°16' W., Gill Cr. 4, Reg. 46, 25 Oct. 1953, (2) 42.0 and 52.5 mm. S.L., D.N., BLBG.... 32°54' N., 77°04' W., Gill Cr. 3, Reg. 61, 10 Aug. 1953, (17) 11.5-42.5 mm. S.L., D.N., BLBG..__32°54' N., 77°04' W., Gill Cr. 2, Reg. 61, 8 May 1953, (3) 10.0-12.5 mm. S.L., D.N., BLBG..,.32°48' N., 78°04' W., Gill Cr. 4, Reg. 53, 27 Oct. 1953, (2) 19.0 and 22.5 mm. S.L., D.N., BLBG.___32°48' N., 78°04' W., Gill Cr. 5, Reg. 53, 16 Feb. 1954, (15) 11.5-26.0 mm. S.L., D.N., BLBG._._ 32°43' N., 76°48' W., Gill Cr. 2, Reg. 62, 8 May 1953, (2) 10.0 and 11.0 mm. S.L., D.N., BLBG.___32°34' N., 77°48' W., Gill Cr. 4, Reg. 52, 26 Oct. 1953, (14) 10.0- 42.5 mm. S.L., D.N., BLBG.___32°34' N., 77°48' W., Gill Cr. 8, Reg. 52, 26 Sept. 1954, (16) 14.0-41.0 mm. S.L., D.N., BLBG.,_.32°27' N., 78°06' W., Combat Sta. 297, 21 Apr. 1957, (3) 14.5-17.5 mm. S.L., D.N., BLBG..,. 32°27' N., 78°06' W., Combat Sta. 297, 21 Apr. 1957, (3) 15.0-17.5 mm. S.L., D.N., BLBG._._32°26' N., 78°43' W., Gill Cr. 7, Reg. 48, 3 July 1954, (3) 6.0-11.5 mm. S.L., D.N., BLBG.___32°24' N., 78°44' W., Gill Cr. 3, Reg. 48, 6 Aug. 1953, (4) 16.5-43.5 mm. S.L., BLBG...- 32°24' N., 78°45' W., Gill Cr. 8, Reg. 48, 25 Sept. 1954, (17) 7.0-23.5 mm. S.L., D.N., BLBG.__.32°11' N., 78°27' W., Gill Cr. 5, Reg. 49, 15 Feb. 1954, (5) 16.0-33.0 mm. S.L., D.N., BLBG..._32°10' N., 78°28' W., Gill Cr. 7, Reg. 49, 4 July 1954, (1) 12.0 mm. S.L., D.N., BLBG.... 31°56' N., 78°10' W., Gill Cr. 8, Reg. 50, 26 Sept. 1954, (16) 14.0-41.0 mm. S.L., D.N., BLBG.__.31°41' N., 80°35' W., Gill Cr. 4, Reg. 36, 21 Oct. 1953, (2) 29.0 and 46.5 mm. S.L., P.T., BLBG....31°41' N., 80°35' W., Gill Cr. 4, Reg. 36, 21 Oct. 1953, (37) 16-48.5 mm. S.L., D.N., BLBG... -31=40' N., 80°20' W., Gill Cr. 8, Reg. 37, 21 Sept. 1954, (10) 24.0-40.0 mm S.L., D.N., BLBG.--. 31°38' N., 80°14' W., Gill Cr. 4, Reg. 37, 22 Oct. 1953, (336) 10.0-37.5 mm. S.L., D.N., BLBG..._31°38' N., 80° 14' W., Gill Cr. 3, Reg. 37, 5 Aug. 1953, (2) 23.5 and 34.0 mm. S.L., D.N., BLBG.-.-31°38' N., 80°15' W., Gill Cr. 7, Reg. 37, 2 July 1954, (28) 8.0-50.0 mm. S. L., D.N., BLBG.-.-31°36' N., 79°51' W., Gill Cr. 8, Reg. 38, 21 Sept. 1954, (7) 17.5-29.0 mm. S.L., D.N., BLBG.... 31°36' N., 79°52' W., Gill Cr. 7, Reg. 38, 2 July 1954, (7) 6.5-21.5 mm. S.L., D.N., BLBG..-.31°34' N., 79°28' W., Gill Cr. 4, Reg. 39, 24 Oct. 1953, (1) 20.5 mm. S.L., M.L.N., BLBG....31°34' N., 79°28' W., Gill Cr. 4, Reg. 39, 24 Oct. 1953, (1) 15.0 mm. S.L., D.N., BLBG.... 31°33' N., 79°27' W., Gill Cr. 8, Reg. 39, 21 Sept. 1954, (29) 9.0-20.0 mm. S.L., D.N., BLBG.--.31°32' N., 79°28' W., Gill Cr. 3, Reg. 39, 5 Aug. 1953, (4) 13.5-24.0 mm. S.L., D.N., BLBG....31°21' N., 80°52' W., Gill Cr. 4, Reg. 35, 21 Oct. 1953, (28) 21.5-54.0 mm. S.L., M.L.N., BLBG.-..31°20' N., 80°53' W., Gill Cr. 8, Reg. 35, 20 Sept. 1954, (2) 21.5 and 27.5 mm. S.L., D.N., BLBG..-- St. Simons Island, Ga., 5 Oct. 1955, (280) 20.5-43.0 mm. S.L., Sn., BLBG.- --St. Simons Island, Ga., 14 Aug. 1955, (12) 9.0-23.0 mm. S.L., Sn., BLBG St. Simons Island, Ga., 26 Dec. 1957, (1) 19.0 mm. S.L., Sn., BLBG.-.. St. Simons Island, Ga., 11 Oct. 1957, (1) 22.5 mm. S.L., Sn., BLBG... -St. Simons Island, Ga., 26 Sept. 1957, (2) 20.5 and 22.0 mm. S.L., Sn., BLBG St. Simons Island, Ga., 27 Aug. 1957, (1) 18.5 mm. S.L., Sn., BLBG.. -.St. Simons Island, Ga., 26 July 1957, (26) 16.5-28.0 mm. S.L., Sn., BLBG.-. -St. Simons Island, Ga., 15 July 1957, (1) 15.9 mm. S.L., Sn., BLBG St. Simons Island, Ga., 16 May 1957, (15) 20.0-.33.0 mm. S.L., Sn., BLBG St. Simons Island, Ga., 2 May 1957, (2) 15.0 and 25.0 mm. S.L., Sn., BLBG St. Simons Island, Ga., 27 Apr. 1957, (3) 13.5-47.0 mm. S.L., Sn., BLBG St. Simons Island, Ga., 16 Apr. 1957, (2) 18.5 and 22.5 mm. S.L., Sn., BLBG.. --St. Simons Island, Ga., 18 Mar. 1957, (1) 27.5 mm. S.L., Sn., BLBG St. Simons Island, Ga., 20 June 1958, (2) 15.5 and 16.0 mm. S.L., Sn., BLBG Jekyll Island Causeway, Ga., 26 Sept. 1957, (3) 17.0-21.5 mm. S.L., Sn., BLBG.- --Jekyll Island, Ga., 6 Aug. 1959, (1) 57.5 mm. S.L., BLBG.- --Jekyll Island, Ga., 18 Feb. 1959, (1) 34.5 mm. S.L., Sn., BLBG Commercial Trawling Area, Brunswick, Ga., 20 Oct. 1955, (1) 59.0 mm. S.L., Tr., BLBG.--.31°00' N., 80°23' W., Gill Cr. 4, Reg. 32, 16 Oct. 1953, (441) 12.5-52.0 mm. S.L., D.N., BLBG.---31°00' N., 80°46' W., Gill Cr. 4, Reg. 33, 16 Oct. 1953, (72) 9.5-46.0 mm. S.L., D.N., BLBG..-. 31°00' N., 80°46' W., Gill Cr. 8, Reg. 33, 15 Sept. 1954, (6) 20.5-32.0 mm. S.L., D.N., BLBG.---31°00' N., 80°23' W., Gill Cr. 7, Reg. 32, 27 June 1954, (36) 9.5- 26.0 mm. S.L., D.N., BLBG.--.31°00' N., 80°46' W., Gill Cr. 7, Reg. 33, 27 June 1954, (6) 30.0-40.5 mm. S.L., D.N., BLBG..--31°00' N., 80°00' W., Gill Cr. 7, Reg. 31, 27 June 1954, (10) 8.5-17.0 mm. S.L., D.N., BLBG.... 31°00' N., 81°08' W., Gill Cr. 4, Reg. 34, 17 Oct. 195.3, (6) 19.O-29.0 mm. S.L., D.N., BLBG.---30°59' N., 79°14' W., to 31°00' N., 79°36' W., GUI Cr. 8, Reg. 29 to Reg. 30, 15 Sept. 1954, (5) 14.0-27.5 mm. S.L., D.N., BLBG.-..30°58' N., 79°38' W., Gill Cr. 7, Reg. 30, 27 June 1954, (2) 13.0 and 40.0 mm. S.L., D.N., BLBG..-. 30°20' N., 79°26' W., Gill Cr. 4, Reg. 28, 16 Oct. 1953, (2) 19.0 and 22.5 mm. S.L., D.N., BLBG.-.. 30° 20' N., FILEFISHES (MONACANTHIDAE) OF THE WESTERN NORTH ATLANTIC 107 80°35' W., Gill Cr. 3, Reg. 25, 28 July 1953, (1) 11.5 mm. S.L., D.N., BLBG._..30°20' N., 80°58' W., Gill Cr. 7, Reg. 24, 26 June 1954, (8) 14.0-32.0 mm. S.L., D.N., BLBG.__-30°20' N., 80°36' W., Gill Cr. 7, Reg. 25, 26 June 1954, (74) 8.5-22.5 mm. S.L., D.N., BLBG...- 30°00' X., 80°10' W., Silver Bay Sta. 476, 18 June 1958, (87) 5.4-38.5 mm. S.L., D.N., BLBG.-_.29°48' N., 80°12' W., Silver Bay Sta. 470, 17 June 1958, (116) 6.2- 21.2 mm. S.L., D.N., BLBG..__29°40' N., 80°23' W., Gill Cr. 4, Reg. 19, 14 Oct. 1953, (1) 25.0 mm. S.L., D.N., BLBG._..29°40' N., 81°08' W., Gill Cr. 8, Reg. 21, 14 Sept. 1954, (6) 15.5-34.0 mm. S.L., D.N., BLBG.._. 29°40' N., 80°45' W., Gill Cr. 8, Reg. 20, 14 Sept. 1954, (46) 7.5-15.5 mm. S.L., D.N., BLBG..._29°38' N., 80°12' W., Combat Sta. 474, 14 Aug. 1957, (4) 10.0-51.0 mm. S.L., D.N., BLBG.^_.29°38' N., 80°09' W., Silver Bay Sta. 471, 17 June 1958, (59) 5.6-33.8 mm. S.L., D.N., BLBG....29°31' N., 80°31' W., Combat Sta. 347, 2 June 1957, (4) 124-155 mm. S.L., Tr. 18 fathoms, BLBG.._. 29°29' N., 80°10' W., Combat Sta. 490, 19 Aug. 1957, (2) 16.5 and 34.0 mm. S.L., D.N., BLBG._-.29°29' N., 80°09' W., Combat Sta. 485, 18 Aug. 1957, (5) 15.5-38.0 mm. S.L., D.N., BLBG._.-29°26' N., 80°08' W., Combat Sta. 315, 27 Apr. 1957, (2) 13.0 and 21.0 mm. S.L., D.N., BLBG....29°22' N., 80°05' W., Silver Bay Sta. 227, 24 Nov. 1957, (3) 11.5-15.0 mm. S.L., D.N., BLBG._._ 29°20' N., 80°04' W., Combat Sta. 316, 27 Apr. 1957, (1) 8.5 mm. S.L., D.N., BLBG.._.29°10' N., 80=19' W., Combat Sta. 336, 1 June 1957, (1) 122 mm. S.L., Tr. 25 fathoms, BLBG.._-29°10' N., 80°19' W., to 29°19' N., 80°15' W., Combat Sta. 336-337, 1 June 1957, (62) 10.6- 28.7 mm. S.L., D.N., BLBG.-..29°07' N., 80°25' W., Gill Cr. 8, 28 Aug. 1954, (1) 22.0 mm. S.L., S.C. of Euihynnus alletteratus, BLBG._ -_29°00' N., 79°48' W., Gill Cr. 4, Reg. 15, 14 Oct. 1953, (1) 16.5 mm. S.L., D.N., BLBG._._ 29°00' N., 80°32' W., Gill Cr. 4, Reg. 13, 14 Oct. 1953, (1) 31.0 mm. S.L., D.N., BLBG.._-29°00' N., 80°10' W., Gill Cr. 4, Reg. 14, 14 Oct. 1953, (32) 17.0-34.0 mm. S.L., D.N., BLBG....29°00' N., 79°26' W., Gill Cr. 4, Reg. 16, 14 Oct. 1953, (23) 13.5-27.0 mm. S.L., D.N., BLBG._..29°00' N., 80°32' W., Gill Cr. 8, Reg. 13, 12 Sept. 1954, (6) 22.5-30.0 mm. S.L., D.N., BLBG.___Port Canaveral Anchorage, Fla., Silver Bay, 22-23 Nov. 1957, (1) 10.5 mm. S.L., D.N., BLBG..-.Port Canaveral Anchorage, Fla., Combat, (1) 20.0 mm. S.L., D.N., BLBG.__-27°40' N., 80°04' W., Gill Cr. 4, Reg. 5, 12 Oct. 1953, (3) 15.0-39.5 mm. S.L., M.L.N., BLBG.-__ 27°20'N.,80°02'W., (7!7/Cr.4, Reg.4, 12 Oct. 1953,(3) 14.0- 47.0 mm. S.L., D.N., BLBG.___27°14' N., 79°50' W., Combat Sta. 462, 29 July 1957, (10) 12.2-14.4 mm. S.L., D.N., BLBG._._26°58' N., 79°40' W., Gill Cr. 4, Reg. 2, 12 Oct. 1953, (9) 13.2-39.2 mm. S.L., D.N., BLBG.__. 26=47' N., 79°53' W., Combat Sta. 459, 28 July 1957, (9) 11.5-61.0mm. S.L., D.N., BLBG.___26°37' N., 79°51' W., Combat Sta. 458, 28 July 1957, (1) 12.0 mm. S.L., D.N., BLBG._..25''16' N., 80°07' W., Combat Sta. 457, 26 July 1957, (12) 10.0-40.5 mm. S.L., D.N., BLBG._ .25°13' X., 80=10' W., Combat Sta. 455, 26 July 1957, (16) 13.0- 51.1 mm. S.L., D.N., BLBG., . ,25=11' N., 79=56' W., Combat Sta. 443, 22 July 1957, (1) 49.5 mm. S.L., D. N., BLBG.._, 25=11' N., 79=56' W., Combat Sta. 443, 22 July 1957, (1) 49.0 mm. S.L., D.N., BLBG.... 25=10' N., 80=02' W., Combat Sta. 438, 22 July 1957, (70) 8.fr-56.4 mm. S.L., D.N., BLBG East coast of Florida, Combat Sta., (5) 111-143 mm. S.L., Tr., BLBG Gulf Stream, (1) 100 mm. S.L., CAS 12824.. ..24=13' N., 81=42' W., Combat Sta. 436, 21 July 1957, (9) 18.0-42.0 mm. S.L., D.N., BLBG... -Key West, Fla., (1) 50.5 mm. S.L., collected by D. S. Jordan, SU 2358 Loggerhead Key, Fla., (2) 81.0 and 97.5 mm. S.L., Tr., UF....Sanibel Island, Fla., 19 Aug. 1959, (2) 36.0 and 42.5 mm. S.L., Sn., BLBG...-Sambel Island, Fla., 17 Aug. 1959, (1) 40.5 mm. S.L., Sn., BLBG Sanibel Island, Fla., 14 Aug. 1959, (26) 20.5-50.0 mm. S.L., Sn., BLBG... .Sanibel Island, Fla., 14 Aug. 1959, (12) 18.6-49.3 mm. S.L., Sn., BLBG....Gasparilla Bay, Fla., 17 Jan. 1958, (1) 145 mm. S.L., CHML..__Placida, Fla., 1 Jan. 1955, (1) 158 mm. S.L., CHML....Englewood, Fla., Mar. 1958, (1) 167 mm. S.L., CHML... .Lemon Bay, Fla., 28 Sept. 1955, (1) 143 mm. S.L., CHML.. ..Tarpon Springs, Fla., Mar. 1930, (1) 151 mm. S.L., UF 4192 Cedar Keys, Fla., 19 June 1949, (7) 32.8-66.6 mm. S.L., UF... .Cedar Keys, Fla., 19 June 1949, (5) 52.0-66.5 mm. S.L., UF..._ Cedar Keys, Fla., 18 Oct. 1953, (2) 22.7 and 27.0 mm. S.L., UF C-10- 1853-4.... Cedar Keys, Fla., 20 Sept. 1953, (2) 96.0 and 108 mm. S.L., UF 2510.... Cedar Keys, Fla., 6 Sept. 1953, (1) 102 mm. S.L., UF C-9-653-2 Cedar Keys, Fla., 6 Sept. 1953, (1) 66.1 mm. S.L., UF C-9-653-4.... Cedar Keys, Fla., 16 Aug. 1953, (2) 61.5 and 83.0 mm. S.L., UF C-8-1653-2.... Cedar Keys, Fla., 16 Aug. 1953, (7) 63.6-88.6 mm. S.L., UF C-8-1653-4 Cedar Keys, Fla., 16 Aug. 1953, (6) 55.5-74.1 mm. S.L., UF C-8-1653-5.-_. Cedar Keys, Fla., 16 Aug. 1953, (6) 56.4-74.9 mm. S.L., UF C-8-1653-5 Cedar Keys, Fla., 16 Aug. 1953, (38) 42.3-90.2 mm. S.L., UF C-8- 1653-4.... Cedar Keys, Fla., 25 July 1953, (3) 55.7-82.5 mm. S.L., UF C-7-2553-2.... Cedar Keys, Fla., 12 July 1953, (6) 61.1-89,6 mm. S.L., UF C-7-1253-4.... Cedar Keys, Fla., 12 July 1953, (1) 81.6 mm. S.L., UF C-7- 1253-2.... Cedar Keys, Fla., 12 July 1953, (1) 44.8 mm. S.L., UF C-7-1253-1.... Cedar Keys, Fla., 12 July 1953, (18) 39.7-91.1 mm. S.L., UF C-7-1253-4.... Cedar Keys, Fla., 27 May 1953, (1) 53.5 mm. S.L., UF C-5-2753-3...- Cedar Keys, Fla., 27 May 1953, (1) 80.8 mm. S.L., UF C-5-2753-2.... Cedar Keys, Fla., 27 May 1953, (8) 34.3-59.9 mm, S.L., UF C-5-2753-3.... Cedar Keys, Fla., 30 June 1954, (1) 73.1 mm. S.L., UF C-6-3054-1 Cedar Keys, Fla., 23 Nov. 1957, (7) 24.0-29.0 mm. S.L., UF....Fort Walton, Fla., Feb.-Aug. 1959, (5) 38.0-41.0 mm. S.L., BLBG.. ..28=44' N., 88=08' W., Oregon Sta. 1583, 20-21 July 1956, (60) 14.9-41.3 mm. S.L., D.N., BLBG Aransas Anchorage, Tex., 7 June 1954, (1) 27.9 mm. S.L., UF 5-10 miles north of San Fernando River, Mexico, 22 Mar. 1947, (1) 53.0 mm. S.L., collected by W. W. Anderson, USNM 155576.... 24=50' N., 92=35' W., Oregon Sta. 2198, 23-24 June 1958, (6) 18.0-57.6 mm. S.L., D.N., BLBG. ...24=05' N., 91=46' W., Oregon Sta. 2196, 22 June 1958, (6) 39.5-48.0 mm, S.L., D.N., BLBG... 22=13' N., 89=43' W., Silver Bay Sta. 404, 12 May 1958, (1) 125 mm. S.L., Tr. 25 fathoms, USNM.... 24=26' N., 81=48' 15" W., Albatross Sta. 2315, 15 Jan. 1885, (1) 82.5 mm. S.L., USNM 143091. ...07=55' N., 108 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 57027' w., Oregon Sta. 2247, 31 Aug. 1958, (1) 69.5 mm. S.L., BLBG Brazil, Albatross, (1) 62.0 mm. S.L., USNM 43319._--No data, (1) 83.1 mm. S.L., CAS 3437. Stephanolepis setifer 34°38' N., 74°46' W., Gill Cr. 2, Reg. 80, 12 May 1953, (1) 39.6 mm. S.L., D.N., BLBG..._33°49' N., 75°59' W., Gill Cr. 2, Reg. 72, 10 May 1953, (1) 18.5 mm. S.L., D.N., BLBG.___33°29' N., 76°40' W., Gill Cr. 3, Reg. 64, 11 Aug. 1953, (1) 12.5 mm. S.L., D.N., BLBG.___33°15' N., 76°23' W., Gill Cr. 2, Reg. 63, 8 May 1953, (1) 17.2 mm. S.L., D.N., BLBG.___33°13' N., 76°55' W., Combat Sta. 290, 20 Apr. 1957, (1) 30.0 mm. S.L., D.N., BLBG..,. 32°40' N., 77°40' W., Combat Sta. 296, 21 Apr. 1957, (1) 43.2 mm. S.L., D.N., BLBG._..32°24' N., 78°45' W., Gill Cr. 8, Reg. 48, 25 Sept. 1954, (1) 20.0 mm. S.L., D.N., BLBG Bermuda, (1) 92.5 mm. S.L., collected by Beebe, USNM 178860.. -_31°42' N., 79°00' W., Gill Cr. 6, Reg. 41, 14 Feb. 1954, (1) 17.3 mm. S.L., D.N., BLBG.... 31°29' N., 78°41' W., Gill Cr. 2, Reg. 40, 5 May 1953,(1) 24.5 mm. S.L., D.N., BLBG...30°20' N., 79°50' W., Gill Cr. 8, Reg. 27, 14 Sept. 1954, (1) 20.5 mm. S.L., D.N., BLBG.... 30° 18' N., 80° 12' W., Gill Cr. 3, Reg. 26, 29 July 1953, (1) 27.4 mm. S.L., D.N., BLBG..--29°48' N., 80°12' W., Silver Bay Sta. 470, 17 June 1958, (1) 34.9 mm. S.L., D.N., BLBG....29°38' N., 80°12' W., Combat Sta. 474, 14 Aug. 1957, (18) 23.0-39.6 mm. S.L., D.N., BLBG....29°28' N., 80°09' W., Combat Sta. 326, 30 May 1957, (1) 32.0 mm. S.L., hooked through eye on trolling rig, BLBG..-.29°28' N., 80°09' W., Combat Sta. 326, 30 May 1957, (3) 34.7-38.2 mm. S.L., D.N., BLBG....29°19' N., 80°18' W., Combat Sta. 343, 1 June 1957, (1) 41.5 mm. S.L., D.N., BLBG....29°19' N., 80°18' W., Combat Sta. 339, 1 June 1957, (1) 37.5 mm. S.L., D.N., BLBG....29°16' N., 80°04' W., Combat Sta. 328, 30 May 1957, (1) 49.5 mm. S.L., D.N., BLBG.... 29°10' N., 80°19' W., Combat Sta. 336, 1 June 1957, (1) 39.0 mm. S.L., D.N., BLBG....29°00' N., 79°26' W., Gill Cr. 4, Reg. 16, 14 Oct. 1953, (1) 29.0 mm. S.L., D.N., BLBG.. -.27° 14' N., 79°50' W., Combat Sta. 462, 29 July 1957, (6) 13.6-36.8 mm. S.L., D.N., BLBG.... 27°00' N., 79°18' W., Gill Cr. 3, Reg. 1, 25 July 1953, (1) 51.5 mm. S.L., D.N., BLBG....26°58' N., 79°40' W., Gill Cr. 4, Reg. 2, 12 Oct. 1953, (5) 17.0-26.5 mm. S.L., D.N., BLBG....26°47' N., 79°53' W., Combat Sta. 459, 28 July 1957, (14) 27.5-58.0 mm. S.L., D.N., BLBG.... 25°10' N., 80°02' W., Combat Sta. 438, 22 July 1957, (2) 40.5 and 56.5 mm. S.L., D.N., BLBG....25°16' N., 80°07' W., Combat Sta. 457, 26 July 1957, (3) 47.5-53.5 mm. S.L., D.N., BLBG....25°13' N., 80°10' W., Combat Sta. 455, 26 July 1957, (1) 39.0 mm. S.L., Tr. 40-50 fathoms, BLBG....25°13' N., 80°10' W., Combat Sta. 455, 26 July 1957, (1) 35.5 mm. S.L., D.N., BLBG.... 28°42' N., 86°36' W., Albatross, (1) 38.5 mm. S.L., USNM 84566.... 24°13' N., 81°42' W., Combat Sta. 436, 21 July 1957, (4) 36.0-54.0 mm. S.L., D.N., BLBG.... 24°05' N., 91°46' W., Oregon Sta. 2196, 22 June 1958, (6) 39.0-49.5 mm. S.L., D.N., BLBG....24°50' N., 92°35' W., Oregon Sta. 2198, 23-24 June 1958, (4) 39.0- 52.5 mm. S.L., D.N., BLBG.... Cuba, (1) 108 mm. S.L., collected by Poey, USNM 9841 Punta Colorado, Cuba, 21 May 1914, (1) 39.0 mm. S.L., Sn., USNM 82562.... St. Lucia, Windward Islands to Cayo Hutio, Cuba, 2 May 1914, (1) 82.5 mm. S.L., Tr. 2-4 fathoms, USNM 82558 Palisadoes, Jamaica, 19 June 1957, (2) 15.0 and 16.5 mm. S.L., UF C-6-1957-1 J.... Jamaica, B.W.I., (10) 62.5-101 mm. S.L., collected by J. S. Roberts, SU 4772 Jamaica, (2) 94.0 and 136 mm. S.L., collected by C. B. Adams, USNM 6066... _ Bizoton Wharf, Haiti, (6) 29.7-49.5 mm. S.L., collected by Beebe, USNM 178066 Haiti, (10) 61.5-106 mm. S.L., collected by Beebe, USNM 178126 Haiti, (1) 50.5 mm. S.L., collected by Beebe, USNM 17861.. ..Fox Bay, Colon, Atlantic Panama, (1) 68.5 mm. S.L., collected by Meek and Hildebrand, USNM 81510 Fox Bay, Colon, Atlantic Panama, (1) 71.5 mm. S.L., collected by Meek and Hildebrand, USNM 81509 16°22' N., 83°31' W., Oregon Sta. 1863, 20 Aug. 1957, (13) 11.3-21.6 mm. S.L., D.N., BLBG....15°57' N., 82°06' W., Oregon Sta. 1935, 15 Sept. 1957, (2) 76.5 and 85.0 mm. S.L., USNM 185267. Atnanses pullus Vineyard Sound, Mass., 3 Sept. 1914, (1) 83.0 mm. S.L., D.N., USNM 85772.... 34°14' N., 76°03' W., Silver Bay, 15 Sept. 1959, (3) 41.2-57.7 mm. S.L., D.N., BLBG.... 31°57' N., 78°09' W., Gill Cr. 3, Reg. 50, 6 Aug. 1953, (2) 42.0 and 46.0 mm. S.L., D.N., BLBG...-30°26' N., 78°20' W., Gill Cr. 5, 21 Jan. 1954, (1) about 42 mm. S.L., S.C. of Coryphaena hippunis Linnaeus, BLBG 30°16' N., 80°21' W., Combat Sta. 70, 31 Aug. 1956, (1) 66.5 mm. S.L., Tr. 22 fathoms, TU 14749.... 30°00' N., 80°10' W., Silver Bay Sta. 476, 18 June 1958, (1) 50.0 mm., M.L.N., BLBG....29°41' N., 80°18' W., Gill Cr. 3, 28 July 1953, (1) 55.5 mm. S.L., S.C. of Coryphaena hippurus, BLBG 29°38' N., 80°12' W., Combat Sta. 474, 14 Aug. 1957, (8) 45.5-92.0 mm. S.L., D.N., BLBG....29°29' N., 80°10' W., Combat Sta. 490, 19 Aug. 1957, (2) 54.0 and 55.5 mm. S.L., D.N., BLBG.-..28°18' N., 79°28' W., Gill Cr. 8, Reg. 8, 12 Sept. 1954, (1) 43.0 mm. S.L., D.N., BLBG.... 28°05' N., 78°24' W., Silver Bay Sta. 446, 10 June 1958, (1) 35.0 mm. S.L., D.N., BLBG....28°00' N., 78°00' W., Gill Cr. 3, Spc. 8, 18 July 1953, (1) 69.5 mm. S.L., D.N., BLBG....27°41' N., 79°40' W., Gill Cr. 8, Reg. 6, 12 Sept. 1954, (1) 42.5 mm. S.L., D.N., BLBG....27°40' N., 79°18' W., Gill Cr. 8, Reg. 7, 12'Sept. 1954, (1) 46.0 mm. S.L., D.N., BLBG....27°02' N., 79°23' W., Gill Cr. 3, 25 July 1953, (1) about 75 mm. S.L., S.C. of Coryphaena hippurus, BLBG..--27°00' N., 79°18' W., Gill Cr. 6, Reg. 1, 25 Apr. 1954, (1) 66.0 mm. S.L., D.N., BLBG.. .-Palm Beach Inlet, Fla., 11 June 1958, (H 158 mm. S.L., UF 7266, RC-6-1 158-2.... Entrance Point, North Bimini, B.W.I., 14 July 1957, (1) 72.0 mm S.L., TU 17801 Settlement Point, Grand Bahama Island, Gill Cr. 8, 29 Aug. 1954, (3) 42.0-49.0 mm. S.L.. D.N., BLBG Nassau, Bahamas, Albatross, (1) 13(i mm. S.L., USNM 38375.. . .26°47' N., 79°53' W., Combnl Sta. 459, 28 July 1957, (1) 72.0 mm. S.L., D.N., BLBG.... 29°29' N., 80°69' W., Combat Sta. 485, 18 Aug. 1957, (3) 50.5-64.5 mm. S.L., D.N., BLBG..--26°11' N., 78°15' W., Gill Cr. 7, 22 June 1954, 1730, (1) 33.0 mm. S.L., S.C. of Katsuwonus pelamis (Linnaeus), BLBG 26°21.2' N., 76°46.5' W., Gill Cr. 3, 23 July 1953, (1) FILEFISHES ( MONACANTHIDAE ) OF THE WESTERN NORTH ATLANTIC 109 :iliout 51 mm. S.L., S.C. of Coryphaena hippurus, liLBG._--26°10' N., 78°13' W., Gill Cr. 7, 22 June 1954, 1722, (1) 44.0 mm. S.L., S.C. of Katsuwonus pelamis liLBG._--26°04' N., 78°08' W., Gill Cr. 9, 15 Nov. 1954, :i) 17.5 to about 35 mm. S.L., S.C. of Katsuwonus pelamis, BLBG.-..26°27' N., 76°44' W., Gill Cr. 7, Std., 13-14 June 1954, (1) 46.5 mm. S.L., D.N., BLBC. ..25°20' N., 77=15' W., Gill Cr. 6, 19 Apr. 1954, (1) about 38 mm. S.L., S.C. of Coryphaena hippurus, BLBG 25°16' N., 80°07' W., Combat Sta. 457, 26 July 1957, (1) 51.0 mm. S.L., D.N., BLBG.._-24°13' N., 81°42' W., Combat Sta. 436, 21 July 1957, (2) 41.5 and 43.5 mm. S.L., D.N., BLBG.___23°40.5' N., 76°50' W., Gill Cr. 7, 19 June 1954, (2) 33.0 and 36.0 mm. S.L., S.C. of Corphaena hippurus, BLBG Tortugas, Fla., (1) 105 mm. S.L., collected by W. H. Longley, USNM 116997... -Fort Myers, Fla., 5 Sept. 1956, (1) 115 mm. S.L., spit up by a grouper, CHML.-__29°26' N., 87°32' W., Oregon Sta. 792, 8 June 1953, (1) 89.0 mm. S.L., Tr. 56-57 fathoms, TU 6064.___26°10' N., 96°25' W., Oregon Sta. 1089, 4 June 1954, (1) 92.5 mm. S.L., Tr. 40 fathoms, TU 10845.... 26°40' N., 92°00' W., Oregon Sta. 1035, 8 May 1954, (14) 61.0-83.0 mm. S.L., Tr. 890 fathoms, TU 10933.. .. 24°05' N., 91°46' W., Oregon Sta. 2196, 22 June 1958, (1) 36.5 mm. S.L., D.N., BLBG._--North of Cuba, (3) 44.0-49.0 mm. S.L., collected by Beebe, USNM 178016.-.. Cabanas Bay, Cuba, 8-9 June 1914, (1; 46.0 mm. S.L., USNM 82566.. ..Cayo Hutia Light, Cuba, 12 May 1914, (1) 85.0 mm. S.L., USNM 82557. ...Cuba, (1) 322 mm. S.L., USNM 32096. ...Cuba, (1) 138 mm. S.L., collected by Poey, USNM 9852.... Jamaica, (2) 127 and 131 mm. S.L., collected by J. S. Roberts, SU 4943.. ..Ocho Rios, Jamaica, 22 June 1957, (1) 78.5 mm. S.L., UF C-6-2257- IJ.... Eaton Hall Cove, Jamaica, 14 June 1958, (1) 63.0 mm. S.L., UF C-6-1458-1J.-.. Jamaica, (1) 325 mm. S.L., collection of Institute of Jamaica, USNM 37694 Jamaica, (1) 288 mm. S.L., collection of Institute of Jamaica, USNM 37693 Port-au-Prince market, Haiti, 19 Dec. 1944, (1) 100 mm. S.L., USNM 132120.. ..Port- au-Prince, Haiti, (5; 87.0-118 mm. S.L., collected by A. Curtis, USNM 133749 Haiti, (1) 128 mm. S.L., collected by Beebe, USNM 178119 Arroyo, Porto Rico, 1899, (1) 138 mm. S.L., collection of USFC, SU 8266 Porto Rico, (1) 142 mm. S.L., collected by C. F. Cole, USNM 162780... -Porto Rico, Fish Hawk, (1) 136 mm. S.L., USNM 126425 Barbuda Island, Leeward Islands, (1) 49.0 mm. S.L., collected by Beebe, USNM 183446 Windward Island, Castries, (2) 107 and 113 mm. S.L., collected by Beebe, USNM 178598... -Port of Fortaleza, Mucuripe, Brazil, Aug. 1945, (1) 182 mm. S.L., SU 52304 Bahia, Brazil, Albatross, (1) 124 mm. S.L., USNM 43323. --.Port of Recife, Brazil, (2) 103 and 123 mm. S.L., SU 52305. ADDENDUM During the course of this study, particular attention was directed to the fact that species of the genus Stephanolepis were not known to occur in tlie Bahama Islands — ^despite the occurrence of both Stephanolepis selifer and S. hispidus in the currents of the Florida Current passing the west side of the Bahamas. After the manuscript was in press, 6 specimens of Stephanolepis were received from James E. Bohlke, Academy of Natural Sciences of Philadelphia, who had recently determined the generic identity of these specimens from the Bahaman collections of the Chaplin Bahaman Shore Fish Program: (1) 113- and 115-mm. males and a 92.5-mm. female from Chaplin Program Station 526, Hatchet Bay, Eleu- thera Island, Bahamas, 3 miles offshore, 30 ft., various stations of the George M. Bowers, April 20 to May 3, 1960. (2) 106- and 113-mm. males from Chaplin Program Station 513B, Hatchet Bay, Eleuthera Island, Bahamas, 30 to 35 ft., collected by the George M. Bowers, Febru- ary 6-13, 1960. (3) 32.5-mm. immature specimen, ANSP 72575, New Providence, Bahamas, collected by C. C. G. Chaplin, 1949. We would expect all these specimens to be Stephanolepis setifer, rather than S. hispidus, because our records indi" cate that S. selifer is a more offshore and insular inhabitant, while we have recorded S. hispidus only from continental waters; but there are inconsistaneies between characters of these specimens and our recorded definition of S. setifer t hat must be pointed out. The 32.5-mm. specimen has D. 29, A. 29, P, 13 on both sides (more pectoral rays than previously recorded for S. setifer); a fairly deep body (58.1% S.L.); and a broken- line effect on the sides, but no spots on the snout or breast. On the basis of the dorsal and anal ray counts and pigment on the side, we would identify this specimen as S. setifer. The 115-mm. male from Sta. 526 in Hatchet Bay has D. 30, A. 30; the 113-mm. male from this station has D. 30, A. 31 (there appears to be a minuet 31st dorsal ray) ; both specimens have 13 pectoral rays on each side. The pig- ment of these specimens is the same as that of the Cuban specimens of S. setifer in fig. 32, which species they un- doubtedly represent. However, the high dorsal and anal fin-ray counts of the 113-mm. specimen indicate that this key character must be qualified, at least in identifj'ing specimens from the Bahamas. The 106-mm. male from Sta. 513B has D. 29, A. 29, P, 12; the 113-mm. male from this station has D. 30, A. 30, Pi 13. Both specimens are faded, and lack spots on the snout and breast, but do have short dim lines on the side, and most probably are S. setifer. The 92.5-mm. female (with large macroscopic eggs) from Sta. 526 has D. 30, A. 29, and P, 12. There are no spots on the snout and breast and no broken lines on the sides; instead large dark blotches are present on the sides. The pigment and relatively large body depth (56.8% S.L.) are more like S. hispidus than S. setifer. However, con- sidering the conflicting characters of distribution and pigmentation, and the intermediate fin-ray counts, we cannot identify this single specimen to species. (Febru- ary 13, 1961.) U.S. GOVERNMENT PRINTINC OFFICE; 1961 O- 566129 UNITED STATES DEPARTMENT OF THE INTERIOR, Stewart L. Udall, Secretary FISH AND WILDLIFE SERVICE, Arnle J. Suomela, Commissioner Bureau of Commercial Fisheries, Donald L. McKernan, Director EMBRYOLOGICAL STAGES IN THE SEA LAMPREY AND EFFECTS OF TEMPERATURE ON DEVELOPMENT By George W. Piavis FISHERY BULLETIN 182 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 Published by the U.S. Fish and Wildlife Serrlce • Washington • 1961 Printed at the U.S. Govemment Printing Office, Washington, D.C. For sale by the Superintendent of I>ocuments, U.S. Government Printing Office Washington 25, D.C. - Price 30 cents Library of Congress catalog cai'd for the series, Fishery Bulletin of the Fish and Wildlife Series : U.S. Fish and Wildlife Service. Fishery bulletin, v. 1- Washington, U.S. Govt. Print. Off., 1881-19 V. in illus., maps (part fold. ) 23-28 cm. Some vols. Issued in the congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies: v. 1^9, Bulletin. Vols. 1—19 issued by Bureau of Fisheries (called Fish Commission, v. 1-23) 1. Fisheries — U. S. SH11.A25 Library of Congress 2. Fish-culture— U. S. 639.206173 crSSef] I. Title. 9-35239 rev 2* CONTENTS Pago I nt roduction 111 Stages in normal development 112 Materials and methods 112 Description of stages 113 Stage 0: Ovulated but unfertilized egg 114 Stage 1: Zygote 114 Stage 2: Two cells 115 Stage 3: Four cells 116 Stage 4: Eight cells 116 Stage 5: Sixteen cells 117 Stage 6: Thirty-two cells 117 Stage 7: Sixty-four cells 118 Stage 8: Full blastula 118 Stage 9: Gastrula 119 Stage 10: Neural plate and groove 120 Stage 11: Neural rod 121 Stage 12: Head 122 Stage 13: Prehatching 123 Stage 14: Hatching 123 Stage 15: Pigmentation 124 Stage 16: Gill clefts 125 Stage 17: Burrowing 126 Stage 18: Larva 127 Review of stages and comparison with earlier studies 128 Development at different constant temperatures 130 Materials and methods 130 Account of individual experiments 132 Development at 45° F 132 Development at 50° F 133 Development at 52.5° F 134 Development at 55° F 134 Development at 60° F 135 Development at 65° F 137 Development at 70° F 138 Development at 75° F 139 Development at 77.5° F 140 Development at 80° F 140 Significance of observations 141 Effect of temperature on development 141 Literature cited 142 ABSTRACT Early embryology of the sea lamprey has been subdivided into 19 stages. The stages are based largely on external morphology, behavior, and organ function. The holoblastic cleavage of sea lamprey eggs exhibited tviro types of third and fourth cleavages, equatorial and meridional. The open blastopore had an ap- parent migration over the surface of the embryo until it became the anus. Gas- trulation resembles teleostean gastrulation in some characteristics and amphibian in others. Lamprey neurulation resembles teleostean neurulation more closely than it does amphibian. Sea lamprey eggs were reared experimentally at 10 constant temperatures (at 5° intervals from 45° to 80° F., inclusive, and at 52.5° and 77.5° F.). No viable, burrowing larvae were produced at any temperature below 60° F. or above 70° F. Optimum temperature was 65° F. which yielded 78 percent survival to the burrowing stage; survival to the same stage was much lower at 60° F. (12 percent) and 70° F. (5 percent). The stage attained before all embryos were dead decreased as the temperatures were shifted in either direction from the 60°-70° F. interval. In general, developmental rate became faster, lengths of stages became shorter, and overlap between stages was lessened as temperature increased. The evidence that sea lamprey eggs can develop successfully only within a limited temperature range suggests that unfavorable temperatures may account for the failure of certain apparently suitable streams to produce larval lampreys. EMBRYOLOGICAL STAGES IN THE SEA LAMPREY AND EFFECTS OF TEMPERATURE ON DEVELOPMENT By George W. Piavis, Fishery Research Biologist The Great Lakes fisheries, the Nation's richest source of fresh-water fishes for both com- mercial and recreational fishing, have suffered depletion of catch and the threat of disaster. The danger has its origin in an increase in abund- ance of the sea lampre\-, Petromyzon marinv^, which is parisitic on and highly destructive of fish. Major goals of the Great Lakes research program of the U.S. Fish and Wildlife Service have been to develop techniques for controlling this menace, to restore the Great Lakes fishery stocks to an economically profitable level of abundance, and to sustain them at that level. Others have outlined in detail the history' of the sea lamprey within the St. Lawrence drainage and their invasion of tlie upper Great Lakes: (Gage, 1928; Creaser, 1932; Hubbs and Pope, 1937; Applegate and Moffett, 1955). The sequence of this invasion was summarized by Applegate (1950) as follows: 1921, Lake Erie; 1934, Lake St. Clair; 1936, Lake Huron; 1937, Lake Michigan; 1946, Lake Superior. In planning an attack against the ever-increas- ing numbers of sea lampreys, practically every phase of their life history has been investigated except that of early embryology. A search of the literature reveals little on the embryology of P. rnarinu-s, Lampetra flumatilis, L. planeri, and Ichthyomyzon unicmipis, the significant predators. Still less has been written concerning the other lampreys. Clear-cut stage designations are lack- ing for all lampre3's. The usual embryological designations for the common early stages of development are mentioned through the gastrula stage, but even tliese lack clear definition. In order to portra}' accurateh' the embryology of P. marinus, a study was undertaken which had as its objective the determination and definition of the various stages of development. XoiE.— Dr. Plavls Is presently Assistant Prolessor of Anatomy, Balti- more College of Dental Surgery, Dental School. University of Maryland, Baltimore I, Md. Fishery Bulletin 1S2. Approved for publication May 4, I96U. In these studies, staging of lamprey development has been considered essential to a better under- standing of the results of the series of experiments on effects of temperature. It was immediately obvious that differences in developmental time periods would result from variations in tempera- tures. In order to place the entire series of temp- erature experiments on a common basis, an accur- ate series of stages was of the utmost importance. A second objective of this study was to deter- mine the range of temperature for development of P. marinus eggs as well as their optimum develop- mental temperature. This work began with a preliminary investiga- tion conducted on a part-time basis during the summer of 1954. Intensive investigations were carried on throughout the spring and summer of 1955 when temperature experiments were under- taken. Confirmatory temperature experiments at 65° F. (18.4° C.) were conducted in the summer of 1956. These latter experiments also provided materials for normal staging. This research was conducted as part of my graduate training at Duke University while employed at Hammond Bay as a fishery research biologist b}- the Great Lakes Biological Laboratory of the U.S. Fish and Wildlife Service. Embryological studies and experimental work on living materials were conducted at Hammond Ba3' where the facilities of the sea lamprey research laboratory were placed at my disposal by Dr. James W. Moffett, Director of the Bureau's Great Lakes Biological Laboratories, and Dr. Vernon C. Applegate, Chief of the Hammond Bay Laboratory. Sectioning and statistical work were conducted at Duke University. The University also provided me with a refrigeration unit for use at Hammond Bay and the other facilities and materials necessary to the investigation. Dr. Edward C. Horn, Duke University, knows of my appreciation for his many criticisms and patient guidance. I wish also to thank the staff 111 112 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE of the Hammond Bay Laboratory, and especially John Howell, for courtesies shown me and the time spent in my behalf. STAGES IN NORMAL DEVELOPMENT MATERIALS AND METHODS Eggs of the sea lamprey were taken from mature, nest-building or spawning lampreys found in either the Trout River or the Ocqueoc River, both tribu- tary to Lake Huron, Presque Isle County, Michigan. Spawning lampreys were seized in a forked grip which placed the lamprey between the forefinger and the middle finger while the thumb anchored the animal. Capture by hand, in this way, minimized the loss of specimens by injury. The lampreys were kept in stream water throughout the transfer from the stream to the central collect- ing point, and to the laboratory where they were tempered and stripped in preparation for fertiliza- tion. To condition the eggs and sperm adequately to the test temperatures the lampreys were tempered in running lake water 2 to 3 hours prior to the initiation of the experiments. Eggs were removed rapidly from female lam- preys, held by two people. One held the anterior end by placing a thumb within the oral disc, thus taking advantage of the cusps to prevent slippage. The other held the tail by means of a pair of pliers. While the lamprey was thus outstretched, a transverse slit was made with scissors in the mid- ventral body wall at a point behind the heart and the liver, i.e., at the anterior level of the ovaries. If the females were ripe, eggs began to extrude from this initial opening, whereupon the lamprey was slit to the vent by a rapid stroke with half-open scissors. Since the eggs are completely free in the coelom of a ripe female, they were allowed merely to flow into a 10-liter battery jar which contained a few liters of lake water from the trough where the animals were tempered. Any eggs entrapped within the coelom or folds of the ovary were removed rapidly by dipping the lamprey into the water with the slit open. This method of remov- ing the eggs was quicker and far superior to the milking procedure used by McClure (1893) and others (including the author) because the time required to clear the female of eggs was limited only by the dexterity of the operators. Further- more, the eggs are not distorted or damaged. Blood was not introduced into the fertilization jar since the slit produced no blood providing neither the liver nor the heart was pierced. The males were grasped in the manner described for the females. The individual who was holding the lamprey by the oral disc forced the milt from within the coelom through the genital papilla in a stream directed over the eggs in the battery jar. Two males were used for each female; four males and two females constituted the usual batch. Immediately after the addition of sperm, the jar was provided with an air bubbler and placed in a trough previously brought to the desired tempera- ture. The entire operation from stripping to placing in the jar was completed within 60 seconds. Twenty to 30 minutes after fertilization the eggs were washed. Washing was repeated at least two more times dm-ing the first hour to insure complete removal of excess sperm. After the fii-st hour the eggs were apportioned among enamel pans (8" X 12" X 2") or, on occasion, 4-inch glass bowls. These containers were covered with glass plates to minimize the accumulation of water-borne debris, and submerged. Circulation of water beneath the glass covers and over the developing embryos was insured by allowing a small area of the pans and bowls to remain uncovered. Care was taken not to crowd the eggs in the containers. Preliminary work had shown that eggs arranged in more than a single layer were highly susceptible to attack by fungus. Development of the embryos took place in the pans or bowls which were held at the desired tem- perature in either constant-temperature troughs or in 20-gallon aquariums. Each insulated trough measured 12 feet by 2 feet by 9 inches and con- tained an inner-water chamber surrounded by a 3-inch outer-water chamber. Spaces beneath the metal divider provided free access between the inner and outer cliambers. Water within the inner chamber was provided with air from con- trollable bubblers. The troughs were equipped with thermostati- cally controlled iieating units and refrigeration units whicii provided temperature control within ±0.5° F. The refrigeration tubing, tlie heating elements, and the thermostat bulbs lay within the 3-inch outer chamber. Water was circulated over the thermal units by a continuously revolving 6-vane water wheel driven by a Ratiomotor. Since water flowed freely between tlie outer and inner chambers, the EMBRYOLOGY OF THE SEA LAMPREY 113 action of the thermal units was transmitted witli little time lag to the inner chamber wliich held tlie container of eggs. Heat for the high temperatures was provided hy Bronwil circulators and heaters used in aquaria filled with lake water and provided with a con- trollable air bubbler. Bowls and pans were used to contain the eggs within the aquaria as in the troughs. Air bubblers and water circulation by the Bronwil circulators provided air for the aquaria. The circulators equipped with a contact- tiiermostat and thermometer can be utilized for any temperature setting from room temperature to boiling. In order to attain the maxiiuum range from this piece of apparatus the aquaria were located within a cool room where it was expected tiiat the ambient temperature would not rise above the desired temperature. Tiie temper- ature variation for the circulator was advertised as ±0.18° F. In actual practice, however, temper- ature variation could not be noticed. The appara- tus just described was utilized for experiments at temperatures of 65° to 80° F. ; the troughs were used for temperatures from 45° to 70° F. Prior to each experiment, troughs and aquaria were washed, air-dried, and refilled with lake water; the thermostat was then set at the desired temperature. Observations of the temperature at 5-minute intervals for a period of 4 to 8 hours and occasional readings in the remainder of a 24- hour interval, assured stabilization at the correct temperature. After the desired temperature was established, Taylor thermographs and hourly readings of total-immersion thermometers (placed on submerged rubber stoppers grooved to receive them) gave a further check. The thermometers were set in such a manner as to be readily visible without liandling. Thermographs were not used with the circulator but tlie temperature was watched for 4 to 12 hours prior to the initiation of the experiment. Because of the small varia- tion in temperature delivered by the circulator, a thermograph record was considered unnecessary except when the ambient temperature was ex- pected to rise above that desired. Furthermore, a submerged thermometer was compared period- ically witli the contact-thermostat and thermom- eter. All sampling was random throughout the experi- nuMits, and in general, tiie procedure varied only slightly from that outhned below. Fertilization was considered zero time; the first sample was taken 20 to 30 minutes after fertilization. There- after, samples were taken at the following hours: 1, 2, 3, 4, .._ 12, 14, 16, 18, 20, 24, 28, 32, 40, 48, 72. After the 72d hour samples were taken at 12-hour intervals until the end of the experiment. In some of the longer experiments, samples were taken each 24 hours after about the 18th day. In addition to the samples taken during the run, all remaining eggs and larvae were kept as a final sample. Samples of specimens were placed in SjTacuse dishes and the gross morphological characteris- tics of the embryos were observed under a binoc- ular dissecting microscope. The microscope was equipped witii a calibrated ocular micrometer with which all measurements were made. Immediately after these observations all samples were placed in Smith's fixative' for 12 to 24 hours, washed in several changes of water during a 24-hour period, and preserved in a 4-percent solution of formalin. This method of fixation and preservation was most satisfactory as judged by the pliabilitj' of the heavily yolk-laden eggs after 3 years' pre- servation. Specimens to be sectioned for microscopic exam- ination were washed overniglit in running tap water, stained in alum-cochineal 16 to 24 hours, then dehydrated and embedded in paraffin. Sec- tions cut at 5 to 10 micra were mounted and counterstained in a 0.5-percent solution of fast green in 95-percent ethyl alcohol and covered permanently. DESCRIPTION OF STAGES The following description of stages was based on materials which were taken from all of the batches of the different temperature series; but greatest emphasis was placed on observations of the 65° F. batches. Most characteristics of embryos of the other batciies were itlentical with those reared at 65° F. Those differences which were observed are pointed out in the discussion of the particular stage. The time intervals listed for each stage include tiie time between the first and last appearance of the stage in tlie samples reared at 65° F. The end-points selected for each stage were established after numerous observations of both '.Solution A: PotiiSSium hlchronuile 0.5 grn., wutor 87.5 cc. .Solution B Formalin 10 cc, glacial acetic acid 2.5 cc. MK solutions .1 and B ImmcJl- atcly before using. 114 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE living and preserved materials. It was considered essential to be able to recognize all preserved stages at a later time as well as living stages at the moment of collection. In addition to the gross morphological end- points, histological examination was used on occasion to define the stage more critically. The criteria selected were such as to include also the physiological differences. Furthermore, since the status of an animal is evidenced in part tbrough its activity and movements, natural movements or activity were incorporated as far as possible into the staging criteria. The method of fixation and preservation proved to be suitable for recognition of practicaUj^ all end-points. The normally transparent prolarvae ^ became opaque in Smith's fixative. When se- lected specimens were fixed and preserved in 4- percent formalin, however, retention of all pig- mentation and transparency aided staging. The formalm, of course, hardened the yolk mass and the notochord enough to prevent the use of these specimens for histological preparations. Stage 0: Ovulated but unfertilized egg Animal-polo depression: Present but just visible. Cellular areas: Nuclear; animal-hemisphere cytoplasm; vegetal-hemisphere cytoplasm. Size: 1.0 ±0.2 millimeters. Ovulated eggs within the coelom are assigned this stage. These creamy-white eggs are sur- rounded by a relatively thin jelly coat which expands when the eggs are shed into water. The stickiness of this coat causes sand grains stirred up by spawners to adhere to the surface of the egg- The surface of the unfertilized egg has a small depression in the egg membrane over the nucleus and extends into the nucleus proper. A demar- cation separates the nucleus and the siu^round- ing cytoplasm. One-third the distance down the animal-vegetal axis the cytoplasm contains another demarcation between the cytoplasm surrounding the nucleus and the remaining cytoplasm. The egg of the sea lamprey is telolecithal in that the egg consists of a relatively large amount of yolk and the nucleus is located at the center of the animal hemisphere. This stage is initiated when the eggs are ovulated into the coelom and ends with fertilization. Stage 1: Zygote (fig. 1) hours 0-2 Animal-pole depression: Increases in diameter and depth; disappears within about 1 hour. Cellular areas: Identical to stage 0. Size: 1.0 ±0.2 millimeters. Fertilization membrane: Appears within 20-30 minutes after fertilization; is retained through stage 13. Figure 1. — ^Lateral view of stage 1, zygote, showing cellular areas. This stage extends from the time of fertilization to the time when the fertilized egg begins to imder- go first cleavage. A meridional section of a fertilized egg (fig. 2) shows the yolk platelets ' Hubbs (1943) defined a prolarva as a "larva still bearing yolk." In the present work prolarva includes stages 14-17. Figure 2. — Meridional secUon of stage I prior to forma- tion of fertilization membrane. EMBRYOLOGY OF THE SEA LAMPREY 115 within the cytoplasm surrounding the nucleus and the remaining cytoplasm. The yolk of the animal and the vegetal areas differs in that the yolk platelets within the vitelline area are much larger than those in the area around the nucleus. The depression at the animal pole seen in stage persists through this stage. The first indication of significant morphological change is a noticeable deepening of this depression which will remain until about the first hour after fertilization. Three distinct external areas indicate the presence of the three internal areas: nucleus; animal-pole cytoplasm; and vegetal-hemisphere cytoplasm. Below the nucleus the cytoplasmic area is demarcated by a band extending approxi- mately one-third the distance down the animal- vegetal axis. The remainder of the egg consists of cytoplasm heavily laden with j'olk. Each of the areas of the egg is visible both externally and internally as early as the unfertilized egg and up to the initiation of first cleavage. Shortly after the animal-pole depression deepens, the fertilization membrane appears. Stage 2: Two cells (fig. 3) hours 2-8 Cellular areas : Visible in the daughter cells. Size: 1.0 ±0.2 millimeters. Cleavage: First furrow appears. Prominent peaks in daughter cells. Holoblastic. Com- pleted within 5-6 hours. The external topography of the 2-celI stage is comparable to that of the zygote in that the nu- clear, and the animal- and vegetal-hemisphere cytoplasmic areas are readily visible in both daughter cells immediately after reconstitution of the nucleus (fig. 3). This stage begins when the animal pole of the zygote begins to furrow and dimple in preparation for first cleavage, and ends at the beginning of the second cleavage furrow and dimple at the animal pole. Cleavage is total, usually slightly unecpial (fig. 4) but occasionally (less than 1 percent) grossly unequal. As mentioned above, cleavage begins as a small furrow; a slight uprising of the cell membrane lateral to the furrow produces the dimpled effect. As cytokinesis progresses the cell membrane expands to a greater and greater extent while the fertilization membrane remains un- changed. The cell membrane continues to expand until the cell reaches the stage seen in figure 3, where the expanded cell membrane can be seen as twin peaks lateral to and above the furrow. As the cleavage furrow progresses meridionally over Fkutre 3. — External views of stage 2, two cells, showing the expanded cell meinl)raiie (right side) and daughter cells showing the cellular areas. 566128 0—61 2 116 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 4. — External views of cleavage extremes of stage 2. the cell, the ceU membrane expands in advance of the progressing furrow. Stage 3: Four cells (flg. 5) hours 8-11 Cellular areas : Visible in four daughter cells. Size: 1.0 ±0.2 millimeters. Figure 5. — Various external views of stage 3, four cells. Cleavage: Second furrow appears: peaks are less prominent than in first division. Holoblastic. Completed within 3 hours. Despite cellular division the three distinct topo- graphical areas are still discernible in each of the four cells. Stage 3 begins with the advent of the second cleavage furrow and dimple which starts in much the same manner as the first; the furrow and dimple are observed at the animal pole to either side of the first cleavage furrow and progress meridionally at right angles to the first cleavage furrow. As in stage 2, the expanding cell mem- branes are observed during cytokinesis. The end-point of stage 3 is the appearance of third cleavage furrow and dimple. Stage 4: Eight cells (flg. 6) hours 10-15 Cellular areas: Visible in meridional type cleavage. Size: About 1.0 millimeter for equatorial divi- sion. Meridional division increases equatorial diameter and shortens meridional diameter. Cleavage: Two types; meridional or equatorial. Holoblastic. Completed within 2 hours. Stage 4 begins with the appearance of the third cleavage furrow, which may be either a double Figure 6. — Polar views of meridional and equatorial cleavages forming stage 4, eight cells. EMBRYOLOGY OF THE SEA LAMPREY 117 meridional furrow or a single equatorial division. The meridional type was described by McClure (1893) for P. marinus eggs held at 6°-8° C. (42.8°- 46.4° F.) or at room temperature, whereas both types of cleavage were found in this study. The equatorial cleavage was not mentioned bj- Mc- Clure, although it predominated in the present study in all experimental batches, regardless of temperature. Embryos formed after meridional cleavage can be distinguished by the flatness at the animal pole. The nuclei are aligned four on each side of the first cleavage plane (fig. 6). In some embrj'os the segmentation cavity can be seen because the embryo splits along the first cleavage plane. The flatness of stage 4 embr>-os formed by meridional divisions produces a large perivitelline space which facilitates removal of the fertilization membrane. The same operating space is not encountered again until stage 9. The embryos formed by equatorial divisions have 4 micromeres resting upon 4 macromeres. In them, the space available for removal of the fertilization membrane is relatively small. The end-pouit of this stage is the appearance of the fourth cleavage furrow. Stage 5: Sixteen cells (fig. 7) hours 13-15 Cellular areas: No longer recognizable from external view. Size : 1 .0 millimeter. Cleavage : Equatorial or meridional, depending upon type in stage 4. Pattern irregular. Com- pleted within less than 2 hours. Stage 5 begins with the appearance of the fourth cleavage, the plane of which varies according to the t>T)e of cleavage which formed stage 4; a third -stage equatorial division, near the animal pole, is followed by a fourth meridional division and vice versa. Cleavage irregularity becomes apparent during this stage; embryos composed of 9 to 16 cells are found and included in this stage. The end-point of stage 5 is reached when the embryo is composed of 17 or more cells. Stage 6: Thirty- two cells (fig. 8) hours 16-19 Size: 1.0 millimeter. Cleavage: Random and indeterminate. Com- pleted within 1 hour. FinuRE 7. — Several views of stage o, sixteen cells, illus- trating differences in size of animal and vegetal cells. Figure 8. — Several polar views of stage 6, thirty-two cells. Embryos were assigned to this stage when 17 to 32 cells were distmguishable. Cytokinesis at this point, however, became mdeterminable so that the fifth cleavage appeared to take place at random. Cell counts were made on all embryos that could not be identified on the basis of relative cell sizes by comparison with both stages 5 and 7. When animal cells arc compared with animal cells, and vegetal cells with vegetal cells, the cells of stage 6 embryos will be about one-half the size of stage 5 colls and approximately twice the size of stage 7 cells. The relative sizes of the animal- 118 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE and vegetal-cells can be ascertained in figure 9, a meridional section of stage 6. The extent of the blastocoel can also be seen in figure 9. The roof of the blastocoel is composed of a single layer of relatively large animal cells whereas large vegetal cells extending from the vegetative pole to the blastocoel constitute the floor. The end-point of stage 6 is reached when the embryo has 32 cells. Figure 9. — Meridional section of stage G. Stage 7: Sixty-four cells (fig. 10) hours 19-24 Size: 1.0 millimeter. Cleavage: Indeterminate. Stage 7 is considered to have begun when the embryo has more than 32 cells. The cleavage continues indeterminable in this and later stages. Recognition of the stage becomes a matter of cell counts and comparison of cell sizes with those of the preceding and succeeding stages. A cursory count of the animal cells is made to get an approxi- mate estimate of the stage to which the embryo should be assigned; the final assignment is based on a combination of cell size and cell count. As cell size diminishes the contour of the embryo becomes smoother. This change can be seen in a comparison of figures 8 and 10. The animal cells are still smaller than the vegetal cells, which have Figure 10. — Polar and lateral views of beginning stage 7, sixty-four cells. since divided and are now only about twice the size of the animal cells. The end-point of stage 7 is reached when the embryo has more than 64 cells. Division of all animal cells along with division of the vegetal cells can be taken as an approximate end-point. Stage 8: Full blastula (fig. 11) hours 24-64 Size: 1.0 millimeter, increasing to 1.2-1.4 milli- meters. Cleavage : Furrows seen on individual cells. Animal hemisphere: Becomes translucent. Blastocoel: Visible through animal cells. Wlien the animal pole cells of stage 7 undergo further cell division as evidenced by further reduc- tion in cell size and the appearance of cleavage furrows, the embryo is considered to have entered stage 8. The most striking external feature of this early phase of the stage is its contours (compare to stage 7) which become progressively smoother as the number of cell divisions increases and cell size decreases. In the final phase of this stage the volume of the embryo increases by some 3 to 5 times. Measure- ments of living specimens made withm a cali- brated ocular micrometer at the initiation of and EMBRYOLOGY OF THE SEA LAMPREY 119 Figure 11. — Polar views of stage 8, full blastula, arranged to illustrate changes in cell size. just prior to this stage were consistently about 1.0 millimeter across as viewed and measured in optical section perpendicular to the animal- vegetal axis as well as along that axis. Subse- quent measurements prior to the appearance of the dorsal lip of the blastopore, the end-point of stage 8, ranged from 1.2 to 1.4 millimeters. The increase in volume is accompanied by translucency of the animal-pole cells so that the outline of the blastocoel becomes visible through approximately one-third the surface of the embryo. The size of the animal cells does not change during the expansion process; the increase of the blastocoel indicates that the expansion is due to blastocoel enlargement. The blastocoel is at its greatest volume at the end of stage 8. The animal cells are still about one-half the size of the vegetal cells. This size relation is maintained as long as the epiboliznig animal cells can be compared externally with the underlj'hig vegetal cells. The end-point of stage 8 is the appearance of the blastopore. Stage 9: Gastrula (fig. 12) hours 64-104 Size: 1.2 to 1.4 niillinieters decreasing to 1 .f) millimeter. Animal hemisphere: Translucent to transparent Figure 12. — Posterior view of stage 9, gastrula, illus- trating the blastopore. to opaque. Opacity extends progressively forward from blastopore. Begins to flatten, forming neural plate. 120 FISHERY BULLETIN OF THE FISH AND WILDLIFE, SERVICE Blastocoel: Progressively obscured by opacity of animal hemisphere. Decreases in volume. Blastopore: Forms as wide arched slit. Hooded as neural plate begins to form. Apparently migrates. The translucency of the animal hemisphere of stage 8 changes to transparency after the appearance of the blastopore. Through the transparent animal hemisphere the underhdng chorda-mesoderm is visible as it undergoes its morphogenetic move- ments. During this period the volume of the blastocoel decreases when the chorda-mesoderm reaches a position two-thirds the distance across the animal hemisphere. The most advanced portion of the chorda-mesoderm is in the mid- sagittal plane; the material in the parasagittal planes lags behind these more advanced cells. As this m.aterial progresses beneath the animal hemisphere, the dorsal lip of the blastopore pro- gressively increases in thickness and begins an apparent migration from its original position at or near the overlap of animal and vegetal cells one-third the distance from the center of the animal hemisphere, toward the center of the animal hemi- sphere. As the chorda-mesoderm moves to its anteromost position in the embryo, the trans- parency of the animal hemisphere decreases, and the decrease in the volume of the embryo continues. The typically circular blastopore of later stages is formed by progression of the animal cells in their epibolic movements to produce a changing pattern to the dorsal lip and, necessarily, the blastopore. At the time of the blastopore's inception the advancing margins of epibolizing cells beneath the blastopore approach each other laterally as a wide-open V with the apex at the center of the dorsal lip of the blastopore. As gastrulation pro- gresses, these V-arranged margins close toward each other until only the vegetal cells below the blastopore remain uncovered. These vegetal cells are covered by epibolizing cells which move upward along the mid-sagittal plane (fig. 12). As the chorda-mesoderm advances farther into the anterior portion of the embryo, the blastopore begins an apparent migration along the mid-sagit- tal plane from its original position to one located at the posterior limit of the mid-sagittal plane in stage 10. Histological comparisons between stages 8 and 9 indicate that this apparent movement re- sulted from reduction in the size of the blastocoel. At this time the dorsal region of the entire embryo begins to flatten and to thicken from the dorsal lip of the blastopore to the anterior region. Stage 9 is marked by the appearance of the dorsal lip of the blastopore, a flat crescentic- shaped furrow within the overlapping line of the animal cells epibolizing over the vegetal cells. Recognition of the stage depends upon locating the blastopore half-way beneath the embryo; the observer must rotate the embryo to find it. The end-point of stage 9 is reached when the flattening process reaches the anterior end of the embryo. Stage 10: Neural plate and groove (fig. 13) days 4-5 Size: 1.1 to 1.3 millimeters. Blastopore: Triangular to ovoid. Reaches its dorsalmost point. Neural tissue: Neural plate forms and thickens. Groove and folds form. Figure 13. — Lateral view of stage 10, neural plate. Stage 10 begins when the flattening of the dorsal ectoderm mentioned in stage 9 has reached the anterior extremity of the embryo (fig. 14). At the same time, the nearly completed blastopore is near or at the dorsalmost point of the posterior area of the mid-sagittal plane of the embryo (fig. 15). Further thickening of the depth of the flattened area follows almost immediately after the flat- tening of the dorsal ectoderm extends from the blastopore to the anterior region. These animal cells are no longer transparent. The ovoid to triangular EMBRYOLOGY OF THE SEA LAMPREY 121 Figure 14. — Dorsal view of stage 10 illustrating neural groove. Figure 15. — Posterior view of stage 10 illustrating blasto- pore and neural plate with neural groove. blastopore is now at the uppermost point of its apparent migration along tlie mid-sagittal plane and the flattened ectodermal cells (neural plate) are at the height of their thickening. Practically all the vegetal cells are covered by tlie epibolizing animal cells. Immediately after the neural plate thickens, the central portion of the plate begins to form a trough, producing a neural groove and fold stage (fig. 16). As tlie groove in the neural plate deepens the "folds" begin to approximate each other and unite. The first actual union occurs in the mid-dorsal region of tlie embryo. This union is the end-point for stage 10. Figure 16. — Cross-section of stage 10 Stage 11: Neural rod (fig. 17) days 5-6 Size: 1.1 to 1.3 miUimeters. Blastopore : Circular. Apparent migration toward ventral surface of embryo. Figure 17. — Oblique lateral view of stage 11, neural tube. Neural tissue: Union of folds middorsally on embryo. Neural tube lacks netirocoel, and thus is a neural rod. Becomes prominent across dor- sum of embryo; circumscribes approximately two- thirds of embryo. In the late phase of this stage the dorsopos- teriorly located, circular blastopore begins an apparent migration toward the ventral surface accompanied by a similar ventral movement of tlie prospective head. Toward the end of tliis stage the circular blastopore is at the venti'al- most point of tlie posterior as contrasted witli its 122 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE earlier position at the dorsalmost point of the posterior. Those movements continue until the nem-al rod occupies approximately two-thirds of the circumference of the embryo. The union of the folds marks the beginning of stage 11. The neural plate begins to round up and appears in external view to have become a neural tube with anterior and posterior neuropores. This appearance is a result of the apparently simul- taneous, progressive union of the folds in anterior and posterior directions from the site of first union, the middorsum. The external morphology of this stage is as deceptive as that of the preceding stage since histological examinations reveal that the neural tube docs not possess a neurocoel; in reality the neural tube is a neural rod as described by Shipley (1885). In the anteromost region the neural rod seems to elevate from the surrounding and under- lying tissue. The end-point of this stage is reached when the anterior region is raised above the globular yolk mass. Stage 12: Head (fig. 18) days 6-S Size: 1.1 to 1.4 millimeters. Blastopore: Below posteriormost point of neural rod, at ventral surface of embryo. Circular. Neural tissue: Very prominent from head to blastopore. Neurocoel over presumptive pharynx. Head: Elevated from yolk mass. Free length approximately 1.0 millimeter. Yolk : Globular as in preceding stages. Stomadaeum: Invagination begins. Measurements of the embryo from the tip of the head to the posteriormost point of the yolk mass along the neural rod gave a range from 1.0 to 1.1 millimeters when the head was just beginning to form and 1 .3 to 1 .4 millimeters when the head was fuUy formed and elevated from the yolk mass. Although sections show that some somites have formed, they are not discernible externally. The blastopore is below the posteriormost point of the neural rod and faces directly ventral during the initial period of this stage. When the head is fully formed, the blastopore faces anteriorly along the ventral surface toward the head. Histological examination proves that the neural rod forms its neurocoel over the presumptive phar\nix only. By this time the neural rod ex- tends along appro.ximately 75 percent of the periphery of the embryo (fig. 19). The rod is prominent and extends from the head to the blas- topore. Figure 18. — Lateral view of stage 12, head. FiGnRE 19. — Parasagittal section of stage 12. Elongation takes place faster in the head region than in the tail region. The anterior part becomes well elevated and protrudes between the lateral swellings. The length of the head and the pre- sumptive branchial region is about 1.0 millimeter at the end of the head stage. The appearance of a head on the exterior of the embryo marks the beginning of stage 12. This change is caused by the sudden increase in length of the neural rod during stage 11 and the sudden expansion of the presumptive pharyngeal cavity both dorsally and laterally to produce an upswelling of the pro- spective head from its original position. EMBRYOLOGY OF THE SEA LAMPREY 123 The yolk mass retains the globular shape it had in the preceding stage. The stomodaeal invagi- nation begins in the ventral portion between tlie lateral swellings. The end-point of this stage is reached when the head region begins muscular activity. Stage 13: Prehatching (fig. 20) days 8-12 Size: 1.4 to 2.5 millimeters. Blastopore: Circular. Much reduced. Lo- cated at anteriormost point of apparent migration. Figure 20. — Lateral view of stage 13, prehatchiiig. Neural tissue: Neural rod now a true neural tube. Prominent above somites from blastopore to head. Head: 2.0 millimeters. Yolk: Obovate, blunt end posterior. Stomodaeum: Deepens and widens. Somites: 5 to 20; not easily distinguished externally. Locomotion: First muscular activity. Move- ment of free embryo, head. The length of the head region has increased from 1.0 millimeter at the beginning to 2.0 milli- meters at the end of this stage. Measurements were taken from the tip of the head to the anterior- most part of the yolk mass. The blastopore, which is situated antero- ventrally lies at its anteriormost position directly opposite the stomodaeal invagination. Stage 13 begins with the advent of muscular activity. These movements initially are mere lateral flexions of the head and "neck" portion of the embryo; up-and-down activity is not apparent. As the embryo enlarges, muscular contraction includes dorsoventral flexion and becomes \n\- 566128 O — 61 3 dulating rather than wagging. With increase in in the embryo's size the perivitelline space be- comes fully occupied, since the embryo, arranged in circular fashion within the membrane, begins to spiral upon itself. During this growth the above-mentioned movements become more forceful and more frequent. Embryos that show movement but still have in- tact fertilization membranes are in stage 13. The endpoint of stage 13 comes when the head pro- trudes through the fertilization membrane to initiate hatching. Stage 14: Hatching (fig. 21) days 10-13 Size: 3.0 to 5.0 millimeters. Blastopore: Minute opening at apex of 90° ventral flexion of posterior. Dubious structure for staging. Figure 21. — Lateral view of several prolarvae of stage 14, hatching, showing posterior curvature, somites, yolk mass, and condition of mouth and nostril. Neural tissue: Neural tube still very prominent above somites. Yolk (gut) : Becomes slender and assumes a spatulate shape. Posterior region has 90° ventral flexion. Anteriormost portion greenish. Stomodaeum: Deep pit located ventrally. Somites: 18-20 to 30-35. Extend from neural tube to gut. Locomotion: Hatching movements. Undula- tion of anterior bodv region onlv. 124 FISHERY BULLETIN OF THE FISH AKD WILDLIFE SERVICE Nostril: Begins as single invagination in mid- ventral line, anterior to the stomodaeum. Transparency: First appearance of ectodermal transparency over pericardinm. Extends pos- teriorly and anteriorlj' in later phases of this stage. Circulatory system: Pericardium visible through transparent ectoderm. Straight tubular heart. Begins beating 40 times per minute. Liver: First indication posterior to pei'icardium in anterior part of gut. Appears in late period of stage. Stage 14 starts when the embryo breaks through the fertilization membrane. The constant move- ment observed in stage 13 finally becomes suffi- ciently strong to extrude the head through the membrane. Further activity enlarges the tear in the membrane and eventually leads to hatching. The pericardial cavity appears as a ventral swell- ing approximately 1.0 millimeter posterior to the tip of the head. It is also present in late stage 13 but is not easily recognized in most embryos. Shortly after hatching the ectoderm overlying the pericardial region and the tissue beneath gradually become transparent. A short time later, the body anterior to the pericardial cavity becomes trans- parent. Through these transparent tissues can be seen the straight tubular heart which begins to pulsate during the eleventh day at 40 beats per minute (at 65° F.). There is no sign of blood in the heart or anywhere in the prolarva. The end-point of stage 14 is reached when melanophores appear on the embryo. Stage 15: Pigmentation (fig. 22) days 13-16 Size: 5 to 6 millimeters. Neural tissue: Brain and tube visible through ectoderm. Gut: Spatulate, changing to cylindrical. Ven- tral flexion of about 10° remains. Anterior face greenish. Stomodaeum (mouth) : Transverse sUt bounded by thickened lips. Opens into oral cavity. Somites: 35 to 50. Locomotion: Undulation of entire body slightly restricted by yolk-filled gut in early stage. Full swimming movements in late period. Nostril: Single, median at anteriormost point on ventral surface. Transparency: Extends anteriorly to branchial region and posteriorly to about two-thirds the length of the prolarva. Circulatory system: Heart becomes S-shaped. Heart walls thicken. Grayish channel forms in midventral gut and turns red as hemoglobin ap- pears. Bilateral channels appear in 15th day. Heart beat, 100 per minute. Liver: Becomes larger and vascularized. Pigmentation: First appears as bilateral melano- phores dorsal to the midbrain. The stomodaeal pit has become a slit opening into the anterior (oral) chamber of the pharynx which is separated from the posterior (pharyngeal) chamber by the velum. The transverse slit is bounded by a thickened ectodermal lip anteriorly and posteriorly. The pharynx has its full comple- ment of 7 visceral pouches. Deftness of swimming increases as the shape of the yolk mass changes from spatulate to cylindri- cal. At the same time the tail straightens from its ventral flexion. The prolarvae progress from the awkward movements of stage 14 to an undulating movement, the smoothness of which depends on the amount of the ventral flexion of the tail. The heart has enlarged and assumes an S-shape Figure 22. — Lateral view of stage 15, pigmentation, showing somites, pericardial area, and condition of yolk. EMBRYOLOGY OF THE SEA LAMPREY 125 as the auricle moves dorsal to the ventricle. Botli auricular and ventricular walls begin to thicken but remain transparent. The rate of heartbeat in- creases from the 40 per minute of stage 14 to 100 per minute at the end of stage 15. Between the 13th and the 14th days a graj-ish channel develops along the midventral line of the cylindrical yolk mass. The channel extends from the anterior portion to the midpoint of the yolk- filled gut and within a short time turns red as hemoglobin is produced. During the 15th day bilateral channels appear along the midlateral line of the yolk mass and then converge to form a single vessel to the heart. Stage 15 is initiated by the appearance of a pair of dorsal melanophores bilateral to the midbrain. vSecond and third pairs appear in sequence immedi- ately posterior to the original ones. Shortly after the appearance of the third pair, a melanophore can be seen above the anterior branchial region. Pigmentation spreads anteriorly and posteriorly from the dorsal pau-s along the neural tube (fig. 23) ; it spreads posteriorly and ventrally from the lateral pair along the line between the yolk and the somites, as far as the anterior limit of opaque or undifferentiated tissue. As transparency pro- gresses, the posterior distribution of melanophores is extended correspondingly. The end-point of stage 15 is reached with the appearance of the gill slits. Stage 16: Gill-cleft (fig. 24) days 15-17 Size: 6 to 7.5 millimeters. Neural tissue: Divisions of central nervous system recognized through ectoderm. Gut: Cj'lindrical yolk mass. Increases in length and becomes more slender. No ventral flexion. Anterior portion greenish. Postanal gut present. Anus forms at persistent blastopore. Mouth: Semicircular slit. Enlarges to hooded mouth. Oral cirri make first appearance. Somites: No longer useful as staging criterion. Locomotion: Larval swimming pattern. Very adept at end of stage. Nostril: Migrates from anteroventral to antero- dorsal extremity. Transparency: Practically entire prolarva be- comes transparent except gut. Circulatory system: Ventricular wall has thick- ened. Heart beat, 150 per minute. Flow of blood readily visible. Liver: Size increases. Distinctly separated from gut. Pigmentation: Melanophores extend along dor- sum and sides of embr^'o from anterior to posterior. Bilateral eyespots appear at end of stage. Gill clefts: First appearance. They become functional. Respiratory rate : 120 per minute. Fins: Caudal and anal fins appear. The mouth continues to migrate anteriorly and Figure 23. — Dorsal view of stage 15 showing condition of pigmentation over the head region. Figure 24. — Lateral view of stage 16, gill cleft, showing gill clefts, somites, pericardial area, and pigmentation over yolk mass. 126 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE the nostril dorsally. The anterior Up of the slit- Hke mouth of stage 15 moves anteriorly at the midline only, producing a semicircular mouth. Migration of the anterior lip and nostril continues until the mouth lies at the anteroventral extremity and the nostril is at the anterodorsal extremity. In later phases of the stage the mouth enlarges and is hooded by the anterior lip. Oral cirri begin to form posterior to the anterior and posterior lips. Movement during this stage is by a rapid undu- lation of the trunk and tail. All ventral flexion of the tail region has disappeared during the trans- ition from stage 15 to 16. The anterior boundary of the auricle lies over the center of the ventricle. The flow of blood through the heart and other parts of the prolarva is seen readily. In the latter period of this stage the bilateral pigmented retina (eyespot) appears anterior and dorsal to the velum. Melanophores which extend down the length of the neural tube and along the yolk-filled gut begin to migrate ventrally from both levels. Melanophores migrate along the myosepta from the neural-tube level and ventrally from the yolk-somite line. The anterior lip and head region also are well pigmented. The appearance of gill clefts marks the initiation of stage 16. They appear and start functioning in order from the first to the seventh. Function- ing of the branchial apparatus can be determined by two means: beating of the velum (between the oral cavity and the first visceral cleft) as it forces water through the pharynx; or, contractions of the gill clefts. The respiratory rate (velum beat) is 120 per minute. The end-point of stage 16 is reached when the prolarvae acquire bilateral eyespots and are about to burrow. Stage 17: Burrowing (fig. 25) days 17-33 Size: 7.5 to 9.0 millimeters. Gut: Cylindrical and yolk-fUled. Postanal gut absorbed. Esophagus visible on left side. Cloaca transparent. Peristaltic movements in hind gut. Lumen of gut opens at end of stage. Mouth: Oval with dorsal and ventral lips. Opens anteriorly. Oral cirri. Locomotion: Swimming movements same as in stage 16. Prolarvae burrow. Nostril: On dorsal aspect, its ultimate position. Transparency: Complete except for yolk-filled gut. Figure 25. — Lateral views of stage 17, burrowing, showing the condition of the mouth and the lips. EMBRYOLOGY OF THE SEA LAMPREY 127 Circulator}" system: Heart walls thicken. Heart beat more than 200 per minute. Flow of blood visually traceable throughout most of prolarva. Liver: Further increases in size. First indica- tion of gall bladder on right side. Pigmentation: Melanophores extensively dis- tributed. Aggregation over pronephric region and caudal end of notochord and neural tube. Respiratory rate: 200 per minute. Pronephros: First detectable externally. The posterior region of the gut becomes trans- parent as the cloaca is formed with the opening of the pronephric ducts into the gut (histological observations). A yolk pellet forms in the gut ante- rior to the cloaca as the result of peristaltic move- ments of the gut. These movements move the pellet toward the anus and then back to the remaining yolk mass. The lips take on the characteristic larval form. The mouth opens directly anteriorly, so that the anterior lip has now become the dorsal lip and the posterior lip the ventral. Stage 17 begins when the prolarvae burrow into the bottom mud. Burrowing is the result of the action of both the tail and the head regions. As the head moves from side to side to create space within the mud, the lashing of the tail drives the prolarva into the mud. Prolarvae of this stage placed in aquaria or beakers first swim near the surface of the water and then suddenly plunge downward with rapid swimming movements until they reach the bottom, when they immediately begin to burrow. Prolarvae of stage 16 merely drift down to and lie on the bottom. Swimming movements do not differ from those of stage 16. The liver is extended farther and the presence of the gall bladder in the later period is marked by its bile-green color. The eyespot or retina is very prominent anterior and dorsal to the velum. Melanophores have spread completely around the gut region to the ventral surface. They have migrated down the lateral surface of the velum and the gill bars, and have completely outlined the branchial basket. The dorsal lip is covered completely with melanophores. A pronephros dorsal and posterior to either side of the pericardial cavity is visible externally because of the presence of much blood. This stage is the first in which the pronephros is visible externally. The end-point of stage 17 is reached when the lumen of the yolk-filled gut is opened. Stage 18: Larva (fig. 26) days 33-40 Size: 9 millimeters and longer. Gut: Lumen completely opened. Yolk ex- truded from gut. Gut tissue becomes trans- parent. Respiratory rate: 200 per minute. Figure 26. — Various views of several stage 18 larvae on a black background. Shown are: condition of the mouth and lips, eyespot, pigment outlining branchial region, liver immediately posterior to the heart, somites, fins, and the gut from the liver to the cloaca. The granular appearance along the gut is caused by pigmentation. The transition from the prolarval to the larval condition (all systems differentiated save genital) is marked by the differentiation of the formerly yolk-filled gut into its definitive form. This change is seen outwardly as the gut becomes transparent at the "stomach" region behind the liver. Transparency progresses posteriorly until the entire gut has differentiated and the digestive system is open from the mouth to the anus. After the pellet described in stage 17 is passed, yolk remaining within the lumen of the gut is extruded constantly from the anus. Stage 18 larvae are between 9 and 10 milli- meters long at the time the gut becomes fully differentiated. They belong to the first larval stage wliich would be equivalent to age-group 0, 11 to 21 millimeters long, of Applegate (1950). 128 FISHERY BULLETIN OF THE FISH AND WTLDLIFE SERVICE REVIEW OF STAGES AND COMPARISON WITH EARLIER STUDIES Staging of lampreys in most early literature lacked precision. Investigators, for example, Shipley 1885, Scott 1887, McClure 1893, Hatta 1914, 1915, who studied early embryology, gave little or no attention to the need for staging. Developmental sequence was described in con- ventional embryological terms. The first serious attempt to show stages of lamprey development was by Damas (1944) in his excellent histological study of L. fluviatilis, con- cerned primarily with tracing development of the head. Each section (comparable to a stage) com- prised a short account of external features and a thorough histological description. His staging series began with embryos possessing 3 somites (section 1). The next section included embryos with 10 somites. (Every somite listed by Damas was not visible in external view, but each was seen histologically.) On the basis of somite numbers Damas was able to describe several stages slightly different from each other which could be described also by staging based on external morphological features. Some of his sections, based usually on intervals of 10 somites, have been translated into stages of the present study for comparison in table 1. Stages described in the present work have been based primarily on changes of external features. Histological observations have been used sparingly and merely for clarification. Damas, who began his study after development had started, omitted the first 9 stages of the present work. Because his criteria were histological, his stages are over- lapped by stages in the present study (table 1). Table 1. — Equaling o/ "sections" of Damas (1944) study with "stages" of the present research Damas (section), Plavis (stage) . . . Damas (section) Piavts (stage) . . . Damas (section) Plavis (stage) . . . I II III IV 10 11 12 12 VI VII VIII IX 13 13 14 14 XI XII XIII XIV 15 15 15 16 V 13 X 14 XV 18 Staging, to be of practical value, should be simply, easily, and promptly recognized by inves- tigators in both living and preserved materials. By far the easiest method of designating stages is one based on morphological characteristics since the resulting demarcations are relatively sharp and distinct and the diagnostic features are visible without elaborate histological preparation. They are natural divisions in a developmental sequence. Particular effort was made in this study to avoid dependence on measurements and counts. Ideally, observation alone should suffice for recognition of stages. Applegate (1950) divided the larval forms of P. marinus on the basis of length and weight into age groups, beginning with age-group 0. He stated that the larvae "... upon leaving the nest have completed their early developmental stages and are perfectly formed but diminutive ammocoetes." These ammocoetes are in reality stage 18 larvae, for specimens from Applegate's collections match specimens from my collections in their external as well as histologic characteristics. Stages, as designated here, can be determined by the naked eye or with a hand lens. Separation of stages 12 and 13 might conceivably cause some difficulty since transition between them is based on muscular activity. It is easy, however, to recognize preserved specimens, for the head region of embryos of stage 13 is displaced to either side of midhne whereas heads of stage 12 embryos are medial. Cleavage in the sea lamprey has several interest- ing characteristics, the first of which is the appear- ance of relatively high prominences lateral to and above the first cleavage furrow. Similar promi- nences occur during second cleavage (stage 3). McClure (1893) noted the unusual cleavage pattern of P. marinus, especially in the third and fourth cleavages. He described tlie third cleavage as meridional and considered this to be the sole type of third cleavage. Actually, both meridional and equatorial types of third cleavage occur, but their relative abundance fluctuated widely irre- spective of temperature. Teleostei, Gymnophiona, Gallus, and other forms exhibit a meridional third cleavage, vvhich, however, is accomplished in mcroblastic division. In the sea lamprey the meridional third cleavage is holoblastic. The cleavage of stage 4 embryos determines the type of cleavage for stage 5, since an equatorial stage 4 is followed by meridional stage 5 and vice versa. Cleavage of stage 6 embryos is indeterminable and is accompanied by a lag in the cleavage of the vegetal cells. After stage 6, demarcation be- tween the animal and vegetal cells is sharp until EMBRYOLOGY OF THE SEA LAMPREY 129 the vegetal cells are covered by the epibolizing animal colls. Blastopore formation, as was observed by Shipley (1885), involves behavior of animal cells and vegetal cells similar to that in amphibians and teleosts. When invagination of animal cells forms the blastopore, the margin of animal cells is arranged latitudinally around the embryo from the blastopore. As the animal cells continue to epibolize, the vegetal cells are covered by animal cells from the anterior and dorsolateral portions of the animal hemisphere and subsequently from the posterolateral and ventral positions in the fashion described for stage 9. Thus, migration of animal cells over vegetal cells in the lamprey matches closely the epiboly of animal cells in amphibians. The dorsal lip does not curve as much, however, in the sea lamprey as in amphib- ians. If the blastopore is considered to be de- fined by the line of epibolizing animal cells, as it is in teleosts, the blastopore of the sea lamprey is at first oval. This shape results from alignment of animal cells lateral to the midsagittal plane. A circular blastopore is formed when animal cells at the ventralmost point of the oval-shaped blastopore cover the oval yolk plug in a ventral- to-dorsal direction along the mid-sagittal plane. The circular blastopore of the sea lamprey apparently migrates toward the center of the animal hemisphere whereas the teleost blastopore migrates toward the vegetative pole as the embryo lengthens. The amphibian blastopore eventually reaches and passes beyond the vegetative pole at a stage equivalent to stage 13 of the lamprey. Because teleost embryos do not extend ventrally over the yolk mass the blastopore does not reach the vegetative pole. The apparent migration of the lamprey blasto- pore might possibly be attributed to the decrease in the vohim(> of the blastocoel during stages 9 and 10. An increase in embryo volume which takes place in stage 8 is apparently due primarily to the incr(>ased volume of the blastocoel. During stage 9, after involution is completed, the volume of the eml)ryo decreases as the result of a decrease in th(> volume of the archenteron. The archen- teron, in reality, is the original blastocoel since th(> blastocoel is not obliterated during gastrula- tion (histological observations) as it is in amphib- ians. In this feature of gastrulation the lamprey closelv resembles teleosts. Since decrease in the size of the archenteron shortens the embr\'0 and since growth of the neural tube in stages 12 and 13 moves the blastopore along the mid-sagittal plane, the blastopore appears to migrate. The present study also corroborates Shipley's (1885) observation that the open blastopore persists, eventually becoming the anus. Histo- logical sections permit tracing the archenteron to an open blastopore through stage 15. During stage 16, the diameter of the blastopore widens and the lips thicken to form the anus. The archenteron can be traced to the anus in stage 16 as it was traced to the blastopore in stage 15. Shipley recognized that the early neural tube did not possess a neurocoel and called it a neural rod, which term has been retained here. Selys- Longchamp (1910) described neural-tube forma- tion in lampreys as intermediate between the keel method of teleosts and neural-fold method of other vertebrates. Among the morphological features promuient during development were the gut, liver, gall bladder, hemoglobin and vascularization, and pigment. The gut opens from the oral cavity to the anus at stage 18. The stomodaeum opens in stage 16, the esophagus in stage 17, the cloaca in stage 17, and finally the portion between esophagus and cloaca at stage 18. The formation of the hver in late stage 14 is indicated by the greenish cast of the anteriormost portion of the gut. Vascularization of the liver occurs during stage 15. The size of the liver con- tinues to increase through stages 15 and 16. A gall bladder forms in stage 17; it is recognizable externally bj^ its accumulation of blue-green bile. Additional changes in stage 15 embryos include the appearance of hemoglobin within the blood channels which had formed in the yolk-filled gut. Blood cells appear first in the midvcntral channel and soon are in all three major channels. Blood formation and the bulk of vascularization take place in stages 15 and 16. Vascularization is ex- tended in stage 17, when practically all major vessels can be traced by following red corpuscles within the transparent tissues. Pigmentation of the embryo begins as two melanophores bilateral to the midbrain. Succes- sive pairs of melanophores appear posterior to the initial one; n(>xt is the appearance of a melanophore above the branchial region. The number of melanophores increases throughout stage 15 when 130 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE they loosely cover the dorsum and form a hne posteriorly from the branchial region between yolk and somites. In stage 16 melanophores extend along the dorsum and sides of the embryo. Bilateral eyespots appear at the end of stage 16. The melanophores become more extensively dis- tributed in stage 17. Aggregations appear over the pronephric region and the caudal end of the notochord and neural tube. The anterior lip, and the head region and the gut become profusely covered with pigmentation. The pattern of pig- mentation changes little in stage 18 except that the pigmented area is extended. Activity in the embryo included locomotory movements, heart beat, and velum beat. First evidence of activity is the movement of the head region, produced by muscles of somites of stage 13 embryos. Greater muscular movements take place during later stages, including the movements that lead to : hatching; poor swimming movements of stage 14; somewhat better swimming in stage 15; the greatly increased deftness of swimming in stage 16; and movements which accomplish bur- rowing at stage 17. Changes in the heart region in stage 14 embryos can be seen through the ectoderm over the peri- cardial cavity as it becomes transparent. The heart begins to beat at a rate of 40 times per minute, increasing to 100 per minute in stage 15, 150 per minute in stage 16, and more than 200 per minute in stage 17. The respiratory system becomes functional at stage 16 when the initial respiratory rate (velum beat) is 120 per minute. This rate increases to 200 per minute in stage 17 and is maintained at this rate into stage 18. In general, lamprey development resembles amphibian development in cleavage (stages 1-8). Lamprey gastrulation resembles amphibian gastru- lation in some respects and teleostean in others. Epiboly is more like that of amphibians than of teleosteans. Invagination and involution of chorda-mesoderm seem similar to that in amphib- ians. Formation of the lamprey archenteron produces a situation comparable to that of a teleost. In both, the developing embrvo is lo- cated on a large yolk mass but separated from the yolk by an archenteron which was the original blastocoel. A major diflFerence between the two is that lamprey yolk is divided whereas teleostean yolk remains undivided. Another similarity to teleostean development is the formation of a solid neural rod which develops a lumen only after neurulation has been completed. DEVELOPMENT AT DIFFERENT CONSTANT TEMPERATURES The primary objective of the experimental rearing of sea lampreys at a series of constant temperatures was to determine the temperature levels at which the eggs were capable of developing into normal, viable larvae. As part of the work, detailed records were kept on : the relation between temperature and progression of development; mortality rate during development ; occurrence and nature of developmental abnormalities. Infor- mation of this type may help explain the failure, noted both in the LTnited States and in Canada, of the sea lamprey to utilize certain apparently suitable spawning streams as extensively as other apparently similar waters. MATERIALS AND METHODS The experiments were conducted at tempera- tures and with the aid of equipment indicated in table 2. Throughout this work, control lots at 65° F. were maintained as an index to develop- ment. Thus, any one control lot could serve efTectively as an indicator to several other experi- ments. Eggs from 2 to 4 females were mixed with sperm from 4 to 8 males. These eggs were then apportioned into containers in the numbers (by actual count) indicated in table 3. Table 2. — Equipment used in experiments on development at constant temperatures Temper- ature (-F.) Temperature-control equipment Tempera- ture (°F.) Temperature-control equipment 45 . . Refrigeration. Refrigeration. Refrigeration. Refrigeration. Heat and refrigeration. 65 Heat, refrigeration 50 70 (Bronwil circulator). Heat. 52 5 75 Circulator. 55 77.5 80- Circulator. 60 Circulator. The sampling schedule differed somewhat among the experiments (as may be seen from later tables that give details for individual samples) but the differences of schedule and the unavoidable occa- sional interruptions of timing were not sufficient to impair comparisons between series or to hamper the description of the progress of development. In the main, the earlier samples were taken at 1- or 2-hour intervals; the time between samples EMBRYOLOGY OF THE SEA LAMPREY 131 Table 3. — Specimens and number of eggs used in experi- ments on development at constant temperatures (Ono group of lamprevs provided the eggs for experiments conducted at 45°, 55°. 7n°, 75°. and 80° and another group for those at 52.5°. 00°, and 77.5°. Eggs from the lots that supplied materials for experiments at 50° and 65° were reared at the one temperature only] Water temperature Number of females Number of males Number of eggs 45 4 2 2 4 2 2 4 4 2 4 8 4 4 8 4 4 8 8 4 8 10,000 10.000 52 5 1.5,000 55 , _ 25,600 60 9.000 65 - - 33.000 70 18.000 13.500 77 5 9.000 80 .1.000 was increased to 4 lioiirs at about 32 hours; later samples (that is, after 2-3 days) were obtained about 12 hours apart; and, finally in the lonpjer experiments sampling was daily. Sampling after stage 14 appeared was biased because the pro- larvae had to be pursued as they became older; thus the samples were non-representative in com- parison with the random samples for all earlier stages. In the main the bias led to over-repre- sentation of live specimens. Several criteria were used to separate live and dead embryos. The most obvious indication of deatli was disintcErration of the embryo- a sepa- ration of cells and subsequent filling of the intra- membrane volume with loosely arranged cells, within an intact membrane. Furthermore, the fertilization membrane became translucent «itli a cloudy cast in contrast to the transparency of a living membrane. This difference was most apparent in later development, stages 7-14; tiie earlier stages (1-6) became vacuolated as they underwent changes after death. Eventually, the membranes of eggs in these earlier stages also be- came cloudy. Another indicator, particularly in early development, was a change of color from tlie creamy white of normal eggs to a brownish tan, accompanied by a fuzzy appearance of the surface. Dying embryos of stages 9-11 possessed a])preci- ably widened blastopores, which in some e.xteiided across the entire diameter of the embryo. The basic data on tlie several temperature experiments are given in the records of number of dead embryos and number of living embryos in various stages of development. In conjunction with these records, information was recorded relative to the elapsed time and tetiiperaturc and remarks were noted concerning the general com- position of each samj)le. The percentage of dead embryos was computed for each sample. Until stage 14 appeared this percentage provided an estimate of mortality up to the time of sampling since all embryos regard- less of time of death were included in the computa- tions. This procedure was possible since all eggs used in any one experiment were recoverable throughout the experiment. Although disinte- gration of dead embryos did occur, the embryonic membrane remained intact to the end of the ex- periment. The information on dead embryos per sample does not include data on time of death, since stages could not be determined for embryos that were decomposing, but it does provide good information on the progression of mortality with development. Because, as has been stated, the quantitative sampling of lots of eggs become biased after the appearance of stage 14 (hatching) and later stages, none of the tables that carry details on individual samples goes beyond the last sample that con- tained stage 13 embryos (the bias starts, of course, with the first sample containing stage 14 speci- mens, but full records for stage 13 appear to be desirable). In the experiments in which stage 14 was not reached, the tabulation ends with the last sample that contained live embryos. Similarly, the tables showing mortality by 1-day intervals end with the records for the first day on which stage 14 appeared, or, if that stage was not reached, with the last day on which samples contained live embryos. Terminal samples are included for those experiments in which some embryos survived the full term of the experiment or in which there was cause to suspect that a few live embryos might still be present. These terminal samples which con- tained all eggs remaining at the end of the experi- ment are considered, for practical purposes to be unbiased. They were affected to some degree by the earlier removal (after the start of stage 14) of samples in which living embrj-os were taken out of proportion to their true abundance, but the numbers were so much greater in the terminal sample than in these earlier biased samples as to minimize the distorting effect. Records from biased samples were of course use- ful for sliowing the time of first appearance and duration of the later stages. 132 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE ACCOUNT OF INDIVIDUAL EXPERIMENTS Development at 45° F. The sea lamprey eggs could not make even a good start toward development in a constant temperature of 45° F. The first two cleavages (stages 1 and 2) proceeded very slowly and all eggs were dead before the third cleavage was completed (tables 4 and 5; fig. 27). Table 4. — Living and dead lamprey embryos per sample and their stage of development at 45° F. {7.2° C.) Sample Num- ber of hours Num- ber dead Stage or living embryos Sample Num- ber of hours Num- ber dead Stage of living embryos 1 2 3 1 2 3 1. 1 3 5 7 9 11 13 15 17 19 21 23 25 67 15 10 13 16 17 11 73 38 82 45 45 62 213 105 94 125 169 120 113 92 233 84 76 14 15 16 17 18 19 20 21 22 23 24 25 27 31 35 39 43 47 51 58 71 82 95 103 91 105 120 123 85 177 207 175 262 322 363 341 — - 23 57 23 32 16 22 8 2 3 4 6 8_. ? 7 4 8 in 9 14 10 33 109 101 63 3 11 ■> 12 9 13 100 80 60 40 20 100 80 60 40 20 iioo ;eo I 60 40 ! 20 J ■100 80 60 40 20 50°F 52.5°F 60°F 15 20 25 30 35 40 DAY OF SAMPLING Table 5. — Mortalities of sea lamprey eggs reared at a constant temperature of 4S° F. {7.2° C.) Time stage span Number of samples Number of embryos Hours Days Alive Dead Total age dead 0-24 25-48 49-72 73-96 97-120 1 2 3 4 5 0-2 2-3 2-3 3 3 12 7 3 2 2 1,667 238 36 5 2 432 763 634 685 579 2,099 1,001 670 690 581 21 76 95 99 '100 1 More than 99.6; actually all embryos (238 specimens) were dead in the sample taken at H6 hours. Stage 1 lasted about 20 hours; stage 2 com- menced at the 19th hour and lasted beyond the 47th hour, when a highly defective stage 3 ap- peared (tables 4 and 6; fig. 28). Many 3-celI embryos were found along with a few abortive 4-cell embryos, in which the cleavage furrow seemed to undergo regression. Throughout the samples, beginning with the sixth at 1 1 hours, the eggs began to vacuolate and to fragment yolk into the perivitelline space, in short, to die and disintegrate. Beginning with the 19th sample, at 47 hours nearly 90 percent 100 80 60 40 20 100 80 60 40 20 100 80 60 40 20 100 80 60 40 20 100 80 60 40 20 65° F I I ' I I I ' I ' 70° F 75°F 77.5°F 80°F 10 15 20 25 DAY OF SAMPLING I I I ' I i I ' I I ' I ' 30 35 40 Figure 27. — Percentage of dead embryos in daily samples from sea lamprey eggs reared at 10 different constant temp- erature levels. The day-by-day records end when the percentage of dead in the samples reached and remained at 100, or with the onset of hatching. A broken line connects the last recorded daily sample with the terminal sample (T). At 77.5° and 80° live embryos occurred in the earlier samples of the first day but none in the later ones. EMBRYOLOGY OF THE SEA LAMPREY 133 Table 6. — Hours at which the first and last specimens of the various stages appeared in samples during each experiment and the duration of stages {in parentheses) stages Constant temperature (F.) at which embryos were reared 45° 50° 52.5° 55° 60° 65° 70° 76° 77.6° 80° I 1-21 (21) 19-51 (33) 47-103 (57) 1-16 (16) 6-25 (20) 19-34 (16) 28-42 (15) 34-48 (15) 40-58 (19) 44-78 (35) 62-126 (65) 114-270 (167) 1-10 (10) 3-22 (20) 13-28 (16) 20-36 (17) 32-44 (13) 40-i8 (9) 48-64 (17) 68-136 (69) 160-338 (179) 338-362 (25) 362^70 (109) 470-570 (101) 1-13 (13) 3-23 (21) 15-31 (17) 23-35 (13) 31-43 (13) 39-47 (9) 47-103 (57) 51-151 (101) 127-271 (145) 211-295 (185) 295-105 (111) 369-525 (167) 1 490-874 ' ( .-) 1-6 (6) 3-10 (8) 13-16 (4) 16-18 (3) 20-22 (3) 22-28 (7) 28-34 (5) 36-89 (54) 100-148 (49) 136-172 (37) 172-220 (49) 208-283 (76) 283-409 (127) 317^09 (93) 352-469 (118) 446-569 (124) ' 694-782 '( ) 1-2 (2) 2-8 (7) 8-11 (4) 10-15 (6) 13-15 (3) 16-19 (4) 19-24 (6) 28-64 (37) 64-104 (41) 104-128 (25) 128-152 (25) 152-200 (49) 200-296 (97) '248- ' ( ) '308- '( ) '363- '( ) 1 405-437 «( .) 1-3 (3) 6-7 (3) 7-11 (5) 11-13 (3) 13-15 (3) 15-17 (3) 17-21 (5) 19-51 (33) 47-82 (36) 82-103 (22) 103-127 (25) 116-151 (36) 151-235 (85) 175-247 (73) 225-307 (83) 295-343 (49) ' 357-633 ' ( ) 1-3 (3) 5 (2) (2) 7-11 (5) 9-13 (5) 13-15 (3) 16-17 (3) 19-43 (25) 43-82 (40) 71-103 (33) 95-116 (22) 103-139 (37) 151-176 (25) 176-211 (37) 211-226 (16) 1 226-307 ' ( ) 1-4 (4) 4-6 (3) 6-10 (5) 10 (4) >13 1-7 2 (7) 5-9 3 (5) 9-11 4 (3) (') 5 C) 6 »13 s 15 7 8 9 10 11 12 13 14. 15 16 - 17 ' Termination of experiment. ' 100-percent mortality subsequent to this hour. J No embryo of this stage found in any sample. ' These blanks cannot be filled properly for these stages were sampled either in or immediately preceding the terminal sample. 65° TEMPERATURE (°F) FincRE 28. — Number of hours required by sea lamprey embryos to reach different developmental stages when reared at variou.s constant temperatures. of the eggs were dead. All eggs obviously were dead in samples taken after the 5th day (table 5). Eggs marked by lesions as described by Damas (1948) did not undergo first cleavage. The highest stage completed in this experiment was stage 2, since stage 3 was not consummated. Development at 50° F. The rate of development was decidedly faster at 50° F. than at 45° F. (table 6; fig. 28) and devel- opment proceeded farther. Stage 9 was reached but few embryos survived to the end of the stage (tables 7 and 8; fig. 27). A single sample con- tained eggs that seemed to be intermediate between stages 9 and 10. No embryos were alive after the 270 hours (sample 65, table 7; table 6, also fig. 27). The duration of each of the first 6 stages fell within the limited range of 15-20 hours (table 6). Stage 7, however, lasted 35 hours, and stage 8 approximately 65 hours, the longest save for tlie last stage reached (stage 9). The greatest period overlap of stages occurred between stages 7 and 8 134 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 7. — Living and dead embryos per sample and their stag i of development at 60° F. {10° C.) Sample Hours Number dead Stage of living embryos Sample Hours Number dead Stage of living embryos 1 2 3 4 5 6 4 5 6 7 8 9 1 Vz 1 2 3 4 6 7 8 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 32 34 36 38 93 140 215 184 133 200 157 132 108 129 132 112 31 42 14 34 - 40 42 44 46 48 50 54 68 62 66 70 74 78 90 102 114 126 137 146 150 163 171 174 182 187 198 211 222 234 246 260 270 3 3 17 14 10 13 25 11 15 14 13 35 71 32 87 42 32 93 43 123 119 118 207 169 232 337 315 283 254 375 435 634 7 5 60 19 69 49 8 5 89 232 187 83 108 41 17 2 35 --- 3 36- 8 5 30 107 163 132 87 13 11 5 2 37 - 5 38.-- 1 1 20 43 34 30 18 27 68 84 139 111 139 151 283 304 271 141 72 47 13 39.-- -.- 7 40 - 41 9 1 42 183 125 271 335 236 238 119 209 77 43.- 44 12 45 - 13 46 14 47 1 6 6 8 11 12 8 10 6 7 6 4 48 16 49 7 17 50 - 12 2 62 127 213 210 262 95 117 72 48 50 137 90 26 51 217 19 52 57 53- - 125 21 54 41 55 210 23 56 135 24 57 - 205 68..- 134 69.-- 106 27 12 48 190 344 453 43 24 60 123 61 82 29 18 23 19 62 25 63 10 31 23 60 80 64.-- 2 65 6 33 3 Table 8. — Mortalities of sea lamprey eggs reared at a constant temperature of 60° F. (10° C.) Time Stage span Num- ber of samples Number of embryos Per- centage Hours Days Alive Dead Total dead 0-24 1 2 3 4 5 6 7 8 9 10 11 12 0-3 3-7 6-8 7-8 8-9 8-9 9 9 9 9 9 9 23 15 6 3 2 2 3 4 2 2 2 2 4,680 2,781 1.258 816 335 306 223 684 229 107 12 6 71 126 91 138 129 125 285 726 652 537 810 1,222 4,751 2,907 1,349 954 464 431 608 1,410 881 644 822 1,228 2 28-48 4 49-72 7 73-96 14 97-120 121-144 28 29 145-168 169-192 193-216—- 217-240 241-264 56 52 74 83 99 265-288 - 99.5 (20 hours) and the next longest between stages 8 and 9 (12 hours). Mortality was light the first 3 days (2 to 7 percent dead; table 8) and was moderate (14 to 29 percent dead) the next 3 days. The death rate increased considerably shortly after the embryos entered stage 9 (gas- trula). Samples on the seventh day contained 56 percent of dead eggs. The percentage had reached 99 and 99.5 percent on the 11th and 12tli days, respectively, and all eggs were dead on the 13th day. Development at 52.5° F. Again, an increase of temperature (this time only 2.5° above that of the experiment described in the preceding section) permitted development at a faster rate and through a greater number of stages (table 6; fig. 28). Heavy mortality started early, however, the death rate increased as devel- opment progressed (tables 9 and 10; fig. 27), and no embryos survived beyond stage 12 (head stage). The early stages (1 to 6) of this experiment pro- ceeded slightly more rapidly than the correspond- ing stages of the 50° F. experiment. The periods of overlap of the earlier stages were nearly equal to those of the 50° F. test, although the length of time for each stage was less at 50° F. than at 52.5° (tables 6, 7, and 9). Stage 9 was prolonged at 52.5° F. over a period of approximately 9 days. During this time the number of deaths increased. The relatively few embryos that developed beyond stage 9 progressed through a short stage 10 and a much longer stage 11 (about 4 days). Stage 12 was reached by approximately 40 embryos, all of which died. The most frequent symptom of death was disintegration of the anterior region of the embryos. Development at 55° F. Development at 55° F. proceeded as far as stage 13 (prehatching). Indeed, the terminal sample included 11 live embryos in this stage (along with some 16,000 dead embryos) but all of them were so defective as to make early death almost certain; they were accordingly classed as EMBRYOLOGY OF THE SEA LAMPREY Table 9. — Living and dead embryos per sample and their stage of development at 62.5° F. {n^" C) 135 Sample Houre Number dead Stage of living embryos 1 2 3 4 5 6 7 8 1 3 5 6 8 10 13 14 16 18 20 22 24 26 28 32 36 W 44 48 54 64 76 89 100 112 124 2 11 15 7 8 21 31 47 38 74 27 37 48 33 39 29 44 49 66 29 51 65 21 46 32 72 50 76 29 23 76 36 31 o 2 4 3 2 1 IS 23 9 12 2 2 3 5 6 7 1 2 1 3 10 12 7 2 2 8 9 10 1 3 3 10 16 13 2 12 13 14 15 16 12 11 2 2 17 18 10 15 4 19 21 18 5 ...... 9 20 13 21 5 21 23 24 25 27 Sample 28. 29. 30. 31. 32 33. 34. 35. 36 37. 38. 39. 40. 41. 42. 43. 44. 45 46 47, 48, 49, 50, 51, 52, 53 Hours 136 148 160 172 814 196 281 230 242 254 269 293 302 327 338 349 362 375 386 419 447 470 494 518 543 570 Number dead 142 37 33 58 57 76 44 70 65 86 40 75 78 53 64 96 85 96 117 44 83 98 68 79 92 Stace of living embryos Table 10. — Mortalities of sea lamprey eggs reared at a constant temperature of 52.5° F. {11.4° C.) Time Stage span Number of samples Number of embryos Per- centage Hours Days Alive Dead Total dead 0-24 25^8 49-72 73-96 97-120 121-144 145-168... 169-192 -. 193-216 1 2 3 4 5 6 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 0-4 3-7 IS 8 8 8 9 9 9 9 9 9 9 9 9-10 10-11 11 11 li' 12 12 12 12 13 7 2 2 2 2 2 2 1 2 2 1 2 2 2 1 1 1 1 1 1 1 389 122 25 29 34 11 10 32 13 22 29 6 16 5 9 8 2 2 4 3 3 1 1 366 289 116 67 104 192 70 115 76 114 151 40 153 53 160 181 70 117 44 83 98 68 79 92 755 411 141 96 138 203 80 147 89 136 180 46 169 58 169 189 72 119 44 87 101 71 80 93 48 70 82 70 75 95 88 78 85 217-240 . . 84 241-264 265-288 84 87 289-312 313-336 337-360 361-384.. 91 91 95 96 385-408 409-132 433-156 457-180... . 97 98 100 95 481-504 505-628 529-552 97 96 99 553-576 99 "dead" in the records for that sample (tables 11 and 12; fig. 27). The incidence of dead embryos in the samples increased rather consistently as development proceeded and was high in tlic later stages. Overlapping was prominent between later stages since individual stages lasted from 4 to 16 days (table 6; fig. 28). Stage 13, the highest stage reached, had lasted 384 hours (16 days) wlien the experiment was finally terminated at 874 liours (ca. 37 (hiys). Stages 12 and 13 overlapped 36 hours. A most interesting feature of this experiment was the condition of embryos that finally reached and remained in stage 13. Although they were in stage 13, they were developing witliin the fertiliza- tion membrane (chorion). The spirally curled embryos developed transparent pericardia, pig- ment spots, and hemoglobin which had a muddy red to brown appearance in contrast to the normal bright pink to red. Some of these embryos finally were released wlien tlicir membranes disintegrated. After release, however, the embryos did not straighten out or develop further but remained in this condition until death. During tliis test several abnormalities were noted in the embryos: enlarged pericardia; straight tubular hearts; shorter but heavier bodies; en- larged gut region; a separation of yolk cells from the gut walls; and failure to straighten from the spirally curled position after removal of the chorion. Development at 60° F. Tlie rate of dcveloiimcnt was much more rapid and the overlap of stages was less at 60° F. than at lower temperatures (tables 6, 13, and 14; fig. 28). The temperature of 60° F. was the lowest at which viabU\ burrowing prolarvae (stage 17) were produced. Mortality was generally less than on corresponding (hiys in experiments at lower temperatures (fig. 27) and abnormalities were relativelv few. 136 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 11. — Living and dead embryos per sample and their stage of development at 65° F. {12.8° C.) Sample Hours N'umber dead Stage of living embry OS Sample Hours Number dead Stage Of living embryos 1 2 3 4 5 6 7 8 9 9 10 11 12 13 1 1 3 5 7 9 U 13 15 17 19 21 23 25 27 31 35 39 43 47 51 58 71 82 95 103 116 127 139 151 163 143 80 91 27 64 32 9 31 175 187 199 211 225 236 247 260 271 283 295 307 319 331 343 357 369 381 393 405 417 432 456 465 490 501 512 525 638 549 91 141 146 133 80 49 54 61 30 47 41 83 47 37 64 71 97 86 63 56 62 48 108 73 97 89 99 148 71 92 26 34 41 27 21 5 6 3 2 2 8 12 53 67 114 92 91 31 66 26 5 32.. ._ 33 34 35. 36 3 4 3 14 16 10 19 13 12 8 5 6 7 37 8 22 48 14 15 47 61 83 81 81 81 62 87 70 80 118 157 151 111 133 51 89 20 226 19 91 65 40 34 26 16 4 38 9 39.... 10 40 41... 42 11 10 14 18 11 15 12 17 12 3 2 12 48 40 35 52 9 13 43 14 44 45 46... 47 16 23 75 25 11 16 17 57 16 25 2 9 5 4 6 9 11 19 12 7 2 3 48 19 52 36 50 50 38 40 28 49. 20 39 77 75 83 81 98 135 34 82 162 """49" 59 68 61 50 21 51 22 52 23 53 54 55 26 6 26 56 12 27 57 58 6 16 29 59 18 30 60.. 12 Table 12. — Mortalities of sea lamprey eggs reared at a constant temperature of 55° F. (12.8° C.) [No sample on 24th day] Time Stage span Number of samples Number of embryos Per- centage Hours Days Alive Dead Total dead 0-24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 25 26 27 28 29 30 31 32 33 34 .35 36 37 37 0-4 4-7 7-8 7-8 7-8 8-9 8-9 9 9-10 »-10 9-10 9-10 10-11 11 11 11-12 11-12 12 12 12 12-13 12-13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 12 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.298 466 327 242 261 224 291 60 71 56 38 27 32 29 27 40 14 15 11 19 37 26 30 15 18 12 5 12 9 10 6 4 6 8 8 3 146 536 268 308 244 140 246 232 279 129 115 77 124 84 135 183 119 110 108 73 186 247 163 113 93 125 104 126 86 108 61 109 103 108 128 105 16,250 1.444 1.002 595 560 506 364 637 292 350 185 153 104 156 113 162 223 133 125 119 92 223 273 193 128 111 137 109 138 95 118 67 113 109 116 136 108 16,250 10 26-48 -. 54 49-72 45 73-96 56 97-120 - - ... 48 121-144 -.- 38 145-168 - 169-192... 193-216. .- 46 79 80 217-240 70 241-264 75 265-288 289-312 313-336... 337-360 74 79 74 83 361-384 82 385-408 409^32 433-456 89 88 91 467^80 481-504 79 83 505-528 90 52ft-552 84 577-600 88 601-624 84 625-648.. 91 649-672 95 673-696.. 91 697-720 721-744 --. 745-768.. . 91 92 91 769-792 96 793-816 817-840 841-864 94 97 865-888.. . . 875 ' 100 I Terminal sample includes 11 live embryos mentioned in text as being incapable of survival. Overlapping of stages was most limited with the exception of stages 1 and 2 (overlap of 3 samples) and the last 2 (stages 13 and 14) (overlap of 7 samples). Stage 13 was longest (127 hours or about 5 days). Stage 8 lasted more than 54 hours and was also represented in 8 samples (tables 6 and 13). Stage 14 was attained within 317 hours (about 13 days for first appearance). Percentages of dead embryos in samples rose from a low of 32 percent during early cleavage stages (1-6) to 88 percent in the terminal sample. The rate of increase was highest in the early stages actually during the first 2 days. Percentage hatch (embryos that survived through stage 14) though not accurately measurable from the biased samples, obviously was good. The relatively few abnormalities during this experiment (appro.ximately 20 percent) took several forms, some of which were similar to those described earlier. Among the more common were enlargement of the pericardial area, and the straight tubular heart which, nevertheless, maintained a regular beat. Other specimens exhibited abnormal curvatures, balloon mouths, or cleft-lip. Abnormal curvatures of the trunk region produced embryos with "C," "J," "O," and "L" shapes. Balloon mouths were caused EMBRYOLOGY OF THE SEA LAMPREY Table 13. — Living and dead embryos per sample and their stage of development at 60° F. (16.6° C.) 137 Sample Hours Number dead Stage of living embryos Sample Hours Number dead Stage of living embryos 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 3 4 6 8 10 13 14 16 18 20 22 24 26 28 32 36 40 44 48 54 64 76 89 143 31 14 8 25 100 112 124 136 148 160 172 184 196 208 220 232 244 259 283 292 317 328 339 362 366 376 409 20 22 26 35 23 36 29 31 55 23 37 68 40 49 45 46 51 36 35 5 IS 13 14 19 5 2 13 20 66 38 26 26 3 27 4 12 11 13 17 22 29 37 24 22 S3 36 31 34 30 28 28 31 17 26 16 31 28 6 7 23 11 5 29 6 30 7 26 25 9 31 7 14 5 9 2 8 32 9 14 29 33 10 .. . . 34 6 12 18 20 11 9 11 17 23 35 12 3 23 11 4 36 13 37 14 38 15 13 19 "'"12' 10 23 26 15 15 13 5 39 16 10 20 4 8 1 5 16 40 17 41 42 4 18 19 43 12 20 44 6 21 45 . . 3 22 46 3 32 23 47 4 11 24. Table 14. — Mortalities of sea lamprey eggs reared at a constant temperature of 60° F. {16.5° C.) [Xo sample on 17th day] Time Stage span Number of samples Number of embryos Per- Hours Days Alive Dead Total dead 0-24.. 1 2 3 4 6 6 7 8 9 10 11 12 13 14 16 16 18 19 20 21 22 23 24 26 26 27 28 29 30 31 32 33 33 0-6 6-8 8 8 9 9-10 9-10 10-11 11-12 11-12 12 12-13 13 13-14 13-15 13-15 13-16 16-16 15-16 16 16 16 16 17 17 17 17 17 17 17 17 17 17 13 7 2 2 2 2 2 2 2 2 2 1 1 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 518 118 30 18 28 39 35 32 20 32 31 25 10 240 218 43 47 42 61 59 60 78 105 89 45 46 758 336 73 65 70 100 94 92 98 137 120 70 56 120 76 31 70 22 20 16 17 42 19 18 24 35 17 12 14 8 16 7 6,401 25-48 . 65 49-72 69 73-96 72 97-120 -. . . 60 121-144.. 145-168 169-192 193-216 61 63 65 80 77 217-240 241-264...- 266-288 289 312-. .- 54 313 336 337-360 - 361-384 409-432 433-466 457-480...- 481-504 - 505-528 629-552 663-576 677-600 601-624 --- 625-648 649-672 673-696 697-720 721-744 745-768 769-792 783 1 795 5,606 88 ' Terminal sample. from a fiiihirc of the stomiulaoiiin to open. An enormous enlargement or ballooning of the pharyngeal region resulted, especially in the anterior portion, which tended to compress the branchial structures and cause them to become malformed and from all indication, non-functional. A midline cleft in the dorsal lip was designated a "cleft-lip" abnormality. Development at 65° F. The rearing of sea lamprey eggs under constant temperature was more successful at 65° F. than at any lower or higher level (tables 15 and 16; fig. 27). Development was rapid (tables 6 and 15; fig. 28), overlapping of stages was practically nil, mortality was low, and abnormalities were few. The details offered in this section are based on an experiment conducted under refrigera- tion. Essentiall}' the same results were obtained in experiments conducted with heat and the circulator. Stages did not overlap at 65° F. with the exception of the overlap of stages 4 and 5 and of stages 13 and 14. The percentage of dead embryos was low in most samples but the day-to-day variation of that percentage (fig. 27) was highly erratic and for several dn,ys the percentage exceeded that of the terminal sample taken on the 19th day. No verifiable explanation can be offered for tliis behavior of the percentages, but it is possible that dead eggs were not distributed evenly through the experimental lot and that certain samples happened to be taken from points at which dead embryos were concentrated. The 78-percent sur- vival to the end of tlie experiment (only 22 percent dead in the terminal sample) was by far the highest 138 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 15. — Living and dead embryos per sample and their stage of development at 66° F. {18.4° C.) Sample Hours Number dead Stage of living embryos Sample Hours Number dead Stage of living embryos 1 2 3 4 5 6 7 8 8 9 10 11 12 13 14 1 H 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 24 28 29 30 31 195 330 112 26 27_ - 28_ _ 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 32 45 53 56 64 69 80 93 104 116 128 142 152 164 176 188 200 212 224 236 248 271 285 296 10 9 22 81 91 82 65 65 231 181 113 284 80 6 118 179 51 107 100 97 38 3 5 6 208 218 227 247 7 3 1 279 224 433 272 287 11 189 304 337 365 96 6 2 1 2 3 6 3 3 12 11 9 14 14 4 14 11 24 19 23 7 g 500 169 376 51 78 161 99 10 1 307 315 303 36 4 194 61 74 12 258 139 231 135 64 14 62 381 274 15 16 17 346 329 97 19 171 170 60 84 39 20 10 19 18 19 20 324 245 "476" 361 329 334 21 43 135 23 129 24 205 26 Table 16. — Mortalities of sea lamprey eggs reared at a constant temperature of 65° F. (.18.4° C) [No samples on 15th and 18th days) Time Stage span Number of samples Number of embryos Per- centage dead Hours Days Alive Dead Total 0-24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 16 17 19 19 0-7 8 8-9 9 9-10 10-11 11-12 12 12-13 13 13-14 13-14 13-15 14-15 15-16 15-17 17 14-17 21 6 4 2 2 2 2 2 2 2 1 2 2 2 1 2 1 6,283 1,926 974 702 335 354 471 366 405 144 108 92 276 130 412 397 86 297 158 197 6,391 2.018 1,250 832 747 751 557 663 563 341 120 302 384 271 31 117 114 15, 273 2 25-48 49-72 73-96 5 22 16 97-120 55 121-144 ... . 53 145-168 169-192 193-216 15 45 28 217-240 58 241-264 - 265-288 289-312 313-336 361-384 385-408 433-456 4381 11, 918 3,355 22 ' Terminal sample. among the experiments and represents unusually good- results for the artificial rearing of fish eggs of any kind. Abnormalities were extremely few, appi'oxi- mately 2 to 5 percent, mostly in the last samples. A few specimens had enlarged pericardia with a straight, tubular heart, and several had yolk separation and the associated hydrocoelus gut. One "twin" embryo was seen. These twins, in early stage 12, possessed a common but somewhat enlarged blastopore (fig. 29). Development at 70° F. The rate of development at 70° F. was only slightly accelerated over that at 65° F., in fact Figure 29. — Posterolateral view of twin embryo.s, showing the common, but enlarged, blastopore. embryos reared at 70° F. reached stages 2 and 4 later than did those reared at 65° F. (table 6; fig. 28), and the limited overlap of stages charac- terized embryos at both temperatures. More pronounced and more significant were the increase of mortality (tables 17 and 18; fig. 27) and the greater number of abnormalities at 70° F. as compared with the "optimum" of 65° F. At 70° F. as in the 65° test, the overlapping of stages was limited to a slight overlap between stages 11 and 12 and a somewhat greater one between stages 13 and 14 (tables 6 and 17). Elsewhere, the progression from one stage into the next was precise. Stage 8 was prolonged over 11 samples covering 32 hours, as against 2 to 3 for all other early stages. EMBRYOLOGY OF THE SEA LAMPREY 139 Table 17. — Living and dead embryos per sam pie and their stage of development at 70° F. {21.1" C.) Sample Hours Numoer dead Stage of living embryos Sample Hours Number dead Stage of living embryos 1 2 3 4 5 6 7 8 8 9 10 11 12 13 14 1 2 3 1 3 5 7 9 11 13 15 17 19 21 23 25 27 31 35 39 43 277 195 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 47 51 58 71 82 95 103 116 127 139 151 163 175 187 199 211 225 235 45 38 65 92 89 325 221 258 44 206 139 195 64 48 25 46 24 24 99 48 24 61 122 160 62 130 84 4 5 50 8 13 43 31 39 42 41 26 40 34 38 28 38 16 74 121 5 31 13 58 6 7 124 56 93 58 61 41 12 69 45 6 53 30 5 9 10 96 76 39 ...... 81 151 113 122 138 123 135 71 35 87 33 15 18 8 1 1 12 13 14 20 15 26 44 17 32 41 Table 18. — Mortalities of sea lamprey eggs reared at a constant temperature of 70° F. (21.1° C.) [Xo sample on 24th day] Time Stage span Number of samples Number of em bryos Per- centage Hours Days AUve Dead Total dead 0-24 25-48 1 2 3 4 5 6 7 8 9 10 U 12 13 14 15 16 17 18 19 20 21 22 23 25 26 27 27 0-8 8-9 8-9 9-10 10-12 11-12 12-13 13-14 13-14 13-15 14-15 15 15-16 16 16-17 17 17 17 17 17 17 17 17 17 17 17 17 12 7 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 2 2 1 1 1 1 1 1,845 825 391 106 166 95 127 293 239 195 414 479 250 334 2,138 1.064 586 520 645 345 461 193 167 160 149 124 118 32 69 91 98 21 68 109 65 59 28 29 5 60 10.819 14 22 49-72 33 73-96 80 97-120 121-144 145-168 169-192 74 72 72 193-216 217-240 241-264 265-288 ^- . 289-312 313-336 337-360 361-384 385-408 409-432 433-456 467-480 ^ 481-504 505-528 529-552 577-600 601-624 - 625-648 634 ' 628 10,291 96 1 Terminal sample. The percentage of dead embryos in the samples rose rapidly to 80 percent on the fourth day and continued above 70 percent through the seventh day after which hatching made the further samples undependable as measured at mortality (table 18; fig. 27). The percentage of embryos dead in the 70°-terminal sample (95 percent) indicates final production a little less than half that at 60° F. (88 percent dead in 60° terminal sample). Abnormalities in this experiment were relatively few (15 to 20 percent of the prolarvae), but more numerous than at 65° F. The abnormalities were similar to those listed for the 60° F. test, especially with respect to body shape. Development at 75° F. No viable, burrowing larvae were produced at a constant temperature of 75° F. Heavy mortality appeared early and persisted throughout develop- ment (tables 19 and 20; fig. 27) with the result that few embryos survived even to hatching. With the exception of stages 2 and 3 the various Table 19. — Living and dead embryos per sample and their stage of development at 75° F. (23.9° C.) Sample Hours .\umber dead Stage of living embryos Sample Hours Number dead stage of living embry OS I 2 3 4 5 6 7 8 8 9 10 11 12 13 14 1 1 3 5 7 9 11 13 15 17 19 2! 23 25 27 31 35 129 200 17 18 19 - 20 21 22 23 24 25 26 27 28 29 30 31 39 43 47 51 58 71 82 95 103 116 127 139 151 163 175 54 60 207 153 268 260 322 188 287 124 202 210 265 176 106 69 17 2 -. .. 25 25 102 186 71 81 88 85 172 191 103 53 103 163 105 36 3 113 4 57 1 40 12 64 81 45 8 5 2 148 113 6 4 23 21 4 7 40 133 8 81 177 ""175" 102 62 89 80 89 202 27 26 14 9 3 37 13 7 10 11 12 13 14 8 1 14 15 11 16 140 FISHERY BXJLLETIN OF THE FISH AND WILDLIFE SERVICE Table 20. — Morlalities of sea lamprey eggs reared at a constant temperature of 75° F. (23.9° C.) Time Stage span Number of samples Number of embryos Per- centage Hours Days Alive Dead Total dead 0-24 25-48 49-72... 73-96 1 2 3 4 5 6 7 8 9 10 11 12 13 0-8 8-9 9-10 9-11 10-12 12 13 13-14 14-15 15-16 16 16 12 7 3 2 2 2 2 2 I 2 1 1 1,585 582 184 79 84 20 22 1,129 745 681 510 411 412 441 2,714 1,327 865 589 495 432 463 325 355 265 94 27 6,600 42 56 79 87 97-120 83 121-144... 145-168.. - 169-192 95 95 193-216 217-240 241-264 265-288 3071 5,600 100 ' Terminal sample. stages were reached more rapidly than at lower temperatures and the overlap of stages was limited (tables 6 and 19; fig. 28). Transition from one stage into the next was rather precise, as in the two preceding experiments. No stage overlapped another in more than 2 samples (tables 6 and 19). Stage 8 was projected over 9 samples covering 24 hours as contrasted with 2 or 3 for each of the earlier stages. Substantial mortality appeared early at this temperature, especially between stages and 8. The percentage of dead embryos had reached 95 percent on the seventh day, and the terminal sample did not include any live embryos. Pro- duction at this temperature was nil, despite the survival of some individuals as far as stage 16. It is conceivable that stage 17 possibly could have been reached since the experiment was termi- nated when the supply of stage 16 prolarvae was exhausted. Were it possible to have allowed development to proceed without sampling, the probability of stage 17 being reached would have been improved. Abnormalities occurred in 35 to 40 percent of the prolarvae. These abnormalities were similar to those already listed except that deformed specimens were especially numerous. E>evelopment at 77.5° F. A constant temperature of 77.5° F. was decidedly above the maximum at which sea lam- prey eggs could develop successfully. Mortality was so great that all embryos were dead after 13 hours and none developed beyond stage 6, 32 cells (tables 6, 21, and 22; fig. 27)' Stage 1 was taken in the 1- to 4-hour samples, stage 2 in the 4- and 0-hour samples, and stage 3 in the 6- to 10-hour samples. Only 4 embryos developed beyond stage 3, and only 3 (stages 5 and 6) were still alive in the sample taken at 13 hours. All were dead in samples from 14 hours to the termination of the experiment at 24 hours. Some stages were reached earlier at 77.5° F. than at 75° F. (stages 2 and 3), but others were attained later at the higher temperatures (stages 4 and 5). See table 6 and ficr. 28. Table 21. — Living and dead embryos per sample and their stage of development at 77.6° F. (25.3° C.) Sample Hours Number dead Stage of living embryos 1 2 3 4 5 6 1 1 3 4 6 8 ID 13 i 10 58 20 64 70 153 67 29 2 3 36 32 4 8 6 . 6 1 7 1 2 Table 22. — Mortalities of sea lamprey eggs reared at a constant temperature of 77.6° F. (25.3° C.) Time Stage span Number of samples Number of embryos Per- centage Hours Days Alive Dead Total dead > 0-24 - -. .. 1 0-6 13 364 708 1,072 66 I All embryos (485) dead in the 6 samples taken after 13 hours. Development at 80° F. Developmient was brief and erratic at a con- stant temperature of 80° F. A few embryos reached the 32-cell stage (stage 6) but most were dead long before that stage was reached (tables 6, 23, and 24; figs. 27 and 28). No signs of cleavage had appeared at the end of the first 3 hours but at the fifth hour practically all of the eggs had begun to dimple, the initiation of first cleavage and stage 2. By the ninth hour all eggs had completed first cleavage and many had started the second cleavage (start of stage 3). At the 11th hour the pattern of cleavage had become rather erratic, since the second cleavage furrow often began while the first was less than half-completed. At times the first cleavage fur- row seemed to regress. Mortality which was slight in the first 2 samples had risen to about 35 percent in the seventh hour. All embryos were dead hi the sample at 13 hours, but the next sample at 15 hours contamed eggs EMBRYOLOGY OF THE SEA LAMPREY 141 that had reached stage 6 (32 cells) but had begun to vacuolate. All subsequent samples had oid}' dead embryos. Table 23. — Living and dead embryos per sample and their stage of development at 80° F. (26.6° C.) Sample Hours Number dead Stage of living embryos 1 2 3 4 5 6 1 3 6 7 9 11 13 15 6 7 85 124 149 165 303 131 107 35 ? 16 131 U 1 4 55 31 <> 6 7 2 Table 24. — Mortalities of sea lamprey eggs reared at a constant temperature of 80° F. '{26.6° C.) Time Stage span Number of samples Number of embryos Per- Hours Days Alive Dead Total deadi 0-24 .- 1 0-6 11 519 1,753 2.272 77 1 Only 2 live embryos were taken after 11 hours and all (914) were dead after 15 hours. SIGNIFICANCE OF OBSERVATIONS Effect of temperature on development The most significant result of the experimental rearing of sea lamprej' eggs at 10 difTerent constant temperatures (ranging from a minimum of 45° F. to a ma.ximum of 80° F.) was the clear demon- stration that successful development through to the production of viable burrowing larvae was possible only within a relatively narrow range. No live larvae were produced at any temperature below 60° F. or above 70° F. Further evidence of the extreme sensitivity of sea lamprey eggs to temperature comes from the much lower survival at 60° F. (12 percent), and 70° F. (5 percent), than occurs at the "optimum" temperature of 65° F. (78 percent). It is to be regretted that experi- ments were not made at 62.5° F. and 67.5° F. to define more clearly the trends within the 60°-70° range, but no further time was available when the importance of tests at these two intermediate temperatures became obvious. Mortality was so heavy at the highest and lowest temperatures that all eggs had died before devel- opment had proceeded beyond very early stages. In general, the highest stage reached increased as the temperature approached the "successful" levels of 60°-70° F. This relationship is brought out by the following listing: Temperature (,F.) Highest stage reached 45°. 3 (4 cells) 50° _ 9 (gastrula) 52.5° 12 (head) 55° 13 (prehatching) 60° 17 (burrowing) 65° 17 (burrowing) 70° 17 (burrowing) 75° 16 (gill clefts) 77.5° 6 (32 cells) 80° 6 (32 cells) Developmental abnormalities were least plenti- ful at 65° F. and increased as the temperature deviated from that value in either direction. In some tests the incidence was high, and the abnor- malities (described briefly in the accounts of the experiments) involved monstrous distortions of the embryos. In general, developmental rate (notably, the time required to reach the various stages) became faster, lengths of stages became shorter, and over- lap betw^een stages was lessened as temperature increased. Some of the exceptions to this state- ment no doubt represent the random variability of the data. Others, as for example, the seem- ingly depressing effect of the highest temperatures on the rate of development in the early stages may reflect a real cause-and-effect relationship. The clear demonstration in the present studies that sea lamprey eggs are capable of full and nor- mal development only within a relatively narrow- temperature range brings out the great importance of controlling temperature at the correct level in developmental studies and experimental research. Consideration of the proper temperature had little place, nevertheless, in past studies of sea lamprey eggs. Authors failed to state the temperatures at which the eggs were reared or reared them at levels at which full, normal development could not be expected. Shipley (1885) did not state the temperature at which his sea lamprey embryos developed and McClure (1893) reared his embryos at 6°-7° C. (42.8°-44.6° F.). Damas (1948) men- tioned temperatures of 12° and 18° C. (53.6° and 64.4° F.) on the development of Lampetra (it is not to be assumed, of course, that the effects of temperature on development are the same for Petromyzon and Lampetra, but certain parallels must be considered highly probable). 142 FISHERY BXJLLETIN OF THE FISH AND WILDLIFE SERVICE Shipley's account of the persistence of the blastopore and his statement that invagination took place at 130 hours suggest that embryos of the lamprey developed at a temperature of about 55° F. It appears then that neither he nor McClure conducted their experiments within the range at which normal development could bo expected. If the relation of temperature to development in Lampetra is similar to that in Petromyzon, Damas' experiments at 18° C. (64.4° F.) should have been at nearly the optimum temperature, but those conducted 12° C. (53.6° F.) were well below the optimum. Thus, certain of the abnormalities (most of them duplicated in the present study, particidarly at the higher and lower temperatures) that he interpreted as the effect of light intensity may actually have been caused by temperature. Although the findings of the present experiments offer the strongest evidence that unsuitable tem- peratures may account for the failure of certain apparently suitable streams to produce larval sea lampreys, a too close application of the results to problems in nature is not advisable. The sea lamprey eggs were reared at constant temperatures in this study, whereas the temperatures in natural streams are subject to diurnal fluctuations, to sub- stantial short-term increases and decreases along with changes of weather, and finally to a longer term, seasonal, upward trend as development pro- ceeds. These fluctuations may have a profound effect on the tolerance of the developing egg. Temperature surely is an important, sometimes a critical, factor in the production of viable larvae in nature, but a good understanding of its opera- tion would require controlled investigations in which eggs develop under fluctuating temperatures which are made to vary much as they do in natural streams. LITERATURE CITED AppLEGATE, Vernon C. 1950. Natural history of the sea lamprey, Pelromyzon marinus, in Michigan. U.S. Fish and Wildlife Service, Spec. Sci. Rept.: Fisheries, No. 55, 237 pp. Applegate, Vernon C, and James W. Mofpett. 1955. The sea lamprey. Sci. Amer., vol. 192, No. 4, pp. 36-41. Applegate, Vernon C, and Bernard R. Smith. 1951. Sea lamprey spawning runs in the Great Lalves, 1950. U.S. Fish and Wildlife Service, Spec. Sci. Rept.: Fisheries, No. 61, 49 pp. Applegate, Vernon C, Bernard R. Smith, and Alber- TON L. McLain. 1952. Sea lamprey spawning run in Great Lakes, 1951. U.S. Fish and Wildlife Service, Spec. Sci. Rept.: Fisheries: No. 68, 37 pp. Augustinsson, K. B., R. Fange, A. Johnels, and E. OSTLUND. 1956. Histological, physiological and biochemical studies on the heart of two cyclostomes, hagfish (Myxine) and lamprey (Lampetra). Jour. Physiol., vol. 131, pp. 257-276, Greaser, Charles W. 1932. The lamprey Petromyzon marinus in Michigan. Gopeia (3), p. 157. Gresticelli, Frederick. 1956. The nature of the lamprey visual pigment. Jour. General Physiol., vol. 39, pp. 423-435. Damas, H. 1944. Recherches sur le developpment de Lampetra fluviatilis. Arch, de Biol., vol. 55, pp. 1-284. 1948. L'influence de la lumiere sur la segmentation et la gastrulation chez Lampetra fluviatilis. Bull. Soc. Roy. Sci. Liege, Seance du 21 Oct., 1948, nos. 7-10, pp. 286-292. 1949. Nouvelles observations sur l'influence de la lumiere sur le developpment embryonaire de Lampetra. Jour. Gytoembryo. Belgo-neerland, Bd. 1949, pp. 96-99. Daniel, J. 1931. Features in ammocoete development. Univ. California Publ , vol. 37, No. 4, pp. 41-52. Gage, Simon H. 1928. The lampreys of New York State — life history and economics. Biological Survey of the Oswego River System, State of New York Conservation Department. Supplement to 17th Annual Report, 1927, pp. 158-191. GUSTAFSON, TrYGVE. 1950. Morphogenetic action of Li ion and chemical basis of its action. Rev. Suisse Zool., vol. 57, pp. 77-92. Hatta, S. 1914. On the mesodermic origin and the fate of the so-called mesectoderm in Petromyzon. Proc. Roy. Soc. London, vol. 88, pp. 457-475. 1915. The fate of the peristomal mesoderm and the tail in Petromyzon. Annot. Zool. Japon., vol. 9, pp. 49-62. HuBBS, Carl L. 1943. Terminology of early stages of fishes. Copeia 1943, No. 4, p. 260. HuBBS, Carl L., and T. E. B. Pope. 1937. The spread of the sea lamprey through the Great Lakes. Trans. Amer. Fish. Soc, vol. 66 (1936), pp. 172-176. HuBBS, Carl L., and Milton B. Trautman. 1937. A revision of the lamprey genus Ichthyomyzon. Misc. Publ., Univ. Michigan Mus. Zool., No. 35, 109 pp. EMBRYOLOGY OF THE SEA LAMPREY 143 Jones, F. R. Hardin. 1955. Photokinesis in the ammocoete larva of the brook lamprey. Jour. Exp. Biol., vol. 32, pp. 492-503. Lennon, Robert E. 1955. Artificial propagation of the sea lamprey, P. marinus, Copeia 1955, No. 3, pp. 235-236. LoEB, Howard A. 1953. Sea lamprey spawning: Wisconsin and Min- nesota streams of Lake Superior. U.S. Fish and Wildlife Service, Spec. Sci. Report: Fisheries No. 97, 36 pp. LoEB, Howard A., and Albert E. Hall, Jr. 1952. Sea lamprey spawning: Michigan streams of Lake Superior. U.S. Fish and Wildlife Service, Spec. Sci. Kept.: Fisheries No. 70, 68 pp. McClure, C. F. W. 1893. Notes on the early stages of segmentation in P. marinus L. Zool. Anzeiger, Bd. 16, S. 367-368; 373-376. MiTCHISON, J. M. 1952. Optical changes in the membranes of the sea urchin egg at fertilization, mitosis and cleavage. Jour. Exp. Biol., vol. 29, pp. 357-362. Sawyer, Wilbur H., and Willard D. Roth. 1954. The storage of biliverdin by the liver of the migrating sea lamprey, P. marinus. Anat. Rec, vol. 120, p. 93. Selys-Longchamp, M. de. 1910. Gastrulation et formation des fuillets chez Pelromyzon planeri. Arch, de Biol., vol. 25, pp. 1-75. Shipley, A. E. 1885. On the formation of the mesoblast and the persistence of the blastopore in the lamprey. Proc. Roy. Soc. London, vol. 39, pp. 244-248. Vladykov, Vadim D. 1949. Quebec lampreys. Department of Fisheries, Province of Quebec, Contribution No. 26, 67 pp. o UNITED STATES DEPARTMENT OF THE INTERIOR, Stewart L. Udall, Secretary FISH AND WILDLIFE SERVICE, Clarence F. Pautzke, Commissioner Bureau of Commercial Fisheries, Donald L. McKernan, Director BLOOD PROPERTIES OF PRESPAWNING AND POSTSPAWNING ANADROMOUS ALEWIVES {Alosa pseudoharengus) By Carl J. Sindermann and Donald F. Mairs FISHERY BULLETIN 183 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 PUBLISHED BY UNITED STATES FISH AND WILDLIFE SERVICE • WASHINGTON • 1961 PRINTED BY UNITED STATES GOVERNMENT PRINTING OFFICE. WASHINGTON, D.C. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington 25, D.C. Price 15 cents Library of Congress catalog card for the series, Fishery Bulletin of the Fish and Wildlife Service : U. S. Fish and Wildlife Service. Fishery bulletin, v. 1- Washington, U. S. Govt. Print. Off., 1881-19 V. in illus., maps (part fold.) 23-28 cm. Some vols, issued in the congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies: v. 1-49, Bulletin. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, v. 1-23) 1. Fisheries— U. S. 2. Fish-culture— U. S. i. Title. SH11.A25 639.206173 9—35239* Library of Congress [59r55blj Paee Introduction 145 Materials and methods 145 Collection of blood samples 145 Determinations of blood properties 146 Comparison of blood properties 146 Hemoglobin 146 Sedimentation rates 147 Erythrocyte fragility 147 Total serum proteins 147 Serum chlorides 148 Serum electrophoretic patterns 149 Summary and conclusions 150 Literature cited 150 m ABSTRACT As part of a general investigation of the potential value of fish blood characteristics to the solution of population and migration problems, a study of modifications induced by environmental and physiological variables has been made. Six blood properties of prespawning and postspawning anadromous alewives (Alosa pscudoharengus) were compared. Changes that could be attributed to fresh-water migration and reproduction were found in only two of these properties ; viz., significant reductions in average serum proteins and chlorides of postspawners. No important differences in average sedimentation rate, erythrocyte fragility, hemoglobin content, or serum electrophoretic pattern were found when fish entering fresh water in May were compared with seaward migrants 1 to 2 months later. Serum electrophoretic patterns were generally similar to those of other elupeoids, with fractious having mobilities comparable with human albumin and human alpha- and beta-globulins, but with little representation in the area of gamma-globulins. Great individual variations in hemoglobin content, total serum proteins, serum chloride, and sedimentation rate were found in alewives both before and after spawning. BLOOD PROPERTIES OF PRESPAWNING AND POSTSPAWNING ANADROMOUS ALEWIVES (ALOSA PSEUDOHARENGUS) Carl J. Sindermann and Donald F. Mairs, Fishery Research Biologists Bureau of Commercial Fisheries Dramatic changes in certain of the blood characteristics of humans and other higher vei-- tebrates often occur under sucli pliysiological stresses as pregnancy or acute disease. It might be anticipated that lower vertebrates, with less precise control of their internal environment, would exhibit equally profound blood changes as the external medium or physiological conditions vary. As an adjunct to serological studies of fishes being carried on at the Bureau of Com- mercial Fisheries Biological Laboratory, Booth- bay Harbor, Maine, it was considered important to assess the extent of environmental and physio- logical influences on blood proiDerties, particularly those which might be related to serological reactions. Erythrocyte antigens, which promise to be of great value in fish population and migration studies, have been examined most extensively in higher vertebrate gi-oups (summarized by Dujar- ric de la Riviere and Eyquem, 1953; Mourant, 1054), where they have been found to be genet- ically determined and unmodified by environ- mental variations. Some evidence for genetic determination of fish erythrocyte antigens has lieen otfered by Hildemann (19.50). Seinim com- ponents, which may also provide information of value to population studies, are in some cases subject to modification by other than genetic factors. For example, antibody production in fishes has Ijeen sliown to vary with external tem- perature (Bisset, 1948) and protein fractions of fisli serum to vary in amount in disease (Sinder- mann and Maii-s, 1958). Because of possible in- Huence of nongenetic factors on serological properties, a stud\' of environmental and physio- logical effects on blood characteristics seemed advisable. Note. — Approved for publication, Maj- 12. 1960. Fishery Bul- letin 183. As part of such a study, this paper is concerned with the manner in which tlie combined stresses of migration from the sea to fresh water and of spawning are reflected in several blood character- istics of the alewife (Alosa pseud oharengus) . The nature and extent of serum changes in pre- spawning and postspawning fish have received particular attention in this investigation, although observations on cellular blood components have been included. MATERIALS AND METHODS COLLECTION OF BLOOD SAMPLES Prespawning and postspawning alewives were sampled in 1958 and 1959 from two separate ]\Laine spawning runs — Damariscotta Mills and "West Boothbay Harbor. Fish were first sampled in May, as they were about to enter fresh water, and again in late June and July, as they were about to re-enter the sea. In addition to the field .samples, pi-espawning alewives taken from both runs were held without food in live cars and sea- water tanks for 2 months before blood samples were taken, to determine the eifect of starvation on electrophoretic characteristics of the serum. The fish were bled by cardiac puncture, using a glass-needle technique developed in this labora- toiy (Perkins, 1957). Blood was collected in screw-top vials as individual samples. Half the samples were collected in vials containing 0.2 milliliter of 6-percent sodium citrate solution, and half were collected in vials without citrate. De- teraiinations of hemoglobin content, sedimenta- tion rate, and erythrocyte fragility were made immediately with the citrated samples. Sera from uncitnited blood samples were decanted after expressing from the clots overnight at 4° C. In- dividual serum samples were kept frozen at —20° C. until determinations of chloride content, total 145 146 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE serum protein content, and elect rophoretic pat- tern wei-e made. Tlie sex and stage of gonad development of each fish were recorded when the blood sample was taken. DETERMINATIONS OF BLOOD PROPERTIES Hemoglobin Content Detennination of the hemoglobin content of in- dividual citrated blood samples was made with the cyanmethemoglobin method, using reagent and standard supplied by Hycel Homione Chemistry Laboratory, Houston, Tex. With a Salili pipette, .02 ml. blood was added to 5 ml. cyanmethemoglo- bin reagent and thoroughly mixed. Contents were transferred after 15 minutes to a cuvette and read colori metrically with a Photovolt Lumetron colorimeter at a wavelength of 530 millimicrons. The colorimeter reading was transferred to a standard hemoglobin curve and hemoglobin con- centration obtained in grams per 100 ml. Sedimentation Rate The rate of settling of erythrocj^es was de- termined with a standard Westergren blood sedi- mentation apparatus. Individual samples of citrated blood was drawn into Westergren pipettes to the 100-millimeter mark and placed vertically in a Westergren rack. The number of millimeters that the erythrocytes had dropped at the end of 1, 2, and 3 houre — the actual sedimentation — was re- corded and multiplied by 2 to make results com- parable with standard tests that use the entire 200- mm. pipette length. Results are expressed in terms of the standard 200-mm. length. Erythrocyte Fragility Sodium chloride solutions ranging from 0.3 to 1.5 percent in 0.1-percent increasing steps were used to test erythrocyte fragility. For each sample, .05 ml. of a 50-percent cell suspension was added to 1 ml. of each saline dilution and readings of "no hemolysis," "partial hemolysis;," or "com- plete hemolysis" were recorded for each tube at the end of 1 hour's incubation at 4° C. Total Serum Proteins The biuret method of Kingsley (1942) was used to determine total serum proteins. This in- volves precipitation with acetone-alcohol followed by addition of biuret reagent. Readings wei-e made with a Photovolt Lumetron colorimeter at a wave length of 530 millimicrons and were plotted against a standard curve prepared with dilutions of clinical chemistry control senmi, supplied by Hyland Laboratories, Los Angeles, Calif., to ob- tain total protein values. Serum Chlorides Determinations of serum chlorides were made by the standard method outlined in the manual for Photovolt Lumetron colorimeter. This involves treatment of serum with tungstic acid, silver iodate, phosphoric acid, and potassium iodide. Readings were made with the Lumetron colorim- eter at a wave length of 420 millimicrons and plotted against a standard curve prepared from known chloride concentrations to obtain serum chloride values. Serum Electrophoresis Elect rophoretic examination of serum was made with a Spinco paper electrophoresis system. Samples were run for 6 hours at 15 milliam- peres, with veronal buffer of pH 8.6 and ionic strength of .05. Pooled human serum was used as a standard with each series. Filter-paper strips were dyed with bromphenol blue and analysed with a Spinco Analytrol densitometer. COMPARISON OF BLOOD PROPERTIES Tests of serum and cellular components of the blood of prespawning and postspawning alewives pro\aded the data presented in table 1. Results from the two Maine spawning nms haA'e been com- bined, since no consistent differences between them were noted. HEMOGLOBIN Variations in hemoglobin content of fish blood have been examined in a number of marine and fresh-water species by several investigators. Black (1955) found that the hemoglobin level of largemouth bass increased after forced exercise, but that of five other fish species did not. Pavlov and Krolik (1936) found that hemoglobin in- creased with the ripening of the sex products, while Naumov (1956) noted that it increased to the time of spawning and then dropped to a very low level. Gelineo (1957) found that hemoglobin values for several species of marine fish were some- BLOOD PROPERTIES OF ANADROMOUS ALEWR'ES 147 Table 1. — Comparison of six blood properties of prespawning and poslspawning alewives [Range In parentheses] Blood property Method of determination Prespawning Number of fish tested Average Postspawnlng Number of fish tested Average Hemoglobin Sedimentation rate... Erythrocyte fragility - Total serum proteins. Serum chloride Serum electrophoresis ' See flg. 4. Cyanmethemoglobin, Lumetron colorimeter Westergren apparatus -- Saline dilutions: 0.3 to 1.5 percent Biuret method of Kingsley (1942), Lumetron colori meter. Lumetron colorimeter Spinco model-R paper electrophoresis system 79 70 60 40 30 121 9.5 g./lOO ml. (4.5-12.5) 4.9 mm. (2.0-9.6) 0.6 percent (0.5-0.7) 5.9 g./lOO ml. (3.9-8.6) 430 mg./lOO ml. (355-458) (I) 9.4 g./lOO ml. (4.0-13.0) 4.7 mm. (1.0-12.0) 0.6 percent. (0.5-0.7) 6.3 g./lOO ml. (2.7-6.9) 395 mg./lOO ml. (302-440) (0 what higher during the pericxl of sexual activity than at other times. Findings in the present study indicated that average hemoglobin content of ale- wives«entering fresh water to spawn is not diflFer- ent from that of the spent fish returning to the sea after spawning. Prespawners had an average hemoglobin value of 9.5 g. per 100 ml. (range, 4.5-12.5), while postspawners had an average of 9.4 g. per 100 ml. (range 4.0-13.0). SEDIMENTATION RATES The settling rate of erythrocytes has wide clini- cal use as an indicator of certain physiological changes. It is higher in human females than males and is greater during pregnancy and in dis- ease. In fishes, Schumacher, Hamilton, and Longtin (1956) found that furunculosis caused a marked increase in the sedimentation rates of brook trout, while Kalashnikov (1939) found that the sedimentation rate increased as the gonads matured. The present study indicated great individual differences in sedimentation rates of both pre- spawning and postspawning alewives (range, 1.0 to 12.0 mm. at 3 hours for 121 fish) . However, no important changes have been disclosed by com- parison of average sedimentation rates of fish entering fresh water to spawn with those of spent members of the same populations leaving fresh water 2 months later. Average sedimentation rates for prespawners were 1.2 mm. at 1 hour, 3.3 mm. at 2 hours and 4.9 mm. at 3 hours; for post- spawners, 1.1 mm. at 1 hour, 3.2 mm. at 2 hours, and 4.7 mm. at 3 hours (fig. 1). Ripe females ex- hibited higher average sedimentation rates than did ripe males (5.4 mm. compared with 4.3 mm. in 3-hour readings), but this difference disappeared in postspawners. ERYTHROCYTE FRAGILITY Another indication of physiological distress is the lowered ability of red blood cells to withstand decreasing osmotic pressure of the siuTounding medium. Fragility of human erythrocytes in- creases in certain diseases. Examination of ale- wife blood disclosed no changes in cell fragility due to the spawning migration. One-hour tests showed that complete lysis occurred consistently at between 0.5- and 0.7-percent saline in both pre- spawning and postspawning fish. TOTAL SERUM PROTEINS The serum proteins of animals have a variety of chemical and physical functions, including their important role in osmotic regulation. Among the invertebrates, wide individual variations (2.2 g. to 10.2 g. per 100 ml.) in total serum proteins of the lobster (Hotnarus americanus) and even greater variations (1.16 g. to 13.75 g. per 100 ml.) in the crab (Cancer mag is te?') were noted by Leone (1953). In fishes, average total proteins of 5.6 g. per 100 ml. for adult salmon (SaJmo salar), and 4.9 g. per 100 ml. for Conger vvlgaris were reix)rted by Drilhon, Fine, and Daoulas (1958). Keys (1933) found that the total serum proteins of eels dropped from 8.4 g. per 100 ml. in sea water to 6.8 g. per 100 ml. in fresh water. The present study of prespawning and post- spawning alewives has disclosed marked indi- vidual variations in total serum proteins — from 3.9 to 8.6 g. per 100 ml. in prespawners, and from 2.7 to 6.9 g. per 100 ml. in postspawners (fig. 2) 148 15- 10 I X CO 5- u CQ Z FISHERY BtJLLETIN OF THE FISH AND WILDLIFE SERVICE ALL PRESPAWNERS ALL POSTSPAWNERS ri , r^ MALE PRESPAWNERS FEMALE PRESPAWNERS 234 56789 10 II 12 3 -HOUR SEDIMENTATION RATE IN MILLIMETERS 13 Figure 1. — Three-hour sedimentation rates for prespawn- ing and postspawning alewives. The average for prespawners was 5.9 g. per 100 ml. while that for postspawners was 5.3 g. per 100 ml. — a reduction in fresh water significant at the 5-percent level with the student 7'-test used for small samples, as well as with the rank sum test. SERUM CHLORIDES Concentrations of various ions in animal body fluids, particularly in invertebrates, may vai^ with environmental and physiological conditions (Prosser, 1950). The closed circulatoiy system of vertebrates probably effects greater ionic sta- bility tlian is true for invertebrates. Numerous studies of teleost ionic regulation in varying exter- nal salinities have been made (reviewed by Fon- taine and Koch, 1950, and Black, 1957). Keys (1933) foimd that seiiim clilorides of eels {An- gtulki angu/Ua) were lower in fresh water than in the sea (480 milligrams per 100 ml. as op- posed to 580 mg. per 100 ml.). Bond, Gary, and Hutchinson (1932) and McFarland and Munz (1958) found that in hagfish {PoU^totrejna stouti) the concentration of blood chloride varied in a linear manner with that of the surrounding me- dimn. Harris (1959) noted a drop in blood chlo- ride from 804 mg. to 683 mg. per 100 ml. when Fundulus heteroclitus were transferred from salt water to fresh. I5n 10 Zooplankton 155 Barnacle setting rate 155 Prorocentrum sp 155 Results 155 Chlorophyll 155 Salinity 155 G. splendens 155 Copper 156 Zooplankton 157 Barnacle setting rate 157 Prorocentrum sp 157 Live car organisms 157 Discussion 159 Summary 159 Literature cited 159 ABSTRACT A large-scale experiment was conducted in a lagoon, off the Galveston Ship Channel in the Gulf of Mexico, to determine the feasibility of using copper ore as a control for destructive plankton blooms. A chemical and biological study was made of the lagoon for a period of 9 months previous to the addition of 60 tons of copper ore. Comparison with a similar studj- made after the addition of the ore revealed that the ore did not have the desired qualities of control; therefore its use for control was not recommended. EFFECTS OF COPPER ORE ON THE ECOLOGY OF A LAGOON By Kenneth T. Marvin and Larence M. Lansford, Chemists, and Ray S. Wheeler, Fishery Research Biologist, Bureau of Commercial Fisheries Sudden, immense increase in the plankton popu- lation has resulted in extensive destruction of commercially important fish and shellfish. Some- times this fish-killing plague is known as "red tide" because of the amber to dull-red discolora- tion of the water. The organism present in frequent occurrences of red tide in OfTats Bayou, Galveston (Tex.) was identified by Gunter (1942), and by Gates and Wilson (1960), as Gonyaulai monilata. Oymno- dinium breve was identified by Davis (1948) as the cause of destructive red tide blooms that have occiured off the west coast of Florida at irregular intervals since, at least, 1844 (Feinstein and others, 1955). Following the outbreak of destructive G. breve blooms in 1946 and 1947, the U.S. Bureau of Com- mercial Fisheries began a study of the organism and the environmental factors limiting or promot- ing its growth, to develop, if possible, a means of controlling or, at least, reducing the occurrence of these lethal outbreaks. The toxic property of copper has been employed elsewhere with varying degrees of success in re- lated occurrences of plankton blooms. Experi- ments in the U.vS. Fish and Wildlife Service Labo- ratory have demonstrated that the minimum amount of dissolved copper lethal to G. breve is about 0.5 microgram atoms per liter (0.03 p. p.m.). An experiment was designed therefore to test the feasibility of using immersed copper ore as a source of copper in lethal concentrations through its re- lease into solutions over a rather long period of time. We wisli to express our appreciation to Mi's. Zoula Zein-Eldin, William Wilson, and Di-s. David Aldrich and Abraiiam Fleminger, who conilucted many of the analyses, and to the Morenco Mining Branch of the Phelps Dodge Copper Corijonitioii for furnishing the copper ore for the experiment. COPPER ORE EXPERIMENT The experiment was designed to determine whether or not immersed ore would affect the flora and fauna. This would be decided by com- paring chemical and biological studies made before and after tlie addition of ore. Questions to be resolved were the foUowhig: (1) Could copper concentration in a body of water be raised to a level lethal to G. breve by the permanent exposure of a reasonable amount of copper ore? (2) Would the copper concentration of the water remain at a constant level? (3) Would the copper liave an adverse effect on other marine organisms? To obtain an estimate of the amount of copper ore needed, laboratory studies were made on the solubility of the copper in various amounts of ore in tanks of sea water. On the basis of the results of these tests, 20 tons of ore seemed a reasonable amount with which to start. Subsequent dosages, if necessary, would be based on the results of the first addition. We used a sulphide ore that con- tained approximately 1 percent copper and 3.5 percent iron. Tiie particle size varied from dust to coarse gravel. The questions were answered (1) by observing the effects of the ore on two indicator organisms." (2) by determining tlie level of the copper con- centration maintained, and (3) by comparing ecological conditions of tlie lagoon before and after the addition of tlie ore. Comparisons were based on gross differences in productivity of tiie water, on significant changes in mortality rates of organ- isms, and on variation in barnacle setting rate. Chlorophyll and zooplankton analyses were used as indicators of productivity. Mortality rate 1 Laboratory pxixrimi'iit.s ilemonstrated that tlie toloranw of these organ- isms to copper was approximately the same as that of (Jymiiorfiiiiiim hrtre. One of thes*'. Prorocrntrum sp.. was placed ii\ the laROon in dialysis bacs. another. Gymnodintum spkndrm. occurred naturally in the lagoon. Note.— Approved for publioUlon SepU'raber 12. 1960. Fishery Bulletin 1 M 153 154 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE studies were made of Oymnodinium splendens that occurred in the water, and of organisms from the live car: 03'sters, Crassostrea virginica; mullet, Mugil cephalus; and two snails, Littorina irrorata and Thais species. Our field activities were conducted in a man- made lagoon located on the eastern end of Galves- ton Island (fig. 1). The lagoon, 1.1 statute miles long, has an estimated volume of about 230 acre- feet and is connected to the Galveston Ship Channel by seven 36-inch cement culverts. Chemical and biological samples were taken at eight stations. Samples were collected three to four times a week until the first ore addition of 20 tons was made and then twice a week. Several months after the final addition of 40 tons, the rate was reduced to once a week and the number of stations sampled to three (stations 1, 4, and 7). During collection trips samples were taken for salinity, copper, chlorophyll, G. splendens, and zooplankton analyses. The water for these was taken with a stainless steel neoprene impeller pump. During collections tlie intake end of the polyethylene connection hose was continuously raised and lowered from a few inches of the bottom to the surface. Thus, the samples were represent- ative of the entire column. A 4-inch Secchi disc reading was also taken at each station. Once or twice a week, depending on weather conditions, a count was made of the organisms in the live cars anchored at each station. At weekly intervals, barnacle-setting plates were suspended 2 feet below the surface at each station. These plates were replaced every week. Before each addition of ore, dialysis bags con- taining cultures of known concentrations of Proro- centrum. sp. were suspended in perforated polyeth- ylene bottles 2 feet below the surface at stations 2, 4, 5, 6, and 7. These bottles were replaced so that some of them remained in the water for 2 days and others for 4 days. The first of these experiments was discontinued 2 weeks after the first ore addi- tion and the second, 1 week after the second addi- tion. ANALYTICAL METHODS Chlorophyll The chlorophyll analyses consisted of estimations of chlorophyll a, b, and c in acetone extracts of plant and animal material. The method em- ployed was that of Riciiards with Thompson (1952) as modified by Creitz and Richards (1955). GULF OF MEXICO Figure 1. — East Lagoon station locations selected for the copper ore experiment. COPPER ORE EXPERIMENT IN A LAGOON 155 Salinity Salinity values were estimated from density- temperature measurements taken with a hydrom- eter calibrated to the nearest 0.2 of a salinity unit (7oo) and a centigrade thermometer calibrated to tlie nearest unit. Gymnodiniutn splendens Two-liter water samples for G. .splendens analyses were placed under fluorescent' lights for about 16 hours. This concentrated the organisms in the upper part of the container. A preliminary examination was made of a portion of the sample taken from the meniscus. If G. splendens were not observed, a zero count was assumed for the sample. If in evidence, the top 200 ml., which contained virtually all of the G. splendens, were carefully siphoned into a flask and thoroughly mixed. Ten 1 -milliliter samples were removed and counted for G. splendens. The average of these was converted to count per liter by multi- plying by 100. An alternative method used for high-count samples was similar except that the 10 portions were taken from the entire mixed sample. The conversion to count-per-Iiter was obtained by multiplying tlie average by 1,000. Copper To determine the copper concentration, we used the method described by Hoste, Eeckhout, and Gillis (1953). This was preferred to that of Chow and Thompson (1952) because the latter method is not so selective for waters of variable pH, such as is found in the lagoon. Further, when coastal and bay waters are aiuilyzed by tliis method, a turbid extract forms occasionally that is difficult to analyze. Zooplankton The zooplankton samples were obtained by I)umpiiig 250 gallons of water through a plankton net of \o. 2 bolting silk. These were diluted to 100 ml., and an aliquot part checked for tlic various types of zooplankton. The count of each was recorded as count per 100 liters by multiplying by the appropriate factors. The size of the I aliquot part varied, depending on the population density of the sample. Barnacle setting rate The barnacle attachment rate was based on the average daily setting-rate on 4-inch square cement plates suspended horizontally 2 feet below the surface at each station. The rate was estimated by averaging the count per square centimeter of eight locations on each plate, and then dividing by the number of days that the plate was sub- merged in the lagoon. Prorocentrum sp. We used Wilson's (1959) dialysis membrane bag method for evaluating the effects of the copper ore on Prorocentrum sp. cultures suspended in the lagoon. Initial and final population estinuites were made by counting the organisms in several 0.01 ml. portions taken immediately before the culture was placed in the dialysis bags and after their removal from the lagoon. RESULTS Chlorophyll The results of the chloropliyll a, b, and c analyses are shown in figure 2. We have placed the November 1958 to April 1959 section of the graph under the corresponding months of 1957 and 1958 to simplify seasonal comparison. The phytoplankton blooms noted during November 1957 and January to March 1959 were reduced during the corresponding months of 1958 and 1959. Wliether or not this was an effect of the ore is not known. The significant fact shown is the con- tinued productivity of the lagoon after the addition of ore. This is indicated by the continuation of chlorophyll concentrations that are representative of a highly productive area (Zein-Eldin, 1959). Salinity Figure 3 shows monthly salinity ranges and averages of the lagoon. All data are based on the average of station salinity values. Gymnodinium splendens The average population of G. splendens in the lagoon from November 1957 to June 1959 is shown ill figure 4. The November 1958 to June 1959 portion of the graph has been placed under the corresponding months of 1957 and 1958 to simplify a comparison of similar seasons. It can be seen that the seasonal occurrence has not been altered by the immersed ore. The January 1959 to April 1959 zero count cannot be considered significant as far as the copper ore experiment is concerned because of the subse(iuent rise that followed the pattern of the previous year. 156 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE CHLOROPHYLL IHOVINO AVERAGE or S DETCRHINATlONS) 5 ■• r - h. - !\ i \ 1 V " i \ / . ^^ 20 TOMS OF ORE AOOEO *0 ^OM S OF o«£ iooeo . <~'~T " MOVEUGEil OECEMBCR jANuMr FEBRUARy MBRCH A HIL - 1957 — -^ .. - ^' '-^ \ "■■ — -l , 1 1 juLV awwsT SEPTEMBEo ocroeER CHLOROPHYLL A Mg/L CHLOROPHYLL B Ug/l CHLOROPHYUl C MSPU/U HOVEMOt OeCXMEn JANUARY FEBRUARY UARCH 1959 Fir.rRE 2. — Average chlorophyll concentration of East Lagoon from station samples. G SPLENDENS (MOVINQ AVERAGE OF 5 DETERMINATIONS) I I I I II I I L JAN MAR MAY JUL SEP NOV JAN MAR MAY 1958 1959 Figure 3. — Salinity of East Lagoon showing monthly averages. Copper Twenty tons of ore were added to the lagoon August 21, 1958. The effect of this addition on the overall copper concentration based on the average of station results is shown in figure 5. A maximum of 0.14 Mg- at. Cu/1. was attained in less than a week but was reduced to the low value shown by the excessive tidal and drainage dilution that accompanied hurricane "Ella". The low iLl 1 L_ NOV OEC JftN FEB MAR APR 1956 1959 _L^.„^ \ \ 1 Fini'RE 4. — Gymnodinium splendens (average counts for all stations). maximum of 0.14 ng. at. /I. indicated that a second and larger addition would be necessary to obtain a copper concentration lethal to G. Ireve. Accord- ingly, on October 21 an additional 40 tons were COPPER ORE EXPERIMENT IN A LAGOON 157 Figure 5. — Copper concentrations (average of all station?) showing approximate minimum lethal level from labora- tory experiments. placed ill the lagoon. Again the results were disappointing. A near lethal concentration was attained but decreased during the next 5 months to about 0.01 Mg-at./l- In some respects, the effects of the ore were similar to those of copper sulphate employed in the Florida red-tide control experi- ments (Rounsefell and Evans, 1958). The rapidly attained maximum levels soon decreased to nomial values for the area. In the Florida control tests, however, lethal levels were reached, and the decline that followed occurred in a matter of days rather than months. Undoubtedly, variation in the hydrography and chemistry of the two areas accounted for much of the difference in maximum levels attained and also in the rates of decline. The water of the lagoon is high in particulate matter (fig. 6), and the copper from tlie ore was assumed to have been adsorbed and made unavail- able by the muds, plankton organisms, and other TURBIDITY (4 -INCH SECCHI DISC) Figure 6.— Turbidity of Ea-st Lagoon. Average Secehi disk readings from stations. material making up I lie particulate matter (Harvey, 1955). Zooplankton A qualitative and ([uantitative study of tiio standing crop of zooplankKm in tiie lagoon was made by Fleminger (1959). The outstanding observation of his work was the broad summer- abundance peak and secondary peak combined with troughs in tlie early spring and autumn (fig. 7). Of particular significance is tlie January to March 1959 secoiulary peak which indicates continued growth of zooplankton populations following the addition of copper ore. Barnacle setting rate This study, conducted by Aldrich (1958a), showed that the adult barnacle population of the lagoon consisted almost exclusively of tlie brackish- water species, Balanux ehurneus. Another brackish -water species found occasionally was Balanm improvisus. The data in figure 8 indicate the seasonal nature of barnacle setting in the lagoon. The outstanding feature is the con- tinuance of tiie seasonal growth pattern after addition of the ore. Prorocentnnn sp. Table 1 shows the initial and final counts of Frorocentnun sp., and copper concentration of the cultures in the dialysis bags used in the lagoon. More than half of these cultures increased in population count. The data indicate that the greatest increase occurred in cultures liaving the lowest initial count. Presumably, tliese had not reached their peak when placed in the lagoon. The copper concentration apparently did not interfere with the population growth of the organism. Tliere was one exception: approxi- mately 60 percent mortality occurred within llie 4-day bags placed at station 4 on October 10. This was the day of tlie second ore addition (table 1), and the day that tlie greatest concen- tration of copper was observed witliin tiie hags. Live car organisms Laboratory experiments (Aldricii, 1958b) con- ducted in conjunction witli the copper ore study indicated tliat tiie snail l.ittorina irroiata was probably most susceptible to copper poisoning. Twenty-four-hour tests demonstrated, however. 158 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE jut FEB HAH UPR Figure 7. — Zooplankton concentration. Average of plankton samples taken at stations 1, 3, and 7. Table 1. — Summary of dialysis bag data showing initial and final counts of organisms and final copper content of bags placed in East Lagoon, after addition of ore Date Days in lagoon Count/.Ol ml. Copper Initial Final (ig.at./l. Aug. 23 4 2 4 2 4 2 4 2 4 2 4 2 4 2 4 2 4 43 21 21 52 52 48 48 32 32 35 35 12 12 24 24 22 22 43 50 48 39 41 42 42 42 140 36 39 37 26 24 26 16 25 0.19 25 . 17 25 . 18 Z7 .28 27 .24 29 .26 29 Oct. 10 -- .21 10 .41 12 12 .32 14 .12 14 .13 16 .25 16 .19 18 - 18 .20 .18 ' About 60 percent of the organisms in the bags at Station 4 were dead. BARNACLE ATTACHMENT RATE IN EAST LACOON rV If 1_ \I Figure 8. — Barnacle attachment in East Lagoon. 100 — 80 — 60 - .0- 2 -/ AUG 6 - AUG 20 1958 100 80 60 ^ 40 a 20 5 AUG 20- SEPT 2 1958 •~20 TONS OF ORE ADDED L_ t- BOp z UJ 60 - o Q. 20 - 0. — to 20 SEPT 23-DEC 31 1958 -40 TONS OF ORE ADDED 40 20 r MOV 12 -DEC. 12 1958 1 r JfiN 15 - FEB 18 1959 (NO MORTALITIES) 1 , 1 , 1 , 1 1 — 1 20 30 DAYS IN LIVE CARS Figure 9. — Mortality of Littorina irrorala in East Lagoon. that the snail could tolerate copper concentrations of 8 jug. at./l. with only temporary minor discom- fort. The mortality rate of this organism for five test periods is shown in figure 9. The snails represented by the August 6 to September 2 periods were a generation older than those used in subsequent experiments. We believe increased age was the cause of greater mortality rate shown. The decrease in mortality following the addition of the copper ore indicates that the ore had little or no detrimental effect on this organism. The relatively small increase of the copper concentration COPPER ORE EXPERIMENT IN A LAGOON 169 in the lagoon apparently did not have a significant effect on the mortality rates of the other species in the live car. DISCUSSION Experiments conducted in our laboratories indi- cate that more copper ore is required to reach the toxic level for Gijmnodinium breve in lagoon water than in Florida coastal water. This is thought to be due to the large amounts of nat- ural chelators and particulate matter present in the lagoon as opposed to the relatively clear Flor- ida waters. On the other hand, the Florida coastal waters receive more tidal flushhig and dilution than the lagoon, and we would expect the maxi- mum level of copper concentration to be less per- manent than that shown for the lagoon in figure 5. Even assuming a toxic level could be reached in the Florida coastal waters, large quantities of ore would have to be added at frequent intervals which would make the cost prohibitive. SUMMARY An analysis of the biological and ciiemical data shows that the copper concentration of the lagoon was not increased to a level lethal to Gymnodinium breve after the addition of 60 tons of ore. The flora and fauna of the lagoon and organisms placed there in dialysis bags and live cars showed no significant effect attributable to the ore. The copper level, after the addition of the ore, in- creased to a maximum that was below the labo- ratory estimate of the level toxic to G. breve (based on Florida sea water) and then dropped to a lower level. These results show that the ore is not capable of maintaining a sufficiently high copper concentration to be considered as a means of controlling red tide outbreaks in waters simi- lar in quality to that of the lagoon. Results of this experiment indicate that copper ore does not have the desired characteristics of a red tide controlling agent, and we recouuuend that it not be used. LITERATURE CITED .^LDRlrn, D.WID V. 19.58a. Barnacle attachment rate.-< at Galveston, Toxa.s. U.S. Department of the Interior, Fish and Wildlife Service, Annual Report of the Gulf Fishery Investigations for the year ending .June 30, 1958, p. 89. 1958b. Toxicity of copper to marine organisms. U.S. Department of the Interior, Fish and Wildlife Service, Annual Report of the Gulf Fishery Inves- tigations for the year ending June 30, 1958, p. 86. Chow, Tsaihwa J., and Thomas G. Thompson. 1952. The determination and distribution of copper in sea water. .lournal of Marine Re.search, vol. 11, No. 2, pp. 124-138. Creitz, Grace I., and Francis A. Richards. 1955. The estimation and characterization of plank- ton populations by pigment analysis. III. A note on the use of "Millipore" membrane filters in the estimation of plankton pigments. Journal of Marine Research, vol. 14, No. 3, pp. 211-216. Davls, Charles C. 1948. Gymnodiniuin brevis sp. nov. a cause of dis- colored water and animal mortality in the Gulf of Mexico. Botanical Gazette, vol. 109, No. 3, pp. 358-360. Feinstein, Anita, A. Russell Ceurvels, Robert F. HcTTON, and Edward Snoek. 1955. Red tide outbreaks off the Florida we.st coast. Marine Laboratory, University of Miami, Report 55-15 to the Florida State Board of Conservation, 44 pp. Fleminger, Abraham. 1959. East Lagoon zooplankton. In Galveston Biolog- ical Laboratory fishery research for the year end- ing June .30, 1959. Bureau of Commercial Fisheries, U.S. Fish and Wildlife Service, Circular 62, pp. 114-118. Gates, Jean A., and William B. Wilson. 1960. The toxicity of Gonyaulax monilala Howell to Mugil cephatus. Limnology and Oceanography, vol. 5, No. 2, pp. 171-174. Gdnter, Gordon. 1942. Recurrent summer fish mortalities on the Texa-s coast. American Midland Naturalist, vol. 28, p. 631. Harvey, H. W. 1955. The chemistry and fertility of sea water. Cambridge, England, University Press, 224 pp. Hoste, J., J. Eeckhout, and J. Gillis. 1953. Spectrophotometric determination of copper with cuproine. Analytica Chimica Acta, vol. 9, pp. 263-274. Richards, Francis A., with Thomas G. Thompson. 1952. The estimation and characterization of plank- ton populations by pigment analyses. II. A spectrophotometric method for the estimation of plankton pigments. Journal of Marine Re.search, vol. 11, No. 2, pp. 156-172. Rounsefell, George A., and John K. Evan.s. 1958. Large-scale experimental test of copper sul- phate as a control for the Florida red tide. U.S. Department of the Interior, Fish and Wildlife Service, Special Scientific Report — Fisheries No. 270, 57 pp. 160 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Wilson, William B. Zein-Eldin, Zoula P. 1959. Evaluating toxicity of dissolved substances to 1959. Plankton pigments in East Lagoon. In microorganisms using dialysis membranes. In Galveston Biological Laboratory fishery research Galveston Biological Laboratory fishery research for the year ending June 30, 1959. Bureau of Com- for the year ending June 30, 1959. Bureau of Com- mercial Fisheries, U.S. Fish and Wildlife Service, mercial Fisheries, U.S. Fish and Wildlife Service, Circular 62, pp. 106-107. Circular 62, pp. 100-102. U.S. GOUERNMENT PRINTING OFFICE I96I UNITED STATES DEPARTMENT OF THE INTERIOR, Stewart L. Udall, Secretary FISH AND WILDLIFE SERVICE, Clarence F. Pautzke, Commissioner Bureau of Commercial Fisheries, Donald L. McKernan, Director VALIDITY OF AGE DETERMINATION FROM SCALES OF MARKED AMERICAN SHAD By Mayo H. Judy FISHERY BULLETIN 185 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 PUBLISHED BY THE U.S. FISH AND WILDLIFE SERVICE • WASHINGTON • 1961 PRINTED BY THE U.S. GOVERNMENT PRINTING OFFICE, WASHINGTON For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington 25, D.C. Price 15 cents Library of Congress catalog card for the series, Fishery Bulletin of the Fish and Wildlife Service : U.S. Fish and Wildlife Service. Fishery bulletin, v. 1- Washington, U.S. Govt. Print. Off., 1881-19 V. In illus., maps ( part fold. ) 23-28 cm. Some vols, issued in the congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies : v. 1-49, Bulletin. Vols. 1^9 issued by Bureau of Fisheries (called Fish .Commission, V. 1-23) 1. Fisheries— U.S. 2. Fish-culture— U.S. i. Title. SH11.A25 639.206173 9-35239 rev 2* Library of Congress crSSeJi CONTENTS Page Int roduct ion 161 Life history 161 Materials and methods 162 Examination of scales 162 Examination of pelvic fins 162 Marked fish recovered 1956 165 Marked fish recovered 1957 165 Marked fish recovered 1958 166 Discussion and conclusion 167 Summary 168 Literature cited 169 ABSTRACT In the fall of 1952, one hiuidred thousand juvenile American shad, marked by removal of the right pelvic fin, were released in the Con- necticut River. Seventeen marked fish were recovered in the river in 1956, 70 in 1957, and 39 in 1958. Ages of the fish, determined from their scales, wei'e 4, 5, and 6 yeare, respectively. These findings were in agreement with known age established from marking and there- fore validate annuli and spawning marks as criteria for age determi- nation of shad. VALIDITY OF AGE DETERMINATION FROM SCALES OF MARKED AMERICAN SHAD By Mayo H. Judy, Fishery Research Biolosist, Bureau of Commercial Fisheries In 1950, tlie U.S. Fish and Wildlife Service, as the primary research agency of the Atlantic States Marine Fisheries Commission, began a study of the American shad (Alosa sapidissinut) on the At- lantic coast of the United States. Objectives of this investigation were to determine the causes for decline in the commercial yield from approx- imately 50 million pounds in 1896 to 8 million pounds in 1950, to detennine conditions favoring recovery, and to provide information for scientific management of the fishery. A necessity for ac- complisliing these objectives was an accurate method of aging shad. Prior to this investigation, techniques for aging shad had been presented by various workers. Leim (1924) determined age by means of winter rings or annuli on scales and established the rela- tion between scale and body length. Borodin (1925) presented a method of reading scales by counting the number of transverse grooves and dividing by 2 to determine the age in years. Bar- ney (1925) found evidence in otolith markings to indicate that age estimates as reported by Borodin were correct, but Greeley (1937) stated that Boro- din's method gave misleading results. Greeley found that Leim's method of age determination agreed with the results of his studies on Hudson River shad. Gating (1953) proposed a method for reading shad scales for total age, age at first spawning, and number of times the fish had previously spawned. Transveree groove counts were used to .separate true from fasle annuli to the fourth an- nulus, and age of fish spawing for the first time was determined by counting the number of an- nuli and adding 1 year for the scale edge. Age of fish spawning for the second or more times was Ndte— Approved for publication December 5, 19B0. Fishery Bulletin 18.0. 590374 — «1 obtained by counting the number of annuli plus the number of spawning marks and adding 1 year for ihe scale edge. Although Gating aged shad with apparent confi- dence he did not establish the validity of his read- ings. LaPointe (1958), using Gating's method, validated the annulus to be a true year mark on scales of fish spawning for the first time. He found that Leim mistook the fresh- water mark for the first annulus, thus causing a difference of 1 year between Leim's age determination and those in his study. Hammer (1942) confirmed that the fresh- water zone was a distinct and measurable scale growth formed while juveniles are in fresh water. In 1952, prior to the completion of Gating's scale study, a marking program was conducted on juvenile shad in the Gonnecticut River. The ob- jective of this program was to recover in future years marked fish of known age, thereby to check the method employed by Gating and to establish a correct method for aging shad from their scales. This was deemed necessary because techniques used prior to this time were subject to question. Data presented in this paper were derived from the scales of marked adult shad recovered 4, 5, and 6 years following the marking program. Scales from these fish of kno^^^l age were studied to de- termine the validity of annuli and spawning marks for age determination. Appreciation is expressed to the Gonnecticut Power and Light Gonipany for use of the Windsor Tiocks Canal System, and to the sliad fishermen and fish dealers of the Connecticut River for their cooperation in this study. LIFE HISTORY Shad range on the Atlantic coast from the St. Johns River in Florida to tlie St. Lawrence River in Canada. It is an anadiomous fish and spawn- ing migrations begin as early as November in 161 162 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Florida and as late as June in Canada. The young remain in the rivers until fall, attaining lengths from 3 to 5 inches, then migrate to sea. Winters are probably spent off the Middle Atlantic, and summer and fall in the Gulf of Maine. After reaching maturity, in 3 to 6 yeai*s, they return to the rivers to spawn. Adult shad native to streams north of Cape Hatteras (N.C.) that survive spawning and other hazards return to the sea and re-enter the rivers to spawn again in successive years. Shad native to streams south of Cape Hatteras die after spawning (Talbot and Sykes, 1958). MATERIALS AND METHODS In the fall of 1952, 100,000 downstream migrant juvenile shad were marked on the Connecticut Eiver in the Windsor Locks Canal, Windsor Locks, Conn. Marking of these fish, which averaged about i inches in fork length, was accomplished by clipping the right pelvic fin close to the body of the fish. Juveniles were trapped, seined, and marked in one level of the canal and then placed in a lower level of the canal and flushed into the main river. Samples of fish were held overnight to obtain an estimate of mortality. Mortality from marking was estimated at 30 percent ; there- fore, it was assumed that 70,000 marked juveniles were returned to the river. The first marked fish were recaptured in the Connecticut River in 1956 from commercial catches and shad passed by the fishway at Hadley Falls Dam, Holyoke, Mass. Subsequent recoveries were made in 1957 and 1958 from commercial and sport catches. Approximately 35,000 shad were ex- amined annually. Fi-om the 1956 collection it was determined that some shad had malformed, or nat- urally missing, pelvic fins. Therefore, in 1957, fish with various pelvic fin abnormalities were collected so that a wide assortment of abnormal fins would be available for comparison with marked fins. The pelvic girdle section of each fish collected was removed, labeled, and preserved. In addition, scale samples were taken and the length, weight, and sex recorded. EXAMINATION OF SCALES Two scales from each fish collected were im- pressed in plastic, using a modification of the method described by Greenbank and O'Donnell (1950). The scale impressions were read on an Eberbach projector, by two biologists using Cat- ing's (1953) method for determining age of shad. Age readings were compared and the results confirmed. In this method the scale edge is counted as a year mark because the last annulus (near scale periph- ery) is frequently eroded during the spawning mi- gration. For example, a shad spawning for the first time (virgin fish) at 4 years of age has 3 annuli on the scale plus the scale edge for a total age of 4 years. After shad spawn and return to the sea, renewed feeding and resumption of growth leaves a characteristic scarlike mark on the scale edge where erosion occurred during the spawning mi- gration (Moss, 1946). This is designated as a spawning mark and is used in place of the eroded annulus, formed prior to spawning, for determin- ing age of "repeater" fish (those spawning for the second or more times) . For example, a 6-year- old rejieater spawning for tlie second time has 4 annuli and 1 spawning mark which, when read to include the scale edge, gives a total age of 6 years. The 4 annuli and 1 spawning mark indicate that this fish first spawned at 5 years of age and was on its second spawning run when captured. EXAMINATION OF PELVIC FINS Examination of the pelvic fin sections indicated that they contained malformed, missing, and marked fins. Malformed and missing fins are often fomid in fish as evidenced from studies by Cable (1956), Code (1950), and Rich and Holmes (1928). Marked fins were characterized by a varied pattern of fin regeneration ranging from no regeneration beyond formation of scar tissue to almost complete, but distorted regeneration. These findings are not unusual since, as reported by Stuart (1958), fin clipping seldom results in a uniform series of marks. From microscopic examination of regenerated marked fins, Stuart found that new gi-owth of fin rays does not extend in a regular manner but commences as a thickened and undifferentiated cap, the comiective and other tissues keeping pace with the gi'owth of the adja- cent rays. The degi'ee and nature of fin regenera- tion was usually dependent on the angle of the cut and the amount of dermal-fin-ray tissue removed during clipping. The pelvic fin section of each shad collected in AGE DETERMINATION OF AMERICAN SHAD 163 Figure 1. — Scale from 6-year-old shad spawning for the second time. (Roman numerals represent annuli, FWZ fresh-water zone, and SM spawning mark. ) this study was X-rayed using a method described by Sutherhiud (1958). Marked fins were identi- fied from radiographs by an enlargement at the distal end of the radial bones extending partially or completely across the area of separation from the dermal fin rays ( fig. 2 ) . This method of classi- fication of marked fish is in agreement with Stuart (1958) wlio found from microscopic examination that a palpable ridge was formed on marked fins at the site of cutting. Marked fins were classified according to the nmnber of fin rays regenerated, regardless of the length of the rays, and placed in tlie following categories: (1) no regeneration — no fin rays; (2) one-third regeneration — one to three fin rays; (3) two-thirds regeneration — four to six fin rays; and (4) complete regeneration — seven to nine fin rays (fig. 2). Pelvic-fin sections were classified as malformed if there was no enlargement at the distal end of the radials (fig. 3B, C). Missing fins were chai-ac- terized by absence of radials or, in some specimens, absence of the entire pelvic girdle (fig. 3A, D), and absence of scar tissue at the site of fin origin. Malformed and missing fins were termed abnormal. From a study of 28 shad collected in 1956, it was determined tlvat 11 had abnormal pelvic fins. These included 4 males and 7 females of which 5 had malformed left or right pelvic fins and 6 had eitlier the left, right, or both pelvic fins missing. These fish ranged in age from 3 to 6 yeai-s. In 1957 fish with a variety of pelvic fin abnor- malities were purposely collected!. Of the 132 shad sajnpled, 62 were classified as abnormal. 164 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 2. — Radiograph of pelvic fin sections from four marlced shad. A — no regeneration, B — one-third, C — two-thirds. and D — complete regeneration. .,J. J a Figure 3. — Radiograph of pelvic fin sections from four .shad with malformed and missing fins (A — right fin and radial fin supports absent, B — right fin with malformed fin rays, C — double malformation of pelvic fins, and D — pelvic girdle absent. ) AGE DETERMnSTATION OF AMERICAN SHAD 165 FiGUBE 4. — Scale from 4-year-olci marked shad spawning for the first time. Tliese included 21 males and 41 females of which 40 had either the left, right, or both pelvic fins malformed, and 22 had either the left, right, or both pelvic fins missing. These fish ranged in age from 4 to 7 years. Of tlie 57 shad collected in 1958, 18 were classi- fipd as abnoranal. These included 9 males and 9 females of which fi had either the left, right, or both pelvic fins malformed and 12 had either the left, right, or both fins missing. These fish ranged in age from 4 to 8 years. MARKED FISH RECOVERED 1956 P'rom a study of pelvic fin sections and radio- graphs it was determined that 17 marked slmd were recovered in 1956. These included 8 males and 9 females of which 3 had no regeneration of the right fin, 5 had one-thiixl regeneration, 4 had two-thirds regeneration, and 5 had complete re- geneiation. Age readings indicated that all marked fish recovered were 4 years old, and spawn- ing for the first time ( fig. 4) . Marked males aA'er- aged 1C.4 inches, fork length, and 2.4 pounds in weight. Marked females averaged 17.9 inches, fork length, and 3.2 pounds in weight. MARKED FISH RECOVERED 1957 From a study of pelvic fin sections and radio- graphs it was determined that 72 marked shad were recovered in 1957. Age determined from scale readings indicated that all but two of these fish were 5 years old. The ages of these two fish were 4 and 6 years. The radiographs and scale samples were re-examined and the above results confirmed. Therefore, on the basis of these find- ings, an error of approximately 3 percent exists either in interpretation of radiographs or in age determination. The sevent}' 5-year-old fish that were marked included 14 males and 56 females of which 11 had no regeneration of the right jielvic fin, 18 had one-third regeneration, 2G had two-thirds re- 166 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 5. — Scale from 5-year-old marked shad spawning for the first time. generation, and 15 had complete regeneration. This group consisted of 56 first- and 14 second-year spawners. Forty-three percent (6) of the males and 14 percent (8) of the females were spawning for the second time. Figures 5 and 6 illustrate representative scales from 5-year-old marked fish spawning for the first and second time. In figure 5 the fifth or last annulus is plainly visible, but since it was laid down just prior to the spawning migration, it is combined with the scale edge and counted as one year. Fish spawning for the second time (fig. 6) had first spawned in 1956 when 4 years old. Marked males averaged 17.6 inches fork length and 3.2 pounds in weight. Marked females averaged 18.8 inches, fork length, and 4.1 pounds in weight. MARKED FISH RECOVERED 1958 From a study of pelvic fin sections and radio- graphs it was determined that 40 marked shad were recovered in 1958. Age determined from scale readings indicated that 39 fish were 6 years old and one was 5 years old. The radiographs and scale samples were re-examined and the above findings confirmed. Therefore, an error of ap- proximately 2 percent exists either in interpreta- tion of radiographs or in age determination. The 39 marked recoveries included 16 males and 23 females, of which 11 had no regeneration of the right fin, 9 had one-third regeneration, 10 had two-thirds regeneration, and 9 had complete re- generation. This group consisted of 13 first-, 18 second-, and 8 third-year spawners. All males (16) and 43 percent (10) of the females had pre- viously spawned. Sixty-two percent (10) of the males and 35 percent (8) of the females were sijawning for the second year, and 38 percent (6) of the males and 9 percent (2) of the females were spawning for the third year. Figures 7 and 8 are representative scales from 6-year-old marked AGE DETERMINATION OF AMERICAN SHAD 167 Figure 6. — Scale from "j-year-old marked shad spawning for the second time. fish spawning for the first and third times.^ Those spawning for the third time had first spawned in 1956 when 4 years old. The second spawning mark was laid down in 1957. Marked males averaged 18.2 inches fork length and 3.6 pounds in weight. Marked females averaged 19.1 inches fork length and 4.5 pounds in weight. DISCUSSION AND CONCLUSION Of the 1'29 fish classified from radiographs as marked, only 3 were in disagreement with age as determined from scale readings. These misclassi- fications, 2 in 1957 and 1 in 1958, were caused by error either in age determination or in inteq)reta- tion of radiographs, which in some cases were difficult to interpret. This minor disagi-eement, approximately 2 percent of the fisii classified as ' Fipure 1 is n represptitative scale from a 6-.vear-oId marked shad spawninc fur tlic second time. These flsh had first spawned In 1957 when 5 years old. marked, was considered insignificant and in no way invalidates the findings of this report. Methods used by Ivcim (192-1), Borodin (1925), Greeley (1937), and Gating (1953), to age shad were all considered in this study. Of these methods, only Gating's proved to be a complete and valid means for determining total age, age at first spawning, and number of times previously spawned. LaPointe (1958) correctly validated the annulus as a true year mark on scales of shad spawning for the fii-st time and he showed that I^im had mistaken the fresh-water mark for the first aitnulus. Therefore, the techniques used by I^eim and Greeley to age shad gave age assessment 1 year greater than the actual age of the fish. Borodin's method, applied to scales of marked shad, gave eiToneous results and could not be justi- fied on the basis of the present study. Age of marked fish collected in 1950, 1957, and 1958 as determined from scale readings, was in 168 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 7. — Scale from 6-year-old marked shad spawning for the first time. agreement with known age established by mark- ing. These findings indicate that the method used to age shad (Gating, 1953) is valid and that annuli and spawning marks are true indicators of age. SUMMARY One hundred thousand juvenile shad from the Connecticut River were marked by removal of the right pelvic fin, in 1952. Tlie objective of this pi'O- gram was to recover marked fish of known age in future years, to validate the use of annuli and spa^vning marks for determining age of shad. Marked fish were first recovered in the Connecti- cut River in 1956 with subsequent recoveries in 1957 and 1958. Fish with marked and abnormal fins were collected in each of these years. Twenty- eight were collected in 1956, 132 in 1957, and 57 in 1958. The pelvic fin section from each fish sampled was X-rayed and classified as follows: (1) marked, (2) malformed, and (3) missing. Marked fins were identified by an enlargement at the distal end of the radial bones at the point of separation from the dermal fin rays. Pelvic fin sections were classified as malformed, when there was no enlargement of tlie distal end of the radials. Missing fins were classified as to the fin or fins af- fected. The number of marked fish collected each year was determined from a study of pelvic fin sections and radiographs. Scales from sampled shad were read for total age and number of times each fish had previously spawned. The 1956 recoveries of mai'ked shad were 4-year-oId fish spawning for the first time. Those collected in 1957 were 5-year-old fish, with I'ecoveries divided between first and second year spawnei-s. Fish spawning for the second time had first spawned in 1956. Marked fish collected in 1958 were 6 years old and consisted of first-, sec- ond-, and third-year spawners. Those spawning AGE DETERMINATION OF AMERICAN 6HAD 169 FiGUBE 8. — Scale from 6-year-old marked shad spawning for the third time. for the second time had first spawned in 1957, and those spawning for the third time had first spawned in 1956. Age of marked shad, as determined from scale readings, was in agreement with known age estab- lished by marking. These findings validate the use of annuli and spawning marks for determining total age of shad. LITERATURE CITED Barney, R. L. 1925. A continuation of Borodin's .scale method of use determination of Conecticut River shad. In Mitchell, P. II., A report of investiRations con- cerning shad in the rivers of Connecticut, Part III, pp. 52-60. Hartford, Conn. Borodin, N. A. 1925. Age of shad Alsoa sapidissima (Wilson), as determined by the scales. In Mitchell, P. H., A report of investigations concerning shad in the rivers of Connecticut, Part II, pp. 46-51. Hartford, Conn. Cable, Louella E. 1956. Validity of age determination from scales, and growth of marked Lake Michigan lake trout. U.S. Fish and Wildlife Service, Fishery Bulletin, No. 107, vol. 57, pp. 1-59. Cating. James P. 1953. Determining age of Atlantic shad from their scales. U.S. Fish and Wildlife Service, Fishery Bulletin, No. 85, vol. 54, pp. 187-199. Code, SIark R. 1950. Cutthroat trout without dorsal fins. Progres- sive Fi.sh Culturist, vol. 12 (2) : pp. 85-86. Greeley, John R. 1937. Fishes of the area with annotated list. In New York ('onservation Department : A biological sur- vey of the lower Hudson waterslunl; supplemental to twenty-sixth annual report, 1936. Part II, pp. 4.5-103. .1. 15. Lyon Company, Albany, N.Y. (iReenbank. Joii.n, and I). John ()"1)onnei.l. 1950. Hydraulic presses for making impressions on fish scales. American Fisheries Society, Trans- actions, vol. 78 (194,S), pp. 32-37. H.wimer. Ralph Curtis. 1942. The homing instinct of the Chesiipeake Bay shad Aluna saiiidissiiiui (Wilson), as revealed by a 170 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Study of their scales. Thesis, University of Mary- land, 4") pp. ( typewritten ) . LaPointe, I)on.u.d F. 1958. Age and growth of the American shad, from three Atlantic coast rivers. American Fisheries Society, Transactions, vol. 87 (1957), pp. 139-150. Leim, a. H. 1024. The life history of the .shad Alosa sapiiUssima (Wilson), with special reference to the factors limiting its abundance. Biological Board of Can- ada, Contributions to Canadian Biology, vol. 2 (new series), No. 11, pp. 161-284. Moss, Douglas D. 1946. Preliminary studies of the shad (Al«sa sapi- (lissima) catch in lower Connecticut River, 1944. Eleventh North American Wildlife Conference, Transactions, pp. 230-239. Rich, W. H., and H. B. Holmes. 1928. Experiments in marking young chino20'E t06.-i Nov. 7 Nov. 17 Nov. 20 Nov. 22 Dec. 6 igss Oct. 27 ...do do I 9.7ft Aug. 19 Sept. 15 Oct. 4 Aug. 25 Aug. 18 Aug. 1 0916-0946 ZT 0921-0950 ZT._... 1000-1036 ZT._ 1008 1039 ZT 0917-0947 ZT 1118-1151 ZT.^... 1153-1246 ZT 0953-1112 ZT _ . 2116-2152 ZT ._ 0352 GMT..- - 1 m. 65 .65 .65 .65 .65 .65 65 65 66 ,56 .366 .56 33 .11 146-72 CN... 119-63 CN.... 122-61 CN... 126-63 CN.... 146-68 CN.... surfice 690 2 1 m 690 8 Do do 1 m. 38 Do do 1 m 966.4 Do do 1 m. 283 8 Do do 47 47 47 35 1 m 1 m 1 m 1 m 1 m 1342 Do . do . 100 CN 380.0 Do do._.-....-. :...l 500 CN 169 Q 922.0 Do.._ do Stranger Scripps Institution of 0-140 Q Institut Francais d'Oceanie. 56-4 2338-2457 LT 0428 GMT 50 cm 1 m. 150 Q() 0-280 Q 616 Horizon Scripps Institution of Japan Hydrographic Office. 2100-2210 ZT 0725-0955 ZT 45 cm 22 5 cm. 0-150 Q 20.71 Kagoshir/ttt- Maru „ Kagoshima University, Japan. 0-50 v.. . 1 ZT, zone time; LT, local time; GMT, Greenwich mean time. 2 CN, closing net: Qt oblique open net; V, vertical open net. COPEPODS FROM EQUATORIAL WATERS OF PACIFIC OCEAN- ITS Table 2. — List of species and number of specimens in each sub-sampfe ("X" indicates the species was found during the examination of other fractions of original sample| Species Hugh M. Smith stranger Oraom Horizon Salsuma Kagoshima 94 132 1 144 1 153 178 30 31 29 63 34 10 32 32 613 2 il 3 21 2 41 1 12 2 1 44 2 2 X 2 1 i 6 1 X X 6 X ^X 6 X 8 6 1 17 X X 1 1 1 4 72 15 57 4 20 1 11 15 X 39 15 6 X 2 11 11 2 1 85 86 2 173 39 45 i i i " 3 48 3 9 X 1 4 2 Vi. Khincalanus cornutus X 1 1 5 X 3 3 1 11 3 4 1 X 1 2 1 1 X 1 X 1 X X 1 ::::;:.::::::.. :....i.... ... 1 3 X 1 1 X 5 1 3 75 30 4 9 14 27 3 15 11 1 3 6 X X X X X 3 X X X X 12 6 18 9 4 25. Euaetideus giesbrechti 1 5 13 2 1 3 3 X 2 X 11 X 4 X 27 E bradvi 28. Chiridius poppei _ _. 2 6 " X 30 G. minor 31. Euchirella bella.-„ X X X 3*' E venusta 33 E pulchra X X X X X X X 37. Pseudochirella sp ' •^ X i . ~'- 39. C. indica .. X X 1 7 1 1 1 X X 2 4 1 1 X 33 8 34 12 5 5 X 2 1 X ■~ 4 2 4 X 3 X X 1 X X 50 Xanthocalanus dilatus n. sp. . , X X 4 X ... 2 4 18 5 1 1 X 26 1 X 17 3 1 7 6 53 S bradyi 14 1 8 4 1 X X 1 1 57. S. teiuiserrata 6 X 9 X 2 2 1 2 4 4 fiO S sp X X 1 1 X fi3 Scottocalanus securifrons X X X X X "x X 66. Temoropia mayumbaensis X 7 X 7 1 S 1 2 13 1 5 12 X X X j^ 1 1 X 1 X X X 74. Centropages gracilis „ 1 X X X 7 X X 1 15 3 X .^ ii 4 X 8 X 1 2 1 X X 1 1 1 1 1 79. Heterorhabdus spinifrons 2 10 43 3 1 X 80. H. papilliger...-- _. 5 19 X 6 12 27 26 9 50 X X X 1 X X 13 X 84. H. ornatus „.„...„. .._... . X 85. H spinirpps . . X 1 X X X X X 88. Euaugaptilus herticus 2 1 1 X ; 1 X 174 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 2. — List of species and number of specimens in each sub-sample — Continued ["X" indicates the species was found during the examination of other fractions of original sample] Species Hugh M. Smith Stranger Orsom Horizon Satsuma Kagoshima 94 132 144 153 178 30 31 29 63 34 10 32 32 613 89. 90. 91, 92. 93. 94. 95. 96. 97. X X i X X X X X X X X X X X X X X X 1 X X X X 99. 100. 101. 102. X X 1 ^ X X 1 X X X X X 3 X 1 1 X 5 X X X X X 1 X 8 15 X 106. 107. X X 1 X 7 X 1 X X 2 22 440 1/16 24.8 31 X X 1 X 109. 110. 1 2 5 ,213 1/256 56.4 27 70 154 1/32 3.7 15 5 216 1/4 2.3 25 X 44 228 1/512 X 30 , 88 1/256 27 2 1/256 27.8 31 X 28 97 1/2 9.3 29 22 134 1/16 3.6 34 70 361 1/64 33.4 27 18 194 1/8 40.8 26 41 133 1/64 &.2 25 23 ,106 1/256 5 Number of specimens in sub-sample.... all Number of copepods per cubic meter.. 55 35 39 2 completing the analysis of the sub-sample, por- tions of the original sample were examined for species not found in the aliquot. The species in each sample are listed in table 2 along with the number of specimens of each found in the aliquot. Representatives of most species from each station were segregated by sex and placed in labeled vials. This material will be deposited in the U.S. National Museum. GENERAL REMARKS Species composition A total of 110 species of calanoid copepods was found in the samples (table 2). Of these, three species belonging to the genera Xantho- calanus, Amallophora, and Scolecithricella are described as new, and three species, Acro- calanus andersoni, Chirundina indica, and Hal- optilus fertilis, are reported for the first time from the Pacific Ocean. The collections examined were obtained from along the Equator from 120° W. to 130° E., a distance of approximately 5,800 miles (fig. 1). Many of the species were widely distributed and occurred in samples collected in the eastern Pacific (east of 170° W.) and those collected in the western Pacific (west of 170° W.). Dis- regarding the small collection made at Kago- shima station 613, the most frequently occur- ring species were Nannocalanus minor, Un- dinula darwini, Clausocalaniis arcuicornis, and Scolecithrix danae. These four species were present in all but one or two samples. A study of table 2 will reveal that there were at least 6 species which were frequently found (4 or more samples) in the eastern Pa- cific, but which were not found in the samples examined from the western Pacific. One of these, Eucalanus subtenuis, was the most abun- dant copepod in three of the eastern Pacific samples and its apparent absence from the western Pacific samples is noteworthy. It is not, however, restricted to the eastern Pacific, as it has been reported from Japan (Fukase, 19.57 ; Tanaka, 1956a) and from the Dutch East Indies (Vervoort, 1946). The other eastern Pacific species have also been reported from these two areas. Equatorial undercurrent samples Four of the fourteen samples examined were obtained from within the equatorial current (Smith 94, 144, 178, 31). The species compo- sition of these samples, when compared with COPEPODS FROM EQUATORIAL WATERS OF PACIFIC OCEAX 175 samples collected from adjacent waters — north {Smith 132), south {Smith 153), and above {Smith 30) and below {Smith 29) the under- current — was not particularly distinctive. Sco- lecithricella tenuiscnata, although apparently absent from one undercurrent sample {Smith 94), did appear in the other three samples and also in two other samples {Sntith 63, Stranger 34) that were collected from depths where the undercurrent could be located, if present, but several hundi'ed miles west of its known limits. Knauss (1959), has indicated that the current may extend as far west as about 160° E. ; in that event the above two samples could have been collected from the current. Even if S. tenuiserrata is typically found within the un- dercurrent, it cannot be considered as an "indi- cator" sensu stricto of the current. In addi- tion to the present samples, this species has been reported from near the Great Barrier Reef (Farran, 1936) and from Japan (Tanaka, 1953). Numerical abundance The number of copepods calculated for those collections in which a current meter was em- ployed, varied from 2.3 to 56.4 with a mean of 21.8 copepods per cubic meter (table 2). These data may be compared to numerical data pre- sented by Brodsky (1952) for the northwestern Pacific Ocean. Using only those collections in which closing nets were used {Smith 94, 132, 144, 153, 178, and 31), a mean number of 26.8 copepods per cubic meter was calculated for depths between approximately 50 and 150 meters in the eastern Pacific Ocean. Brodsky's data included numerical abundances for seven different vertical levels. For the 50 to 100-m. level he gave a figure of 5,040 calanoids per cubic meter and for the 100 to 200-m. level, 320 per cubic meter. Nine species were pres- ent at the former level and ten at the latter level. In comparison, there were no less than 25 species in any of the aforementioned eastern Pacific closing net samples. TAXONOMY For each species I have given references to its occun-ence in the Pacific Ocean, with the exception of those early records summarized by Giesbrecht and Schmeil (1898) and those cited by Vervoort (1946, 1957). The former refer- ence has usually been omitted in the following sections. I have given measurements for sev- eral individuals of a species and usually at all stations where it occurred. The total-length measurements are from the tip of the forehead to the end of the furca, without regard to the telescoped portions of the abdominal segments, and are recorded in mm. unless otherwise indi- cated. Except for PseudochireUa. copepodids which could not be assigned to a species are not mentioned. Included under Remarks are certain taxo- nomic or ecological notes and usually a few brief statements on the diagnostic characters of the species. It is hoped that the specific characters mentioned and the figures presented for each species will be useful to others making identifications of tropical calanoid copepods. All illustrations were made with a camera lu- cida. Family CALANIDAE Calanns tenuicornis Dana, 1849 Pacific records: Vervoort, 1946. Also, Yamada, 1933; Johnson, 1942; Mori, 1942; Davis, 1949; Brodsl->\ 'tyvi FISHERY BULLETIN 187 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 PUBLISHED BY UNITED STATES FISH AND WILDLIFE SERVICE • WASHINGTON PRINTED BY UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON, D.C. For sale by the Superintendent of Documents, U.S. Government Printing OfiBce, Washington 25, D.C. Price 30 cents The series, Fishery Bulletin of the Fish and Wildlife Service, is cataloged as follows : U.S. Fish and Wildlife Service. Fishery bulletin, v. 1- Washington, U.S. Qovi. Print. Oft'., 1881-19 V. in illus., maps (part fold.) 23-28 cm. Some vols, issued ia Oie congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies: v. 1-40, HuUetiii. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, V. 1-23) 1. Fisheries— U. S. 2. Fish-culture— U. S. I. Title. SH11.A25 Library of Congress 639.206173 [59r.55bl] 9—35239* CONTENTS Page Introduction 247 Review of the fishery 247 Methods of collecting data 249 Estimating egg abundance 249 Effect of temperature on length of incubation period 250 Estimates of egg abundance and seasonal-regional distribution of spawners 252 Boundaries of the spawning area 257 Sources of error and bias in sampling eggs 258 Retention of eggs by the nets 258 Sampling of the vertical range 258 Variable distribution of eggs 258 SampHng of the horizontal range of spawning 261 Effects of temperature on spawning 261 Survival of the larvae 265 Regional estimates of abundance of larvae 265 Growth rate of larvae 269 Sources of error and bias in sampling larvae 271 Retention of larvae by the nets 271 Sampling of the vertical range 272 Variable distribution of larvae 272 Sampling of the horizontal range 273 Avoidance of net 274 Summary 275 Literature cited 275 Appendix 277 A. Fish egg incubator 277 B. Staging scheme of jack mackerel eggs 278 m ABSTRACT Distribution and abundance of eggs and larvae of the jack mackerel, Trachurus symmetricus (Ayres), and survival of the larvae are described, utilizing quantitative data collected on monthly cruises of the California Cooperative Oceanic Fisheries Investigations during 1951-54. Spawning in the period 1951-1954 occurred from Washington to Magdalena Bay, Baja California. In each of the 4 years it began in February and ceased by October. The peak month of spawning was March in 1951 and May in 1952, 1953, and 1954. About 30 percent of the spawning occurred during the peak month. Estimates of egg abundance varied by less than a factor of 2 during the 4 years studied. The effect of temperature on the rate of development of eggs was investigated. Regression statistics are given \ for the developmental rate. Reliability of the regression was checked by direct observation of developing eggs at controlled temperatures. The annual estimates of survival for 1952, 1953, and 1954 indicate a reasonably constant survival of month-old jack mackerel larvae in these years. The growth rate of jack mackerel larvae was approximated from data derived by direct observation of developing jack mackerel larvae under laboratory conditions and was described by two successive logarithmic growth curves. The second curve originates at yolk sac absorption and has the lesser slope. Survival data may be broken into two periods: the first period, concurrent with the fast growth period, is characterized by poor sur- vival and may be the critical period; in the second, survival is much better and growth much slower. ABUNDANCE AND DISTRIBUTION OF EGGS AND LARVAE AND SURVIVAL OF LARVAE OF JACK MACKEREL (TRACHURUS SYMMETRICUS) By David A. Farris, Fishery Research Biologist Bureau of Commercial Fisheries The purposes of this study are to delimit both spatially and temporally the spawning of the jack mackerel, Trachurus symmelricus (Ayxes) 1885, and to estimate the abundance of the eggs and the survival rate of the larvae. Quan- titative data collected on monthly cruises of the California Cooperative Oceanic Fisheries Investi- gations, 1951 through 1954, are utilized in the study. Data derived from the stud}^ of eggs and larvae may give insight into the present abundance and future fluctuations of the adult population, and estimates of larval mortalitv' may aid in predict- ing future recruitment to the fishery. With knowledge of the fecundity, estimates of egg abundance may be used to ascertain the present size of the adult population. These data may also be compared vni\\ phj'sical, chemical, and other biological data gathered by the California Cooperative Oceanic Fisheries Investigations. To accomplish the stated purposes of this study, the following information was needed: 1. Boundaries of the area occupied by devel- oping eggs and larvae. 2. Seasonal distribution of the eggs and larvae within those boundaries. 3. Quantitative depth distribution of the eggs and larvae. 4. Relation between temperature and rate of development of eggs and larvae. The author appreciates the help and encourage- ment given him by E. H. Ahlstrom, D. E. Wohlschlag, and John C. Marr during the course of this study; the assistdnc* of O. E. Sette in the preparation of the manuscript; and the valuable advice of Bruce Taft in the preparation of the statistical portion of this paper. George Mattson prepared most of the figures. Also, without the help of members of the Cahfornia Marine Re- search Committee and its cooperating agencies and the staff of the South Pacific Fishery In- vestigations ' this study could not have been undertaken. The laborious proofreading was done by Mrs. Paula K. Farris. REVIEW OF THE FISHERY The carangids most commonly found in the area surveyed by the California Cooperative Oceanic Fisheries Investigations are listed by Barnhart (1936) and Fowler (1944). The family is largely tropical or subtropical in its distribution, Tra- churus symmetricus being a notable exception. Only three members of the family are taken in any numbers by the California Cooperative Oceanic Fisheries Investigations: yellowtail, Seriola dor- mlis Gill; Mexican scad, Decapferu.•= ° \:' 1 •^ ° k 1 1951-1954 »°o ofe 1 ^o o A^- 1 o 6° „ fCAPE MENDOCINO 1 'o ° \:v ' >^o \:- 1 <.o ° *^^^- V"' ^ 60," B%AN FRANCISCO ^^ ^i""" 0%': >f ""^ ^°°°% < O ° "^v o vbO O C^... ^ ^^° 0^"- ° n ° J^POINT CONCEPTION ^ o ^ -°° \°°\»°\ o """^.o-^r, DIEGO -3^- ^'^ o " o ° ^(.^o ° „ oKi. a. ^^~ *° o o J) o o %■•. 5l feV ° 10 'o ° ° ot- " ) \' O ^ " — 1 i 1 1 1 Figure 1. — Chart of stations occupied by the California Cooperative Oceanic Fisheries Investigations 1951-1954. DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 249 1956: p. 8 and 12). The catch fluctuations since 1947 arise primarily from three causes: (1) Avail- ability of sardines and Pacific mackerel, (2) fluc- tuations in demand for canneil jack mackerel, and (3) availability of tlie jack mackerel. Since jack mackerel are packed as "substitute sardines," catch data do not necessarily reflect the size of the adult population. Eventual independence of the industry is assured, however, by the world's in- creasing need for cheap protein and bj^ techno- logical advances in the packing of this species. Jack mackerel are taken with a variety of gear; however, more than 99 percent are taken with purse seines and lampara nets. The operation of this gear is described by Scofield (1951). Some jack mackerel are taken b}' sport fishermen using live bait. In 1953, the sport catch was unusually large, nearly 200,000 fish being taken (Fitch, 1956: p. 27). However, this catch amounts to less than 2 percent of the commercial catch. E.xcept for a minor amount used in the fresh-fish market, the commercial catch is used for canning. Jack mackerel are principally packed sardine style, usually in tall, 1 pound cans. A small part of the catch is packed in other ways. To date, the catch of jack mackerel has not un- dergone any sustained decline. Therefore, all the previously mentioned data take on an added signif- icance when one considers that manv fishery in- vestigations are initiated after the industry has experienced a decline in the number of catchable fish (Walford, 1948). By noting the variations in the strength of spawning, stock size, mortahty, et cetera, now while fishing mortality is relatively low, future observations under conditions of higher fishing mortality should permit the assessment of the effects of man on the population size of the jack mackerel. METHODS OF COLLECTING DATA Since this study of jack mackerel constitutes but part of a larger and more comprehensive studv of the ecology of pelagic fishes off the coast of Cali- fornia, the methods used are those originated by the Bureau of Commercial Fisheries Biological Laboratory at La Jolla and adopted by the stafl"s of the California Cooperative Oceanic Fisheries Investigations. These methods iiave been planned to maximize the amount of information obtainable from this ecological province. The methods used in collecting and processing these data, with a summary of the previous year's work are found in reports of the California Marine Research Committee, the sponsoring organization of the California Cooperative Oceanic Fisheries Investigations, for 1950, 1052, 1953, 1955, and 1956. More detailed explanations are given by Alilstrom (1948 and 1953), and in the following discussion. The station pattern and numbering sj^stem are described in Station Positions of the California Co- operative Sardine Research Program, prepared bj' the Scripps Institution of Oceanography and the U.S. Fish and Wildhfe Service (1952). The sta- tions laid out in lines occupied during the period 1951-54 are shown in figure 1. The exact location of each station (at each occupancy) during 1951- 54 is given by Staff, South Pacific Fishery Investi- gations (1952, 1953, 1954, and 1955). A plankton- net tow (Ahlstrom, 1953), Nansen bottle cast, and bathj'thermograph cast are made at each station. Temperature data are obtained from the reversing thermometers on Nansen bottles and from bathy thermograms. The obliquely hauled plankton net is retrieved from a depth of approximately 140 meters (200 meters of wire out) at an average rate of 20 meters of ware per minute. The angle from the vertical of the towing wire is kept as close as possible to 45 degrees. A current meter placed in the mouth of the net measures the volume of water strained. The sample obtained is preserved in its entirety in a buffered formalin solution, and these preserved samples are subse- quently examined for fish eggs and larvae. The numbers of jack mackerel larvae and localities in which they were taken during 1952-54 are given in Ahlstrom (1954a) and Ahlstrom and Kramer (1955, 1956). The numbers of jack mackerel eggs and localities in which they were taken in 1951-54 are given by Farris (1958). ESTIMATING EGG ABUNDANCE The method used to estimate egg abundance has been described in detail by Sette and x\hlstrom (1948) and Ahlstrom (1954b). The monthly esti- mates of egg abundance are obtained from the relation — n Cm=^ CtW,t, where C\/ = the monthly estimate of egg abun- dance 250 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE n=the number of stations considered Cj=the average number of eggs spawned per day at the ith station^ Wi= the weighting factor for space in standard area (i.e., units of 10 m^ of sea surface) ti=the time factor which is eciual to one-half the time from the preceding occupancy of the station pUis one- half the time to the succeeding occupancy. An annual estimate of abundance is obtained by summing the monthly estimates for the entire year. The eggs are identified and the number belong- ing to each species is recorded by station. The count for each station is adjusted so that all re- sults are expressed as the number of eggs under a standard area which is 10 square meters of sea surface (Ahlstrom, 1953). This standardized haul value (c'«) is the product of the raw count times a standard haul factor, which is derived for each haul by dividing 10 by the average volume of water strained per meter of depth for the entire water column. EFFECT OF TEMPERATURE ON LENGTH OF INCUBATION PERIOD The length of the incubation period (dt) is dependent on temperature of the water mass in which the eggs are developing, and may be pre- dicted if the temperature coefficient for the rate of development is known. The effect of temperature on the rate of develop- ment of jack mackerel eggs has been derived by two methods. In the first method, the eggs were taken from the sea shortly after they had been spawned, placed in a fish egg incubator (see appen- dix A) and observed at 4-hour intervals until they hatched. The temperature of the water in the incubator was maintained at about 14° C, the temperature of the sea water from which the eggs were collected. The observed hatching time was 108.5 hours. This experiment was repeated 1 year later at 15° C, with an observed hatching time of 84 hours. The second method, which is indirect, was de- veloped by Ahlstrom (1943) for the Pacific sardine and was also successfully used by Gamulin and Hure (1955) for sardines in the Mediterranean Sea. A series of arbitrarily chosen but precisely defined morphological stages is selected. Such a series of stages is described for jack mackerel in appendix B. The jack mackerel eggs from station samples are separated into stages and counted (Farris, 1958: table 2, p. 7-11). Several successive days of spawning are indi- cated by the stage frequency modes present in each sample. A mode is interpreted as repre- 1500 1800 2100 2400 0300 TIME OF COLLECTION 0600 0900 1200 ' a is derived as follows: Qi=cJ V • ^ X \X \ X "■ \ VIII \ :• N. V N. • X X \ r^ • • •■" \ \ •^ 126 92 84 75 ^j 68^ in 61 fe t- 55 ? 50 Q 45, 40' o I 12 13 14 15 16 17 18 TEMPERATURE "C. Figure 3. — Relation between temperature and rate of development for three stages of jack mackerel eggs. The close agreement between incubation period predicted and actually observed lent confidence to the reliability of the indirect method. The temperature-dependent incubation period (di) is used to compute the average nimiber of eggs spawned per day at the tth station(C() in the follow- ing manner. The standard number of eggs at the ith station (c'i) is divided by d,. After computing the estimate of the average number of eggs spawned per day at the ith station of 10 square meters, the number of eggs is inte- grated over space to the siu-rounding stations by an area factor (w,). The sample is then weighted on the basis of the area it represents. The boundaries of an area are formed by the perpen- dicular bisectors of lines drawn to the stations immediately surrounding the one under considera- tion. The time factor (it) is derived by taking the number of daj's from the previous occupancy of the station to the occupancy of the station immedi- ately succeeding the one under consideration and dividing by 2. The products {ctWftt) are summed 580553 O — 61- 252 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE for the month to give an estimate of total monthly egg abundance {Cm)- ESTIMATES OF EGG ABUNDANCE AND SEASONAL- REGIONAL DISTRIBUTION OF SPAWNERS Using the method of Ahlstrom (1954b) pre- viously described, the monthly estimates by region were obtained for the 4-year period, 1951-54 (tables 2-5). Spawning does not occur uniformly over the area throughout the spawning season and the seasonal-regional variation is discussed in the following section. The estimates of egg abundance revealed that during this 4-year period, the highest annual esti- mate (1951) was less than twice that of the lowest (1954). Furthermore, no trend was apparent from 1951 through 1954. Spawning averaged 666 trillion eggs per year, with a range of 873 trillion to 462 trillion. The seasonal distribution of spawners was in- ferred from the monthly estimates of egg abun- dance and the number of eggs spawned per month expressed as a percent of the annual total for each year. Although the percentages have been car- ried to hundredths, no statistical significance should be attached to these postdecimal places which merely serve to indicate trace amounts of spawning. These figures are given in the last row in tables 2 through 5. Table 2. — Estimated number (in billions) of jack mackerel eggs in survey area, 1951 [Cruise numbers in parentheses. No eggs taken during cruises 6101 and 5U0 to 6112] Feb. (5102) Mar. (5103) Apr. (5104) May (5105) June (5106) July (5107) Aug. (5108) Sept. (5109) Annual total Percent of total for all areas North of Point Conception: Lines 40-57 1, 365 22, 235 2,907 206 4,272 101, 643 60-77 8,616 8,099 62, 487 Total and percent _. 8,616 8,099 62, 487 23, 600 3,113 105,915 12.13 Southern California: Line 80 44, 060 50, 497 2,476 47 3,863 7,915 4,932 1,228 53 12 106, 288 8,795 41,234 42. 463 138. 742 76, 184 83 85 - -- 40, 344 890 7,334 2,071 87 2,829 8,930 9,706 33, 804 18,977 30. 784 22, 298 9,568 1,973 13, 712 3,107 90 48, 338 30, 972 10, 412 2,559 15 93 - - - Total and percent 55, 819 170, 151 64,963 66, 660 31, 639 14, 199 10, 348 27 413. 706 47.40 Northern Baja California: Line 97 - - 6,379 1,970 10.564 7,469 14, 487 15,014 19, 735 16, 490 6,486 18. 484 3,614 12,964 3,761 3,708 7,421 21, 484 1,442 907 16 8 60, 286 73, 810 38, 307 307 27, 215 100 103 105 307 107 - - 11,282 6,162 7,348 2,433 Total and percent ._ 8,349 43, 802 57.391 35,932 22, 866 28,906 2,656 24 199.925 22.91 Upper central Baja California: Line 110 20,092 17,053 19. 439 6,756 7,263 3,746 8,754 10, 209 890 424 66, 862 37, 764 445 32, 464 22, 839 113 11.'^ 445 117 20. 430 291 8,530 4,781 3.248 2,318 256 14, 993 120 341 115 Total and percent 57, 866 39, 506 16, 575 34,212 1,231 984 150. 374 17.23 Lower central Baja California: Line 123 42 42 63 28 560 30 451 369 204 41 101 430 49 213 40 18 16 3 1,040 153 615 843 49 127 130 133 137 Total and percent. 175 1,410 825 253 34 3 2,700 .31 Southern Baja California: Line 140 165 39 165 39 143 147 150 Total and percent _ ___ 204 204 .02 Grand total 64,168 7.35 272, 198 31.19 171,886 19.69 127, 991 14.66 151, 457 17.35 67, 969 7.79 17 104 .",1 S72. 824 Percent' 1 96 0. ni 1 100.00 1 1 1 ' Hundredths of a percent are used so that trace amounts of spawning may be Indicated (see text above) . DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 253 The regional (north-south) distribution of spawning fish was inferred from the regional dis- tribution of eggs. The number of eggs found in a region was expressed as a percent of the total for all regions (i.e., annual total). These figures are given in the last column in tables 2 through 5. The lines comprising each region are as follows : Region 1 2 3 4 5 6, Area Northern California.. Southern California Northern Baja California Upper central Baja C-.ilifornia. Lower central Baja California, Southern Baja California Lines 40-77 80-93 97-107 110-120 123-137 140-157 A slightly different approach was used to elucidate the offshore-inshore distribution. Un- like the preceding section, only selected stations on selected lines were used, because the selected data were more quickly and easily handled, and because these particular selected stations had the most regular coverage, having been occupied almost every month throughout the 4 3-ears studied. Only those lines and stations that are multiples of 10 were used (e.g., lines 40, 50, and 60, but not 63, 67, or 73, and stations 40, 50, and 60, but not 45, 55, or 65), e.xcept for the most inshore stations. The standardized nimibers of eggs Table 3. — Estimated number (in billions) of jack mackerel eggs in survey area, 1952 [Cruise numbers In parentheses. No eggs taken during cruises 5210 and 5211] Jan. (5201) Feb. (5202) Mar. (5203) Apr. (5204) May (5205) June (5206) July (5207) Aug. (5208) Sept. (5209) Annual total Percent of total for all areas North of Point Conception: Lines 40-57 5,278 49 9,541 49 15,636 60-77 817 Total and percent 5,278 9,590 817 16,686 2. &4 Southern California: Line 80 161 1,993 107 2,769 6,789 22, 624 9,290 3,262 793 31,602 16, 186 3.135 36. 717 122. 608 36. 692 83 85 3,108 27 87 982 3,076 11,713 30,880 30, 302 9,727 4,855 20.144 2,944 90 7,625 54,153 905 2,013 8,293 3,034 3,001 2,262 109 93 Total and percent . . . . 7,786 57, 051 15, 877 80.467 59,867 16, 676 6,828 2.398 246, 940 41 64 Northern Baja California: Line 97 33 1,813 1.631 11, 532 1,691 12, 735 34, 173 26, 325 10, 914 12.380 10,821 24, 782 8,632 4,440 1,428 22 058 4,839 1,368 27 1,600 127 8 83.666 72.632 37 169 38,488 25, 573 100 103 105 173 29, 562 1,278 7,475 107 8,899 12,323 4,323 15 13 2,019 42, 725 15, 704 80,311 60,306 18,823 34, 372 1,410 1,748 257. 418 43.40 Upper central Baja California: Line 110... 63 325 18, 950 1.350 26. 376 1,944 4,877 2,672 751 98 61,017 6,389 3,713 7,783 113 115 117... 61 1,313 132 1,777 539 4,145 2,841 623 140 25 120 . . 1,762 22, 209 33,004 10, 913 989 25 68,902 11.63 Lower central Baja California: Line 123 1.300 627 93 1,978 239 4,044 93 127. 130 . . 133 137 1,300 620 1,978 239 4,137 .69 Southern Baja California: Line 140 143 147 150 Total and percent .00 Grand total 2,019 0.34 60.511 8.62 75, 817 12.78 119,017 20.07 175, 755 29. 63 95, 110 16.04 61, 627 10.39 9,080 1.53 4,146 0.70 693, 082 100.0 Percent '. 1 Hundredths of a percent are used so that trace amounts of spawning may be indicated (see p. 252). 254 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE (instead of estimates of abundance) for these selected lines and stations were summed by 2- month intervals from February to July. The bimonthly totals for the lines were divided by the bimonthly totals for the entire area to give the bimonthly percentage of eggs found on the line. To estimate the offshore-inshore move- ments, stations 100 and seaward (offshore) were grouped together, stations 70-90 (intermediate) were grouped together, and stations 60-shore (inshore) were grouped together for each line. The station groups were summed bimonthly and the percentage of the bimonthly total for each of the three station groups was computed (tables 6, 7, 8, and 9). Table 10 presents a comparison of the estimates of relative regional abundance obtained from the standardized numbers of eggs at selected stations on selected lines and the regional distribution based on estimates of abun- dance using all data. Since the north-south regional distribution indicated by the selected data agreed with the north-south distribution indicated by all the data, no distortion was anticipated in using the selected data to reveal the offshore-inshore distribution. Table 4. — Estimated number {in billions) of jack mackerel eggs in survey area, 19BS (Cruise numbers in parentheses. No eggs taten during cruises 6301, 6309, 6311, and 5312) Feb. (5302) Mar. (5303) Apr. (5304) May (5305) June (5306) July (5307) Aug. (5308) Oct. (5310) Annual total Percent of total foraU areas North of Point Conception; 60-77 12 6,075 1,947 1,619 9,663 Total and percent . 12 6,076 1,947 1,619 9,663 1.31 Southern California: 3,783 5. 512 1,378 22, 633 2, 835 162 12.813 60, 199 14,006 3,774 6,639 590 3,285 1,241 12, 776 6,570 5,402 3,252 1,769 7,016 10, 588 288 18 55 378 91 36.048 13, 876 4.012 17, 922 74, 393 38, 844 83 _ _ 85 87 . 90 47 6 93 10, 673 112, 638 27,305 33, 597 830 52 185, 095 26.14 Northern Baja California: Line 97 14, 348 4,508 4.327 2,700 6,722 19, 560 6,664 21, 840 36,880 14, 660 16,905 4,666 8,939 4,601 602 1,232 619 181 65, 915 52. 232 49, 587 72. 527 113.2.56 103 105 2,196 70, 331 107 66,907 19, 528 14, 200 10, 606 2,016 Total and percent 16, 544 74,839 79,656 66, 582 82, 646 28, 702 4. .168 181 3.63, 617 48.02 Upper central Baja California: T inp no 9,585 6,671 27, 732 738 59,051 17, 658 5,847 1,376 1,228 3,412 2,250 721 1,122 106. 815 29, 576 117 4,239 1,743 1,964 7,505 3,779 6,090 12, 936 2,324 247 231 124 21,363 18, 829 120 - Total and percent 21, 238 30, 434 87, 993 26, 249 7,211 3,326 1,122 176, 573 23.99 Lower central Ba]a California: Line 123 6,701 1,974 1.327 81 1,022 140 41 8 6,782 2.996 1,467 41 8 127 130 133 137 Total and percent _ 10, 002 1,292 11,294 1.63 Southern Baja California: Line 140 . . 143 147 . . - 160 Total and percent _ _ 37, 782 5.13 105, 273 14.30 178, 334 24.22 214, 471 29.13 124. 528 16.92 67, 572 9.18 7,939 1.08 233 0.03 736, 132 99.99 ' Hundredths of a percent are used so that trace amounts of spawning may be indicated (see p. 262). DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 255 Table 5. — Estimated number (in billions) of jack mackerel eggs in survey area, 1954 [Cruise numbers in parentheses. No eggs taken during cruise 5412) Jan. (5401) Feb. (5402) Mar. (5403) .^pr. (5404) May (,M05) .Tune (5406) Julv (5407) Aug. (5408) Oct. (5410) Annual total Percent of total for all areas North of Point Conception: 82 7,697 82 31,034 13, 610 9,685 42 Total and percent 13, 610 7,779 9,685 42 31, 116 6.73 Southern California: 43 165 17, 090 6.529 24,012 14, 898 25,822 3,185 11.430 3,894 3,822 8.166 13,838 9.797 3. 666 4.905 1,010 4,937 1, 239 221 .12 9.825 6.053 31 25 1,596 874 16 28,044 41,095 13,203 32,263 26, 749 35, 787 83 85 87 90 93 Total and percent 47, 839 63,051 41.382 22,327 2, 526 16 177, 141 38.31 Xorthem Baja California: fane 97 8,034 878 7,846 40. 461 2.178 22, 602 21,729 7,505 4.428 10,523 11,424 11.690 985 915 L632 7,289 256 539 48,454 89, 560 22,900 100 103 105 107 13,887 1,220 4,029 3,717 1,998 24,851 Total and percent 22,799 51, 705 55, 865 30,092 15,588 8,921 796 185. 765 40.18 Upper central Baja California: Line 110 148 208 846 365 4,137 6.451 4.301 6,622 4.861 1,820 538 2,978 70 40 18, 592 14,793 113 115 117 464 208 924 2,008 585 4.547 2,011 4,671 392 33 4,017 11, 826 120 Total and percent 148 1,518 4,710 13,684 16, 615 9,040 3, 440 73 49,228 10.65 Line 123 168 H 1) (1 n (1 II 124 915 2,160 2.984 12, 526 207 U 4 12. 650 1,126 2.160 2.984 168 127 130 133 137 Total and percent 168 6,183 12, 733 4 19,088 4.13 Southern Baja California: n 143 147 150 Total and percent 316 0.07 24,317 5.26 56, 415 12.20 123, 671 26.73 136. 101 29.44 73. 793 15.96 44,373 9.60 3.436 0.74 16 462.338 100.00 Percent ' .- ' Hundredths of a percent are used so that trace amounts of spawTiing may be indicated (see p. 252). Spawning in 1951 began in Fobruarv. About 7 percent of the total number of eggs for the season were spawned during this month. Spawn- ing rose to a peak of more than 30 percent of the total in March, and then gradually declined until June, when an increase occurred. Spawning de- creased thereafter, being negligible in September. Spawning during February and March was centered about 150 miles offshore in region 2. During the ne.xt 2 months spawning was more general and no compact center was observed. Spawning had reached its widest distribution (most of the eggs being taken between lines 00 and 120) ill April and May, and tiie center liud moved inshore. During June and -July tlie center was dispersed and offshore. Spawning in 1952 began in January, when less than 1 percent of the total number of eggs were spawned, and rose to a peak of about 30 percent in May. It then declined to less tlian 1 percent in September and ceased altogether by October. The center of spawning was about 120 miles farther south during February and March than during the same period of 1951. Once again it was in the intermediate area. During April and May, spawning became heavier in region 3 and there was a strong inshore movement. During tlie final 2 months the spawners were grouped to 256 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 6. — Relative north-south, inshore-offshore distribu- tion of jack mackerel eggs by 2-nionth intervals for 1951 [Standard haul totals] Table 8. — Relative north-south, inshore-offshore distribu- tion of jack mackerel eggs bij 2-month intervals for 1953 (Standard haul totals] Line Offshore stations (100- seaward) Inter- mediate stations (90-70) Inshore stations (60-shore) Total Percent February-March: 60 2,796 1,506 91 7.790 9.013 955 477 1.658 1.484 907 10, 063 74 12,243 12,003 1,953 10, 630 74 70 - go _. 33.3 90 32.6 100 5.3 110 - 28.6 120 - .2 Total 4,392 11.9 18. 235 49.6 14, 176 38.5 36, 803 100.0 100.0 jr ercent April-May: 60 1,673 2,502 3,430 589 647 847 14 27 629 1.832 201 526 309 160 2,976 7,312 2,506 638 14 1.600 3.291 8.238 8.102 3.677 1,694 .1 70 _ 6.0 80 12.4 90 --- 31.0 100 30.0 no 13.8 120 - 6.4 Total 9,588 36.0 3.537 13.3 13.491 60.7 26.616 100.0 99.7 Percent June-July: 60 1.490 137 40 945 15 222 2,613 1,100 284 1,134 765 648 6 1,414 664 2,399 802 420 66 1,718 4,164 1,804 3,628 1,961 1,185 704 11.3 27.6 80 - - 11.9 23.9 100 12.9 7.8 120 4.6 Total 2.627 17.3 6,766 44.6 5,761 38.0 15, 154 99.9 99.9 Table 7. — Relative north-south, inshore-offshore distribu- tion of jack mackerel eggs by 3-month intervals for 1962 [Standard haul totals] Line Offshore stations (100- seaward) Inter- mediate stations (90-70) Inshore stations (60-shore) Total Percent February-March: 60 (') (') 11 (') 0) (') (') (') (') 377 6,665 4,528 4 142 (>) (') 2,027 2 0) (') 388 6,665 6,555 6 142 (') 70 (') 80 2.8 90 48.5 100 47.7 110 <0.1 120 1.0 Total 11 0.1 11,716 86.2 2,029 14.7 13. 766 100.0 100.1 April-May: 60 223 206 82 (') (') 3,635 1,724 5,495 293 4,805 6,107 667 159 223 8.645 7.913 6.062 452 70 80 1.0 90 37.1 100 34.0 110 26.0 120 1.9 Total— 510 2.2 11,147 47.8 11,638 60.0 23.295 100.0 100.0 Percent June-July: 60 6 429 3,496 74 161 (') (') 535 270 2,235 3,766 601 398 36 1,098 950 1,208 665 13 576 699 6,828 4,790 1,970 1,063 13 3.6 70 4.4 80 42.8 90 30.0 100 12:4 no 6.7 120 - . 1 Total 4,164 26.1 7,806 49.0 3,970 24.9 15, 939 100.0 100.0 Percent _ . _ Line Offshore stations (ICO- seaward) Inter- mediate stations (90-70) Inshore stations (60-shore) Total Percent February-March: 60 (') (1) (') (') (') (') (') (■) (1) 1,551 1,138 (') (') (') 4,445 4,399 228 (') (') 5.996 5,637 228 (>) 70 (') 80 --- 90 100 --- 51.0 110 47.1 120 - 1.9 Total 2,689 22.9 9,072 77.1 11,761 100.0 100.0 Percent April-May: 60 (') 1,149 1,726 (1) (') 0) 2,452 6,631 680 6.437 403 8 18 298 805 8.721 3,679 8 3,619 7.654 1.385 14,168 4,082 70 <0.1 80 11.7 90 24.8 100 4.5 110 45.8 120 13.2 Total -- 2,874 9.3 14, 603 46.9 13, 529 43.8 30, 906 100.0 100.1 Percent June-July: 60 (■) 86 21 (') 161 (0 (>) 116 703 423 962 (') 17 96 61 667 364 3,456 662 10 96 262 1,381 787 4,559 662 27 1.3 70 3.4 80 18.0 90 10.3 100 69.4 110 7.3 120 .4 Total 257 3.3 2,211 28.8 6,206 67.8 7,674 99.9 100.1 ' Region not occupied. Table 9. — Relative north-south, inshore-offshore distribu- tion of jack mackerel eggs by 2-month intervals for 1954 [Standard haul totals] Line Offshore stations (100- seaward) Inter- mediate stations (90-70) Inshore stations (60-shore) Total Percent February-March: 60 (') (') (') 0) (') 0) (') 0) 0) 4,167 12 1,207 186 (') 0) 6,374 185 12 (■) 70 (1) 80 90 100 --- 96.6 110 3.3 120 0.2 Total 4,179 75.0 1,392 25.0 6,571 100.0 100.0 rerceni April-May: 60 43 0) (') (') 987 (') (') 180 746 2.326 2.051 1,206 351 423 40 19 616 1,161 3,308 36 223 785 2. 345 2. 667 3,344 3,659 469 1.7 70 80 17.4 90 100 110 27. 1 120 1.030 7.6 7,282 54.0 6,170 38.3 13, 482 99.9 100.0 June-July: 60 60 0) 347 (') 0) (') (') 816 48 743 404 135 92 270 784 936 1,931 367 302 153 866 832 2.026 2,336 602 394 423 11.7 70 80 90 100 110 120 - Total 397 5.4 2,508 34.0 4,473 60.6 7,378 100.0 99.9 ' Region not occupied. Region not occupied. DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 257 Table 10. — Comparison of estimates of regional distribution using selected stations on selected lines and all data, 1951-54 Feb- Selected stations All Line ruary- March April- May June- July stations (percent) Total Percent 1951: 60 14 1,718 1.732 12.20 } 12. 13 70 1,600 4.164 5.764 7.34 80 12.243 3.291 1.804 17.338 22.07 22.77 90_- 12.003 8.238 3.628 23.869 30.38 31.53 100 1.953 8. 102 1.951 12.006 15.28 16.00 110 10.530 3.677 1,185 15.392 19.59 14.61 120 74 1.694 704 2.472 3.15 2.95 Total— 78.573 100.01 99.99 1952: 60 (>) 576 576 1.09 1 2.64 70 (") 699 699 1.32 80 -. 388 223 6,828 7,439 14.04 14.78 90 6.665 8.645 4,790 20,100 37.93 40.95 100 6.555 7.913 1,970 16, 438 31.02 29.31 110 6 6,062 1,063 7,131 13.46 10.31 120- - 142 452 13 607 1.15 2.01 52,990 100.01 100.00 19.W: 60... (') 96 96 0.19 1 1.31 70... (') 8 262 270 .54 80 3.619 1,381 5,000 9.91 9.76 90.-- 7,654 787 8.441 16.77 24.34 100 5,996 1.385 4, 5.59 11.940 23.72 39.07 110 5,537 14.158 562 20. 257 40.24 21.43 120 228 4.082 27 4,337 8.62 4.09 Total -. 50,341 99.99 100.00 1954: 60.-- (") 223 866 1,089 4.12 } 6.73 70 (') 785 832 1,617 6.12 80 . 2,345 2,026 4,371 16.54 24.79 90 2.667 2,335 5,002 18.92 24.01 100 5.374 3.344 502 9,220 34.88 29.70 110 185 3.659 394 4,238 16.03 8.09 120 12 459 423 894 3.38 6.69 Total 26,431 99.99 100.01 1 Hundredths of a percent are used so trace amounts of spawning may be indicated (see p. 252). ' Region not occupied. the north in region 2 and were farther offshore in the intermediate area. Spawning in 1953 began in February and rose to a peak of nearly 30 percent of tlie total in May, and then declined. Only small nxmibers of eggs were taken in October. Early spawning was centered in region 3, with a fair amount in region 2. The center of spawning was about 240 miles farther south than the early spa^vning of 1951, and most of the spawning was inshore. From this center, spawning moved both north and south and somewhat offshore to the intermediate area. During the final 2 months, the center of spawning was again in region 3 and inshore. The monthly distribution of spawning for 1954 is very similar to that for 1953. Spawning ex- tended from January to October and reached a peak of nearly 30 percent of the total in May. Early spawning was centered in region 3 about 150 miles offshore (intermediate area) and no eggs were taken off California. In the next 2 months the heavier concentrations extended north to region 2 and south to the northern edge of region 4, with some inshore movement. During tlie final 2 months the center shifted northward to region 2 and inshore. The distribution and relative abundance of jack mackerel eggs for 1951-54 are illustrated in Farris (1958: figs. 3-6). Tlie remarkable similarity of monthly distri- bution of spawning in 1952 through 1954 (peak month. May) is illustrated in figure 4. The year g .10 I9SI © i952 / \! ^ \ e 1953 • 1954 /a r ^ ^ N 1 y Y 1 N U^ ^ FEB MflR APR MAY JUNE JULY AUG SEPT OCT Figure 4. — Proportion of annual spawning of jack mackerel, by months, 1951-54. 1951 appears to have been an anomalous year with an early peak of spawning in March. The proportion of spawning that occurred during the peak month of each year, including 1951, was approximately three-tenths of the total for the year. Furthermore, over seven-tenths of the spawning for any year occurred during the first 5 months. Of the 4 years studied, 1951 had the highest proportion of the annual total number of eggs in region 1. High proportions of the annual total eggs were found in regions 3 and 4 during 1953, with a small proportion being taken in regions 1 and 2. BOUNDARIES OF THE SPAWNING AREA The northern and western boundaries of jack mackerel spa\vning during August and September were established by expedition Norpac, an exten- sive study of the north Pacific in 1955 conducted by the California Cooperative Oceanic Fisheries Investigations and other agencies (Ahlstrom, 1956: p. 39; Ahlstrom and Kramer, 1957: p. 55). These boundaries may be less certain than the others because the study of the area was more limited in time. The eastern and southern 258 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE boundaries were established from data collected on regular survey cruises of the California Coop- erative Oceanic Fisheries Investigations (Farris, 1958). Spawning areas are approximately bound- ed by the 26th parallel on the south, the 45th parallel on the north, the coast of North America on the east, and the 150th meridian on the west. SOURCES OF ERROR AND BIAS IN SAMPLING EGGS These data and subsequent interpretations are subject to errors inherent in the collection proce- dures. The types of error investigated and evaluated were (1) completeness of retention of eggs by the nets; (2) completeness of sampling of the vertical range; (3) sampling error owing to a variable distribution of eggs in space and time; and (4) incomplete sampling of the horizontal range of jack mackerel spawning. RETENTION OF EGGS BY THE NETS The eggs are fully retained by the net once they are in it, because the plankton nets have a stretched mesh of 0.5-0.7 mm. (Ahlstrom, 1953), and the spherical eggs range in diameter from 0.9-1.1 mm (Ahlstrom and Ball, 1954). It would appear, therefore, that no eggs are lost through the mesh of the sampling net. SAMPLING OF THE VERTICAL RANGE Investigation of the vertical distribution of jack mackerel eggs (Ahlstrom, 1959: table 7) with a horizontally towed closing net reveals that most of the eggs occur in the upper 40 meters of water. Jack mackerel eggs have rarely been taken below 100 meters, and never below 140 meters. The bulk of the eggs have occurred above the thcrmocline. Since plankton hauls are routinely made from a depth of 140 meters, which has always included the thermocline depth, it seems likely that the vertical distribution of jack mackerel eggs is completely sampled. VARIABLE DISTRIBUTION OF EGGS The distribution of jaclt mackerel eggs is vari- able with respect to both time and space. An illustrative example is given in figure 5. The standard numbers of jack mackerel eggs for line 97 have been plotted by station for 4 months. 5,000 45 STATIONS Figure 5. — Standard numbers of jack mackerel eggs fovmd on line 97 during 4 months. The average number of eggs per station by month is given below. Year: April May 1953 181 1,116 1954 762 143 The average station on hne 97 in May 1953 contained six times as many eggs as the average station in the preceding month. Assuming that this change was rectihnear in time, the estimate would be altered considerably, depending on which day of the month the sample was taken. Spatial variability is indicated by the data for April 1954, in which a change of 1 order of magni- tude occurs within 20 miles. There is at least one such combination of adjacent stations having as great a change in the distribution of jack mackerel eggs for each cruise illustrated. The grid of stations occupied is too coarse except for fairly rough estimates of egg abundance. Al- though more frequent sampling of more closely DISTRIBtTTION OF EGGS AND LARVAE OF JACK MACKEREL 259 spaced stations is highly desirable, such samphng cannot be effected, since it would raise the current cost of sampling prohibitively. The monthly samphng of the Cahfornia Coop- erative Oceanic Fisheries Investigations grid of stations has all but precluded the simple assess- ment of the error associated with these estimates of abundance. Although the construction of the proper statistical model was not mthin the scope of this investigation, I was able to make an estimate of the error arising from the practice of hnear interpolation of egg numbers in time and space. This calculation was possible because in 1953 and 1954 stations which were only 20 miles apart were occupied, and the samples contained jack mackerel eggs. In 1952, a few stations containing jack mackerel eggs were occupied in late March. The errors arising from stratification in space (i.e., spacing sampling stations 40 miles apart) and time (i.e., spacing sampling cruises 1 month apart) were considered. Standard numbers of eggs for stations 20 miles apart- — obtained by linear inter- polation of values obtained from stations 40 miles apart^ — ^were compared with values actually ob- served. In hke manner, standard numbers of eggs for stations sampled at intervals of one-half month^ — obtained by linear interpolation of ob- served values from stations sampled at monthly intervals — were compared with the values actually observed. The differences should be zero if no error arises from linear interpolation through space and time. Since the average difference cannot be expected actually to equal zero, owing to sampling variability, the 95-percent confidence limits for both estimates were computed. These limits should include zero. The error arising from spacing the sampling stations 40 miles apart was estimated using data from selected stations on selected lines. (Stations having no eggs were not used.) These stations were placed 20 miles apart. An estimated value was given for every other station by a process of linear interpolation of values between the remain- ing stations which were 40 miles apart. These interpolated values were then compared with the actual number of eggs found at the stations and the difference (Aj) calculated. (A(=obscrved standard number of jack mackerel eggs minus estimated number of jack mackerel eggs.) The deltas were averaged to give A, the average differ- 5S0553 O— 61 3 ence between the observed standard number of jack mackerel eggs and the number calculated by linear interpolation. The frequency distribution of A< was plotted and appeared to be normally distributed (fig. 6). [— r ! rn J" — 1 -600 -400 -200 200 400 600 800 A, Figure 6. — Frequency distribution of A, (the diflFerence between observed and estimated numbers of jack mackerel eggs). Therefore A( was considered to be a normally dis- tributed random variable with mean A and variance 5^. statistic Value Description A 7.8 518.956 720 74.2 94 Average difference. s'Ai. S^i Variance of the individual diflerences. Standard deviation of the Individual diflerences. »A -. Standard deviation of the mean. Number of diflerences observed. -137.2$ AS 152.8 The average difference is close to zero. The wide confidence limits indicate the high variability associated with any one observation. The A( were compared with the size of each ith haul to determine whether the differences were related to population size (fig. 7). No relation was evident and it was concluded that the A< were not related to population size. The error arising from stratification in time was estimated by considering the standard numbers of jack mackerel eggs taken at selected stations during cruises spaced about 2 weeks apart (March, late March, and April, 1952). By linear inter- polation of the number of eggs taken at a station between the March and April occupancies, an estimate of the number of eggs that should occur at the station during the late March occupancy was obtained. This estimated number was then 260 FISHERY BtlLLETDSr OF THE FISH AND WILDLIFE SERVICE Figure 7. — Relation of A,- (the difference between ob- served and estimated numbers of jack mackerel eggs) and the number of eggs observed at the ith station. compared with the number actually observed and the difference (A,) noted. (A( = the standard num- ber of jack mackerel eggs estimated by linear interpolation minus the number actually ob- served.) The differences were averaged to give a mean difference (A) between the estimated num- ber of eggs and the actual number of eggs. The average difference (A) was minus 11, with a vari- ance of 3370. The 95-percent confidence hmits on A are minus 47 to plus 25. Although the individual errors arising from the practice of linear interpolation of jack mackerel eggs in time and space were high and variable, the average error tended toward zero. I concluded that for a large number of samples (i.e., inter- polations) the error arising from linear interpola- tion of the number of eggs in time and space was negligible. A further indication of irregularities in the spatial and/or temporal distribution would be the nonconcurrence of eggs and larvae in the sampling areas. In table 11 the occurrences of eggs and larvae, by regions, are compared. In region 1, Table 11. — Occurrences of jack mackerel eggs and larvae, by month and region, 1952-54 Region 1 ' Region 2 Region 3 Region 4 Region 5 Region 6 Date Sta- tions occu- pied Sta- tions with eggs Sta- tions with larvae Sta- tions occu- pied sta- tions with eggs sta- tions with larvae Sta- tions occu- pied sta- tions with eggs Sta- tions with larvae Sta- tions occu- pied Sta- tions with eggs sta- tions with larvae Sta- tions occu- pied sta- tions with eggs sta- tions with larvae sta- tions occu- pied Sta- tions with eggs Sta- tions with larvae 1952: January 28 27 25 37 50 58 38 22 22 21 22 5 4 14 31 36 31 10 6 3 6 12 28 42 22 8 8 1 19 20 20 31 43 58 32 17 17 15 16 3 7 8 24 35 34 27 3 5 1 12 26 37 42 25 9 3 2 22 30 25 36 40 42 36 30 19 19 21 7 17 30 21 9 1 11 13 25 18 11 25 18 21 33 31 30 27 15 15 15 14 1 6 5 3 3 10 7 3 2 Febraary 14 April 18 23 34 46 15 21 18 18 4 14 2 3 4 1 May June _ _ July.-- August September October November. . December... Total 193 20 8 350 137 130 288 146 167 320 85 78 244 14 25 14 1953: January... _ 34 35 36 61 63 56 38 38 18 18 19 28 9 41 31 21 11 2 s 1 3 8 23 20 16 6 I 20 17 17 41 38 43 20 20 2 9 30 35 33 14 8 4 13 21 22 34 8 3 16 24 29 38 37 40 21 21 24 21 8 11 31 28 19 8 3 5 12 18 21 23 5 2 4 13 19 31 35 35 31 19 20 8 11 1 10 9 13 March _._ _ April 20 30 26 24 19 18 9 6 2 2 1 June _ July August September October 16 2 6 14 1 November December 17 21 16 Total... 119 34 6 444 115 79 249 133 111 292 108 90 233 19 21 13 1954: January .__ 2 2 2 18 19 36 19 22 8 13 8 1 2 9 6 2 42 37 35 61 64 48 37 39 16 33 31 13 10 13 23 20 11 13 22 24 32 39 42 39 20 19 11 15 29 34 25 9 6 3 19 26 34 8 27 31 48 62 52 51 22 22 1 6 8 24 26 17 6 2 4 19 19 29 22 3 23 22 36 38 38 35 20 20 I 8 9 1 2 1 16 8 6 1 18 March April.... May June July... August September... October 2 38 2 1 22 23 20 1 November December 36 16 26 19 18 Total. . 122 30 19 417 105 81 275 129 133 354 88 96 271 19 34 36 > See page 2S3. DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 261 there are more occurrences of eggs than of larvae, while in region 5, the converse is true. This may indicate southern transport of the larvae by the California Current. The high proportion of stations occupied in region 2 at which eggs or larvae were taken is a further indication (see also tables 3, 4, and 5) that this is the region of maximum spawning activity. A comparison of total occurrences of eggs with total occurrences of larvae is interpreted as indicative of a distribution more regular than that encountered for other pelagic eggs and larvae (e.g., sardines). SAMPLING OF THE HORIZONTAL RANGE OF SPAWNING To determine the proportion of the total annual spawning which might be missed by failure to extend the sampling far enough seaward, the proportion of eggs taken beyond stations 90 (the usual seaward extent of sampling) was computed (see column 100 — seaward in tables 6, 7, 8, and 9). This areal proportion was multiplied by the pro- portion of annual spawning that occurred during the appropriate 2-month interval (tables 2, 3, 4, and 5) to give a bimonthly estimate of the pro- portion of annual spawning which might be missed by failure to extend the sampling suffi- ciently seaward (table 12). It would appear that at least 21 percent of the annual spawning has occurred seaward of stations 90, and that a portion of the eggs has been missed in those years when monthly sampling was not extended beyond that point. Table 12. — Proportion of jack mackerel spawning occurring seaward oj station 90, by 2-month intervals, 1951-54 Month 1961 1952 1953 1954 0.05 .12 .04 0) 0.01 .07 (■) 0.05 .01 (') April-May 0.04 .01 .21 .08 .06 .06 1 Region not occupied. Data from the previously mentioned Norpac indicate that the proportion of eggs spawned in northern waters (north of line 40) may be between 1 and 2 percent of the annual total. This pro- portion is minimal, since spawning occurs in periods other than that covered by Norpac. This estimate was determined by estimating the total spawning in the region for the period August-early September (8,655 billion oggs) and comparing that figure with the estimated number of eggs spawned in 1954 (464,452 billion) and 1951 (874,322 billion). It is therefore inferred that an appreciable amount of annual spawTiing may occur west of stations 90 but a lesser amount takes place north of line 40. EFFECTS OF TEMPERATURE ON SPAWNING Temperature may have at least two effects on the jack mackerel. It influences the rate at which the eggs develop (see p. 250), and it may well govern when and where the adults spawn. To determine the effect of temperature on the spawning jack mackerel, temperatures at 10-meter depths were tabulated by 0.5° C. intervals for all stations where jack mackerel eggs were taken in 1951 through 1954 (table 13). The temperature at tjiis depth is generally representative of the strata in which jack mackerel eggs are most abundant. These data were examined also for seasonal effect by dividing the year into an early (January-May) and a late (June-December) period. The effect of temperature on the size of haul was also examined by dividing the samples into two categories: hauls containing 1-100 eggs and hauls containing 101 eggs and more. A seasonal trend toward higher spawning temperatures in the late summer with a greater temperature range was indicated (table 14 and figure 8). The data were then tabulated by 14 25 15:25 TEMPERiTURE *C Figure 8.— Percentage of early (January-May) stations, of late (,Iune- December) stations, and of the total stations having jack mackerel eggs, shown by 0.5° C. temperature intervals measured at 10- meter depth. 262 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 13. — Temperature distribution at 10 meters of stations having jack mackerel eggs, hy season and size of sample, 1951-S4 [Id 0.6° C. intervals] Temperature 1951 1952 1953 1954 Total Percent of Early Late Total Early Late Total Early Late Total Early Late Total Early Late Total total 1-100 eggs: 10.00-10.49 10.50-10.99 11.00-11.49 _, 11.60-11.99 12.00-12.49 12.60-12.99 13.00-13.49 13.50-13.99. 14.00-14.49 14.50-14.99 15.00-15.49 16.50-15.99 16.00-16.49 16.50-16.99 17.00-17.49 17.50-17.99 18.00-18.49 18.50-18.99 19.00-19.49 19.50-19.99 20.00-20.49 20.50-20.99 1 2 3 3 9 10 6 4 10 15 17 14 9 6 6 2 2 1 4 2 5 3 6 3 11 11 17 8 12 6 3 6 2 1 1 2 3 3 7 5 14 13 12 7 21 26 34 22 21 12 8 7 2 1 1 1 1 3 7 4 4 13 6 11 8 2 1 1 1 1 1 5 4 2 3 6 7 5 15 14 20 3 6 1 1 1 1 1 1 2 6 7 9 7 10 20 11 26 22 22 4 5 1 1 1 1 4 1 4 2 1 6 7 5 10 4 8 4 1 1 1 1 1 4 4 6 7 5 6 10 6 10 4 6 5 5 3 6 3 2 1 1 6 1 2 8 6 6 13 12 10 20 10 18 8 7 6 5 3 5 3 2 1 1 1 3 1 2 1 10 11 17 13 17 13 8 1 1 3 1 4 3 6 3 6 1 9 12 6 6 7 6 2 6 4 1 2 6 1 1 6 4 16 14 23 14 26 25 14 6 7 6 2 7 4 1 5 2 6 8 7 15 24 27 24 50 38 53 39 20 8 6 1 1 3 5 6 10 11 18 16 20 17 34 23 51 38 44 20 20 16 10 10 7 2 2 8 7 12 18 18 33 39 47 41 84 61 104 77 64 28 25 17 10 11 7 2 0.2 1.1 1.0 1.7 2.5 2.5 4.6 6.6 6.6 6.7 11.7 8.6 14.5 10.7 9.0 3.9 3.5 2.4 1.4 1.5 1.0 .2 Total 115 106 221 61 96 167 69 93 162 99 86 185 334 381 716 99 7 lOlH- eggs: 10.00-10.49 10.50-10.99 11.00-11.49 11.50-11.99... 12.00-12.49 12.60-12.99 13.00-13.49 13.50-13.99... 14.00-14.49 - 14.50-14.99 16.00-15.49 15.50-15.99.. 16.00-16.49 16.60-16.99 17.00-17.49 17.50-17.99 18.00-18.49 18.50-18.99 19.00-19.49 19.50-19.99 20.00-20.49 20.60-20.99 2 1 4 7 19 14 29 23 20 6 1 1 1 1 2 2 8 11 3 3 4 9 6 4 3 4 1 3 3 6 15 30 17 32 27 29 12 6 4 4 1 5 2 8 26 8 22 14 7 2 1 1 6 2 1 10 7 7 6 9 6. 7 1 1 1 1 10 4 9 35 15 29 20 16 8 8 1 1 1 ! 3 2 2 6 13 21 25 24 7 5 2 1 1 2 3 6 6 11 8 8 4 8 5 3 2 4 2 3 2 4 9 19 27 36 32 15 9 10 6 3 2 4 2 6 3 14 24 23 22 13 13 1 2 2 5 6 2 4 1 3 1 1 1 4 8 8 20 26 27 23 16 15 4 1 1 1 6 5 17 18 54 84 85 91 54 31 6 3 2 1 5 11 18 24 21 25 20 26 21 21 15 8 2 5 1 2 3 7 10 28 36 78 105 110 111 80 52 27 IS 8 2 5 1 2 .4 1.0 1.5 4.1 5.3 11.4 15.4 16.2 16.3 11.7 7.6 4.0 2.6 1.2 .3 .7 .1 .3 Total . . 127 61 188 95 64 169 112 71 183 121 32 153 456 228 683 100.1 Table 14. — Summary: Distribution of temperatures at 10 meters of stations having jack mackerel eggs, by season and size of sample, 1951-54 [In0.6°C. intervals) Early season Late season Season total Temperature 1-100 eggs 101-t- eggs Total Percent of total 1-100 eggs lOH- eggs Total Percent of total 1-100 eggs 101-F eggs Total Percent of total 10.00-10.49 1 6 2 6 8 7 15 24 27 24 50 38 63 39 20 8 6 1 1 1 6 5 17 18 64 84 85 91 54 31 6 3 1 5 2 7 14 12 32 42 81 108 135 129 107 70 26 11 6 1 1 1 .6 .3 .9 1.8 1.5 4.1 5.3 10 3 13.7 17.1 16 3 13.6 8.9 3.3 1.4 .6 .1 .1 1 3 6 6 10 11 18 16 20 17 34 23 51 38 44 20 20 16 10 10 7 2 2 1 6 11 18 24 21 2b 20 26 21 21 16 8 2 5 1 2 1 3 5 8 U 16 29 33 44 38 59 43 77 69 66 35 28 18 16 11 9 2 2 .5 .8 1.3 1.8 2.6 4.8 5.4 7.2 6.2 9.7 7.1 12.6 9.7 10 7 6.7 4.6 3.0 2.6 1.8 1.5 .3 2 8 7 12 18 18 33 39 47 41 84 61 104 77 64 28 25 17 10 11 7 2 3 7 10 28 36 78 105 110 111 80 52 27 18 8 2 5 1 2 2 8 7 16 25 28 61 75 126 146 194 172 184 129 91 46 33 19 15 12 9 2 0.1 10.60-10.99. .6 11.00-11.49.. .5 11.50-11.99.. 1.1 12.00-12.49... 1.8 12.50-12.99 2.0 13.00-13.49 4.4 13.50-13.99.. 6.4 14.00-14.49 8.9 14.50-14.99 10.4 15.00-15.49. 13.9 15.60-15.99 12.3 16.00-16.49.. 13.2 16.50-16.99. 9.2 17.00-17.49 6.5 17.60-17.99 3.3 18.00-18.49. 2.4 18.60-18.99 1.4 19.00-19.49.. 1.1 19.60-19.99 .9 20.00-2049 .6 20.50-2099 .1 Total 334 455 789 100.0 381 228 609 lOOO 716 683 1,398 100.1 DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 263 1 1 1 1 y 1954 " / /- 1953 — - / ^.1952 x'^l95l - ..'"'> N.^ \ s y /^ " \ / - ^- 1 1 1 \ Table 16. — Average temperalures at 10 meters during jack mackerel spawning, by season and size of sample, 1951- 54 Figure 9. — -Monthly median temperatures at 10 meters, at which jack mackerel eggs were spawned. monthly intervals and the median temperatures at 10 meters computed for each month. The monthly median temperatures are shown in figure 9 and indicate that spawning generally occurs at lower temperatures in spring than in late summer. The within-year median temper- ature shift is greater than the between-year temperature variation (see below). The data from table 13 are summarized in tables 15 and 16. A difference in optimum spawning temperature between small hauls (1-100 eggs) and large hauls (101 eggs and more) is noted (fig. 10). The reason for the difference is that a large proportion of the large hauls were T.^BLE 15. — Summary: Distribution of temperatures at 10 meters of stations having jack mackerel eggs, by size of sample, 1951-64 [In 0.5° C. Intervals] Temperature 1-100 eggs 101-1- eggs Total samples Percent o( total 10.00-10.49 -■ . 2 8 7 12 18 18 33 39 47 41 84 61 104 77 64 28 25 17 10 11 7 2 3 7 10 28 36 78 105 110 111 80 62 27 18 8 2 5 1 2 2 8 7 15 25 28 61 75 125 146 194 172 184 129 91 46 33 19 15 12 9 2 1 10.60-10.99 11.00-11.49 5 11.50-11.99 1 1 12.00-12.49 12.50-12.99 2 13.00-13.49 13.50-13.99 5 4 14.00-14.49 14.50-14.99 10 4 15.00-15.49 13 9 15.60-15.99 12 3 16.00-16.49 13 2 16.50-16.99 17.00-17.49 6 5 17.50-17.99 3 3 18.00-18.49 18.50-18.99 1 4 19.00-19.49 _ 19.50-19.99 y 20.00-20.49 20.50-20.99 \ Total 715 683 1,398 100.2 Sample size and spawning season Mean Median Mode 1951: 1-100 eggs: Early 15.5 15.8 15.5 16.0 16 Late Anniml avAfftgA 15.7 15.5 16 101+ eggs: Early 15.2 15.4 15.0 15.0 15 Late Anniifti average 15.3 16.0 15 1952: 1-100 eggs: Early 15.3 16.0 1.5.0 16.0 Late 17 15.7 16.0 16 101+ eggs: 15.3 15.9 16.0 16.5 14 6 Late .- 14 6 15.6 16.5 14.5 1953: 1-100 eggs: Early - 14. S 15.9 14.5 15.5 15 Late - 16 Annual average 15.3 16.0 15 101+ eggs: 15.1 16.1 15.0 15.5 15.0 Late - - --- 15.0 15.6 15.0 15.0 1954: 1-100 eggs: Early - 15.5 16.6 16.0 16.6 16.0 Late 16.5 Aimual average 16.0 16.0 16.0 101+ eggs: 15.2 16.2 15.0 16.0 14.5 Late 14.0 15.2 15.0 16.0 All years: 1-100 eggs: 15.2 16.0 15.5 16.0 16.0 Late 16.0 15.7 15.6 16.0 101+ eggs: 15.2 15.7 15.0 16.0 15.5 Late - - 15.0 Annual average 15.4 15.5 15.6 taken during the peak of the season (April and May), and have therefore a more restricted temperature distribution. The early, late, and annual distributions of temperature at a depth of 10 meters, by 0.5° C. increments, for all stations occupied in 1951-54 are given in table 17. These differ from the distribu- tion of temperatures at which jack mackerel were taken in two ways: their temperature range is greater, and they show less tendency to cluster about a central value. 264 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 1 1 1 1 4/ \ 1 TO 100 \-IOI a OVER J -^ k 14,25 1625 TEMFtRATURE »C Figure 10. — Percentage of stations having jack mackerel eggs, grouped by 0.5° C. temperature intervals measured at 10-meter depth for stations with hauls of 1-100 eggs, with hauls of 101 eggs and more, and for all stations where eggs were taken. All jack mackerel data were then combined for each year and the mean, median, and modal tem- peratures at 10 meters were computed (table IS). The mean and median temperatures for the 4 years are remarkably constant (15.5° C.) with about 60 percent of the annual spawning occurring within 1° of the median and mean. Less than 40 percent of all the stations occupied during the spawning season (February through July) had temperatures within 1° of 15.5° C. (table 19). The constancy of annual temperature medians and means would indicate a sharp temperature opti- mum for spawning were it not for the within-year temperature shift. The within-year temperature shift suggests that a physiological temperature optimum for jack mackerel is a function of many environmental factors, such as condition of the fish and availability of food, to mention two possibilities. Furthermore, if temperature is the controlling factor, spawning should occur more or less imiformly throughout the area having the optimum temperature. A temperature of 15.5° C. is usually present in the waters off California or Baja California, but spawning occurs only during spring and summer. This may indicate that the length of day has some regulatory effect on spawning. Table 17. — Summary: Distribution of temperatiires at 10 ineters, by season, at stations occupied, 1951-S4 [In 0.5° C. intervals] Temperature 1951 1952 1953 1954 Total for all years Perrent Early Late Total Early Late Total Early Late Total Early Late Total of total 8.00-8.49 - 1 2 6 12 13 18 13 36 50 50 42 71 66 63 53 47 36 21 9 3 6 1 2 1 2 2 3 8 4 10 7 9 16 19 33 30 49 51 52 45 43 49 45 40 30 29 18 21 10 8 8 8 6 9 8 8 2 8 8 15 2 4 10 10 22 20 27 20 61 69 83 72 120 117 115 98 90 85 66 49 33 35 18 22 12 9 10 8 6 9 8 8 2 8 8 15 1 3 5 7 6 5 19 31 35 49 59 70 02 61 53 43 30 24 7 8 4 12 5 5 3 2 1 1 1 11 9 20 14 21 26 31 20 27 31 46 36 50 50 57 39 41 33 30 18 13 17 15 11 6 3 6 4 6 1 n 1 4 16 16 26 19 40 67 66 69 86 101 108 96 103 93 87 63 48 41 34 30 18 22 IS 13 7 3 7 4 6 1 3 8 7 10 18 37 52 69 69 78 72 75 60 66 36 29 21 21 6 8 7 2 1 2 4 1 3 4 6 9 5 12 21 21 37 42 36 54 41 50 38 43 39 33 28 29 16 12 9 8 4 6 4 3 1 1 2 1 2 4 11 11 15 27 42 64 80 80 115 114 HI 114 106 85 67 64 60 39 36 36 18 13 11 12 9 6 4 3 1 1 2 1 2 2 2 2 7 13 19 35 46 64 93 94 94 96 61 43 31 15 18 11 11 5 1 1 1 1 1 1 6 3 6 6 11 14 16 21 30 29 43 39 65 54 66 39 55 36 18 37 31 11 10 5 5 6 3 8 2 1 1 1 3 8 5 13 19 30 49 62 85 123 123 137 135 126 97 87 64 73 47 29 42 31 12 11 6 6 6 3 8 1 1 2 6 22 45 46 88 100 161 206 259 338 406 407 479 464 429 355 328 262 226 173 132 125 80 67 63 37 27 21 Ifi 24 15 9 3 12 10 17 0.02 s.-W-s.gs .04 9.00-9.49 . 11 9.50-9.99 .40 10.00-10.49 . - .83 10.5010.99..- - .85 11.00-11.49 _. ... 1.62 11..50-11.99 1.84 12.00-12 49 2.96 12.50-12.99 . . . 3.79 13.00-13.49 4.76 13.50-13.99 - 6.22 14.00-14.49 7.48 14.50-14.99 7.49 15.00-15.49 8.71 15.,50-15.99 8.35 16.00-16.49 7.89 16.50-16.99 6 53 17.00-17.49 6 03 17.50-17.99.. 4.82 18.00-18.49 4.16 18.50-18.99 3.16 19.00-19.49 2.41 19.50-19.99.. 2.30 20.00-20.49 1.47 20.50-20.99 1.23 21.00-21.49... .98 21.50-21.99 .68 22.00-22.49 .50 22.50-22.99 .39 23.00-23.49 .29 23..';0-23.99 .44 24.00-24.49. .27 24.50-24.99 .16 25.00-25.49. .06 25.50-25.99 .22 26.00-26.49. .18 26.60 and over .31 Total 624 717 1,341 611 692 1,303 744 619 1,363 765 669 1,434 6,441 99.95 DISTRIBtTTION OF EGGS AND LARVAE OF JACK MACKEREL 265 Table 18. — Annual mean, median, and modal temperatures at which spawning occurred, 1951-54 [At 10 meters; In °C.] Year Median and mean temperature Percent spaw-ninE within 1° of median Mode 1951 _,- 15.5 15.5 15.5 15.5 58 65 60 63 16.3 1952 - -- 16.3 1953 16.3 1954 15.3 Table 19. — Summary: Distribulion of temperatures at 10 meters, at all stations occupied, February through July, 1951-54 [In O.S'C. intervals] 1951 1952 1953 1954 Temperature Total Per- cent Total Per- cent Total Per- cent Total Per- cent 8 00-8 49 2 7 7 10 16 20 16 35 50 61 46 77 76 80 61 48 35 36 17 5 10 5 5 3 1 4 1 1 0.27 .95 .95 1.36 2.18 2.72 2.18 4.76 6.80 8.30 6.26 10.49 10.35 10.90 8.30 6.53 4.76 4.90 2.31 .68 1.36 .68 .68 .41 .14 .54 .14 .14 1 4 13 11 20 16 34 37 60 50 60 87 75 67 76 69 53 32 18 12 6 8 3 3 2 4 1 1 0.12 .49 1.60 1.35 2.46 1.97 4.18 4.55 6.15 6.15 7.37 10.70 9.23 8.24 9.35 8.49 6.52 3.94 2.22 1.48 .74 .98 .37 .37 .25 .49 .12 .12 4 10 10 15 25 40 61 63 61 86 79 76 87 77 49 41 43 29 15 17 11 3 4 3 1 1 0.44 1.10 1.10 1.65 2.74 4.39 6.70 6.92 6.70 9.34 8.66 8.34 9.55 8.35 5.38 4.50 4.72 3.18 1.65 1.87 1.21 .33 .44 .33 .11 .11 1 1 3 8 5 12 19 25 38 51 63 92 92 102 105 97 79 62 32 35 25 16 15 7 2 1 1 0.10 8 50-8 99 9 00-9.49 .10 9 50-9 99 -- - .30 10 00-10.49 .81 10.50-10.99... .61 11 00-11.49 1.21 11 50-11.99 1.92 12.00-12. 49 2.53 12 50-12.99 3,84 13.00-13.49. 5.16 13.50-13.99... 6.38 14 00-14.49 9.30 14.50-14.99. 9.30 15 00-15.49 10.31 15.50-15.99 10.62 16 00-16. 49 . . . 9.81 16 50-16.99 7.99 17 00-17. 49 6.27 17 50-17.99 3.24 18.00-18.49 3.54 18.50-18.99 2.53 19.00-19.49 1.62 19.50-19.99 1.52 20. 00-20. 49 .71 20 50-20.99 .20 21.00-21.49 21 50-21 99 .10 22.00-22.49 22 50-22 99 23.00-23.49 .10 23.50-23.99 24.00-24.49 24.50-24.99 25 00-25.49 Turner (1948: p. 351) says, "The reproductive rhythms of the female, as in the male, are influ- enced by numerous factors in the external environ- ment as well as by physiologic factors conditioning the internal enwonment." Since httle is known about the internal factors governing the spawning of fishes, no comprehensive explanation for the variation in the distribution of spawning jack mackerel can be given at this time. It is con- cluded that temperature is important, but not the controlling factor in spatial-temporal distribution of spawning jack mackerel. SURVIVAL OF THE LARVAE The method used by Ahlstrom (1954b) to de- termine the survival of larvae has been retained so that interspecific comparisons might more easily be made for fish occupjing the area surveyed by the Cahfornia Cooperative Oceanic Fisheries In- vestigations. All larvae of a species were with- drawn from a station sample and measured. The measurements are grouped into size classes and adjusted by the standard haul factor (p. 250). These standardized counts are integrated over time and space (p. 251) and adjusted for growth. The products are summed for the year to give an estimate of abundance of the size class. The decline in abundance of successively larger size classes provides an estimate of survival. REGIONAL ESTIMATES OF ABUNDANCE OF LARVAE The regional estimates of abundance by size class for jack mackerel larvae are given in table 20 for 1952, table 21 for 1953', and table 22 for 1954. These tables are summarized in tables 23 and 24, and the annual estimates of abundance and survival are given in table 25. It wall be noted that the curves shown in figure 11 derived from this table are very similar, suggesting that the number of larvae surviving a 30-day period has 1,000 100 10 01 .01 001 10 20 30 40 50 60 DAYS OF DEVELOPMENT Figure U. — Abundance curves of jack mackerel larvae to age 57 days, 1952-54. 266 FISHERY BULLETESr OF THE FISH AND WILDLIFE SERVICE been relatively constant. In 1952, of the 593 trillion eggs estimated to have been spawned, only 780 billion larvae are estimated to have sm-vived at the end of 30 days. In 1953, 736 trillion eggs were estimated to have been spawned and 850 billion larvae are estimated to have siu-vived, while in 1954, 462 trillion eggs were estimated to have been spawned with 830 billion larvae the estimated survivors at the end of the first month of life. This indicates an average survival (for the first month of life) of a little more than 1 larva per 1,000 eggs spawned. Table 20. — Regional distribution of jack mackerel larvae, by month and size class, 19S2 [In billions] Area and month 2.0 mm. 2.5 mm. 3.0 mm. 3.5 mm. 4.0 mm. 4.5 mm. 6.0 mm. 5.76 mm. 6.75 ram. 7.75 mm. 8.76 mm. 9.76 mm. 10.75 mm. 11.76 mm. 12.75 mm. 13.75 mm. 14.75 mm. Northern California Oines 40-77) : April 650 242 42 561 1,246 1.637 1,641 29 107 36 57 38 9 12 48 53 May - July Total 834 1,807 3,078 136 93 47 60 63 Southern California (lines 80-931 : January 311 2,156 134 3,130 1,640 285 307 480 230 3.068 1,287 8,654 2,727 2,145 761 503 2.334 2.916 8,116 6,355 2,691 1,260 140 15 468 1.026 841 778 1,098 24 7 268 2,337 682 1,206 391 30 23 261 1,351 183 926 704 103 1,429 146 766 668 33 409 17 252 553 38 33 71 235 9 12 16 95 7 12 14 8 8 25 17 April July September October . ._ _ Total 8,443 19,375 22, 717 4,242 4,827 3,425 3,111 1,302 348 122 19 14 16 26 17 Northern Baja California (lines 97-107): January 363 7.089 3,374 83 1,962 27 26 2,825 6,478 6,506 610 3,612 1,284 68 15 3,699 3.749 6,944 2,930 7,523 2,366 161 1,361 383 677 629 2,103 283 73 425 960 7 506 1,241 189 264 205 292 168 284 1,238 77 205 15 661 359 264 379 51 18 14 22 325 70 204 277 16 82 28 147 50 9 28 11 83 36 16 103 8 68 6 6 February April May July 37 August September October Total --- 12,898 21,324 27,261 5,409 3,591 2,469 1,641 914 316 158 119 66 11 37 Upper central Baja California Oines UO-120): 5,410 628 3,610 985 258 162 112 5,393 1,466 1,634 138 236 161 634 127 78 300 66 396 49 69 421 19 338 37 130 284 22 185 45 86 99 33 138 20 18 58 32 11 6 11 17 23 11 20 15 20 23 April June — July Total .__ 6,038 4,905 8,642 1,236 860 946 622 308 117 40 31 15 43 Lower central Baja California Qines 123-137): 83 49 448 25 122 48 1,179 441 97 8 66 60 34 45 29 21 45 6 4 22 April -_, May June- Total 83 644 1,766 123 79 50 46 10 22 28,296 48, 056 63, 463 11,146 9,450 6,936 5,479 2,687 781 320 169 102 42 26 17 43 37 DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 267 The slope of the abundance curve is not con- stant. Survival during the first 9 days is verj' low but after this initial period, it is better. The inconsistency in the estimates of larvae older than 40 days is most likely due to sample variation. These estimates are made from small numbers of observations and have therefore a large error associated with them. Although these data are in some respects quite limited, they are comparable with the survival data collected by Sette (1943) for Atlantic mack- erel and Ahlstrom (1954b) for Pacific sardine. Table 21. — Regional distribution of jack mackerel larvae, by month and size class, 1963 [In bUUoDS] Area and month 3.0 mm. 3.5 mm. 4.0 mm. 4.5 mm. 5.0 mm. 5.75 mm. 6.76 mm. 7.75 ipm. 8.75 mm. 9.75 mm 10.76 mm. 11.75 mm. 12.76 mm. 13.75 mm. 14.75 mm. Northern Calltomia (lines 40-77): Tnne 316 175 64 66 51 76 40 40 15 17 10 1201 July August , Total 555 117 lis 40 15 17 10 201 Southern California (lines 80-93): February 2,846 1,067 5,017 2,086 883 32 113 515 620 125 114 11 252 426 410 51 11 126 211 362 315 65 491 27 36 4 61 360 261 19 50 181 68 2 67 20 10 33 22 18 30 I 59 March, April May _ .Time July August September October Total 11,931 1,487 1,150 1,025 609 705 301 67 30 65 48 69 Northern Baja California (lines 97-107): 2,712 996 1,512 1,249 3.621 213 151 274 223 89 188 358 22 161 697 87 261 280 70 64 17 971 96 346 337 8 10 409 87 138 488 18 46 69 23 155 206 20 18 69 17 65 208 6 20 20 41 11 12 41 13 62 6 10 6 I April May . July .. . . October Total . . . 10,454 183 680 864 1,224 2,446 104 1,154 1,620 145 9 140 104 1,785 1,186 491 405 118 68 16 7 Upper central Baja California (lines 110- 120): 58 62 57 292 9 155 6 78 51 11 267 37 29 30 186 4 24 14 18 15 8 64 24 25 69 16 16 40 40 35 13 6 16 15 44 21 14 8 16 23 13 20 22 March April May July September , October Total 5,501 478 398 290 374 246 136 181 48 81 21 61 13 20 22 Lower central Baja California (Unes 123- 137): May , 463 646 16 60 12 18 6 8 20 7 9 July 1,109 76 36 28 7 9 OrftTifi tntAl 29,550 3,312 3,319 3,168 2,191 1,468 852 557 146 152 28 61 61 20 81 268 FISHERY BIJLLETIN OF THE FISH AND WILDLIFE SERVICE Table 22. — Regional distribution of jack mackerel larvae, by month and size class, 1954 [In billions] Month and area 3.0 mm. 3.5 mm. 4.0 mm. 4.5 mm. 5.0 mm. 6.75 mm. 6.75 mm. 7.75 mm. 8.75 mm. 9.75 mm. 10.75 mm. 11.75 mm. 12.75 mm. 13.75 mm. 14.75 mm. Northern California (Hnes 4(}-77): May 269 2,061 578 210 150 222 141 67 108 17 67 116 63 164 97 46 41 July Total . 2,908 210 372 198 108 84 169 164 97 46 41 Southern California (Unes 80-93): April 4,099 6,649 21,445 455 39 133 241 525 3,122 4 77 130 3,371 6 84 59 2,146 15 63 31 2,491 10 9 29 1,034 37 5 15 18 478 16 1 5 222 22 10 4 13 June July September ._- -.. 32,720 3,892 3,583 2,304 2,595 1,114 528 227 36 13 Northern Baja California (lines 97-107): February.. . 175 2,424 6,022 4,666 2,704 741 61 186 707 428 301 14 31 97 353 203 303 17 18 371 456 102 79 555 278 78 194 9 85 88 38 103 6 11 9 46 21 42 7 6 30 16 6 10 21 26 5 April July 10 Total 15, 693 1,666 991 1,008 1,114 331 124 58 31 31 10 Upper central Baja California (lines 110- 120); 820 2,437 1,058 1,312 1,710 27 37 214 76 373 162 39 15 321 235 125 96 22 139 64 39 60 84 4 139 92 5 12 3 4 35 April .Tnnp, July : 7,364 901 792 324 319 17 7 35 Lower central Baja California (lines 123- 137); 198 270 70 74 17 9 19 10 18 17 10 6 8 2 5 13 Februai y April June _ . - . Total. 612 90 10 14 2 6 13 Grand total 69, 297 6,759 5,748 3,834 4,150 1,618 833 484 164 72 31 41 10 Table 23. — Regional distribution of jack mackerel larvae, by size class, 1952-54 [In billions] Year and region 3.0 mm. 3.5 mm. 4.0 mm. 4.5 mm. 5.0 mm. 5.75 mm. 6.75 mm. 7.75 mm. 8.75 mm. 9.75 mm. 10.75 mm. 11.75 mm. 12.76 mm. 13.75 mm. 14.75 mm. 1952: Northern Cahfomia (lines 40-77) Southern California (lines 80-93) Northern Baja California (lines 97-107).. Upper central Baja California (lines 110-120) 3,078 22. 717 27, 261 8,642 1,765 136 4,242 6,409 1,236 123 93 4,827 3,591 860 79 47 3,425 2,469 945 50 60 3,111 1,641 622 45 63 1,302 914 308 10 348 316 117 122 157 40 19 119 31 14 66 22 16 11 15 25 17 43 37 Lower central Baja California (lines 123-137) Total _ 63, 463 11,146 9,450 6,936 5,479 2,587 781 319 169 102 42 26 17 43 37 1953: Northern California (lines 40-77) Southern California (lines 80-93) Northern Baja California nines 97-107).. Upper central Baja Cahfomia (lines 110-120) 655 11,931 10, 464 5,501 1,109 117 1.487 1,154 478 76 115 1, 150 1,620 398 36 40 1,026 1,785 290 28 15 609 1,186 374 7 17 705 491 246 9 10 301 405 136 201 57 118 181 30 68 48 55 16 81 21 61 48 13 20 59 22 Lower central Baja Cahfomia flines 123-137) Total 29, 550 3,312 3,319 3,168 2,191 1,468 862 667 146 162 28 61 61 20 81 1964: Northern California aines40-7?) Southern California (lines 80-93) Northern Baja California (lines 97-107). Upper central Baja California (Imes 110-120) _ 2,908 32, 720 15, 693 7,364 612 210 3,892 1,666 901 90 372 3,583 991 792 10 198 2,304 1,008 324 108 2,695 1,114 319 14 84 1,114 331 17 2 169 528 124 7 5 164 227 58 35 97 36 31 46 13 13 31 41 10 Lower central Baja Cahfomia nines 123-137) Total 59, 297 6, 759 5,748 3,834 4,150 1,648 833 484 164 72 31 41 10 DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 269 Table 24. — Percentage of each size class of jack mackerel larvae, occurring in each region, 19SS-S4 Year and region 3.0 mm. 3.5 mm. 4.0 mm. 4.5 mm. 5.0 mm. 5.75 mm. 6.75 mm. 7.75 mm. 8.75 mm. 9.75 mm. 10.75 mm. 11.75 mm. 12.76 mm. 13.75 mm. 14.75 mm. 1952: 4.9 35.8 43.1 13.6 2.8 1.2 38.0 48.6 11.1 1.1 1.0 50.9 37.8 9.1 0.8 0.7 49.4 35.6 13.6 0.7 1.1 66.7 30.0 11.3 0.8 2.0 50.2 35.2 11.8 0.4 44.8 40.5 14.7 38.1 49.4 12.5 10.9 71.0 18.1 13.7 05.1 21.2 39.0 26.7 34.3 100.0 100.0 100.0 Northern Baja California Olnes 97-107)... Upper central Baja California (lines 110- 120) 100.0 Lower central Baja California (lines 123- 137) Total 100.2 100.0 99.6 100.0 99.9 99.6 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1953: Northern California (lines 40-77) 1.8 40.4 35.4 18.6 3.8 3.5 44,8 34.9 14.4 2.3 3.5 34.6 48. 8 11.9 1.1 1.3 32.4 56.3 9.2 0.9 0.7 27.8 54.1 17.0 0.3 1.1 48.0 33.4 16.8 0.6 1.2 35.4 47.5 15.9 36.1 10.3 21.1 32.5 20.4 46.5 32.9 36.9 10.3 53,8 26.3 73.7 100.0 79.4 20.6 100.0 72.6 Northern Baja California (lines 97-107)... Upper central Baja California (lines 110- 120) 27.4 Lower central Baja California (lines 123- 137) Total 100.0 99.9 99.9 100.1 99.9 99.9 100.0 100.0 99.8 100.0 100,0 100.0 100,0 100.0 100.0 1954: Northern California (lines 40-77) 4.9 65.2 26.4 12.4 1.0 3.1 57.4 24.6 13.3 1.3 6.6 62.4 17.3 13.8 0.2 5.2 60.1 26.2 8.4 2.6 62.6 26.9 7.7 0.3 5.4 72.1 21.4 1.0 0.1 20.3 63.4 14.9 0.7 0.6 34.1 47.0 11.7 7.2 69.2 22.0 18.8 64.5 17.7 17.8 100.0 100.0 Northern Baja California Olnes 97-107)... 120) 100.0 Lower central Baja California (lines 123- 137) Total 99.9 99.7 100.2 99.9 100.1 100.0 99.9 100.0 100.0 100.0 100.0 lOO.O 100.0 Table 25. — Annval estimates of abundance and survival of jack mackerel eggs and larvae, 1952-54 [ In billions] Category Duration Average (days) age (days) 3.6 1.8 1.0 4.1 1.0 6.1 1.0 6.1 4.8 9.0 4.2 13.5 il 17.4 21.0 5.9 25.5 6.0 31.0 4.3 35.7 3.8 39.4 3.6 43,3 3.1 46.7 2.9 49.7 2.6 62.4 2.4 64.9 2.3 67.2 Size range (mm.) Abim- dance Survival per 100.000 eggs 1953 Abun- dance Survival per 100,000 eggs Abtm- dance Survival per 100.000 eggs Eges Larvae: 2.0 mm.. 2.5 mm.. 3.0 mm.. 3.5 mm.. 4.0 mm.. 4.5 mm.. 5.0 mm._ 6.75 mm. 6.76 mm. 7.75 mm.. 8.75 ram. 9.76 mm. 10.76 mm 11.75 mm 12.75 mm 13.76 mm 14.76 mm 1. 90- 2. 25 2. 26- 2. 75 2. 76- 3. 25 3. 26- 3. 75 3. 76- 4. 25 4. 26- 4. 75 4. 76- 5. 25 5. 26- 6. 25 6. 26- 7. 25 7. 26- 8, 25 8. 26- 9. 25 9. 26-10. 25 10. 26-11. 25 11.26-12.25 12. 26-13. 25 13. 26-14. 25 14.26-15.25 593. 100 28.300 48, 100 63.600 11, 100 9.400 6,900 6,500 2.600 780 320 170 100 42 25 17 43 37 736, 100 462. 300 4.711 8.109 10, 706 1,871 1,648 1,163 927 438 131 63 28 16 7 4 2 7 6 29,600 3.300 3,300 3.200 2.200 1.600 860 660 150 150 28 60 60 20 81 4.007 448 448 434 298 203 115 76 20 20 3 8 8 2 11 59.300 6.800 6.700 3.800 4.100 1.600 830 480 160 71 30 12.827 1,470 1,232 821 886 324 179 97 34 15 GROWTH RATE OF LARVAE Tlie growth rate of jack mackerel larvae was obtained by direct observation of material taken from station 70 on line 97 (97.70) during March of 1957 (Farris, 1959). Eggs were taken from the sea with a plankton net, sorted according to develop- mental stage, placed in jars of fresh sea water, and observed daily until they hatched. After hatching the larvae were measured daily until they died. No attempt was made to feed the fish. The measurements were averaged daily and plotted as log average length against days (fig. 12). Such a plot suggested that growth during the first 3 days was more rapid than for the next 4 daj's. The yolk sac was absorbed and the eyes became pigmented on the sixth day of larval life. The growth rates for the first 3 daj^s (section A) and for the following 4 days (section B) were then compared by regression analysis (table 26), where X=days past hatching, F= length in miUimeters, log Y=a+bX, Si)=standard deviation of slope, and .Sy/i= standard deviation of sample points 270 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE .2- DINE ANCHOVY JACK MACKEREL 2 DAYS 3 PAST HATCHING Figure 12. — Growth curves of jack mackerel, Pacific sardine, and northern anchovy. (Open circles indicate complete absorption of the yolk.) from the line. The comparison of the two slopes indicated that the initial relative growth rate is about five times that of the later relative growth rate. Table 26. — Regression statistics for the relative growth rate of jack mackerel larvae Section A: days 1 through 3 Section B: days 4 through 8 --- 0.328 .486 0.0C7 .013 0. 0091 .0017 Syh 0.01)10 .0184 1.3 3.1 0.614 .687 The relative growth rates of the other two species (Pacific sardine, Sardinops caerulea, and northern anchovy, Engraulis mordax) illustrated in figure 12 are similar to the growth rate shown for jack mackerel. In addition, the relative growth rate of the sardine as determined by direct observation (6 = 0.018) is in good agreement with the relative growth rate (6 = 0.019) determined by Ahlstrom (1954b) using an indirect method (Farris, 1959: p. 33). Ahlstrom, working with preserved material, was able to follow length- frequency modes through time. Although growth has been described by two log curves instead of one, either would have served as well for estimating survival. The abundance of a size class is given by the following equation : C= i=l c'tWftt where C= estimate of total abundance of larvae of size class C(=standard number of larvae belonging to the size class at the ith station ■W(=area factor proportional to area of the t'th station <(=time factor equal to one-half the time from preceding ocupancy plus one-half the time to succeeding occupancy d<= duration of size category in days, i.e., the number of days used by larvae to grow from the lower-size boundary to the upper-size boundary of the size class 71 = number of stations considered. The effects of 1- and 2-phase growth on mortal- ity estimates were compared by recomputing estimates of abundance for sardines given by Ahlstrom (1954b: p. 133). In recomputing abun- dance of yolk-sac larvae, the fonnula used was — Duration of size category (days) = log r'-log V 0.081 ' where 0.081 is the log increase in length per day of the larvae, I' is the lower boundary of the size-class interval, and I" is the upper boundary of the size-class interval. The duration of size category for the remaining size categories of larvae is given by — logZ"-logr 0.018 The results of the recomputation are given in table 27 under the heading "double phase," and may be compared with Ahlstrom's figures. The slight differences in average age for given size may be due to shrinkage of the larvae upon preservation in formalin. Since relative growth rates derived from labo- ratory observations on the first 5 days in the life of a sardine could be extrapolated and reconciled DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 271 with field observations, it was assumed that the extrapolation could be made for jack mackerel, too. An analysis of successive length-frequency diagrams has not been used because the time interval between successive cruises is too great for the estimation of a rapid growth rate. Sec- ondly, any changes in the survival of larvae between cruises would influence the length fre- quencies; hence the changes would alter the estimation of the growth parameter and thereby alter the estimates of annual survival. A growth rate derived from direct observation has the advantage of being independent of variation in survival. T.\BLE 27. — Estimates of abundance of young sardine, using single phase and double phase growth curoes, 1950 (Single phase after Ahlstrom (1954b)) Category Single phase: Egg - - Yolk-sac larvae. Larvae: 4.75 mm 5.75 mm 6.75 mm 7.75 mm 8.75 mm 9.75 mm 10.75 mm Double phase: Egg Yolk-sac larvae: 3.25 mm 4.75 mm Larvae: 5.75 mm 6.75 mm 7.75 mm 8.75 mm 9.75 mm 10.75 mm Size range (mm.) 2. 26- 4. 25 4.2&- 5.25 5. 26- 6. 25 6.26- 7.25 7. 26- 8. 25 8. 26- 9. 25 9. 26-10. 25 10.26-11.25 2. 26- 4. 25 4. 26- 5. 25 5.26- 6.25 6.26- 7.25 7. 26- 8. 25 8. 26- 9. 25 9. 26-10. 25 10.26-11.25 Duration (days) 3.0 3.5 4.8 3.9 3.3 2.9 2.6 2.3 2.1 3.0 1.9 6.1 4.2 3.5 3.1 2.7 2.4 2.2 Average age (days) 1.5 4.8 13.2 16.8 20.0 22.7 25.1 27.3 1.5 3.9 7.5 12.1 15.9 19.2 22.1 24.7 27.0 Estimated abundance 285, 676 11,850 10. 778 5,590 6.197 5,931 4,834 3,738 2,880 285, 676 21,829 10,144 5,191 5.843 5. 648 4.655 3,582 2,749 The regression statistics of section A (table 26) were used to estimate the duration of the various size categories of jack mackerel through the 3.0-nim. size class and the regression statistics of section B for all size classes thereafter. The duration of the size category (in days) through the 3,0-mm. size category is given by — \ogl"-\ogl' 0,067 where /" is the upper boundary' of the size class, and /' is the lower l)oundarv. The duration of the size category (in days) for all remaining size categories is given by — \ogl"-\ogl' 0.013 The average age for any size category is given by summing the duration of size class for shorter size classes and adding one-half the duration of the size category under consideration. For ex- ample, the average age of the 4.5-mm. size cate- gory is obtained hy summing the durations for categories, eggs through 4.0 (15.6) and adding one half of 3.7. The average age is 17.4 days. The coincidence of the absorption of the yolk- sac and the inflection of the survival curve is tentatively interpreted as follows : Basic mortality rates of pelagic fish eggs and yolk-sac larvae are high owing to factors intrinsic to the eggs and inherent in the species; most of those which are unfit have died before absorbing their yolk or die shortly thereafter; the survivors beyond the critical stage now survive at a higher rate because they have successfully negotiated the change in nutrition (i.e., from yolk to copepod eggs and nauplii). A more comprehensive expo- sition of this hypothesis is given by Farris (1960). SOURCES OF ERROR AND BIAS IN SAMPLING LARVAE In the section on sources of error in egg sampling, some of the more obvious sources of error and bias were examined. These same sources of error were examined in sampling pro- cedures for larvae. In addition, the avoidance of the net by the larvae might be added as a source of error. RETENTION OF LARVAE BY THE NETS Incomplete retention by the net of some small size classes of larvae becomes a serious problem. Estimates of abundance were made for the 2.0- and 2.5-ram. size classes of larvae in 1952 (table 25). These estimates are lower than the estimated abundance of the 3.0-mm. size-class larvae, indicat- ing that the smaller size classes were undersampled. Larvae less than 3.0 mm. in length, meeting the mesh head on, are able to pass through and do not appear in our collections in proportion to their true abundance. Estimates for these size classes were not made in 1953 and 1954. Figure 4, from Ahlstrom and Ball (1954) suggests that the head depth may be 10 to 20 percent greater tlinn the body deptii at the pectoral. If this is so, the maximum depth of the 3.0-mm. larvae is greater than the maxinuini mesh opening of tlie sampling net (0.55 mm. after shrinking 272 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE and 0.7 mm. before shrinking). As the larvae grow longer, their bodies grow increasingly deep and they are therefore retained within the sampling net. Evidence that larvae 3.0 mm. and longer would be retained once they are in the plankton net is given in table 28. Table 28. — Standard length and body depth at pectoral of jack mackerel larvae [Measurements in millimeters] Average standard length Average depth at pectoral ? S' 0.461 3 2 0.62 3 8 0.S2 4.2 - 1.03 4.8 -- 1.23 1 Data from Ahlstom and Ball (1964, p. 216-216). SAMPLING OF THE VERTICAL RANGE The vertical distribution of larvae was examined for both the total range and quantitative diurnal differences in abundance at different depths. The data (table 29, after Ahlstrom, 1959: p. 119) indicate that almost all the larvae were found above 100 meters and that no larvae were found below 140 meters. Since net hauls are routinely made from a 140-meters depth, the net passes through the entire water column which can be expected to contain jack mackerel larvae. The possibility of quantitative day-night differ- ences (because of diurnal vertical migration) was analyzed by examining the vertical distribution of samples taken during the day and at night. These ciaiparable day -night series were completed for several stations. The 2-, 8-, and 19-meter strata contain proportionally more larvae when sampled in the day than at night. The data indicate that there may be some slight tendency for the larvae to migrate to the surface during the day ; however, it is more likely that there was a change in the water being sampled at station 94.80 (5403) between the day and night series. Both sardine and anchovy larvae are known to be distributed somewhat deeper in the daytime (negative photo- taxis) than at night (Ahlstrom, 1959), as are many species of plankton organisms, so that a positive phototaxis would need to be better documented than is possible from our data. No definite diurnal variation in vertical distribution is evi- denced. Table 29. — Standard number of jack mackerel larvae taken in day and night hauls, by depths [Stations in parentheses; after Ahlstrom, 1959] Aver- age depth (m.) Day hauls Night hauls Depth range (m.) Cruise 4106 (94.37) Cruise 4106 (94.47) Cruise 5206 (90.28) Cruise 5306 (93.50) Cruise 5403 (94.80) Total larvae taken Percent of total Cruise 4106 (94.37) Cruise 4106 (94.47) Cruise 5206 (90.28) Cruise 5305 (93.50) Cruise 6403 (94.80) Total larvae taken Percent of total 2-3. 2 8 19 28 41 66 72 106 138 215 285 •1 •0 2 •0 2 •0 92 86 24 2 2 2 2 •0 92 88 24 2 6 2 2 42.8 40.9 11.2 .9 2.3 .9 .9 •0 4 •3 12 13 •0 1 4 •0 26 5 7 6 8 •0 38 17 9 10 7 8 42.7 6-10 19.1 16-21 24-31 10.1 11.2 33-45 7.9 46-60 9.0 63-79 92-112 127-150 200-239 276-291 . Total 1 2 2 210 216 99.9 7 26 4 52 89 100.0 •Region of the thermocline. VARIABLE DISTRIBUTION OF LARVAE The difficulties in assessing the variability of these estimates have previously been discussed in the section on sources of error in sampling eggs. Since the proper statistical model is not known at this time, no good estimate of error can be given. It was shown in the case of eggs that for a large number of interpolations in space and time the error tended toward zero. The error of the esti- mates of abundance for the larvae is probably as great as the error of the estimates for egg abund- ance, and very likely it is even greater. This may be particularly true for the estimates of abundance of the older size categories which are based on fewer than 10 observations. DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 273 SAMPLING OF THE HORIZONTAL RANGE It was shown that incomplete samphng of the entire range of jack mackerel spawning introduced an error of at least 21 percent of the estimated total egg population. This error appears to be reduced for the estimates of larvae. A very small propor- tion of the 30-day and older larvae is taken sea- ward of stations 90 (tables 30, 31, and 32). These tables were constructed using the standard numbers of jack mackerel larvae (6.75 mm size-class) for all stations sampled during the year. The data were grouped by 2-month intervals about selected sampling lines and further grouped to reflect the offshore-inshore distribution of the larvae as was done with the eggs (see p. 253). The data seem to indicate that almost all the 30-day-old larvae are to be found in an area bounded by line 80 on the north, line 110 on the south, stations 90 on the west, and the coast on the east. Some larvae are found both to the north and south of the main area. This area is far more restricted than that exhibited by the distribution of the eggs (see above) . Table 30. — Relative north-south, inshore-offshore distribu- tion of 6.75-mm. (30-day old) jack mackerel larvae, by 2-month intervals, 1962 [Standard haul totals] Table 31. — Relative north-south, inshore-offshore distribu- tion of 6.76-mm. (30-day old) jack mackerel larvae, by 2-month intervals, 1353 [Standard haul totals] Stations Total Lines 100- seaward 90-70 60-shore Percent February-March: 60 (') (') (') (') (') (') (') (') 2.7 (') (') 8.3 (') (■) 11.0 (') (') 100.0 70 80 90 100.. . no 120 Total 2.7 24.6 8.3 75.5 11.0 100.0 100.0 Percent... April-May: 60 (') C) (') (') 9.4 19.2 9.0 8.2 10.2 9.4 27.4 19.2 16.8 48.9 34.3 70 80 90 100 110 120 Total 37.6 67.2 18.4 32.8 56.0 100.0 100.0 Percent June-July: 60 (') (1) 42.4 47.1 3.6 3.1 63.3 54.9 6.9 95.7 102.0 9.5 3.1 45.4 48.4 4.5 1.5 70 80 90 100 110 120 Total 96.2 45.7 114.1 64.3 210.3 100.0 99.8 Percent Stations Total Lines 100- seaward 90-70 60-shore Percent February-March: 60 (') (') 4.9 (') (') CO (1) (1) (') 8.4 (') (') 3.8 8.1 5.1 (') C) 4.9 3.8 16.5 5.1 (') 70 (') 16 2 80 90 12.6 100 64 6 110 16.8 120 Total 4.9 16.2 8.4 27.8 17.0 66.0 30.3 100.0 100.0 April-May: 60 (') 0) (') (■) 6.9 18.4 62.6 24.3 52.6 70 80 90 100 31.6 110 68.6 120 Total ... 5.9 7.7 71.0 92.3 76.9 100.0 100.1 June-July: 60 (') (') (') (') 12.1 35.2 32.6 11.8 3.0 36.8 18.6 58.9 10.3 15.1 72.0 51.2 70.7 10.3 70... 80. 6.9 90 32.9 100 23.3 110 32.2 120 4.7 Total 91.7 41.8 127.6 58.2 219.3 100.0 100.0 Percent * Region not occupied. 1 Region not occupied. There are several possible explanations for the relatively restricted distribution of jack mackerel larvae. Four interactions of ocean current and fish survival are discussed as possible explanations of the observed distribution. Uniform current flow, diflerential survival. — In this model, it is assumed that the southerly flowing California Current has a uniform velocity, i.e., it has no eddies or countercurrents, and the larvae have a differential survival. The eggs spawiied in some part of the Current will survive at a much higher rate over those deposited in other parts. Under the conditions of this model, the older larvae (survivors) would occur in regions south of the one in which they were spawned, never north of the spawning region. If the current, though of uniform velocity, is very slow, the southward displacement will be slight and therefore very hard to measure. Uniform, current flotr, uniform survival. — If the California Current flowing south contained larvae which survived equally well in any part of the current, one would expect the older larvae to 274 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Table 32. — Relative north-south, inshore-offshore distribu- tion of 6.75-mm. {30-day old) jack mackerel larvae, by 2-month intervals, 1954 [Standard haul totals] Stations Total Lines 100- seaward 90-70 60-shore Percent Febraary-March: 60 (■) (') (') (•) (0 (') (') (') 6.3 (') (') (') C) 5.3 (') 70 (') 80 90 100 100.0 110 120 Total --- 5.3 100.0 5.3 100.0 100.0 Percent April-May: 60 (') (') (') (') (') 4.6 3.0 3.2 5.9 6.2 8.7 15.8 6.6 4.6 9.2 11.9 21.7 6.6 70 80 8.4 90 17.1 lOO 22.1 110 40.1 120 12.3 Total 16.6 30.8 37.3 69.2 63.9 100.0 100.0 Percent June-July: 60 - o 2.8 (') (') (') (') 3.0 10.5 7.7 64.8 2.9 49.8 66.0 13.8 3.1 3.0 10.6 60.3 110.8 16.7 3.1 1.6 70 -- 6.1 80 29.6 90 64.2 8.2 110 1.6 120 Total --- 2.8 1.4 78.9 38.6 122.7 60.1 204.4 100.0 100.0 ' Eeglon not occupied. occur always to the south of the spawning area. If one considers the special conditions of low current velocity and high larval mortality, a sampling problem becomes apparent. Consider: A small subarea X on the periphery of the spawn- ing area contains a thousand eggs. At the end of a month, one larva has survived. Subarea Y in the center of the spawning area contains a million eggs. Under conditions of uniform sur- vival a thousand larvae would survive at the end of a month. The sampler has a much better chance of obtaining larvae from subarea Y than from subarea X. The failure in obtaining month- old larvae from peripheral areas would lead to a conclusion of better survival in the center of the spawning area, though in fact, survival was uniform over the entire area. Differential current flow, uniform surmval. — Under conditions of this model, the California Current would have an average transport to the south, but some parts of the water would move faster than others, and eddy currents would be present. The larvae, although surviving at a uniform rate, would not appear to do so because of the postulated concentrating mechanisms. A high proportion of the larvae would be found in the eddies. Diflerential current flow, diflerential suruival. — Using this model, it is virtually impossible to predict the distribution of the older larvae, as that distribution depends on the special conditions of the current and survival. Smce the average flow, as determined from dynamic topography of the California Current, is slow (about 0.2 knot) and since it contains water flowing both slower and faster accompanied by eddies, no definite conclusion may be reached concerning differential survival, although survival appears to be better in some regions (table 33). Table 33. — Annual regional summary of distribution of jack mackerel eggs and month-old larvae, 1952-54 [Survival at the end of 1 month Is given for each region] Year and region Eggs (bilUons) Percent of total Larvae ' (bilhons) Percent of total Survival per 100,000 eggs 1962: 1 16,686 246, 940 257,418 68,902 4,137 2.64 41.64 43.40 11.63 0.69 348 316 115 44.8 40.6 14.7 2 1.4 3 - - 1.2 4 1.7 6 6 Total 693,082 100. 00 779 100.0 1953: 1 9,653 185,095 353, 517 176, 573 11, 294 1.31 25.14 48.02 23.99 1.53 10 301 405 136 1.2 35.4 47.5 16.9 1.0 2 1.6 3.. 4 1.1 .8 6 Total 736, 132 99.99 852 100.0 1954: 1 ... 31,116 177, 141 185, 765 49,228 19,088 6.73 38.31 40.18 10.66 4.13 169 527 124 6 5 20.3 63.4 14.9 .7 .6 6.4 2 3.0 3 .7 4 .1 5 __ _. .3 Total 462, 338 100.00 831 99.9 I 6.76-mm. size class; estimated to be 1 month old. AVOIDANCE OF NET When Ahlstrom (1954b) computed the mortality of sardines, he found that a correction was neces- sary to account for the larvae which dodged the sampling net. He demonstrated dodging in a relative way by examining the ratio of average number of sardine larvae per night haul (when presumably the larvae cannot see the net) to the average number of larvae per day haul. When he computed the night/day ratio for each size class he found that the ratio increased with size. The DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 275 6 8 10 _ SIZE OF LARVAE IN MM. Figure 13. — Ratio of the average number of jack mackerel larvae per night haul to the average number taken per day (N/D), by size class, 1952-54. implication was that as the larvae grew larger and stronger they were better able to dodge the net. This increase in ratio with size is true not only of the sardine, but also of the anchovy, the hake, the Pacific mackerel, et cetera (Ahlstrom, personal communication; Bridger, 1956). It is, indeed, surprising to find that jack mack- erel do not behave in this manner. All jack mackerel data for 3 years have been combined in figure 13. The N/D ratio for the 7.75-mm. size class has been deleted because one of the samples, which was unusually large, prevented comparison of this size class with the other size classes. A least-squares line has been fitted to the data Y=a + bX, where X=size in millimeters and Y=N/D ratio. The regression statistics are a=1.06; 6 = 0.006; ^^.= 0.196; X=6.0; T=1.10- S6=0.057. The slope of the regression is not significantly different from zero. The interpretation placed on these data is that jack mackerel do not dodge the net despite their apparent ability to do so. The eyes are pigmented and presumably func- tional about the time the yolk is absorbed (at approximately 3.5 mm.), and larvae at yolk-sac absorption are capable of movement. Ahlstrom (personal communication) believes that jack mackerel larvae can swim as well as the sardine larvae. SUMMARY The distribution and abundance of jack mack- erel eggs is described for 4 years, 1951 through 1954. The early survival of jack mackerel larvae is described for 1952, 1953, and 1954. The data were obtained from monthly cruises during which an average of 150 stations was occupied. Jack mackerel spawned in an area bounded by the 26lh parallel on the south, the 45th parallel on the north, the west coast of North America on the east, and the 150th meridian on the west. Most of the spawning occurred in a more restricted area, the boundaries of which varied from year to year. Eggs were mainly confined to the upper 40 meters of water. Spawning usually began in February, reached a peak in May, and ceased by October. The peak of spawning in 1951, which occurred in March, is considered abnormally early. The temperature coefficient for the rate of egg development was derived by a regression of log hours of development on temperature in degrees centigrade. Jack mackerel eggs kept under con- trolled temperature conditions (in an incubator) hatched at the time predicted by the derived temperature coefficient. The estimates of egg abundance for 1951, 1952, 1953, and 1954 are 8.7X10'*, 5.9X10'*, 7.4X10'*, and 4.6X10'*, respectively. The survival at the end of a 30-day period for 1952, 1953, and 1954 was 131, 112, and 179 larvae per 100,000 eggs spawned, respectively. The variation was considered insignificant. An in- crease in surAaval rate during the second v>-eek of larval life was noted. The relative growth rate of jack mackerel larvae was approximated from observations on laboratory populations. The relative growth dur- ing the first 3 days is more rapid by a factor of 5 than the relative growth of the succeeding 4 days. The onset of the slower growth is correlated in time with yolk-sac absorption. An area which is bounded by line 80 on the north, line 110 on the south, stations 90 on the west, and the coast of California and Baja Cali- fornia on the east has been shown to contain almost all the month -old larvae. LITERATURE CITED Ahlstrom, Elbert H. 1943. Studies on the Pacific pilchard or sardine (Sardtnops caerulea). 4. Influence of tempera- ture on the rate of development of pilchard eggs in nature. U.S. Fish and Wildlife Service, Special Scientific Report No. 23, 26 p. 1948. A record of pilchard eggs and larvae collected during surveys made in 1939 to 1941. U.S. Fish 276 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE and Wildlife Service, Special Scientific Report No. 54, 76 p. 1953. Pilchard eggs and larvae and other fish larvae, Pacific coast— 1951. U.S. Fish and Wildlife Serv- ice, Special Scientific Report: Fisheries No. 102, 55 p. 1954a. Pacific sardine (pilchard) eggs and larvae and other fish larvae, Pacific coast — 1952. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 123, 76 p. 1954b. Distribution and abundance of egg and larval populations of the Pacific sardine. U.S. Fish and Wildlife Service, Fishery Bulletin No. 93, vol. 56, p. 83-140. 1956. Eggs and larvae of anchovy, jack mackerel, and Pacific mackerel. Progress report of the Cali- fornia Cooperative Oceanic Fisheries Investigations, 1 April 1955 to 30 June 1956, p. 33-42. Sacramento, State Printer. 1959. Vertical distribution of pelagic fish eggs and larvae off California and Baja California. U.S. Fish and Wildlife Service, Fishery Bulletin No. 161, vol. 60, p. 107-146. Ahlsteom, Elbert H., and Obville P. Ball. 1954. Description of eggs and larvae of the jack mackerel (Trachurus symmetricus) and distribution and abundance of larvae in 1950 and 1951. U.S. Fish and Wildlife Service, Fishery Bulletin No. 97, vol. 56, p. 209-245. Ahlstrom, Elbert H., and David Kramer. 1955. Pacific sardine (pilchard) eggs and larvae and other fish larvae, Pacific coast — 1953. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 155, 74 p. 1956. Sardine eggs and larvae and other fish larvae, Pacific coast— 1954. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 186, 79 p. 1957. Sardine eggs and larvae and other fish larvae, Pacific coast— 1955. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 225, 90 p. Barnhart, Percy Spencer. 1936. Marine fishes of southern California. Uni- versity of California Press, Berkeley, California, 209 p. Bridger, J. P. 1956. On day and night variation in catches of fish larvae. Journal du Conseil, International pour I'Exploration de la Mer, vol. 22, no. 1, p. 42-57. California Marine Research Committee. 1950. California Cooperative Sardine Research Pro- gram, Progress Report — 1950. Sacramento, State Printer, 54 p. 1952. California Cooperative Sardine Researc h Pro- gram, Progress Report — 1 January 1951 to 30 June 1952. Sacramento, State Printer, 51 p. 1953. Progress Report, California Cooperative Oceanic Fisheries Investigations, 1 July 1952- 30 June 1953. Sacramento, State Printer, 44 p. 1955. Progress Report, California Cooperative Oce- anic Fisheries Investigations, 1 July 1953-31 March 1955. Sacramento, State Printer, 52 p. 1956. Progress Report, California Cooperative Oce- anic Fisheries Investigations, 1 April 1955-30 June 1956. Sacramento, State Printer, 44 p. Clothier, Charles R., and Edward C. Greenhood. 1956. Jack mackerel and sardine yield per area from California waters, 1946-47 through 1954-55. Cali- fornia Department of Fish and Game, Fish Bul- letin No. 102, p. 7-16. Farris, David A. 1958. Jack mackerel eggs — Pacific coast, 1951-1954. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 263, 44 p. 1959. A change in the early growth rate of four lar- val marine fishes. Limnology and Oceanography, vol. 4, no. 1, p. 29-36. 1960. The effect of three different types of growth curves on estimates of larval fish survival. Journal du Conseil International pour I'Exploration de la Mer, vol. 25, no. 3, p. 294-306. Fitch, John E. 1956. Jack mackerel. Progress Report of the Cali- fornia Cooperative Oceanic Fisheries Investiga- tions — 1 April 1955 to 30 June 1956. Sacramento, State Printer, p. 27-28. Fowler, Henry W. 1944. The fishes: in the results of the fifth George Vanderbilt expedition (1941). Academy of Nat- ural Science of Philadelphia, Monograph No. 6, p. 57-583. Gamulin, T., and P. J. Hurb. 1955. Contribution a la Connaissance de I'Ecologie de la Ponte de la Sardine (Sardina pilchardus Walb.) Dans I'Adriatique. Acta Adriatica, Institut za Oceanographiju i Ribarstvo Spht FNR Jugoslavija, vol. 7, no. 8, 23 p. Roedel, Phil M. 1949. Jack mackerel. In The commercial fish catch of California for the year 1947 with a historical review 1916-1947. California Department of Fish and Game, Fish Bulletin No. 74, 267 p. SCOPIELD, W. L. 1951. Purse seines and other roundhaul nets in Cali- fornia. California Department of Fish and Game, Fish Bulletin No. 81, 83 p. ScRipps Institution op Oceanography and U.S. Fish and Wildlife Service. 1952. Station positions of the California Cooperative Sardine Research Program. Scripps Institution of Oceanography, La JoUa, California, SIO Reference 52-04, 5 p. Sette, Oscar Elton. 1943. Biology of the Atlantic mackerel (Scomber scombrus) of North America. Part I: Early life history, including the growth, drift, and mortality of the egg and larval populations. U.S. Fish and Wildlife Service, Fishery Bulletin No. 38, vol. 50, p. 149-237. DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 277 Settb, Oscar E., and Elbert H. Ahlstrom. 1948. Estimation of abundance of the eggs of the Pacific pilchard (Sardinops caervlea) off southern California during 1940-41. Journal Marine Re- search, vol. 7, no. 3, p. 511-542. South Pacific Fishery Investigations. 1952. Zooplankton volumes off the Pacific coast, 1951. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 73, 37 p. 1953. Zooplankton volumes off the Pacific coast, 1952. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 100, 41 p. 1954. Zooplankton volumes off the Pacific coast, 1953. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 132, 38 p. 1955. Zooplankton volumes off the Pacific coast, 1954. U.S. Fish and Wildlife Service, Special Scientific Report: Fisheries No. 161, 35 p. Turner, C. Donnell. 1948. General endocrinology. W. B. Saunders Co., Philadelphia and London, 604 p. Walford, Lionel A. 1948. The case for studying normal patterns in fishery biology. Journal Marine Research, vol. 7, no. 3, p. 505-510. APPENDIX A. FISH EGG INCUBATOR Description When estimating the total abundance of pelagic fish eggs it is necessary to know the rate of develop- ment. Since this parameter varies with tempera- ture it has been customary in the past to compute a regression of log hours of development against temperature in tenths of a degree centigrade. This indirect method, while very accurate, is extremely laborious and time consuming. A more direct method was desired, and \vith this need in mind, an incubator was designed for use at sea. Since most of the biological material is taken well offshore, and since hatching times are rela- tivelj' short, it is necessary to work at sea. Some of the many technical problems that are peculiar to sea work and their solutions are described here. The pitch and roll of ships cause delicate equip- ment to be damaged quite easily, and the need for sturdy construction is readily apparent. Instru- ments such as this one should be portable, since they have a limited use and cannot be left aboard research vessels indefinitely. A compromise be- tween sturdy construction and portability of an egg incubator was effected by resorting to a double- box construction, using marine pljnv'ood. The temperature-sensing and temperature-control de- vices were of a mechanical nature. A heavy duty, stainless steel sensing element enclosing a mercury colmnn was employed. The mercury column ac- tivated a mechanical linkage in the thermostat, which in turn opened and closed an electrical switch. The circulating pump and cooling device were remote from the incubator itself, but con- nected by garden hoses (appendix fig. 1). The RUBBER GASKET THERMOSTAT INFLOW TEMPERATURE ,.««tl» ^ SENSOR WATER BATH JAR RACK OUTFLOW CLAMP LOCK PUMP a COOLING UNIT Appendix Figure 1. — Fish egg incubator. separation of units contributed to the portability of the instrument. Corrosion from the salt air which attacks most metals was controlled by using a synthetic resin paint on all exposed parts of the incubator. Other structures were concealed in corrosion resistant housings. Temperature control (± 0.2° C. of selected tem- perature) was obtained by using the main water mass in the water bath as a heat reservoir. As the water mass warmed, the change in temperature was recorded by the sensing element and the water routed through the cooling mechanism. It was believed that temperature changes in the incuba- tion chambers were small, because the main water mass was so large, by comparison, that a large amount of heat would have to be transferred before anj^ appreciable temperature change in the incu- bation chambers would occur. 278 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE The eggs were incubated in pint or quart mason fruit jars. The pint jars had glass tops held in place by a spring clamp; the quart jars were wide- mouthed and equipped with plastic lids. Procedure for use While the vessel is still at the dock, the incu- bator and cooling mechanism are secured to the deck. The water bath is filled with fresh water (which causes less corrosion in the cooling system than salt water) and the circulating and cooling motors are started. Operability of the thermostat is checked. Once the vessel is in the collecting area, the thermostat is set so that the water in the water bath assumes the same temperature as the sea. Later on, at the time the biological sample is re- ceived, a bathythermograph reading is taken and the thermostat adjusted. The fruit jars are filled with sea water from the area and the samples placed in them. A small air space is left at the top of the jars to aid aeration of the sample. The jars are then placed in the jar rack and locked into place. From then on the operation is automatic, although temperature readings should be taken frequently to assure that the incubator is operating properly. General specifications The outside dimensions of the incubator were 32 inches by 32 inches by 35 inches deep. The inside dimensions, the perimeter of the water bath, were 26 inches by 26 inches by 26 inches. Both the inside and outside boxes were made of 5^-inch marine plywood fastened together with brass screws. The space between the boxes was filled with an insulating material. The inside box was lined with copper sheeting that had been soldered at the joints and was therefore watertight. The jar rack also was made of copper. The individual chambers of the bottle rack were 4 inches square and accommodated either pint or quart wide- mouthed jars. The copper lining was pierced by three holes: one for inflowing water, one for outflowing water, and one for the temperature-sensing element. The lid was fastened by two heavy metal strap hinges and tliree clamp locks. A sponge-rubber gasket prevented leakage. The water in the water bath was circulated by a Jabsco pump and cooled by a Temprite cooler. The temperature control was maintained by a Partlow thermostat. Two j4-inch plastic garden hoses with brass fittings connected the pump and cooler with the water bath. All exposed surfaces were painted with green plastic paint. B. STAGING SCHEME OF JACK MACKEREL EGGS Stage I. — Unfertilized eggs or fertilized eggs be- fore cell division. Stage II. — Begins when the first cell becomes visible on the yoke and ends at the completion of blastodisk formation (at about the 256-cell stage). Stage III. — Starts at the completion of blastodisk formation and terminates when the germ ring has migrated to its greatest diameter (half-way up the egg). Stage IV. — Begins as the germ ring moves up- ward over the greatest diameter and ends when the germ ring lies over the oil globule, prior to blastopore closure. Stage V. — Begins at blastopore closure and ter- minates when the tail bud starts to separate from the yolk. Stage VI. — Begins when the tail bud becomes free of the yolk and ends when the caudal one-eighth of the body is free of the yolk. Stage VII. — Begins when the caudal one-eighth of the body is free of the yolk and ends when the caudal one-quarter of the body is free of the yolk. A finfold is visible. Stage VIII. — Begins when the caudal one-quarter of the body is free of the yolk and the tip of the tail approaches the chin. The tail portion of the embryo begins to rotate out of the em- bryonic plane and the finfold is moderately wide. Stage IX. — This stage is characterized by the tip of the tail laterally approaching the head. The oil globule comes to lie in the anteroventral portion of the yolk sac. The finfold is wide and fuUy formed. This stage terminates when the embryo hatches. Disintegrate. — Includes all jack mackerel eggs whose internal structure is such that staging is impossible. DISTRIBUTION OF EGGS AND LARVAE OF JACK MACKEREL 279 Appendix Figure 2. — Stages of jack mackerel egg development. V. S. GOVERNMENT PRINTING OFFICE: 1961 O - 580553 UNITED STATES DEPARTMENT OF THE INTERIOR, Stewart L. Udall, Secretary FISH AND WILDLIFE SERVICE, Clarence F. Pautzke, Commissioner Bureau of Commercial Fisheries, Donald L. McKernan, Director DISTRIBUTION AND ABUNDANCE OF SKIPJACK IN THE HAWAII FISHERY, 1952-53 By HERBERT H. SHIPPEN H> FISHERY BULLETIN 188 From Fishery Bulletin of the Fish and Wildlife Service VOLUME 61 Published by the U.S. Fish and Wildlife Service • Washington • 1961 Printed at the U.S. Government Printing Office, Washington, D.C. For sale by the Superintendent of Document^ U.S. Government Printing Office, Washington 25, D.C. - Price 20 cent* Library of Congress catalog card for the series, Fishery Bulletin of the Fish and Wildlife Series: U.S. Fish and Wildlife Service. Fishery bidletin. v. 1- Washinglon, U.S. Govt. Print. Off., 1881-19 V. in illus., maps (part fold.) 23-28 cm. Some vols, issued in the congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies: v. 1-49, Bulletin. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, V. 1-23) 1. Fisheries— U. S. 2. Fish-culture— U. S. i. Title. SH11.A25 Library of Congress 639.206173 ir55efi 9-35239 rev 2* CONTENTS Page Hawaii skipjack fishery 281 Purposes of this study 281 Utilization of data 282 Choice of unit of fishing effort 282 Sources of error 286 Error in determination of fish size 286 Error in estimation of fishing effort 287 Other sources of error 287 Conclusions on sources of error 287 Results and discussion 287 Geographical distribution of 1952 and 1953 catches 288 Catch per unit-of-effort in the Hawaii skipjack fishery 288 Size of skipjack and pounds caught per unit of effort 289 Distribution of 1952 and 1953 catches by biweekly periods 292 Catches for the entire fishery 292 Effects of small craft warnings 292 Movements of skipjack within the fishery 292 Large skipjack 294 Small skipjack 294 Oahu skipjack fishery, 1952 and 1953 294 Statistical comparisons 294 Relationships between catch statistics 294 Use of catch records to determine population composition 295 Fluctuations in the catch of large skipjack in the Oahu region 296 Size composition 297 Conjecture 297 Summary 300 Literature cited 300 ABSTRACT Commercial catch records of the Hawaii sliipjack fishery for 1952 (a poor year) and 11)53 (a good year) are summarized by area and time of catch and by size composition. A unit of fishing effort and its appropriateness are discussed. Geographical distribution of the catch and effort is determined and the two years are compared. Movements of skipjack throughout the fishery are analyzed. The usefulness of the raw catch and the catch per unit of effort as indexes of abundance are considered, and some conjectures as to the nature of the population supporting the fishery are offered. DISTRIBUTION AND ABUNDANCE OF SKIPJACK IN THE HAWAII FISHERY, 1952-53 By Herbert H. Shippen, Fishery Research Biologist, Blreau of Commercial Fisheries A study of the environmental factors that may influence the availability of the skipjack {Katsu- wonus peJamis) to the Hawaii fishery was begun by the staff of the U.S. Fish and Wildlife Service Biological Laboratorj' (Honolulu, Hawaii). Be- cause the index of availability is to be based on records of commercial skipjack landings, an analy- sis of these records is an essential part of this study. HAWAII SKIPJACK FISHERY The skipjack, or aku, is the most important com- mercial species of fish in Hawaii, both in terms of quantity landed and dollar value. The 11 million pounds caught and sold for $1,260,000 in 1956 con- stituted about 70 percent of the total catch of marine species and 40 percent of the value received by Havraii fishermen during that year. Most of the catch is canned, but a small amount, estimated at less than 10 percent, is sold fresh. June (1951) and Yamashita (1958) have de- scribed the fishery in some detail. Since World War II, the skipjack fleet has consisted of approxi- mately 15 to 20 sampans based in Honolulu, with a few boats based at the islands of Kauai, Maui, and Hawaii. A sampan usually carries a crew of 8 to 15 men. The fishermen rely on the presence of flocks of wild birds to locate skipjack schools. The fish are caught on pole-and-line after being attracted to the boat by chumming with live bait. The fishery is seasonal with large catches gener- ally occurring in the summer and small catches in the winter months. Catches have fluctuated widely in recent years (fig. 1). Tlie skipjack taken weigh from 2 to 30 pounds. The most sought after size is the 17- to 22-pound fish, known to the fishermen as "season fish.'' Brock (1954, p. Note. — Approved for publication February 24, 1961. Fishery Bulletin 195. 96) estimates these to be either in their second or third year of life. The reason for the seasonal fluctuation in the catch appears to be the migra- tion of season fish into and out of the area of the fishery, but the direction and significance of this migration in the life history of the species are largely miknown. 12 10 e 6 4 2 n 1 - 1 1 — L 1 — — — — - 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 FiGUEH 1. — Annual Hawaii skipjack catch, 1948-58. PURPOSES OF THIS STUDY The purposes of this study are (1) to examine the i-aw catch data and I lie catch per unit of effort as measures of the apparent abundance of skip- jack; (2) to search the data for differences Ije- tween good and poor years in the fishery; (3) to study movements of skipjack within the fishery during the course of the season ; (4) to examine the distribution of pounds of skipjack caught, catch per unit of effort, total effort, and size composi- tion of the catch throughout the fishery. 281 282 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE I wish to thank the staff of the Hawaii Division of Fish and Game who collected the fish-catcli re- ports that form the basis of this study. Vernon E. Brock and Tamotsu Shimizu made their data available for study. Additional information was received from Saul Price of the U.S. Weather -Bu- reau who furnished the data on small craft warn- ings. Peter Wilson of Hawaiian Tuna Packei's, Ltd., was instrumental in obtaining the logbooks from two fishing sampans; and Dr. Robert Riff en- burgh suggested certain useful statistical procedures. UTILIZATION OF DATA The fish-catch reports (fig. 2), completed by the fishermen, were used in this study. Ite^is in these reports are treated as follows : Thme of catch. — The interval from the begin- ning of 1952 througli 1953 was divided into bi- weekly periods (table 1). Catch reports were grouped by periods according to date of landing. Table 1. — Biweekly periods in 1952 and 1953 Period 1952 1953 1 Jan. 1-12 Dec. 28-Jan. 10. o Jan. 13-26. Jan. 11-24, 3. Jan. 27-Feb. 9... Jan. 25-Feb. 7. 4 Feb. 10-23 Feb. 8-21. 6 Feb. 24-Mar. 8 Mar. 9-22 Feb. 22-Mar. 7 6 Mar. 8-21. 7 Mar. 23-April 5 Mar. 22-April 4. 8 _._ April 6-19 April 6-18. 9 --. April20-May3 April 19-May 2. 10 May 4-17 May 3-1 ft. 11 May 18-31 - May 17-30. 12 June 1-14 May 31-June 13. 13 - - June 15-28. June 14-27. U... June 29-.Tuly 1 2 - June28-.ruly 11. 15 Julv 13-26 July 12-25. 16 17 Aug. 10-23 Aug. 9-22. 18 .\ug. 24-Sept. 6 ... Aug. 23-Sept. 5. 1<) Sept. 7-20 Sept. 6-19. 20-.. Sept. 21-Oct. 4 Sept. 20-Oct. 3. 21 Oct. 5-18 Oct. 4-17. 22 Oct. 19-Nov. 1 Oct. 18-31. 23 Nov. 2-15 Nov. 1-14. ?4 Nov. 16-29 Nov. 15-28. 25 Nov. 30- Dec. 13 Nov. 29- Dec. 12. 26.. Dec. 14-27 Dec. 13-26. Area of catch. — -The catch reports were sorted and reported according to statistical area (fig. 3). For reasons discussed under Sources of Error, the statistical areas have been summarized in terms of zones and regions (fig. 4) . Pounds caught. — This figure was used exactly as recorded in the catch reports. Average size of skipjack caught. — The total weight was divided by the estimated number caught to arrive at the average weight per fish in the catch. Catches were then classified according to the following categories: (1) small fish (aver- age weight 10 pounds or less) , (2) large fish (aver- age weight greater than 10 pounds) , or (3) catches for which no size estimate was possible, because the number of fish caught was omitted from the report. Estirnate of total number of skipjack caught in each size group. — A simple proportion, utilizing the known weights and numbers, was used to esti- mate the numbers of small and large skipjack in the total catch. For example, if the summary of data from the fish-catch reports for a particular region and period yields the following informa- tion : Weight and number offish Small skipjack Large skipjack No size data Total Pounds.. 30,000 6,000 50, 000 2,500 40,000 120,000 Number • Unlmown. then, the estimated total number of small skipjack e,,gHt IS ^^«^lg») =9,000, and the esti- oO,000 mated total number of large skipjack caught is (2,500) -(120,000) 80,000 = 3,750. Unusable f.^h-catch reports. — A small number of reports was set aside and not used, except to accumulate gross totals of pounds caught. If a report fell into one or more of the following cate- gories, it was classified as unusable: (a.) no sta- tistical area was given on catch report, or area number given did not appear on Division of Fish and Game Chart (fig. 3) ; (6) several statistical area numbers were given so that assignment of the catch to any single zone or region was impossible ; (c) several trips were apparently grouped on one catch report so that estimates of fishing effort would be erroneous. All other reports were considered usable. CHOICE OF UNIT OF FISHING EFFORT The fish-catch report gives no direct informa- tion on the amount of effort. There are no data to indicate the number of fishermen making the catch, the time in terms of scouting and fishing, the number of mireported trips with no catcli, or any of the other factors which might be pertinent. The fish-catch reports provide, insofar as the de- termination of effort is concerned, a listing of SKIPJACK IN HAWAII FISHERY 283 dates on which fish were unloaded from the vessel. It is from this list, and other data, that fishing eilort was estimated. Eaoli usable catch repoi-t was assumed to de- scribe tiie results of a single trip of tlie ves.se]. Each boat has an official number-of-crew, which is reported to the U.S. Customs ( Yamasliita, 1958, table A-1 ) . Tliis figure, a constant for each vessel, was assigned as a weight to each usable catch re- port to represent the amount of effort expended in TERRITORY OF HAWAII BOARD OF COMMISSIONERS OF AGRICULTURE AND FORESTRY DIVISION OF FISH AND GAME FISH CATCH REPORT Name of Permittee Boat Permit No. Nome of Boat FG No- Type of Fishing J Fishing Gear FORM C 1 5 B 93859 lOM StTS-7-51 Ai rea of Catch 1 Date of Landing. Mo. 19 (See Fisheries Chart No. 2) J Ooy SPECIES CAUGHT 1 No. CAUGHT LBS. CAUGHT LBS. SOLO VALUE* Aku (Skipjack) 002 Ahi (Yellowfin) (Shibi) 003 Ahipalaha (Albacore) (Tombo-shibi) 004 < z J E apanese Blueiin (Block Tuna) (Maguro) 005 lig-eye (Menpachi-shibi^ ("Bluefin") 006 Kawakawa 007 Striped Marlin 009 lU Black Marlin 010 O 4 Short-nose Marlin 107 Silver Marlin 108 i- Broadbill Swordfish Oil Au lepe (Sailfish) 012 Mahimahi 013 One 014 ! 11 BAIT REPORT DATE TAKEN TIME TAKENt | LOCALITY TAKEN QUANTITY TAKEN QUANTITY USED DAY NIGHT Nehu 41 buckets buckets lao 42 buckets buckets Opelu 20 fish fish Sardines 07 1 pounds * Value represents the amount of money Received by the fisherman for total pounds of fish sold. Do not record price per pound. t Check one to indicate whether baiting was done at doy or at night. Applies to liveboiting only. The above reports are true, correct, and complete to the best of my knowledge and belief. Signature Port of Landing , Permitlee or Authorized Agent Island a Figure 2. — Hawaii Division of Fish and Game, Pish Catch Report (1950-54). 284 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE ,.. 9»C 997 9»S 9»» 640 641 642 9*5 9»e 991 9ffO 999 999 597 99« 470 469 4«9 5T9 900 99 1 SIZ 993 994 999 463 464 469 466 467 ST 3 ^ - _^''' "' N ,, ^ 9>i ^ \ S74 X "' - ^€^^^^' 111 46i 461 460 499 496 372 971 3T0 SC9 366 .J \ i^'"''^£^~- j ,„ 496 \ «9 J 49T 3., 362 S6S 38 4 969 366 38 T ' ^ ''. **o -' ■■'■T7.. ;3fc,.„ 426 '... 360 399 1 •«■"■• \ '"^ fA ,., 499 -^SSB 397 396 S59 33. Z92 "' "~^ ^ 599 423 *y<."" 429 332 9S9 see "^ 32 998 99T 999 S60 961 .,. V 49S 22 421 .33 .., ,3, 29> '*^^?*^ \| SSI 9S0 94 9 949 94T 449 449 4 50 451 491 ^"^ \ »z' >c S^^'" \ 248 24T 942 943 944 949 94G 44T 446 443 444 44J 330 949""-^- i 123 3i9 \^I9T / 198 124 199 240 241 242 440 441 442 949 J« 947 L..3 194 193 132 34a 198 \Ji^-... ^:^s;^ r '^ " '" \ 191 i FISHERIES CHART NO 2 Ik kcpl on hoard ill fithinE hiuu lu <<.»i^ A /: \ A • A C • • A • r>A ' A • 1 • A \ • A A 1 • * A • A • A • AY ^MWa A A • A • :2lT^ ^ A 1 ) A A A • A • A • 1 L/a • V. ^ ~ 1 1 1 A IL A A ZONES REGION 11 HW OCEANIC 11 ( :W OFFSHORE OAHU lltlW HAWAII 41 (4W MAUI SI t SW INSHORE OAHU 61 t 6W KAUAI lit SI LEEWARD OAHU 2W4SW WINDWARD OAHU - Jl.:W.S145W OAHU "1" INDICATES LEEWARD IT INDICATES WINDWARD A z/. iLV^ A ( V • \ ^ \\ HA WAN \n. ) V \,/K^ y 1 j V • ty \ y^ 1 v^ • '9^2 '^STATISTICAL AREAS t I953J REPORTING CATCHES , , Figure 4. — Hawaii skipjack fishery fishing zones and regions and the extent of fishing in 1952 and 1953. making the catch. The number of fishermen is used as a factor in the computation of fisliing effort because it seems reasonable that in pole-and-line fishing the efficiency of a vessel is more or less di- rectly related to the number of men hooking fish. No adjustments were made for ditl'erences in trip time or for deviations from the official number of fishermen. Inasmuch as a fish-catch report is re- quired only if fish are caught, the imit of effort employed in this study is the productive fisher- man-trip. Thus, if a vessel with a regi.stered crew of 10 men reported a catch of 20,000 pounds, the effort is considered to be 10 units and the catch per unit of efi'ort is 2,000 pounds. If two or more catch reports were combined, the sum of the catches was divided by the sum of the efi'ort to oht ai n t he catch per unit of effort. To gain some knowledge of the reliability of the productive trip as a factor in tiie unit of efi'ort, the logbooks of two Honolulu-based skipjack fish- ing sampans were analyzed to determine the ratio of productive trips to the actual number of days spent fishing; i.e., the time spent in scouting for and catching skipjack. The results of this analysis appear in table 2. Boat .1 is typical of the fleet as a whole in that it makes freciuent trips of sel- dom more than a single day. Boat B. on the other hand, is probably the most atypical in the fleet since it ventures far afield and may remain at sea for as many as 4 days, especially when skipjack are relatively scarce. The ditl'erences between the two boats are apparent in the number of trips per biweekly period (col. 7) and the number of trips per day's fishing (col. 10). 596560 O -62 -2 286 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Each year is divided into the more productive and less productive parts; for convenience these are called summer and winter, respectively (rows 1 and 2 for each boat and year). The number of trips per biweekly period (col. 7) and number of day's fishing per biweekly period (col. 9) are greater during summer, but the ratio of trips to day's fishing (col. 10) is not markedly different in the two seasons, although trips are somewhat longer in winter. Nonproductive trips (col. 5) occur with greater frequency (col. 6) in winter than in summer, and there is a tendency for the number of productive trips per day's fishing to be greater in summer. Differences in the number of productive trips per day's fishing are also apparent between years, as in 1953 there were generally more productive trips per day's fishing than in 1952. To summarize the performances of the two boats, it appears that the ratio of productive trips to the nvmiber of day's fishing is greater during times of good fishing and smaller during times of poor fishing. Since trips are shorter when the fish- ing is good and there are also fewer nonproductive trips, the actual effort in terms of day's fishing will usually be underestimated dui"ing the periods of poor fishing as compared witli periods of good fishing. Variations from the official number of crew will also affect the accuracy of the estimate of fishing effort. The official number is a maximum and variations will usually mean that fewer than the official number are aboard. In the Hawaiian skip- jack fleet, boats ordinarily carry the maximum number of crew during the summer season, after which some men leave to find other employment. In this study, since the official (maximum) num- ber of crew has been used throughout the year as a weight for the individual trip, the "fisherman" factor in the productive fisherman-trip is probably overestimated during the times of poor fishing. Thus, the biases in the productive fisherman-trip between times of good and poor fishing tend to cancel each other because during the winter season and years of generally poor fishing, longer trips and the increased frequency of nonproductive trips cause an underestimation of the actual time spent fishing while, at the same time, the actual number of fishermen on the boat is likely to be fewer than the official number. The converse will hold true during the summer season and in years when good fishing attracts the maximum number of fishermen to the fleet. Information is not available to permit the ex- amination of the actual variations in the number of fishermen and the extent to which they offset the bias introduced by nonproductive trips, but if boat ^4 is assumed to represent the average situa- tion, the number of productive trips per day's fish- ing (col. 11) appears to be about 10 percent greater in summer than in winter. Since the crew of the average skipjack boat is about 10 men, the absence of one of these men on the average during the winter season represents a 10-percent overestima- tion of the number of fishermen. Thus, the ab- sence of one fisherman per boat during the winter season would be sufficient to equalize the bias in the productive trip factor introduced during the winter season. SOURCES OF ERROR Unreported catches or forms containing incom- plete or inaccurate information are an obvious source of error. Yamashita (1958, p. 258) esti- mates that the reported portion of the 1952 catch included di percent of the pounds, but only 88 per- cent of the trips, indicating a bias in favor of the reporting of large catches. Since small catclies are most likely to occur in the slack part of the year, there may be a tendency for an estimate of the fish- ing effort, which is a function of the number of trips, to be correspondingly reduced. Inaccurate information is difficult to detect without data from other sources with wliicli to compare the catch records. On the basis of inter- view records, Yamashita (1958, p. 258) estimates that only 45 percent of the statistical areas indi- cated in the 1952 catch reports were reasonably accurate. By means of bi-oad geographical divi- sions to summarize the data (fig. 4), it is assumed that the effects of such erroneous information will be minimized. ERROR IN DETERMINATION OF FISH SIZE Dividing the total weight caught by the esti- mated number as indicated in the catch report, yields the average size of fish caught, but provides no indication of the range or variability of sizes. Since the entire catch is assigned to eitlier the small or large category on the basis of the average SKIPJACK IN HAWAII FISHERY 287 weight per fish, a certain amount of error will re- sult from mixed catches of small and large fish; this error should disti-ihute itself more or less ran- domly, however, so that neither size group is consistently favored. ERROR IN ESTIMATION OF FISHING EFFORT There is no way to determine from the catch records the actual effort, i.e., the fisherman-days whether productive or not, put forth on a skipjack boat. In this study only positive fishing results (catch' reports) are available, and the productive fislierman-trip is of necessity used in lieu of the fisherman-day. Sources of error in the productive fisherman-trip have been discussed in the section, Clioice of the Unit of Fishing Eifort, and on the basis of the performance of two skipjack boats for which logbooks are available, it appears to be a reasonable substitute. OTHER SOURCES OF ERROR The weight of the catch of skipjack taken in tlie Hawaiian live-bait fishery is affected by complex factors which present sources of error that are difficult to estimate. Among these factors are variations in bait supply, response of skipjack to chum, behavior and niuuber of birds in the flocks which serve to locate schools, the size and behavior of the skipjack schools, selection by the fishermen, and probably several others. Yamashita (1958, p. 270) has discus-sed the problem of ascertaining the influence of variations in bait supply on the skipjack catch in terms of annual production and suggests that in certain years, when skipjack have been plentiful, the availability of bait may be a limiting factor in the fishery. Royce and Otsu (1955) have investigated many aspects of Ijeliavior of skipjack schools and birds; Yuen (1959) has studied the response of skipjack to live bait. In the present study no attempt has been made to evaluate the sources of error introduced by the factors considered alx)ve. Information available is not adequate to discern which of these may be important at any particular time. It seems reason- able that most of these factors act relatively in- dependent of one another so that over a period of time their combined effects should not introduce bias. However, it is just as plausible that at cer- tain times several of these elements may act in unison resulting in considerable deviation from the normal state. The investigation of the role of these factors in the fishery awaits a more sophisti- cated study than is attempted here or is possible with the present sources of information. CONCLUSIONS ON SOURCES OF ERROR None of the sources of eri-or appears to be so ex- tensive as to destroy the usefulness of the catch report as the basis for a study of distribution and abundance. Some of the sources of error tend to reduce the bias introduced by othei'S. With re- spect to time, geogi-aphy, and size, the categories employed in this study have deliberately been made broad. Were the study concerned with only a few vessels, very short time periods, or several size groups, the probability of error would be in- creased, but as only the most general of categories are used, the influence of error on the results should be slight. RESULTS AND DISCUSSION There are small discrepancies between the official total catches for 1952 and 1953 as listed by Yama- shita (1958, table 2) and the totals obtained in the present study (table 3). These differences amount to 1 percent and are probably the result of catch reports, which were turned in too late to be in- cluded in official summaries and to records lost or misplaced during the interval of storage. The pi-oportion of unusable data in 1953 was greater than in 1952 (table 4). largely because of the poor Table 3. — Comparative data from 2 studies of the Ha- waiian skipjack catch for lHoZ and 1953 Year Pounds skipjack caught Difference Percent Yamashita i Shippen » difference 1952 7, 291, 851 12, 059, 406 7, 390, 882 11,928,965 99,031 -130,441 1.3 1953 —1.1 1 Source: Yamashita (195S, table 2). ' Figures adjusted to correspond with calendar year. Table 4. — I'sahility of 1952 and 195S catch report data 1952 1953 Pounds Percent Pounds Percent Usable - 7,270,990 105, 463 98.6 1.4 11,345,013 598,391 95.0 Unusable 5.0 Total 7,376,443 100.0 11,943,404 100.0 288 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE reports from the master of one sampan who con- sistently summarized his catches by weeks throughout much of the year. GEOGRAPHICAL DISTRIBUTION OF 1952 AND 1953 CATCHES Desjiite the large differences in total landings in 1952 and 1953, the geographical distribution of catch and effort (table 5, tig. 5) is much the same. In both years the leeward Oahu and Hawaii re- gions furnished approximately 50 and 16 percent of the total catch, respectively, and the oceanic region and Maui were relatively unimportant with less than 8 percent in the aggregate. The com- bined windward Oahu and Kauai regions con- tributed about 25 percent of the total catch in each year, but in 1953 a much larger proportion of this came from windward Oahu. Within the vicinity of Oahu, the distribution of effort appears to be related to the distance from the home port ; the amount of effort expended in the zones decreases as their distance from Hono- lulu increases. This is probably because of the fragile nature of the nehu {Stoleplionis pur- fureus), the most important bait species. CATCH PER UNIT OF EFFORT IN THE HAWAII SKIPJACK FISHERY If the regions of the fishery from Hawaii in the southeast to Kauai in the northwest are arranged in sequential order (fig. 6), there is some sugges- tion of an increasing catch per unit-of -effort in the direction of Kauai, but the inequities in the distri- bution of effort and certain known differences in the local fisheries make it doubtful that the ap- parent trend is of biological significance. The Hilo (Hawaii) fishermen usually make short trips and land each day's catch on the day it was made, Table 5. — Geographical distrihiitlon of the 1952 and 1953 usable catch data [See fig. 4 for location of zones; C/E= catch/effort] 1952 1953 Regions and zones Catch Effort C/E Catch Effort C/E Pounds Percent Units Percent Pounds Percent Units Percent Oceanic: IL 183, 729 128, 458 2.6 1.8 244 348 1.2 1.8 763 369 274, 358 191,531 2.4 1.7 439 236 1.7 0.9 625 IW -- 812 Total 312, 187 4.3 592 3.0 627 465,889 4.1 676 2.6 Hawaii: 3L 252, 659 881, 673 3.5 12.1 1,073 2,869 6.4 14.6 236 307 48.5,317 1,288.967 4.3 11.4 1,572 3,676 6.2 14.4 309 3W 361 Total 1,134,332 15.6 3,942 20.0 288 1,774,284 16.6 5,248 20.6 338 Maui: 4L __ 115. 113 139, 939 1.6 1.9 119 463 0.6 2.4 967 302 25,257 179, 291 0.2 1.6 62 452 0.2 1.8 407 4W.. 397 Total. 255,052 3.5 582 3.0 438 204, 648 1.8 614 2.0 398 Inshore Oahu:' 6L 2,259,734 719, 780 31.1 9.9 7,350 1.952 37.3 9.9 307 369 3,305.638 1, 742, 242 29.7 15.3 9,116 3,266 36.7 12.8 369 5W 634 Total 2, 979, 614 41.0 9,302 47.2 320 6, 107, 780 46.0 12,382 48.5 413 Offshore Oahu:* 2L 1, 486, 181 157, 854 20.4 2.2 3,519 361 17.9 1.8 422 437 2, 666, 150 416,346 23.4 3.7 6,022 674 19.7 2.6 629 2W.. 616 Total . 1,644,035 22.6 3,880 19.7 424 3, 071, 496 27.1 5,596 22.3 539 Oahu region subtotals:* 3,745,915 877, 634 61.5 12.1 10, 869 2,313 55.2 11.7 345 379 6,077,707 2,212,112 53.6 19.6 14,242 3,990 65.8 16.6 427 W 554 4, 623, 549 63.6 13, 182 66.9 351 8,289,819 73.1 18,232 71.5 466 Kauai: 6L 284, 845 661,025 3.9 9.1 560 844 2.8 4.3 509 783 217, 722 392, 751 1.9 3.5 373 462 1.5 1.8 6W 850 Total. . . . 945,870 13.0 1,404 7.1 674 filO, 473 5.4 835 3.3 731 Grand total 7,270,990 100.0 19, 702 100.0 369 11,345,013 100.0 25, 504 100.0 445 *The Oahu region includes inshore and offshore Oahu, 2L, 5L, 2W, and 5W. For 1953, a few additional catches were made across zone boundaries within the Oahu region. SKIPJACK IN HAWAII FISHERY 289 Figure 5. — Geographical distribution of catch and effort in the Hawaiian skipjack fishery, 1952 and 1953. whereas the trips by Honolulu-based fishermen to the vicinit_y of Kauai are longer than a single day ; therefore, tlie differences in tinie-of-trip between the two areas are probably significant. Siniilarl}-, catch per unit of effort tends to in- crease with increasing distance from shore ( fig. 7) . Only a fraction of the total effort was expended in the oceanic region as compared with the effort inshore, and a few good catches of large fisli may have produced an index far out of proportion to the actual apparent abundance. Royce and Otsu (1955, p. 18), however, report sighting more tuna schools per day's scouting beyond 19 miles from shore than were seen within 19 miles of shore. SIZE OF SKIPJACK AND POUNDS CAUGHT PER UNIT OF EFFORT There is a positive correlation (fig. 8) between the average size of skipjack caught in zones of the fishery during the year (table (i) and correspond- ing catch per unit of effort (table 5). Zones with less than 5 percent of the total annual effort are not included in the analysis because they ai'e un- likely to represent fishing conditions througliout the year. This correlat ion appears to substantiate the observation that tlie larger skipjack usually can be caught more efliciently than the smaller, up to the size at which individuals must be gaft'ed in landing and the efficiency drops. 290 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE HAWAI I MAUI LEEWARD WINDWARD OAHU OAHU KAUAI i 1 1 1 1 A / / \ / / ^>^ LARGE SKIPJACK / P ' \ / / / ' \ / / ' N ^ / / "h^ / / / ^X^^^^,_^^ ALL SKIPJACK / _ / ^^ ' — .^ _--o •l^^ ^ ° ^C... " ,• ° — " -o '■" \ SMALL SKIPJACK 19S2 1 I 1 1 20.0 55.2 1 1 1 1 1 / 800 - / / ri / / jP o LARGE SKIPJACK / y^ o GOO — / -rf y^ — ft ^-'''^^"-^—-^-'"^^ \- ^< 1 a. 400 — ^ ALL SKIPJACK ..■•" a. . — " I ° o.. , ""T SMALL SKIPJACK 1953 20.6 3.3 2.0 55.8 15.6 PERCENT EFFORT Figure 6. — Pounds caught per unit of effort by island regions. Table 6. — Average weight of skipjack caught in each zone of the fishery 1952 1953 Zone 1962 1963 Zone Average weight (pounds) Average weiglit (pounds) Average weight (pounds) Average weight (pounds) IL 15.6 7.9 •9.0 17.2 ♦4.8 •4.8 18.2 14.7 •13.9 16.9 •7.7 •7.5 4L_ 4W _. 6L 14.0 11.9 •8.2 •9.2 6.7 12.8 12.1 13.0 *11 2 IW 2L 2W 3L 5W. _. 6L •12.8 11 6 3W 6W.. 8.1 •Indicates zones receiving more than 5 percent of the total fishing eflort during the year. Each year the small fish are caught in Hawaii (zones 3L and 3W) ; the larger fish are from in- shore windward Oahu (5W), and offshore leeward Oahu (2L). The association of small fish with 1000 INSHORE OAHU OFFSHORE OAHU OCEANIC 1 1 1 800 / / / / LARGE SKIPJACK / '\ • \^ • / ^^ ALL SKIPJACK " •, o - 1 600 O ? 400 5 o-"^ - 3 200 Q 1 SMALL SKIPJACK 1952 1 1 - LARGE SKIPJACK ALL SKIPJACK SMALL SKIPJACK 1953 I 48 5 22 3 2 6 PERCENT EFFORT Figure 7. — Pounds caught per unit of effort by distance from shore. ' 1 1 1 1 2L y ^ « 1952 500 o 1953 2L VX — ^y^ — u. 300 3W ^ o t 3L6L 5°L i a: •& u 200 - /"^ 3L — Y= 116.4 +28 85(X) - o r=t.90 100 ' 1 1 1 P<.0I 1 1 1 ) 2 4 6 S 10 12 14 16 AVERAGE WEIGHT OF SKIPJACK Figure 8. — Regression of catch per unit of effort on average weight of fish caught. SKIPJACK IN HAWAII FISHERY 291 Hawaii may be explained by the nature of tlie fishery there, wliich is based hirgely on semi- resident popidations of small skipjack. Other populations of small fish are known to occur in inshore areas of leeward Oahu (oL) , and these are usually exploited wjien the large skipjack are in low abundance and produce the intermediate aver- age weight for zone 5L. Zones oW and 2L are more remote from Honolulu than is 5L, and it seems probable tliat the fishery in fliese zones may be biased in favor of ])er-iods when lai'ge skipjack are available, which could account for llie rela- tively greater average weight per fisii in these zones of the hshei'y. I S 10 S 1952; 7.376,443 POUNDS O 1953: 11,943.404 POUNDS 1953 DATES FOR BIWEEKLY PERIODS ARE ONE DAY DISPLACED EXCEPT AT THE BEGINNING OF THE YEAR WHEN THEY ARE TWO DAYS DISPLACED FROM 1952 W\ '*^ 1 13 27 10 24 9 23 6 20 4 18 I 15 29 13 27 10 24 7 21 5 19 2 16 30 14 12 26 9 23 8 22 5 19 3 17 31 14 28 12 26 9 23 6 20 4 18 I 15 29 13 27 JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I 1 ! 1 1 1 1 1 1953 1 1 \i\m\ 1 1 1 1 1 1 nnni 1 ~ Figure 9. — Catches (pounds) of the Hawaii skipjack fishery, 1952-1953, by biwivkly i)eriods. Periods of small craft warnings are shown below. 292 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE DISTRIBUTION OF 1952 AND 1953 CATCHES BY BIWEEKLY PERIODS Catches for the Entire Fishery Only during July and August (periods 14-17) and briefly in December (period 25) did the 1952 skipjack catches reach the magnitude of the 1953 totals (fig. 9). In particular, the spring of 1953 and, to a lesser extent, the autumn months pro- vided much larger catches than occurred in 1952. Effects of Small Craft Warnings There were 49 days with small craft warnings during 1952 and 17 during 1953 (fig. 9), but only 6 of these fell in the interval from April through September in 1952 and none in 1953. The effects of rough weather, thei-efore, appear to be rela- tively minor in comparison with the seasonal fluctuation in the availability of skipjack. The immediate effects of poor weather may be indicated by the relatively small catches made dur- ing periods 1, 2, 5, 6, 23, and 24 of 1952, each of which had several days with small craft warnings. Period 25, on the other hand, shows an increase in catch despite 5 days of poor weather. MOVEMENTS OF SKIPJACK WITHIN THE FISHERY Local changes in the skipjack catch within the entire range of the fishery may result from changes in the amount of fishing effort or from changes in availability. The latter may include horizontal movement of the fish into or out of a particular area or, within an area, a change in the vertical distribution or behavior such that the catch rate by live bait fishing is affected. The records of the fishery, however, provide no means by which one or the other cause may be determined, and it is therefore assumed for the purposes of this discus- sion that all changes in the catch are caused by movements of fish from one area to another. Thus, errors, if any, are likely to be on the side of postulating a horizontal movement of fish when there has been a change in vertical distribution, behavior, or fishing effort. This approach seems to be the most reasonable one, because tagging ex- periments show that individual skipjack travel the length and breadth of the fishery, while knowledge of changes in availability and fishing effort, par- ticularly if nonproductive, remains quite limited. Large Skipjack After an interval of low abundance throughout the islands during the early part of 1952, large skipjack (fig. 10) appeared simultaneously in small numbers in leeward Oahu and Kaiu^i in period 8. In period 10 the fish arrived in wind- ward Oahu and Hawaii. This sequence suggests an approach from the west. In period 12, a con- centration centered in windward Oahu occurred; it appears to have shifted northward to Kauai by periods 15 and 16. In jieriod 17, however, the catches of large skipjack ceased in Hawaii and began to dwindle in Kauai, but at the same time the largest catches of the year were being made in leeward Oahu. All these changes seem to indicate that the large skipjack had returned to the leeward side of the island chain. The gradually diminish- ing catches from Kauai and leeward Oahu in periods 18 through 20 indicated the withdrawal of season fish to the westward. After period 20, the numbers of large skipjack in the catch returned to the state of low variable abundance which char- acterizes the off-season of the fishery. During the interval from period 25 (1952) until period4 (1953) the number of large skipjack taken in all regions of the fishery was uniformly low, a condition typical of the winter season. In period 5, however, a sharp increase occurred in the catch of large skipjack in the leeward Oahu region. To judge from the variation in average weights (fig. 12), these fish were 1952-season fish, being some- what heavier than 1953-season fish which entered the fishery in period 9. These 1952-season fish ap- peared in the catches during periods 5, 7, 8, and 9 and were the cause of the apparent early begin- ning of the "season" in 1953 (fig. 9). In period 9 of 1953, the season fish were present throughout most of the fishei-y (note the declining average weights in periods 9 and 10, fig. 12) but the large catches in leeward Oahu in periods 10 through 12 suggest that the direction of the ap- proach of the main body of fish was from the lee- ward. As in the previous year, a peak occurred early in the season in windward Oahu (period 12, 1952; period 13, 1953), and in succeeding periods the fish dispersed southward to leeward Oahu and Hawaii where large catches were made in periods 15-17. Following the excellent catches of period 17, the best of the year, a gradual decrease in catch occurred, and by period 23 the season was over. SKIPJACK IN HAWAII FISHERY 1952 1953 293 KADAI ■■- ■■■I IllL KADAI I - WINDWARD OAHU .1 III.I . WINDWARD OAHn J MIL LEEWARD OAHD Jhjlljllj lULul LEEWARD OAHD ll IL - HAWAII - . ..1 1 1 1 1 [ 1 1 1 1 1 1 1 1 1 1 1 1 1 5 10 15 20 25 HAWAII . Illlllli I . BIWEEKLY PERIODS BIWEEKLY PERIODS PiGUEE 10. — Number of large skipjack caught (1952-53) ; biweekly periods. Certain features common to both years are to be noted: (1) the approach of the hirge fish at the start of the season, apparently from the leeward ; (2) the concentrations on the windward and lee- ward sides of Oahu in June and August, re- spectively ; and (3) the final disappearance of fish to the leeward. Differences in the 2 yeai-s are as follows: (1) the appearance in the early part of 1953 (period 5) of large skipjack, and (2) the direction of movement of the season fish between the time of the windward Oahu peak catches (periods 12-13) and tlie leeward Oahu peak (period 17). In 1952 the fish went northward to Kauai and thus close to the limit of the fishery. In 1953 they returned to leeward Oahu and Hawaii to remain well within the range of the Honolulu and Hilo based vessels. In general, the movements of large skipjack, as indicated by their occurrenc<^ in the connnercial catch, do not suggest an orderly migration along the island chain. The reason for this may be in the direction of approach of the migrating schools, 294 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE which appears to be perpendicular rather than parallel to the barrier formed by the islands. The relative speed with which schools travel, which may reach 15 knots (Eoyce and Otsu, p. 18), may be such that a period of 2 weeks is too long to dis- cern a migi-atoiy pattern within an area as small as that encompassed by the Hawaiian skipjack fishery. Small Skipjack Examination of the numbers of small skipjack caught and the corresponding catch per unit-of- effort data revealed no discernible pattern of move- ment within the fishery. Large skipjack are the prime objective of the fishery and the smaller sizes are usually taken as a second choice. The number of small skipjack in the catch, therefore, tends to be a function of the number of large fish available. The relationships between the numbers of large and small fish in the catch are discussed under the section, Size Composition. OAHU SKIPJACK FISHERY, 1952 AND 1953 STATISTICAL COMPARISONS The comparative statistics for 1952 and 1953 in the Oahu region are summarized in table 7. In almost every way, 1953 reflects the greater avail- ability of skipjack than in the previous year, as evidenced by (1) a mucli larger catch, (2) almost half as much fishing effort, and (3) a larger catch per unit of effort. The number of small skipjack caught (4b) and pounds caught per unit of effort (3b) are not markedly different between years, but the catch of large fish, both in absolute numbers (4a) and on a relative basis (3a), is considerably gi-eater in 1953. Most of the differences between the 2 years can be attributed to the abundance of this size group in the fishery. Independently of size considerations, the num- ber of fish taken per biweekly period in 1953 was larger than the corresponding number in 1952 (5 and 6) . The relative abundance of large and small skipjack in the 2 years is indicated by the nimiber caught per unit of effort (7) wliicli is greater for the small fish in 1952 and for the large fish in 1953. The importance of the abimdance of large fish to the success of the fishery may be measured by the comparative number of biweekly periods in the various categories (8). Table 7. — Comparative statistics for the Oahu fishery, 1952 and 1953 1. Total pounds caught (Percent of total for skipjack fishery). a. Large skipjack only b. Small skipjack only 2. Total productive elTort _-_ a. Large skipjack, percent h. Small skipjack, percent 3. Pounds caught per unit of effort, all usable catches a. Large skipjack only, ___ b. Small skipjack only 4. Total number of flsh caught a. Large skipjack only b. Small skipjack only 5. Average number caught per biweekly period 6. Median number of skipjack caught in each year for the 26 biweekly periods. 7. Number of fish caught per unit of effort _. a. Large skipjack only b. SmaU skipjack only 8. Number of biweekly periods with — a. more than 300,000 pounds catch. b. more than SOO units Ashing ef- fort c. more than 400 pounds catch per unit of effort 19S2 1953 4,623,549 8, 715, 958 (63) (73) 2, 058, 921 6, 366, 336 2, 564, 628 2, 349, 622 13, 182 19, 169 34 65 66 45 351 465 462 533 270 261 521, 500 677,000 123, 600 330, 500 398, 000 346, 500 20, 577 26, 038 18, 044 26, 833 40 36 28 31 46 40 5 14 5 11 5 14 1953/1952 3.09 .92 1.45 1.30 1.16 .97 1.30 2.67 .87 1.27 1.43 .88 1.11 .87 2.80 2.20 2.80 RELATIONSHIPS BETWEEN CATCH STATISTICS The population indices derived from the catch reports are the raw catch and the relative catch; i.e., catch per unit of effort. Either index may be in terms of weight or number of fish and may be calculated for the entire catch or for limited cate- gories. Since 1954, however, no information on the number or size of fish in the catch has been in- cluded in the catch report, so the only indices which may be considered for recent years in the fishery are the pounds caught and the catch per unit of effort without respect to size categories. The biweekly statistics for pounds caught, catch per unit of effort, and effort within the Oahu region for 1952 and 1953 are plotted in figure 11. Of particular interest here is the relation be- tween the raw catch and the catch per unit of effort, for if the two show essentially the same variation, there is little or no advantage to be gained in employing catch per unit of effort as tlie index of apparent abundance. It is obvious from figure 11 that there is much similarity in the fluctuations of all three indices; each has a seasonal variation on which lesser fluc- tuations are superimposed. Additionally, there is a secular trend from 1952 to 1953. The catch curve tends to change gradually and peak sharply, while the effort curve changes rapidly at the start and close of the season, with little trend during midyear. SKIPJACK IN HAWAII FISHERY 295 1 1 1 I 1 I I T' r 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I ! 1 1 1 1 1 1 1 1 I 1 1 1 1 1* Mil 1 1 I I I I 1 I 1300 • POUNDS CAUGHT {IN THOUSANDS) 1200 - -- - EFFORT (NO. OF FISHERMAN TRIPS) ii 11 1 - 1100 - — - CATCH PERUNITOFEFFO«T(POUNDS) 1 * ' 1 * '1 1 \ 1 \l II '1, "^ - 1000 900 t 800 700 / i 600 - / / »/■■ A » A / / 1 1 rA X \ \ \ \ - 500 400 _ 1 1 1 1 I A. \\ \ \ \ :/ V / f k t \ * \ V''-.- 300 - 1 1 /7 \i V\ ; r J !■■■:■ '■ \ ' / ^ ( r / - 200 / \ \ Y ...; f -■" ■. \ I-/ \; •■••J - 100 % 'i''~(- •!■•■ ! 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r"T 1 I'-'r 1 1 1 1 1 1 \ 1 1 I 1 1 1 1 1 1 1 1 ! 1 1 1 1 1 1 3 5 7 9 II 13 15 17 19 21 23 25 1 3 5 7 9 II 13 15 17 19 21 23 25 1 BIWEEKLY PERIODS - 1952 J L -BIWEEKLY PERIODS- 1953 - 1 F^GUKE 11. — Oahu region catch statistics, 1952 and 1953, biweekly periods. The results of these tendencies in catch and effort curves on the catch per imit-of-effort curve are as follows: (1) During midyear when effort tends to be constant, the catch per unit of effort will closely follow the fluctuations in catch. (2) During the onset and decline of the season, effort is changing more rapidly than catch and the catch per unit of effort will change at an intermediate rate. Random fluctuations in the catch per unit- of-effort at this time may be somewhat at variance with those in the catch, as occurred in the interval from period 24, 1952, to period 3, 1953. Correlation analysis between raw catch and catch per unit-of-effort data yields a coefficient of + 0.92. It was necessary to use rank correlation methods (Snedecor, 1956, p. 190) because the dis- tribution of catches is skewed toward small catches. In order to determine the amount of agreement between the random fluctuations in the two indices, the first differences were correlated and yielded a coefficient of +0.79. Conventional methods were used here because the first differ- ences are distributed more normally tiian the original series. The probability that correlation coeflScients this large would occur by chance is less than 0.01. The variate-difference technique (Kendall, vol. II, p. 387-390) was used to obtain an estimate of the variance in the random component of each index compared to the variance of the original series. The values obtained for the random com- ponents were 24 percent for the raw-catch series and 18 percent for the catch per unit-of-effort series. Since it exhibits a smaller random com- ponent, the catch per unit-of-effort series appears to be somewhat more reliable than the catch as an index of slapjack availability, but only slightly so. In summary, the raw catch is almost as accurate as the catch per unit of effort in indicating the seasonal variation in skipjack abundance in the Hawaiian fishery. During the middle of the year the raw catch is in good agreement with the ran- dom fluctuations in catch per unit of effort, but during the off-season of the fishery, when effort and catch are either declining rapidly or are at a low level, random fluctuations in catch per unit of effort may not vary in agreement with fluctuations in catch. For most puqjoses, total catch would appear to be as useful an indicator of availability as the catch per unit of effort, especially in prob- lems where the seasonal trend in the fisheiy is apparent. USE OF CATCH RECORDS TO DETERMINE POPULATION COMPOSITION The catch statistics of the Hawaiian fishery are the only continuous source of information which provides a means of assessing the nature of the skipjack population which supports the fishery. Inferences about this population must be made 296 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE with caution, however, because the fishei-y is limited geographically to the immediate island area (fig. 4). Furthermore, the catches are also influenced by availability, fishing effort, and selec- tivity on the part of the fishermen, which may not be constant throughout the year. Brock (1954, p. 100-103) ha? shown, by means of sex ratios, that the availability of female skipjack is not constant. He suggests that spawning activity may be the crucial element. The cyclical nature of fishing effort has been shown previously (fig. 11), and the fact that the fishermen are selective in the schools they fish is common knowledge. These biases appear to have an annual cycle and between-year comparisons may not be affected by them to the extent of within-year comparisons. Fluctuations In the Catch of Large Skipjack In the Oahu Region The growth rate of Hawaiian skipjack has been studied by Brock (1954, pp. 96-97) by means of length frequency distributions. During the sum- mer there are two distinct modes, one at about 45 cm. (4 pounds) and one at about 70 cm. (18 pounds) } The mode of large skipjack represents season-fish and presumably a smaller number of the previous year's season-fish. It is this mode that is considered here as large skipjack. The time of year when a mode, which Brock assimies to be a year class, passes through the weight ( 10 pounds) which separates small and large fish in this study, is apparently winter or early spring. During the period from May to October, it may be assumed with reasonable certainty that the small skipjack are a year younger than the large skipjack. In 1952 there appears to be little consistency in the average weight for large skipjack (fig. 12), which fluctuates widely from one biweekly period to the next. By contrast, 1953 has an interval from May 3 through October 3 (periods 10-20) with a regularly increasing weight for large fish. The rate of this increase, 0.25 pound per week, is in agreement with Brock's curve for skipjack growth, which yields a linear weight increase of 0.25 pound per week. The coincidence of the sea- son of gi-eatest productivity (as indicated by the > ConverslonH of length (millimeters) to weigh (pounds) were made accordilng to the formula: Log welght= — 8.2755-|-3.34913 loc total length. Figure 12. — Catch of large skipjack in the Oahu region, 1952 and 1953. number of large fish caught) with the interval of uniform weight increase suggests that in 1953 a single population of season skipjack was available to the fishery, but the erratic fluctuation in the numbers taken indicates that variations in this availability were quite marked. Period 5 of 1953, with its unusual numbers of very large fish, must consist of skipjack greater in size than the 1953-season fish. It seems probable that these very large skipjack are 1952-season fish, which were present only briefly that year. The relatively high average weights during other periods of early 1953 imply that 1952-season fish may have been generally present during that time. In early 1952, on the other hand, few of the previ- ous year's season fish were present, as judged by the average weights during periods 4 to 7. SKIPJACK IN HAWAII FISHERY 297 Size Composition Small skipjack are usually sought by the fisher- men only wlien they are unable to locate larger fish. One would expex't. therefore, that the occur- rence of small fish in the catch would be invei-sely related to the presence of the larger skipjack, and this does seem to be the situation. The numbers of large and small skipjack taken in the Oahu fishery in each biweekly period of the 2 years under stud,v is plotted in figure 13. In 28 biweekly periods (1952, 8-20; 1953, 8-22) when large skipjack were generally pi-esent, a tabulation was made to see how frequently the changes in the number of one size group were associated with similar or opposite changes in the other. Opposite trends, e.g., the number of large skipjack decreases from the pre- ceding period while the number of small skipjack increases, occur in 22 periods while similar trends occur in 6. The probability of obtaining such a distribution if the numbers of large and small fish in the catch fluctuate independently of one another is less than 0.01. Size composition appears to be important in tlie determination of the general level of catch (fig. 14) and the pounds caught per unit-of -effort (fig. 15). Both increase rapidly with an increasing proportion of large fish up to a ratio of 1 large fish for 1 small fish. Above this ratio the total catch continues to increase at a fairly rapid rate, but catch per unit of effort increases at a slower rate. The likelihood of catches of large numbers of indivduals seems to be loosely linked with the size composition. During the interval included in this study, the largest numbers of skipjack were taken either when small fish were especially numerous and very few large fish were available or when large skipjack were in a majority (fig. 16) . When small fish outnumber the large, but are less than 10 times as numerous, there seem to l>e factors that work against the capture of a large number of individuals. These factors, if they exist, are prob- ably related to the distribution of the various size groups in the population which supports the fish- ery. A hypothesis concerning the structure of this population is offered below. Conjecture In order to account for the variations in appar- ent abundance of particular size groups in the catches, it is necessarj' to hypothesize a skipjack population consisting of at least three and pos- sibly one additional element. In the approximate order of their importance to the success of the fishery in 1952-53. these are as follows: Group A : "Season fish," approximately 17-22 pounds in, weight, which Brock assumes to be in either their second or third year of life. This group is migratory. 1 — I — I — r J 1 1 I I L_l 1 I I I I J I I I I I I BIWEEKLY PERIODS FiQUKE lii. — Estimated numbers of large and small skipjack taken in the Oahu fishery, 1952-53, by biweekly periods. 298 FISHERY BULLETIN OF THE FISH AND "WILDLIFE SERVICE 1 1 1 1 I 1 1 1 1 • 13 •12 1 1 1 1 • 17 » 19S2 BIWEEKLY PERIODS • 1953 BIWEEKLY PERIODS • 18 • 16 ,--•'' ■''' lie -415 y 9» / '^'tlO /' ^22 '' "° 2y /^*'' ^23 ^•25 •" 19 ^26 •2" 7 1 1 1 1 1 AI7 •II 1 • 15 il2 2 4 6 2 3 7 II NUMBER OF LARGE / SMALL SKIPJACK IN THE CATCH 15 19 23 Figure 14. — Size composition and the total catch in the Oaliu skipjack fishery, 1952-53. Group B : Nonmigratoi-y small fish, which during the summer months are about 4 to 8 pounds in weight. These fish occur in semi-permanent ag- gregations wliich are to be found in certain locali- ties, usually near shore, where presumably oceano- graphic conditions are suitable for the concentra- tion of food organisms. These fish, according to Brock's hypothesis, are a year yomiger than the season fish. This group serves as the main source of supply for some of the fishermen, but in gen- eral it functions as a reserve supjily where most of the fishermen can use their bait, when larger fish are not available. Group C: Large migratory skipjack, 28-32 pounds, which may be a year older than the season fish. This group seems to have been abundant dur- ing the first part of 1953 and accoimts for the ap- parent early begining of the season in that year. Group D: Migratory small skipjack. The ex- istence of this group is not well established ; how- ever, the large numbers of small fish which appear suddenly in the fishery in periods 14, 15, and 20 of 1952 suggests that there may be, a migratory group of small fish as well as the semiresident group. In figure 16, catches of large skipjack number- ing in excess of 15,000 were all made when season fish (group A) were apparently dominant in the fishery; the large catches of small skipjack, those in excess of 40,000, are presumed to result from the presence of migratory small fish (groui^ D). The sharp decline in the number of periods with catches of small skipjack greater than 25,000 may indicate that this number is about the maximum number of nonmigratory fish (group B), which are available during a biweekly period. Except for period 17, 1953, the number of small fish in the catch declines as the number of large fish in- creases, which is consistent with the assumption that the number of small fish caught is inversely related to the availability of large fish. The extra- large migratory fish (group C) are distinguished by their greater average weight relative to the season fish, and not, at least during 195.2-53, by their unusually large numbers. At the time of SKIPJACK IN HAWAII FISHERY 299 900 1 1 1 ' 1 1 1 1 ' 1 1 ltS2 BIWEEKLY PERIODS 800 • 1|$| BIWEEKLY PERIODS »I7 700 " •17 •ly • 12 • 13 600 • 18 •" 118 ^,"' ,,-''"' 500 4<5 / •21*20 400 — /tig*!-* / *20 ,' • 10 •M 112 300 •24 200 12-422 A9 100 1 1 1 1 1 1 1 1 1 2 •■» -6 I 2 3 7 n 15 19 23 NUMBER Of LARGE / Small skipjack in THE CATCH Figure 15. — Size composition and catch per unit-of-efifort in the Oahu sljipjack fishery, 1952-53. 1 1 1 1 1 1 1 50 • 1(S2 PERIODS WITH catch] • 19SS >75.000 POUNDS J 40 •17 •12 *" 30 •'5 20 • 1' ~ •« - 10 ^* 1.3 1* " - °c •" «• ;/22 8 »"" 1 120 1.5 *"• 1 1 1 10 ' '" 20 30 40 50 60 70 e( NUMBER OF SMALL SKIPJACK IN CATCH (IN THOUSANDS) Figure 16. — Size composition and the number of sliipjack caught in tlie Oahu fishery, 1952-5;i. their appear; nee in tlio catch during periods 5 and 7-9, 19.53 ( fig. 1'2) , they were associated witli fairly large numbers of the nonmigi-atory small fish. The fact that the three migratory groups seem to occur at different times suggests tliat there is little overlap in their distribution, but the catch records do not show this witli certainty, for (see Sources of Error) the average weight of fish in the catch is a none too adequate index of size com- position. The actual distribution of small skip- jack is not defined by their appearance in the catch. It is possible that during the periods when the catch consists predommantly of large fish, small skipjack are also available, but have been rejected by the fishermen. In order to obtain in- formation as to the quantity of small fish actually present, a method such as maintenance of log- books in which the fishermen could record their observations of all fish sighted, whether fished or not, would be requii-ed. The catch records provide little means of deter- mining the relations between the four groups of skipjack which seem to make up the population exploited by the Hawaii fishei-y. It is quite probable that the small fish in both the migi-atory and semiresideut groups furnish recruits for the season fish. The season fish of a given year may be the large fish of the next year. The long-term recoveiT of tagged specimens would appear to offer the best means of ascertaining the relations between these different groups of fish. 300 FISHERY BULLETIN OF THE FISH AND "WILDLIFE SERVICE SUMMARY 1. The staff of the Honolulu Biological Labora- tory is trying to determine the environmental con- ditions which influence the local availability of skipjack. Commercial catch records are a source of information. 2. Methods and sources of error are considered. Fish catch reports for 1952 (a poor year) and 1953 (a good year) were summarized by areas of the fishery and biweekly fishing periods. The unit of fishing effort, the productive fisherman- trip, is discussed. 3. The distribution of catches and effort in the 2 years was generally similar, with leeward Oahu contributing one-half the catch. Hawaii, wind- ward Oahu, and Kauai fell well below leeward Oahu in productivity, while Maui and the oceanic region contributed insignificant proprotions. 4. Pounds cauglit per unit of effort increased from southeast to northwest in the fishery and from inshore to offshore, but these trends may re- sult from differences in the fishery rather than to distribution of fish. 5. There was a positive correlation between the average weight per skipjack caught in various zones of the fisliery and catch per unit of effort. 6. Catches (in pounds) daring the fishing periods of 1953 were, with few exceptions, larger than those made during the corresponding periods of 1952. 7. In comparison with the seasonal trend in the fishery, the effects of rough weather (as indicated by periods of small craft warnings), were un- important. 8. Large skipjack, from their appearance in the catches, seemed to have arrived first in lee- ward areas, and at the end of the season they last appeared in catches from leeward areas. In June and August of both years, concentrations of sea- son fish occurred in windward Oahu and leeward Oahu, respectively. 9. The numbers of small fish taken by the Oahu fishery in 1952 and 1953 were approximately equal, but almost three times as many large fish were caught in 1953. In the Oaliu region, there was almost one and one-half times the fishing ef- fort in 1953 in comparison with 1952, and a much larger proportion was directed toward catching large skipjack. 10. Catch, effort and catch per unit of effort indexes have similar seasonal variations. The positive correlation between catch and catch per unit-of -effort is so close that there is little to be gained in using the catch per unit-of -effort as an index of apparent abundance in the fishery. 11. During the middle of 1953, the average weight of large skipjack increased at 0.25 pound per week, the growth rate for Hawaii skipjack estimated by Brock. This suggests that fish of the same age were constantly available to the fishery during this period. 12. The number of small skipjack in the catch varied inversely with the number of large fish. 13. A hypothesis for the structure of the skip- jack population supporting the fishery is offered. The population has four groups: (1) season fish and (2) extra-large fish, both of which are migra- tory, and (3) a semiresident and (4) a migratory group of smaller skipjack. LITERATURE CITED Brock, V. E. 1954. Some aspects of the biology of the aku, Katsu- ivonus pelamis, in the Hawaiian Islands. Pacific Science 8(1) : p. 94-104. June, F. C. 1951. Preliminary fisheries survey of the Hawaiian- •Line Islnnds area. Part III. The live-bait skip- jack fishery of the Hawaiian Islands. U.S. Fish and Wildlife Service, Commercial Fisheries Review 13(2) : p. 1-18. Kendall, M. G. 1951. The advanced theory of statistics. Vol. II. Third edition. Charles Griffin & Co., Ltd., London, England. 521 p. RoTCE, W. F., and T. Otsu. 1955. Observation of skipjack schools in Hawaiian waters, 1953. U.S. Fish and Wildlife Service, Special Scientific Report — Fisheries No. 147, 31 p. Snedecor, G. W. 1956. Statistical methods, 5th edition. The Iowa State College Press, Ames, Iowa. 534 p. Yamashita, D. T. 1958. Analysis of catch statistics of the Hawaiian skipjack fishery. U.S. Fish and Wildlife Service, Fishery Bulletin 134, vol. 58, p. 253-278. Yuen, H. S. H. 1959. Variability of skipjack response to live bait. U.S. Fish and Wildlife Service, Fishery Bulletin 160, vol. 60, p. 87-106. U. S. GOVERNMENT PRINTING CFICE : 1962 O -596560 INITED STATES DEPARTMENT OF THE INTERIOR, Stewart L. Udmll, S^crttmry FISH AND WILDLIFE SERVICE, Clarence F. Pautzke, Commissioner Bureau op Commercial Fisheries, Donald L. McKernan, Director ABUNDANCE AND AGE OF KVICHAK RIVER RED SALMON SMOLTS B^ Okra E. Kekns, jr. ^^mch FISHERY BULLETIN 189 From Fishery Bulleriii of the Fish and Wildlife Seryice VOLUME 61 [Conlribution No. 84, College of Fisheries, University of Wushington] r PUBLISHED BY UNITED STATES FISH AND WILDLIFE SERVICE • WASHINGTON • 1V61 PRINTED BY UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON, D.C. For sale by Superintendent of Documents, U.S. Government Printing Office, Washington 25, D.C. Price 20 centa Library of Congress catalog card for the series, Fishery Bulletin of the Fish and Wildlife Service : U. S. Fish and Wildlife Service. Fishery bulletin, v. 1- Washington, U. S. Govt. Print. Off., 1881-19 V. in illus., maps (part fold.) 23-28 cm. Some vols, issued in the congressional series as Senate or House documents. Bulletins composing v. 47- also numbered 1- Title varies: v. 1-49, Bulletin. Vols. 1-49 issued by Bureau of Fisheries (called Fish Commission, V. 1-23) 1. Fisheries— U. S. 2. Fish-culture— U. S. I. Title. SH11.A25 639.206173 9-35239* Library of Congress [59r55bl] CONTENTS Page Introduction 301 Determining relative smolt abundance 302 Fyke-net design 302 Fyke-net site - 303 Fyke-netting procedure — 304 Fishing season _- — ^ 304 Daily fishing period — — 305 River conditions and fyke-net efficiency 305 Water velocity 305 Channel changes - 306 Water depth -._ _ _ 306 Debris 306 Determining smolt age 307 Sampling procedure 307 Length-frequency samples 307 Scale samples 308 Abundance of smolts — 308 Timing of migration 309 Hourly index catches 309 24-Hour fishing 310 Smolt catches at adjacent net sites 310 Smolt catches at distant net sites.— 311 Age composition of smolts - 311 Smolt sizes at adjacent net sites ._ 312 Smolt sizes at distant net sites _ 313 Verification of fresh-water age.. 313 Summary _ 314 Literature cited 315 Appendix _ _ 316 III ABSTRACT Standardized methods are described for use of an index fyke net to deter- mine annually the abundance of smolt mig-rations from the Kvichak River system, Bristol Bay, Alaska. Details of the fyke-net construction are pre- sented. The fyke-netting pi-ocedure and the fishing- season are discussed as well as some of the more important river conditions for an effective fyke-net site: water velocity and depth, channel changes, and debris. Annual indices of smolt abundance for the years 1955 through 1959 are presented. Com- parisons in the timing of migration between years are included. Results of tests to determine the variation between sizes of catch and age composition of fish taken in nets fished side by side and up and down river are described. Information is presented showing that large parent escapements produced large smolt migrations and small escapements produced small smolt migra- tions. A sampling program of fyke-net catches to determine age composition is described. Age composition has been based on length-frequency and scale studies. Age composition for each year is presented. Smolt age composition has been compared with fresh-water age composition of returning adults. IV ABUNDANCE AND AGE OF KVICHAK RIVER RED SALMON SMOLTS Bv Orra E. Kerns, Jr., Senior Fisheries Biologist University of Washington Two essentials to managing a fishery for red .salmon, Oncorhijnchus nerka (Walbaum), are knowledge of the abundance and of the ages of the smolts as they leave a river system on their way to the sea. When the abundance and ages of the smolts are related for a number of years to the number and ages of the returning adults, predictions can be made of the size of subse- quent adult returns. These data are particu- larly important in the Kvichak River, since it is the largest red salmon producing system in Alaska. Specific objectives of red salmon smolt studies in the Kvichak River system were (1) an index of the abundance, and (2) the age composition of the entire migration. Our method of assessing smolt abundance is based on the catch of a single fyke net. Gear and fishing effort expended are kept constant, but the fishing site is changed slightly to pro- vide uniform water depth and velocity. Com- bined daily fyke-net catches for the season pro- vide an index of total smolt migration (Burgner, 1958). An index of abundance is not as desirable as an enumeration of the total migration, which has been explained by Foerster (1929), Krogius and Krokhin (1948), and the International Pa- cific Salmon Fisheries Commission (1955), but a total enumeration in the Kvichak River sys- tem has not been practical. Suitable gear has not been developed to cope with the width, depth, and velocity of the river. Therefore, we have located and operated the fyke net in such a manner that we think it reasonable to Note. — The author is presently with the Fisheries Research Insti- tute. Collesre of Fisheries. University of WashinKton, Seattle 5, Washington. Approved for publication, April 5. 19G1. Fishery Bulletin 189. assume a con.stant (but unknown) ratio of the fyke-net catch to the total migration. Samples for the determination of age com- position of the smolt migrations were taken from the fyke-net catches. Age composition has been based on smolt length-frequency and scale studies. The Kvichak River drainage basin covers nearly 8,000 square miles. Included are two major lakes, Iliamna Lake, which is 80 miles long, 20 miles wide, and 1,115 square miles in area ; and Lake Clark, which is 50 miles long, 4 miles wide, and 143 square miles in area (fig. 1). Iliamna Lake is about 50 feet above sea level and Lake Clark is about 220 feet above sea level (U.S. Army Corps of Engineers, 1957). From 1947 through 1954 studies of age com- position and sex ratio of the Naknek-Kvichak commercial catch and spawning-ground escape- ments were conducted annually by staff mem- bers of the Fisheries Research Institute, Uni- versity of Washington. In the spring of 1955, at the request of Alaska Salmon Industry, Inc., systematic observations of red salmon runs in the Kvichak River system were initiated by the Institute under contract with the U.S. Bureau of Commercial Fisheries. Expanded investiga- tions since 1955 were designed to measure mor- talities at various points in the red salmon life history (Thompson, 1953). The Kvichak River program began under the general direction of Dr. W. F. Thompson ; project leader since 1956 has been H. D. Smith. The smolt enumeration program was supervised in 1955 by Dr. R. L. Burgner, in 1956 and 1957 by D. W. Linn, and in 1958 and 1959 by the author. Records and preliminary unpublished 301 302 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Figure 1. — Kvichak River system, Bristol Bay, Alaska. reports of the Kvichak smolt studies are on file at the Fisheries Research Institute. Credit for the development of most of the gear and methods used in this study is due Drs. W. F. Thompson, R. L. Burgner, and Ted S. Y. Koo, of the Fisheries Research Institute. The developmental work was conducted before 1951 at Mosquito Point, on the Nushagak River system, and was financed by Alaska Salmon Industry, Inc. Unpublished reports were written by Dr. R. L. Burgner and D. W. Linn for the Kvichak smolt studies of 1955 and 1956, and the author has used freely and without reference the in- formation from these reports. Dr. 0. A. Mathisen critically reviewed this paper and made helpful comments on the pre- sentation of the data. The manuscript was edited by Drs. W. F. Royce, Ted S. Y. Koo, R. L. Burgner, and J. P. Harville, and H. D. Smith and J. F. Roos. Appreciation is extended to all Fisheries Re- search Institute staff members, past and pres- ent, permanent and temporary, who were in- volved in collection of data. DETERMINING RELATIVE SMOLT ABUNDANCE Fyke-Net Design The rigid frame of the fyke net used in this study measured 4 feet by 4 feet. The body of the net tapered from the frame to a single rec- tangular funnel 2 inches by 10 inches at the throat (fig. 2). A second funnel of the same dimensions was located in a detachable cod-end section, which facilitated emptying the catch. The net had two wings, each 10 feet long and 4 feet deep, with appropriate cork and lead lines. The two wings were held open by the force of the river current and two connecting spacer lines allowed the net to fish a consistent 9-foot wide section of the river. The net was made of ABUNDANCE AND AGE OF KVICHAK RIVER RED SALMON SMOLTS cork line 1/2" mesh in throat and cod end S'dio. metal hoops 2 X 10^ mefal"^ rectangles 303 10 ft. -^K- 4 ft. -X^ 2 72 ft. ^ K-2 ft. ^fe- 2 ft. -Si Figure 2. — Fyke net used in Kvichak River for assessing red salmon smolt abundance, 1955 through 1959. knotted cotton webbing ; the wings and body of 1-inch mesh (stretched measure) and the re- mainder of the net of 1 o-inch mesh. The fyke net is shown anchored in place in the river in figure 3. Fyke-Net Site Smolts contributed by both Iliamna Lake and Lake Clark must descend the Kvichak River on their way to the sea. The upper 4 miles of river from Iliamna Lake to Kaskanak Flats offer the <.^^;._"^^^''^ "^^•^N^***-^-'^--^!;*!^!^ Figure 3. — Fyke net anchored in fishing position in the Kvichak River. ( I'hotii liy C. D. Becker. I 304 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE best fyke-netting sites. Here the river varies in width from about 400 feet to 1,000 feet, with depths to more than 15 feet and surface water velocities to at least 8 feet per second. The bottom consists mostly of gravel less than 4 inches in diameter. The riverbanks ai-e steep because of the erosion that occurs nearly every year during high water. The river downstream from the Kaskanak Flats is affected by tides, causing variable water velocities unfavorable for fyke netting. The index site (site B) used for smolt enu- meration was approximately 4 miles down- stream from the outlet of Iliamna Lake. Two other sites designated A and C were 1/2 n^ile and 2 miles downstream from the outlet (fig. 4). At the index site the river forms two chan- nels, 570 feet wide and 225 feet wide. The fyke net was set on a submerged, slightly sloping Scole in fee) ^—^ — I — t-H 1000 2000 Koskanok Flots - Figure 4. — Fyke-net sites A, B (index), and C, Kvichak River. (Map by D. W. Linn.) gravel bar near the centei- of the main channel. This gravel bar extends at the same depth for at least 100 yards above and below the index site. Fyke-Netting Procedure The fyke net was set before 2200 hours each night and tended from a 20-foot skiff by two or more men. To check the net, the cod end was first raised out of the water by one man. In this position the throat of the net was closed to the passage of fish. The cod end with the catch was detached by removing a single locking pin and an empty cod end immediately put in place and lowered into the water to continue fishing. The second man pulled the release cord or zipper on the cod end and spilled the catch into a large weighing basket that was immersed in a tub of water. This weighing basket was then removed from the tub and hung on a spring balance of 40-pound capacity, suspended from a weighing stand in the skiff. The weighing basket was allowed to drain about 10 seconds before the weight of the fish and the time of catch were recorded. During peak migration the fish weighed about 20 pounds at a net check. Immediately after being weighed, the fish were returned to the river. Elapsed time for a net check was less than 30 seconds. The number of fish in a 1-pound sample was usually counted four times an hour and the number was used for conversion of total weight of fish to total number. To avoid excessive mortality during periods of heavy migration, the net was checked as the fish accumulated. In an extreme instance in 1958 it was necessary to check the index fyke net nearly twice a minute. When the migra- tion was very light, the net was checked every hour. Fishing Season The experimental fishing season during this study began in the spring before any smolt migration takes place and continued until only a few fish were caught each day. The smolt migration started after ice breakup (fig. 5) and following a rapid rise in water temperatures in Iliamna Lake. ABUNDANCE AND AGE OF KVICHAK RIVER RED SALMON SMOLTS 305 Figure 5. — Ice floes in the Kvichak River. (Photo by C. D. Becker.) Daii> Fishing Period The smolt index was based on the total catch of the fyke net from 2200 to 0100 hours. This 3-hour period was selected on the basis of ex- perience in the Nushagak system, where most of the smolts leave the lake during the darkest hours. The same nightly concentration of mi- gration was found in the Kvichak River during 24-hour periods of fishing, as shown later. River Conditions and Fyke-Net EfiSciency Water velocity. — Desirable surface-water ve- locity for fyke netting in the Kvichak River is 3 feet per second or more. This velocity is pre- sumed necessary to prevent size selectivity, or larger fish evading the net. Net selectivity in- fluenced by water velocity in the Kvichak River during 1955 was illu.strated by two nets fished side by side at site A. The net in faster water, 3.1 feet per second, caught a greater poundage and larger-sized fish than the net in water of 2.4 feet per second velocity (fig. 6). A second test for size vselectivity caused by similar water velocity was also made at the index site in 1955. The length frequencies of smolts taken during this test, in which two nets were fished side by side, are shown in figure 7. Both nets, fished in a current velocity of approximately 3.5 feet per Site A, - 2.4 ft/sec. velocity, 107 fish Site Ao- 3. 1 ft/sec. velocity, 98 fish 80 90 100 FORK LENGTH no 120 N MILLIMETERS 130 Figure 6. — Length frequencies of smolts captured in high and low velocity waters near site A from 2150 to 2210 hours, June 2, 1955. (Frequencies are smoothed by moving averages of threes.) 306 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Index Site, 68 fish 80 90 100 no 120 FORK LENGTH IN MILLIMETERS 130 Figure 7. — Length frequencies of sniolts captured in waters of similar velocity (approximately 3.5 feet per second) near index site from 2200 to 2212 hours, June 4, 1955. (Frequencies are smoothed by moving aver- ages of thi'ees.) second, captured fish of comparable size. Water velocities at the index site were measured care- fully with a Gurley current meter at least once each year during this study and varied little (table 1). Table 1. — Mkldepth water velocities at the index fyke-net site on Kvichak River Year Date Velocity (ft. Aec.) 1956... 1957... 1958 1959 June 21. June 21 . . Mav 17 May 29 3.5 3.6 3.5 3.4 Channel changes. — Changes in river depth or width in the vicinity of the index fyke-net site could radically affect the migration path of smolts and thus influence fyke-net catches. However, no appreciable changes in the contour of the river bottom at the index site have been detected on maps prepared each year by the method of plane table mapping and sounding (fig. 8). Water depth. — Water depth for effective fyke netting in the Kvichak River is fixed at from 3.5 to 4 feet. At a greater depth some fish escape over the wings and center of the net, since the net must rest on the river bottom at all times to prevent fish passing beneath it. Changes in water depth of the Kvichak River have followed the same pattern each year of this study. Water depth is at a minimum in spring and reaches a maximum in late summer or early autumn. Increased water discharge results from melting snow and glaciers or from rainfall. The annual range and increase in water depth as measured at the Igiugig gaging station during the period of fyke netting are shown in table 2. Fluctuations in water depth have been proportionate between the gaging and fyke-netting sites each year. Changes in water depths necessitate shifting the fyke net periodi- cally to optimum depth during the smolt migration. Table 2. — Annual water-depth ranges and increases in the upper Kvichak River Year Observation dates Comparative range (in inches) Increase (in inches) 19.55., 19.56 1957 1958.. 1959 May 29 to June 30 May 18 to June 30. . . Mav 8 to June 30 May 4 to June 30. . . May 23 to June 30 . . 11 to 24 13 to 26 to26 7 to 37 to 14 13 13 26 30 14 Note. — All water depths are based relatively on the lowest record- ing (in 1957), which is assiRned the value of 0. The permanent water level Kage is about 1 mile downstream from the outlet of Iliamna Lake. Debris. — Net-clogging debris in a river can be a variable factor in operation of fyke nets. Efficiency of clogged nets was tested by Dr. Koo and the author in the Ugashik River sys- tem of Bristol Bay in 1956 (table 3). On even- numbered nights one net was used during the entire fishing period, and on odd-numbered nights a clean net was substituted midway in the fishing period at 2000 hours. Tests on nights when the net was not replaced indicated Left flivef Bonk Index Fyhe Net Sire 3' 6" ,g^ ""^^■^TT^...^^^^ max. 13' l" . _. n,_. ■rr: jyiO^. 8' \>; Mav31 June 2 4 624 1,281 13 384 160 1 l,6,i6 6S6 2,005 1 .-)40 4 221 1,200 390 I 38 3 206 3.096 1.722 86 140 199 1 63 492 1.417 8 5 2 720 357 15 909 23 155 7,726 4 . 34 1 3 2 3 1 58 25 12 1 2 37 68 1 6 8 3.460 G 10 8 49 7 18 19 5 16 322 10. . 12 14 895 70 1,492 16 389 37 2.463 12.3 5.083 26 6 1,838 9.6 5.249 27.4 3.079 16.1 1,003 5.2 150 0.8 277 1.4 19.142 Percent 100.00 1 Net change occurred at 2000 hours. that the fyke-net catch decreased gradually with clogging for a period of 5 hours, after which few fish were caught. On nights when the clogged net was replaced with a clean net, the greater efficiency showed in an increased catch. The most troublesome form of debris in the Kvichak River was a colonial diatom (Gom- phonema sp.) which drifted in ribbonlike streams from Iliamna Lake whenever a mod- erate to strong wind prevailed down the lake. Debris of terrestrial origin, grass and leaves, became an occasional nuisance late in the sea- son with the higher water levels. During periods of abundant debris it was necessary to change the nets frequently or to clean them while in fishing position. DETERMINING SMOLT AGE Samplinji Procedure From 1955 through 1957, representative 2- pound samples of smolts were taken from the fyke-net catches in approximate proportion to the size of the migration. On nights of heavy migration several samples were taken, and on nights of light migration one or no sample was taken. In 1958, 2-pound .samples, taken in i -_>- pound lots each 15 minutes, were collected dur- ing each hour, provided adequate numbers of fish were available. The sampling procedure in 1959 was similar to that of 1958 except that 1- pound samples rather than 2-pound samples were taken. The fish included in all samples were taken randomly from the weighing basket to pi'event size selection resulting from possible stratification. The samples of live fish were transferred to separate containers, which were 10-gallon milk cans or boxes placed along the riverbank in slow-moving water. The milk cans were partly screened on the sides and the boxes were made of i/K-inch saran screen. During the first 3 years of the investigations, .samples were processed each morning; during 1958 and 1959, within 5 hours of capture. The change was made in 1958 to prevent mortalities that occurred in the earlier years from holding the fish overnight. l-en^th-Frequenc\ Samples The two important age groups of smolts in the Kvichak River, those spending one or two winters in fi-esh water, can nearly always be separated by length alone. Therefore, all fish in each sample were measured and the length- frequency method of age determination was employed. 308 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE Fish in groups of 10 to 15 were anesthetized in urethane ' or chlorotone, measured from the tip of the snout to the fork of the tail, and returned to a container of fresh water to revive. Each length was tallied together with informa- tion identifying the sample. At a later date all length-frequency tabulations for 1 day were weighted according to the magnitude of the fyke-net catches during the daily 3-hour index period and combined in a composite season sample. Scale Samples Ages determined from length frequencies were verified from scale samples. Scale samples were taken from fish used in the length-fre- quency measurements, and the fish selected for scale samples were immediately preserved in 5- percent formalin. Several weeks after preser- vation, the fish were remeasured and the scale samples removed. Shrinkage of preserved speci- mens was adjusted by a shrinkage factor deter- mined by measuring individual fish before and after their preservation. Shrinkage from live length varied from 3 to 7 percent, depending on the length of the fish, or about 4 millimeters. Four to eight scales were taken from each fish from immediately above or below the lateral line and between the dorsal and adipose fins. These scales were mounted in a spread pattern on a 1- by 3-inch glass slide. Scales from eight fish were mounted on a slide, with the length of each fish recorded on the slide label. The scales were covered with a second glass slide and the two slides taped together. All scales were studied to determine number of annuli and amount of spring growth since formation of the last winter annulus. Spring growth does not appear on the scales of smolts that migrate soon after lake ice breakup, but it becomes apparent about midway through the migration season and the growth increases dur- ing the summer. ABUNDANCE OF SMOLTS The annual Kvichak River smolt index was based on the number of fish captured in a single fyke net fished each year of the study under similar fishing conditions and for the duration of the migration. This method was designed to detect fluctuations in the number of smolts from year to year. The number of smolts and cal- culated index values for the 5 years of the study are presented in table 4. The daily smolt catches each year are shown in appendi.x tables 1 through 5. Table 4. — hidices of smolt abundance in the Kvichak River Year Number of smolts Index value * 193.T 214,000 64,000 25,000 ^gi.'i.OOO 1,64.3,000 19,56 S S 19.57 lfl.)8 . . 19.59 . 1.3 100 85.9 1 The total number of smolts caught in 1958 has been arbitrarily assigned the base value of 100.0. Some adjustments of each year's total catch have been necessary. For the first 3 years of the study, some smolt catches from secondary net sites A and C were included to obtain the index value. The use of these sites was neces- sary because ice in the river prevented fishing at the index site early in the season and because the biologists who did the counting were in- volved with other duties late in the season. Dur- ing 1955 and 1957, the catches from net-sites A and C were included when the two nets were contributing less than 7 percent of the season's catch. During the period of the ice flow in 1956 a substantial migration was detected at site A ; and, consequently, an evaluation of these catches in terms of principal index-site catches was necessary. This evaluation was made on the basis of simultaneous fishing at the index site and site A for 4 days, from June 7 through June 10 (table 5). The ratio of catches of site A to those of the index site for the 4 days was Table 5. — Si'tnultancoits smolt catches at site A and at the index site, 1956 1 Urethane has not been used e.xtensively since reports of its carcinogenic elTects were published. Catch at— Date Site A Index site June 7 - . - .lune 8 .luiip 9 .liuif 10 874 16 8 9.449 4.127 1.851 115 Total ... 898 15,542 ABUNDANCE AND AGE OF KVICHAK RIVER RED SALMON SMOLTS 309 0.06:1.00, and this ratio was applied to the catches obtained at site A before June 7 to esti- mate the index catches for this early period. The estimated index catches for 1956 ai-e shown in appendix table 2. During each year's smolt migration .some hours and days of fishing were missed because of ice in the river or failure of the fyke-net anchors. Estimates of the number of fish pass- ing the net site during hours not fished have been calculated on the basis of the average catch of the preceding and following hours. Estimates for days missed have been made by averaging the catches of the preceding and fol- lowing days. Estimates for fishing hours missed have never exceeded 8 percent of the season's catch, and estimates for days missed have never exceeded 2.5 percent of the season's catch. The maximum estimate of hours missed was for the 1955 run and the maximum esti- mate of days missed was for the 1957 run. During the peak of the 1959 migration, the cod end was placed on the net for 5 minutes of each 15 minutes fished and occasionally 5 min- utes of each 30 minutes fished. The catches were multiplied by 3 or 6, as appropriate, for estimates of the 15- and 30-minute periods. This subsampling reduced handling of the fish and consequent injury to them. The estimates from the intermittent fishing in 1959 are con- sidered reliable because of the homogeneity of catches noted in 1958 during periods of con- tinuous net checking at the peak of migration. To test this homogeneity for 1958, all combi- nations of every third 5-minute catch (for the estimate of the 15-minute periods of fishing in 1958) were compared with the total of all con- secutive 5-minute catches (total of 98), and the maximum error was found to be about ±0.5 percent. When each combination of every sixth 5-minute catch (for the estimate of the 30 min- ute periods) was related to the total catch, the maximum error was less than ±4.0 percent. Timing of Migration The timing of the Kvichak River smolt mi- grations is depicted by the annual cumulative catch curves (fig. 9). Each spring the migra- tion started near the final day of ice flow in the Kvichak River (table 6) and continued through 10 20 JULY Figure 9. — Cumulative daily smolt catches at the index site in the Kvichak River, 1955 through 1959. June or early July. The 1958 smolt migration is the earliest on record and corresponds to the early breakup of lake ice for that year. Table 6. — Final day of ice flow in the Kvichak River Year Date l