BULLETIN OF THE FLORIDA STATE MUSEUM , BIOLOGICAL SCIENCES Volu.me 3 Number 3 A REVISION OF THE FISHES OF THE SUBFAMILY ARGENTININAE Daniel M. Cohen <4 lf' % '0 \.4,9.14~f...~ UNIVERSITY OF FLORIDA Gainesville 1958 The numbers of THE BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES, will be published at irregular intervals. Volumes will contain about 300 pages and will not necessarily be completed in any one calendar year. WILLIAM J. REA{ER, Editor ROLAND F. HUSSEY, Associate Editor All communications concerning purchase or exchange of the publication should be addressed to the Curator of Biological Sciences, Florida State Museum, Seagle Building, Gaihesville, Florida. Manuscripts should be sent to the Editor of the BULLETIN, Flint Hall, University of Florida, Gainesville, Florida. Published 2 July 1958 Price for this. issue $1.14 A REVISION OF THE FISHES OF THE SUBFAMILY ARGENTININAE DANIEL M. COHEN 1 SYNOPSIS: A systematic revision of the Argentininae divides the subfamily into two genera, Argentina with seven species and subspecies and Glossanodon with five species. Comments are given on habits and distribution, and· brief dis- cussions of the swimbladder and the spiral valve are included. The following new taxa are presented: Prosoarchus, new subgerius; Glossanodon (Glossanodon) polli, new species; Glossanodon (Prosoarchus) pygmaeus, new species; Argentina elon- gata australiae, new subspecies. , INTRODUCTION The subfamily Argentininae as herein understood consists of argen- tinid Bshes formerly referable to the g6nus Argentina. This subfamily is worldwide in distribution, and its members, as adults at -least, are rest*idteel to the edges and slopes of the continental shelf, in contra- distinction to the Other fishes of the suborder Argentinoidei which are bathypelagic. The Argentininae thus form a closely related group ecologically as well as morph6logically This paper contains comments on various aspects of argentinoid morphology pertinent to the classification of the Argentininae, keys to the taxa, generic diaghoses; species descriptions. and- synonymief, and the information available on the habits and distribution 'of the species included. The following new taxa are proposed in this paper: - Prosoarchus - new subgenus Glossanodon (Prosoarchus) pygmaeus - new · species. . Clossanodon (Glossanodon) polli- new species Argentina elongata australiae - new subspecies. Methods and Terminology All counts and measurements were made on the left side of the specimen whenever possible. Measurements were taken with a pair of 140-millimeter needlepoint dividers, measured on a Keuffel and Esser 300-millimeter rule, and recorded to the nearest tenth of a millimeter. i The author is Assistant Professor of Biological Sciences at- the University of Florida, is Associate Of the Florida State Museum, and serves as curator in ichthy- ology for the University of F16rida Collections. Manuscript submitted 20 May 1958.-ED. .94 BULLETIN FLORIDA STATE MUSEUM Vol. 8 Measurements over 200 mm. on the larger specimens of Argentina silus were taken with a pair of sharp-pointed 600-mm. dividers, meas- ured on a meter stick, and recorded to the nearest tenth of a millimeter. The latte'r measurements are probably not as accurate as the former. All calculations were made on a Monroe calculator. Unless otherwise noted the data for measurements are given as the arithmetic mean followed by the range in parentheses. The data from counts are given as the modal number followed by the range in parentheses. In the measurements listed below the most anterior point is taken as the tip of the upper jaw. Standard length is measured to the crease at the hypural fan when the tail is bent. Preanal length is taken to a vertical through the anterior end of the anal fin base. Pre- ventral length is taken to a vertical through the anterior end of the ventral fin base. Predorsal length is taken to a verticdil through the anterior tip of the dorsal fin base. Prepectoral length is taken to a vertical through the anterior end of the pectoral fin base, Head length is taken to a Vertical through the posterior tip of the opercle. Max- illary length is taken to a vertical through the posterior tip of the maxillary. Snout is taken to a vertical through the anterior margin of the eye. Eye is the horizontal diameter of the comea (not to be confused with the orbital diameter; in Argentina and' Clossanodon the anterior cir- cumorbitals extend far forward, and the area between the anterior edge of the eye and the posterior edge of the circumorbitals is filled with the hyaline tissue of the adipose eyelid). Head depth is taken at the greatest depth of the· head, immediately anteriar to the nape. Depth of caudal peduncle is taken at the point of least depth. Depth at dorsal fin is taken at the anterior end of the dorsal fin base; when the dorsgl fin base is slightly elevated above the dorsum, the measure- ment is tiken anterior to. the rise. Interorbital is the shortest dis- tance across the top of the skull between the lateral edges of the bony orbit. Maxillary width is taken at the greatest width of the maxillary, close to the posterior end. Vertebral counts do not include the urostyle. In dorsal and anal fin ray counts the double ray at the end of each fin (the last ray is closely appressed to its anterior neighbor) is counted as two, Other terms used are self-explanatory Or defined when used. The osteological terminology follows that of Chapman (1942). The following abbreviations are- used throughout the text in re- ferring to fish collections: BM Universitet I Bergen,» Zoologisk.Museum CKM Collection of Dr. Kiyomatsu Matsubara 1958 COHEN: SUBFAMILY ARGENTININAE 95 CNHM Chicago Natural History Museum DMW Doniinion Museum, Wellington, New Zealahd MCZ Museum 6f Comparative Zoology, Harvard College NFRS, Newfoundland Fisheries Research Station RCB Mus6e Royal du Congo Belge ROMZ Royal Ontario Museum of Zoology SIO Scripps Institutibn of Oceanography SU Natural History Museum, Stanford University TU University of Tokyo. UCI,A University of California at Los Angeles LF University of Florida Collectidns UM University of Miami USNM United States National Museum Acknowledgments The following individuals were mostgenerous in lending or ~ in giving me specimens: Tokiharu Abe, Tokyo University; Giles W. Mead, Stewart Springer, and , Robert Wilson of the U.S. Fish and Wildlife Service; Hans Brattstrom and Johan Willgohs of the Univer- sitet I Bergen; John C. Briggs of the University of Florida; Carlos Bas and F. Garcia del Cid of the University of Barcelona; Pierre Desbrosses of the Institut Scientifique et Technique des P@ches Maritimes 'in' Paris; J. A. F. Garrick of Victoria University College, Wellington, New Zealand; Carl L. Hubbs and Boyd Walker of the University of Cali- fornia; Jozica Karlovac of the Institut za Oceandgrafiju u Splitu in Y.u- goslavia;' Kiyomatsu Matsubara of Kyoto University; John Moreland of the Dominion Museum, New Zealand; Ian Munro and Dudley Kurth of C.S.I.R.O. Division of ,Fisheries, Australia; Max Poll of the Mus6e Royal du Congo Belge at Tervuren, Belgium; Alfredo Ramalho, Esta~ao de Biologia Maritima, Lisbon; C. Richard Robins of the Uni- versity of Miami; William C. Schroeder and (Mrs.) Myvanwy M. Dick of the Museum of Comparative Zoology at Harvard; W. B.. Scott -of the Royal Ontario Museum of Zoology; Leonard P..Schultz of the U.S. National Museum; H. J. Squires of the Fisheries Rehearch Board of Canada; A. Vedel Taning of the Marinbiologisk Laboratorium, in Del- mark; Enrico Tortonese of the Genoa Museum; and Loren P. Woods and (Mrs.) Marion Grey of·the Chicago Natural History Museum: Each of the followinghas sent me unpublished information on fishes of t~is group: Elbert H. Ahlstrom of the U.S. Fish and Wildlife Service, Hans Tambs-Lyche of the Universitet I Bergen, Carl L. Hubbs, H. J. Squires, and Robert Wilson. James Ballike of the Academy of N~tural Sciences of Philadelphia; Anne C. Cohen of Gainesville, Florida; Arni Fridrikss6n of the Conseil 96 BULLETIN FLORIDA STATE MUSEUM Vol. 8 International pour L'Exploration de la Mer; Lisa Lanz of Palo Alto, , California; and Norman J. Wilimovsky 6f the U.S, Fish and Wildlife Service have helped in various ways. Warren Freihofer of Stanford University generously made some histological preparations. I am , particularly indebted to George S. Myers and Margaret H. Storey of Stanford for their help during this study, most of which was carried on at Stanford. Figures 3,4,5, and 9 were preparedby Margaret G.' Bradbury. Figure 7 is reproduced through the courtesy of the Dominion Museum, Wellington, New Zealand. THE 'SPIRAL VALVE PROBLEM Kendall and Crawford (1922) and Chapman (1942) state that Argentina sphyraena and A. silus posse5s an apparently functional spiral valve. Chapman uses the presence of this structure as a charac- ter in defining the suborder Opisthoproctoidei (= Argentinoidei). I have investigated the so-called "spiral valve" in as many species as possible. The following description is based on specimens of Argentina sphyraena, of which I had abundant material for dissection. On other species I made briefer observations. As these showed considerable variation between species, the findings are included in the systematic section under the species descriptions. For slightly less than half its length the intestine is a thin-walled, almost· transparent tube. The inner wall is slightly elevated into a large number of zigzag fieshy ridges (the zickzack Grundnetze of Jacobshagen, 1937) giving the inner surface a labyrinthine appear- ance. These ridges are closer together and more abundant along the dorsal third of the gut and cause the wall to appear considerably thicker in this area. A second area of the intestine, slightly shorter than the Brst, is characterized by the presence of rings of tissue (Ringfalten) which project out into the gut. The Brst four or five of these rihgs are truly spiral in nature, while the remainder are simple circles. The struc- tures are easily visible through the walls of the empty gut. In no instance was the center of the gut observed to be occluded, so that although the inner surfac6 of the gut is greatly increased, the distance that food must travel is not increased. It seems highly probable, how- ever, that these rings of tissue hanging into the cavity 6f the gut slow the passage of the food considerably. A specimen from Portugal 165 mm. in standard length has 15' rings, the first five of which are each connected with the next by a septum running aiong the roof of the gut. 1958 COHEN: SUBFAMILY ARGENTININAE 97 A similar connection is present between rings 10 and 11. The posterior ring is the most strongly developed. It blocks off a greater section of gut than do any of the more anterior ones. In pushing material along the gut with a probe to clear it, the absence of spiral motion is notable, as is the valvelike action of the posterior fold in slowing the passage of the· gut contents. The rings are about 1.5 mm. apart, and the in- tervening areas are filled with the labyrinthine ridges mentioned above. In some instances the ridges run out onto the rings. In smaller specimens the rings are considerably less developed, and the intestine is reJatively longer compared to the body length than - in larger ·fishes. For example, in a fish with standard length of 165 mm. the length of the intestine from the junction of the posterior- most pyloric caecum to the anus was 40 percent of the standard length; in a specimen 128 mm. in standard length the intestine was 47.5 per- cent of the standard length; in an example 115 mm. in standard length the intestine was 49.6 percent of the standard length. Although more data should be collected to test the hypothesis, it may- well be that in A. sphyraena the increase in internal surface area in .the intestine caused by strong development of Ringfalten is an adaptation to counter the relative,decrease in linear dimension caused by allometric growth. Posterior to the Ringfalten area and for a distance equal to less - than half the length of the Ridgfalten area, the intestine is thin-walled ,and transparent and carries a reduced humber of labyrinthine ,ridges On the basis of histological investigations Jacobshagen interprets the Ringfalten as being homologous with the Grundnetze, and he distinguishes both structures from the Spiralfalten or true spiral valve, although he states that the Ringfalten can appear valvelike. The true spiral valve he restricts to the Chondrichthys, sturgeons, gars, and Amia, while he lists the Ringfalten as occurring in Salmo, Coregonus, Thymallus, Clupea, Alosa and perhaps others. Chirocentrus, which some ichthyologists have mentioned as possessing a spiral· valve, dif- fers in having its rings extend all the way to the anus; Of the above genera it resembles Coregonus more than any others in having its rings in a spiral and very close together. Of the species in which I examined the gut (listed below), the rings of A. sphyraena are perhaps most similar to those of Salmo gairdneri, differing only in having a greater number of more strongly ,developed rings: I have found dis- tinct Ringfalten in the intestines of the following species; Salino gaird- neri (148.5 mm. SL, SU 49265, San Mateo Co., California), Thymallus arcticus signifer (156 mm. SL, SU 483-18, Lake Alexander Netting Sta- tion, Susitna Basin, Alaska), Coregonus sardinella (172 mm. SL, SU , 49119, Barrow, Alaska) and Chirocentrus dorab (145.5 mm. SL, SU 98 BULLETIN FLORIDA STATE MUSEUM Vol. 3 80689, Singapore). The Ringfalten of Coregonus are very similar to those of Coregonus maraena as figured by Jacobshagen (1987, fig. 528), and the Ringfalten of Chi(ocentrus are identical with the figure of C. dorab given by Goodrich (1909, fig. 77). I have examined tire follow- ing species of osmerid fishes and found the Ringfalttn to be either lacking or barely discernable: Spirinchus starksi (118.5 mm. SL, fish market, Monterey, California) and Mallotus villosus (132.5 mm. SL, Barrow, Alaska). If Jacobshagen's ideas are correct (and I might add that I consider - certain aspects of his theories Open to question) there is an·unfortunate confusion in terminology., As noted above, he has restricted the use of the Spiralfalten to the structure present in Chondrichthys and cer- tain archaic fishes. Yet several of the isospondylous fishes, which ac- cording·to Jaeobshagen's interpretation have only Ringfalten, have the Ringen partially, as in Argentina sphyraena, Or wholly, as in Chiro- centrus dorab, formed in a spiral. - This has given rise to the term «vestigial spiral valve" in the British and American ichthyoJOgical literature. This term implies homologies with the Spiralfalten and - should be used with great caution. A further complication is the situation which Kendall and Crawford (1922) rep6rt in Argentina silus in which a functional "spiral valve," presumably ddrived from _ the Ringfalten, has developed, but which I have been unable to verify. (See s6ction on A. silus.)' - Further consideration of this problem is beyond the scope of the present work. The reader is referred to the works of Jacobshagen (1987) for additional references and a summary of his theories on the homologies of the intestinal structures. It seems safe to conclude that .the presence of a spiral valve cannot be used in defining th5 sub- family Argentininae nor.indeed any other argentinoid group. THE SWIMBLADDER The Argentininae differ from other Isospondyli except the Stomia- toidei (Jones and Marshall, 1958) in possessink a swimbladder that lacks a connection with the gut. These fishes agree, however·, with the Salmonidae and Cyprinidae (Jones and Marshall, 1958) and Spir- inchus starksi, an osmerid, in lacking a rete. I have made,observations on as many species as possible and find general agreement among the species of Argentina and Glossanodon. although specimens of Glossan- odon I dissected were usually in such a poor state of internal preserva- tion that it was dimcult to make out many details. The following 1958 COHEN: SUBFAMILY ARGENTININAE 99 description is based on specimens of A. sphyraena. Variations are noted in the systematic section under each species. AV, AG K/ / \ 1 / 1 1 li' / 1 pc \\1 HPV 1 DV ' IN PV ST Figure 1.-Diagram of Jeft side of body cavity of Argentina sphyraena. Mesenteries and peritoneal membranes are not shown. AC, anterior chamber of swimbladder; AV, anterior vein from swimbladdur to kidney; DV, veiri draining distal end of swimbladder; HPV, hepatic portal vein, IN, intestine; KI, kidney; PC, posterior chamber of .swimbladder; PV, poste*or vein from swimbladder to kidney; ST, stomach. The swimbladder (see fig. 1) is a two-chambered structure lying d6rsal to and separate from the gut cavity. The anterior chamber is an elongate, fusiform structure with its bluntly rounded anterior end situated above the posterior end- of the stomach and its posterior end at about the level of a vertical from the posterior tip of the ventral fins. The thiek-walled outer layer of the anterior chamber is heavily impregnated with a brilliant silvery pigment which has drawn com- ment from ichthyologists since the days of Willughby. The internal la-yer is a thin-walled, transparent membrane lying closely appressed to the inner walls of the outer layer. The posterior chamber, which has heretofore gone unnoticed in this genus, is a small, thin-walled diverticulum which lacks silvery pigment. It originates at the posterior end of the anterior chamber and extends posteriorly for a distance equal to about one-fifteenth the length of the anterior chamber. At this point the posterior cham- ber curves sharply downwards and then runs anteriorly for at least twice the distance of its own posterior extension, from the anterior chamber. The forward extension becomes much smaller at its. distal end and finally terminates as a narrow blind sac. The wall of the distal end of the bladder (see Bg. 2) continues anteriorly carrying a vein which drains the disfal end of the posterior I00 BULLETIN FLORIDA STATE MUSEUM Vol. 3 chamber. Slightly distal to its ori- gin the vein descends through the - ,-p-'.. --/ i peritoneum and runs forward in the dorsal mesenteries parallel to the DV I PC hepatic vein before joining that Figure 2.-Diagram of distal vein. A second vein runs from theend of posterior chamber Of swim- bladder of Argentina situs showing most posterior part of th6 posterior the origin of the vein that drains chamber and, as far as I have been this area. able to discern, enters the renal portal system. In five A: sphyraena I dissected, the vein entered the left kidney, while in one specimen it entered the right kidney. At the anterior end of the anterior cham- ber a pair of veins run from the sides of the bladder to the kidneys. Lack of material properly preserved for investigating the soft anatomy prevented reliable observations on the arterial supply to the swim- bladder; however, many blood vessels, usually invested with consid- erable amounts of fat, course over the surface of the swimbladder and send branches into the' interior. An Observation of some interest is that the striking silvery pigment which colors the swimbladder of Argentina sphyraena, A. silus, and A. striata is absent from the swimbladders of the, other species of the genus. Furthermore, I have not found it in any of the species of Glossanodon whose swimbladders I was able to examine. As far as I have been able to discern, the reason for the lack of pigment in some species is not due to age,- size, or time of year when the specimens Were caught, or to the method of preservation. I have examined speci- mens of Argentina elongata australiae that are a bright silver along the sides of the body and the head, areas which would be leached out at once by a preservative, while the swimbladder had only a faint irides- cent sheen Over its surface and no thick deposition of pigment at all. Neither does the character seem to be geographical, for specimens of Argentina sphyraena from the Adriatic were found to be brightly pig- mented with silver, while the swimbladder of an Adriatic specimen of Glossanodon Zeioglossus lacked silvery pigment entirely. The above reasons lead me to assume this character is genetically controlled, and I have used it taxonomically, although alway5 in company with other characters. SUBFAMILY ARGENTININAE GILL Argentininae Gill, 1868: 14 (in part). 1898b : 181 (name only). Argentinidae Gill, 1898a : 112 (in part). Goode and Bean, 1895 : 51. Regan, 1918 :.289 (in part), Jordan, 1928 : 125. Berg, 1940 : 428. Chapman, 1942 : 116. 1958 COHEN: SUBFAMILY ARGENTININAE 101 Type genus Argentina Linnaeus, 1758., These are argentinid fishes that may be distinguished from 6ther Bshes in the Argentinidae chiefly in having supratemporals lateral te and in contact with the parietals and fused to the pterotics, in having a mesocoracoid, and in having the pectorals inserted on the ventro- lateral contours of the body. The fishes of this subfamily vary in body shape from almost sub- cylindrical to rather strongly laterally cdmpressed.. Many of. the spe- cies, perhaps all, have the body molded by the scales into an elongate, angular box. A cross-section shows a polygon of from 6 to 10 sides, depending on what jpart of the body the, section was taken through and perhaps which species as well. From its greatest width immediately behind the head, the body tapers to the caudal fin. The eyes are nontubular; adipose eyelids are present. The eyes are on the sides of the head: which slope inward vent~ally and thereby afford the fish a ventral as well as a lateral field of vision. " The lachrymal is n6t in contact with the eye, but is set forward a distance approximately equal to the width of the pupil. The space' between the anterior edge of the eye and the posterior edge of the lachrymal is occupied by hyaline tissue which projects posteriorly as the adipose eyelid. A crescent of white tissue of unknown function lies above the dorsal or p6sterodorsal segment of the iris. Some workers feel that this structure may be luminescent. I have examined the tissue under ultfavidlet radiation with.negative results. The palatines, head of vomer, ceratobranchials of 5th gill arch and suprabranchials of 4th gill arch all bear teeth. Lingual,and dentary teeth may be present or absent. The snout is longer than half the diameter of the eye. The nares are placed at the level of the upper third of the eye and are usually slightly closer to the eye than to the snout. A small flap is present on the posterior nostril. The branchiostegals are 4 to 6, and the gill membranes are sepa- rate. The ventral arm of the preopercle and the interopercle extend forward to the level of the anterior edge of the eye. The lateral line originates slightly below the dorsal angle of the opercle and runs straight back to the base of the caudal fin. The scales are large; one taken from between the lateral line and the ventral fin origin is larger than the pupil of the eye. A scalesheath covers the proximal portion of the folded ventral fins. The dorsal Sh origin is closer to the ventral Bn origin than to the pectoral fin origin. The distance from the origin of the ventrals to the 6rigin of the anal - 102 BULLETIN FLORIDA STATE MUSEUM Vol. 8 is equal to or greater than the distance from the origin of the ventrals to the origin of the pectoralk. A dorsal adipose fin occurs over the anal fin base. In many of the species the larvae and postlarvae have a row of dark, transverse bands on the sides. In subadults and in adults of some species these are reduced to dusky blotches along the dorsal half of the body, while in adults of other species the bands are entirely absent. Adults and adolescents have a broad silvery stripe aIong the side and a considerable amount of silvery pigmentation on the opercles and other parts of the head. Ontogenetic changes in color pattern may perhaps be correlated with changes in habitat for, so far as known, the y-oung are pelagic while adults live close to the bottom, generally on the continental shelf or slopes. Distribution ih worldwide along the margins of continents, though no species have been recorded as yet from the west coasts of South America, from the southern coasts- of Asia,. or from Antarctica. Records of seven fossil species referred to Argentina appear in the literature, all described from otoliths. Sites at which fossils have been discovered are the lower Tertiary of southern England (Frost, 1988), early Tertiary of southern Rumania, middle Oligocene of Germany, Miocene of Czechoslovakia, Miocene of Austria and Pli6cene of Italy (Weiler, 1950). Although the general shape of the otoliths suggests that Argentina-like Bshes were Dresent at the times and places men- tioned above, considerable care should be exercised in differentiating species and genera of this group on the basis of otoliths alone. In examining series of otoliths from Argentina sphyraena I noted varifi- tions of a magnitude equal to some of those used in describing fossil species as new. Glossanodon may be an older. group than Argentina, as the former has a more complex maxillary and premaxillary and the general tendency in the argentinoids is towards a reduction of these elements. Furthermore, Glossanodon has evolved to the,point where two distinct subgeneric types are recognizable, while the species of Argentina have not diverged subgenerically. In conflict with the primacy of Glossano- don is the fadt that Argentina has reached the west coast of North America and the coasts of Australia and New Zealand, both areas from whence Glossanodon is unrecorded. Key to the Genera of Argentininae. la. Vomerine and palatine teeth in a continuous band of closely spaced, posteriorly directed teeth; vomerine teeth located less than 1958 COHEN: SUBFAMILY ARGENTININAE_ 103 i P i b a Figure 8.-(a) Head of vomer dissected from Glossanodon semifasciatus. (b) Head of vomer dissected ffom Argentina elongata elongata. 9 6 I'a ' Figure 4.-(a) Tongue dissected from Glossanodon semifasciatus. (b) Tohgue dissected from Argentina sphyraena. 104 BULLETIN FLORIDA STATE MUSEUM Vol. 8 the length of a tooth behind the anterior edge of the head of the vomer; tongue teeth located on· the front and lateral edges of the tongue; medial ends of the maxillaries sefarated by at least one- half the width of the broad, underlying mesethmoid (figs. Sb, 4b, 5b).............. Argentina p. 104 lb. Vomerine and palatine teeth not in continuous band of closely spaced, posteriorly directed teeth; vomerine teeth located three to eight tooth-lengths·behind the anterior edge of the head of the vomer; tongue teeth when present located at the front of the tongue, not along the lalteral edges; medial ends of the maxillaries touching in the midline or separated by less than one-fourth the width of the narrow, underlying mesethmoid (figs. Ba, 48, Sa) ........ Glossanodon p. 148 ./ 6 Figure 5.-(a) Dorsal view of alizarin preparation of snout of Glossanodon semifasciatus. (b) Dorsal view of alizarin preparation of snout of ArgUntina striata. GENUS ARGENTINA LINNAEUS Argentina Linnaeus, 1758 : 315 (type species Argentina sphyraena Linnaeus, 1758, by monotypy) Situs Reinhardt, 1883:11 (type species Salmo situs Ascanius, 1775, by absolute tautonymy) Goniosoma C6sta, 1844, in Costa 1886-44 (type species Argentina sphyraena Lin- naeus, 1758, by monotypy). Acantholepis Kr0yer, 1846 : 97 in Kr0yer, 1846-49 (type species Salmo silus As- canius, 1775, by monotypy).. DIAGNOSIS.-Argentinine fshes that have their maxillaries separated medially by a distance at least one half the width of the broad, under- lying mesethmoid. The premaxillary is simple and does not articulate with the medial portion of each maxillary, which is not enlarged. Dis- tal to its median end, each maxillary sends down a single prong (not 1958 COHEN: SUBFAMILY ARGENTININAE 105 shown in figure) which artieulates with the posterior face of each pre- maxillary The premaxillaries do not send up prongs to articulate with the maxillaries. The palatine and vomerine dentition is in the form of a rather reg- ular, continuous arch of small, closely spaced, posteriorly directed teeth.'The palatine and vomerine teeth are difficult to distinguish separately without dissection of a cleared and stained specimen. The vomerine teeth are located less than the length of a tooth behind the head of the vomer. The tongud teeth lie along the lateral edges of the b6ne as w-611 as at· the anterior margin. ..» Figure 6.-De Sphyraenae parva sive Sphyraena, secunda specie (From Rondelet, 1554) HIsToR¥.-What is probably the lirst record of Argentina was pub- lished in 1554 when Rondelet described and figured De Sphyraena pal'ua sive Sphyraena, secunda specie, or the "hautin" from the Medi- terranean. The flat dorsum and the shape 9~ the head in the Bgure (see fig. 6, a reproduction of Rondelet's figure) combined with the general description leave little doubt as to the identity of Rondelet's specimen. Unfortunately, as Cuvier (1815) points out in a review of the early history of the genus, Rondelet's description is marred by sev- eral imperfections. First of all, an adipose fin is neither indicated in the figure nor mentioned in the brief description; second, the anal fin is too long; and third, the lines which represent the posterior edge of the pre6percle have been continued below the ventral border of the head, and they appear. to represent a structure hanging from the opercle. These several inaccuracies continued to plague authors for many years to come. Whether or not Belon mentioned Argentina I have been unable to verify. Artedi's (1738) synonymy of the genus lists an "hautin"' of Belon, but Cuvier states Belon did not mention it. Gesner (1558: 1061) follows Rondelet in his description of the fish and he also reproduced Rondelet's Bgure. Willughby (1686: 229) gives a more thorough de- scription than does Rondelet. He heads his section on the lish as "Pisciculus Romae Argentina dictus, and he further states it ii the Sphyraena secunda specie of Rondplet and Gesner. He gives fin counts , 106 BULLETIN FLORIDA STATE MUSEUM Vol. 8 and mentions details of internal anatomy. He notes the absence of teeth in the jaws and their presence on the tongue, thus rnaking it fairly certain that he examined a specimen of Argentina sphyraena. Willughby is also the first to comment on the use of the silvery pigment of Argentina as a coloring agent in the manufacture of artificial pearls. He also copied Rondelet's Egure. Cuvier states that Ray (1710) copied Willughby, abridging his de- scription slightly. The next mention of Argentina is by Peter Artedi in 17,88. In part 8, page 8, he gives a brief description of Argentina, which he recognizes as a separate genus in his order Malacopterygii. In part 5, page 17 he gives a synonymy of the form. Linnaeus first lists Argentina in the fourth edition of the "Systema Naturae" (1744). Cuvier states that Linnaeus followed Artedi. In- deed, it seems doubtful that Linnaeus ever had the opportunity of ex- amining a specimen of Argentina. As Ctinther (1899) points out, most of Linnaeus' fish specimens came from three major sources. The first of these was Scandinavia. Though Argentina is indeed found in Scan- dinavian waters, Linnaeus does not record the genus as occurring there. The second source was a collector who sent German freshwater Bshes to Linnaeus. These collections could certainly not have included Argentina, a strictly marine form. The third ef Linnaeus' major sources of fish was Alexander Garden, of Charleston, South Carolina. Although Argentina has been taken off South Carolina, this area could not have produced the Argentina referred to in the 1744 edition of "Systema Naturae," for Klauber (1948) informs us that Dr. Garden did not send collections to the great Swedish naturalist until the early 1760's. Gunther makes no mention of Argentina in his 1899 catalog of Linnean specimens in the collections of the Linnein Society in Lon: don. L6nnberg, however, in listing the Linnean types at Upsala men- tions (1896 : 25) that the type specimen of a species of Argentina has been lost. Lonnberg goes oil to state that the original description of the species was given in the "Museum Adolpho-Fridericianum" (1746) by Linnaeus, was later published in "Amoenitates Academicae" (1749) entitled "Museum Principis." After examining the description "in Museum Principis" there seems little doubt that Cuvier was correct in identifying this nominal species of Argentina as a pike. Cuvier goes on to state that in the sixth editi6n of "Systema Naturae" (1748) Linnaeus incorporates in the genus Argentina the pike, with 10 branchiostegal rays, and A. sphyraena which, according to Linnaeus, has 6 branchi6stegal rays. Gronovius (1754, vol. 1 : 6) lists an Argentina. Although he cites Arte'di, it seems doubtful that the specimen Gronovius examined at 1958 COHEN: SUBFAMILY ARGENTININAE 107 this time was congeneric with that of Artedi as Gronovius mentions the presence of teeth on the jaws, palate, and tongue, while Artedi finds them present on only the palate and tongue. In volume 2 of the same work (1756) -Gronovius attributes to Argentina a species from Surinam, which Cuvier considers an anchovy. The ninth editi6n of "Systema Naturae" (1756), Cuvier tells us, also contains two species of Argentina. One of these is A. *hyraena, which Linnaeus states possesses 6 branchiostegal rays, while the other is the pike of the sixth edition which Linnaeus lists as having 10 branchiostegal rays. As a generic character, however, Linnaeus gives 8 branchiostegal rays. He also states that presence of teeth in both jaws is a generic character, although they are absent in A. sphyraena. Finally, in the tenth edition 6f "Systema Naturae" (1758), Linnaeus removes the pike and leaves Argentina sphyraena standing as the sole representative of the genus. He does this, however, without emending, his definition. of the genus, so that among the generic characteristics are teeth in the jaws and 8 branchi-ostegal rays. Neither of these fea- tures, of course, is characteristic of A. sphyraena. In his brief synony- my and description of the species Linnaeus cites Artedi, whose de- seription was of the true.A. sphyraena, and Gronovius, whose descrip- tion is discussed above. Linnaeus als6 quotes Ray concerning the swimbladder and the use of the silvery pigments as coloring agents for aftifcial pearls. Thus considerable confusion attended the pre-Linnaean identity of Argentina and c6ntinued in the works of certain post-Linnaean au- thors as well, largely because of the authority carried by the writings of Gronovius. Indeed, fishes belonging to groups such as Elops, Albula, and Maurolicus as well as anchovies and pikes continued to be ' ascribed ta the genus for some time. Key to the Species of A,gentina la. Branchiostegals 6, scales with spines. 2a. Lateral line scales 52 (50-54); gill rakers 9n lower arm of first arch 8 (7-10) . . . . Argentina sphuraena p. 108 2b. Lateral ·line scales 67 (64-69); gill rakers on lower arm of first arch 18 (11-15) ........ Argentina silus p. 117 lb. Branthiostegals 5, scales without. spines. Sa. Gill rakers on lower arm of first arch 17 (14-21); swimbladder impregnated with silvery pigment . . Argentina sialis p. 125 36. Gill raker5 on lower arm of first arch 6 (5-10), swimbladder not impregnated with silvery pigment .......4 108 BULLETIN FLORIDA STATE MUSEUM Vol. 8 4a. Pectoral rays 19 ( 18-20) ; approximately 45 teeth on vomer ..: 2... Argentina striata p. 129 4b. Pdctoral rays 15 (18-17); approximately 25 teeth on vomer . .. . . . . . . . . Argentina elongata p. 188 Argentina sphyraena Linnaeus Argentina sphyraena Linnaeus, 1758 : 815, in part (original description, type 10- cality: "Mari infero" cites Artedi who gives "Mare Inferum ad Romam etc. as locality, holotype: unknown). Britnnich, 1768 : 78 (description, Marseilles). Gmelin, 1778 : 1894, in part (description) Walbaum, 1792:44, in part (de- scription). Walbaum, 1793 : 17 (synonymy). Lac6p6de, 1808 : 866, in part (description, synonymy, distribution). Sonnini de Manoncourt, 1804 : 861 (description). Rafineslue-Schmaltz, 1810 : 85 (Sicily, common names). Cu- vier, 1815 : 228, pl. 2 (description, relationships, nomenclature, poor Egure) Cuvier, 1817 : 164 (description). Risso, 1826 : 462 (description, synonymy, Nice). Oken, 1886 . 831 (description, synonymy, Mediterranean). Swainson, 1839 : 287 (description). Bodaparte, 1846 : 25 (synonymy, Mediterranean). Gunther, 1866.: 203 (description, synonymy, Mediterranean). Steindachner, 1868 : 782 (description, synonymy, Cadiz). Canestrini, 1870 : 129 descrip- tion, synonymy, common names, spawning time, Italy). Collett, 1875 : 171 (description, A. hebredica synonymized with A. sphyraena, Christianafjord, Stavanger). Doderlein, 1878 : 56 (Sicily). Collett, 1879 : 92 (Norway). Giglioli, 1880 : 12' (in part, variation in tongue dentition, sytionyinized 'With A.. ·leioglossa, Nice, Naples, Messina, Spalato). Day, 1881 : 78, pl. 4 (de- seription, synonymy, nomenclature, distribution). Day, 1880-84 : 186, pl. 125 (description, synonymy, nomenclature, distfibution). Moreau, 1881 : 554, fg 208 (description, synonymy). Reguis, 1882 : 175 (description). Faber, 1888 : 224 (common names, Adriatic). Gunther, 1887 : 218 (in part, synonymy, dis- tribution). Gunther, 1888 : 220 (northwest coast of Scotland, 32 and 87 fms.) Lilljeborg, 1891 : 692 (description, synonymy, distribution along Norwegian coasts). Carus, 1893 : 555 (description, distribution in Mediterranean, com- mon names). Holt and Calderwood, 1895 : 861 (descripti6n, summary of Irish records, in skate st6mach 375-500 fms. off Achill Head, Ireland). Smitt, 1895·: 917, fig. 230) (description, synonymy, distribution in Atlantic and Med- - iterranean, good figure). Goode and Bean, 1895 : 51 (in part, description, synonymy, distribution). Carman, '1899 : 408 (listed). Collett, 1903·: 168 (Nojn*egian records, food, parasites). Schmidt, 1906 : 10, pl. 1, figs. 14-23 (descriptions of larvae and postlarvae, comparisgns with A. silus, distribution in Atlantic). Fage, 1907 : 75 (listed). Ehrenbaum, 1909 : 351, fig. 188 (eggs . , - and larvae, after Schmidt, 1906). Fage, 1910 : 8, fig. 1 (larva, near Monaco, 20 meters). Holt and Byrne, 1910 : 8 (listed). Fowler, '1912: 556 (Italy). Murray and Hjort, 1912 : 894, 447 (west of British Isles; 146 and 148 m.). Ninni, 1912 : 181 (common names, markets of Venice iii winter). Meek, 1916 : 143, fig. 52 (spawning time, distribution map). Schmidt, 1918 .6, fig, 1, chart 1 (young stages, comparison with A. leioglossa, distribution in Atlantic and Mediterranean). D6 Buen, 1919 : 216, 815 (Spanish references, distribu- tion, common names, abundantat Cadiz). Johnsen, 1919 : 56 (Norwegian records, comparison with A. SIU) Jenkins, 1925 : 248 (description, distribu- , 1958 COHEN: SUBFAMILY ARGENTININAE ,109 tion, eggs). De Buen, 1926 : 49 (synonymy, Spanish references, Spanish dis- tributibn). Smith, 1928 : 88 (Irish Sea, Pt. Erin). Kyle and Ebrenbaum, 1929, XII. F 46, fig, 27 (description, distribution, common names). Scheur- ing, 1929 : 658, ~g. 88 (distribution, after Meek). Schnakenbeck, 1931, Bg. (description, distfibution, common names). Sanzo, 1983 26, figs. 24-81, pl. 2 (development, comparison with development of A. leioglossa). Brun- nelli and Bini, 1934: 743 (Montecristo, Italy). De -Buen, 1985 : 47 (synony- my,, Spanish distribution). Valiani,· 1940 : 125 (comparison of stomach con- tents and size of specimens from Nice and the Tirrenean Sea). Chapman, 1942, figs. 1-8 (osteology, relationships of family). Lozano Rey; 1947: 127, fig. 25 (descriptioh, distribution, common names). Solian, 1948: 185, figs. (description, Adriatic). Karlovac, 1949 : 9 (postlarvae in Adriatic), Ton tonese and Trotti, 1949 : 84 (Mediterranean distribution, common names, Italian references). Bruun and Pfaff, 1950 : 28 (Skagerrak, Norwegian and Swedish coasts). Went, 1951 : 197 (Ireland, coast of county Dublin). Pa- lombi and Santarelli, 1958: 12, fig. (description, common names). Andria- shev, 1954 : 120 (description). Tambs-Lyche, 1954= 8 (fjords in vicinity of Bergen, 195-200 m.). Kotthaus and Kreft, 1957 : 171 (west and southeast coasts of Iceland). Salmo sphymena Bloch and Schneider, 1801 : 414 (description). Argentina silus funior Risberg, 1885 : 7 (original description, type locality: ChrisM- ana, holotype: unknown). Oamerus hebredicus Yartell, 1838 : 108 (original description, type,locality: Bay of Rothsay, Isle of Bute, Hebrides, holotype: unknown). Bonaparte, 1846 : 24 (Britain). Yarrell, 1860 : 16, fig. (reprint of original ddscription). Couchs, 1865 : 297, pl. 238 (description, nomenclature, Orkneys, plate after Bloch and Schheider, copied from Ascanius' figuke of. Salmo silus) Coniosoma argentinum Costa, 1844 in Costa 1836-44, pl. 36 (nomen nouum pro Argentina sphgraena Linnaeus, description, synonymy, distribution, nomen- clature, common .haines, Naples). Argentina cut)ieri Valenciennes, 1848 : 418 (nomeh nouum pro Argentina sph[/raena Linnaeus, description, Malta, Malaga, Baleakic Islands). Bertin, .1940 : 807 (lists paratypes in Paris Museum). Argentina yarrelli Valenciennes, 1848 : 418 (nbmen nouum pro Osmenis hebred- icus Yarrell, description, distribution, nomenclature). Bertin, 1940 : 807 (lists , , paratypes in Paris Museum). Argentina hebredica Nilsson, 1855 : 474 (description, synonymy, Scandinavia, com- parison with A. sphuraena), Yarrell, 1859 : 300, fig. (description, nomencla- ture). Gunther, 1866 : 208 (description, synonymy, coasts .of Norway and Scotland). Brito Capello, 1881 : 35 (synonymy, Lisbon). Garus, 1893 : 555 (description, distribution). Diagnosis.-A. sphyraena differs from all other species of Argentina except A. silus in having 6 branchiostegal rays (other species have 5) and in having spines on the scales. It can be distinguished from A. silus by a lower lateral line scale count (50 to 54 in A. sphyraena, 64 to 69 in A. silus), fewer gill rakers on the iower arm of the first arch 110 BULLETIN FLORIDA STATE MUSEUM Vol. 8 (7 to 10 in A. sphyraena, 11 to 15 in A. silus), lower vertebral count (49 to 52 in A. sphyraena, 65 to 69 in A. silus), and a lower pectoral count (12 to .15 in A. sphyraena, 15 to 18 in A. silus). A. sphyraena can also be differentiated from A. elongata and A. striata by the silvery pig- ment impregnating its swimbladder. COUNTS AND MEASUREMENTS.-Based on 73 specimens. D. 11(10- 12); A. 13 (11-15); pectoral 14 (12-15); ventral 11 (10-12); gill rakers on ~ first.arch 4 (8-5) + 1 (0-1) + 8 (7-10); lateral line scale rows 52 (50-54); standard length 70.9-210.5 mm. (the largest specimens recorded in the literature are examples up to 82.0. cm. noted by Kotthaus and Krefft, 1957); preanal 82.3 (78.0-84.8); preventral 51.6 (49.0-52.1); pre- dorsal 44.9 (41.2-47.I); prepectoral 28.9 (25.4-82.8); head length 28.3 (24.0-81.2); snout 9.1 (7.4-10.6); eye 8.8 (7.2-10.3); maxillary length 6.5 (5.3-7.9); maxillary width 1.9 (1.4-2.4); depth at dorsal fin 18.3 (9.7-14.2); width behind head 10.6 (9.2-12.2); head depth 18.2 (11.8-14.4); inter- orbital 7.4 (6.5-8.5); depth of caudal peduncle 5.7 (4.5-6.5); vertebral count of one North Sea specimen 52, a specimen from Portugal 49 and twb from the Mediterranean 49 and 50; pyloric caeca 9 for a specimen from Portugal, 7-12 in four examples from the Mediterranean (Smitt, 1895 summarizes the observations of several authors, the counts ranging from' 5 to 20); branchiostegals 6 (Chapman, 1942 records 7; however, I have found only 6, even in cleared and stained specimens). DESCRIPTION.-Body elongate, Iaterally compressed, with the greatest depth in some specimens immediately behind the head, in others at the dorsal origin, tapering to the caudal peduncle. A lateral view sh6ws the upper profile of the head descending in a gentle curve which is broken by the dorsal margin of the eye. The ventral proBle rises more gently from the posterior edge of the.preopercle to the union of the articular and the quadrate from whence it rises more rapidly. Chapman (1942) has _described the interorbital space as "depressed to form a broad, V-shaed trough." In specimens I have examined the frontals between the supraocular canals constitute a broad flat area. The canals themselves are slightly raised above the flat surface in larger, examples, while in smaller specimens the canals are almost flush with the surface. Lateral to the canals the frontals arch upwards to accommodate the, top of the eye. The dorsal edge of the maxillary lies tucked under' the lachrymal in smaller specimens and extends to the jugal in larger ones. The posterior edge of the maxillary lies closer to the eye than to the snout. The epidermis covering the ven- tral edge of the posterior portion of the maxillary bears minute papillae which extend in a narrow band onto the upper lip where they meet a j similar band from the other 5ide. Both the upper and lower jaws are 1958 COHEN: SUBFAMILY ARGENTININAE 111 r6unded, and the upper jaw projects slightly. A small knob is present below and slightly behind the curved symphysis of the lower jaw. The palatines and the head of the vomer bear a band from 2 to-4 teeth wide composed of small, closely spaced, posteriorly directed teeth. There are about 50 teeth on each palatine and about 85 on the vomer. 'The tongue is armed with 5 to 8 strong, recurved teeth. - Tongue teeth were present in all the specimens I examined. (Giglioli (1880) examined a series of specimens which he said varied consider- ably in the dentition of the tongue. His comments lead me to believe he bxamined a collection of Argentina sphyraena mixed with (Nos- sanodon leioglossus, a species in which the tongue teeth are small and easily overlooked.) Sensory papillae were not observed on the tongue. The ceratobranchial of the Sth gill arch bears 11 small, conical teeth. Two oval-shaped patches of short, conical teeth, about 14 teeth on the anterior patch and 16 on the posterior patch, are present at the anterior end of the 4th suprabranchial. The gill rakers are medium-sized, widely spaced, triangular in shape and compressed. The 4th raker forward from the angle on the lower arm is equal in length to about one-sixth of the interorbital distance. The rakers on the upper arm are' fewer in number and slightly smaller than those on the lower arm. The pectoral fins usually originate in line with a vertical from the posterior tip of the opercles, although the 6rigin may be slightly ahead of or slightly behind this line. The posterior end of the pectoral fin base forms an angle of 20 to 85 degrees with a plane parallel to the venter. The distance between the bases of the innermost rays of each pectoral Bn is approximately equal to .the least depth of the caudal peduncle. The longest dorsal ray is equal to the depth of the body at the dorsal fin origin, and the longest anal ray is less than the least depth of the caudal peduncle. The tip of the pectoral fn extends to a point midway between the anterior base of the pectoral and the anteribr base of the ventral. The ventral fin is equal in length to the pectoral Bn. The scales bear minute spines which are barely discernible by run: ning a finger forward over the body of the fish. The weakness of these spines is in marked contrast to the well-developed scale spines of A. silus. The scales of both species have been illustrated by Smitt (1895: 913, 917). The sckles seem to play an important part in main- taining the shape of the body. They act as a true exoskeleton by form- ing an elongate, 10-sided box. The shape is evident only -in the few specimens I have examined with nearly complete squamati6n. In specimens without scales the sharply defined angles slump away and leav6 the cross section oval. 112 BULLETIN FLORIDA STATE MUSEUM Vol. 3 The peritoneum is silvery on its outer surface, While the inner sur- face is heavily Recked with dark pigment, appearing almost solid black in some places, as is the dorsal section of the stomach. The ventral sec- tion, is more lightly colored; the duodenum is lightly flecked and the remainder of the gut has only an occasional sprinkling of melano- _ phores. The posterior curve of the stomach is about midway between the level of the pectoral insertion and the level of the ventral insertion. The outer.surface of the eurve bears only a small protuberance. The longest pyloric caeca extend back to the curve of the stomach in larger specimens, not as far in smaller ones. The gut runs in a straight line from the distal end of the stomach to the anus. Specimens fr6m Portu- gal contain considerable amounts of fat investing the mesenteries, sur- rounding the pyloric caeca and pancreatic tissue, and adhering to the face of the swimbladder. Examples from Barcelona are thinner and practically free of fat. Many of·the Spanish fish are heavily parasitized by roundworms. The swimbladder is heavily impregnated with silvery pigment. The anterior end of the bladder lies above the posterior curve of the stomach and the posterior end is at about the level of a vertical from' the posterior tips of the ventrals. The color varies in the alcohol s]>ecimens I have examined. This variation seems to be dife to locality, length of time since the specimen was collected, and type of initial preservation. A recently collected example from Lofoten, Norway, has a uniform light brown ground color ventrally with the dorsal half of this ared lightly Recked with dark brown spots. An iridescent sheen is present in the skin of the en- tire light brown area. A narrow, silvery band runs along the lateral line; above this band is a dark brown horizontal band extending dor- sally to the dorsolateral angle of the body. The entire dorsum of the fish is a light brown color similar to the area below the lateral line. The caudal is heavily flecked with dark brown, especially at the upper and lower corners of the caudal peduncle, while the lateral half of the pectoral fins and the entire dorsal and the ventral Bns are Hecked to a lesser degree, and the anal and medial half of the pectoral are un- colored. The opercle is transparent except for a sprinkling of large chromatophores. It also is faintly iridescent. The eheeks are brown; the top of the head is transparent posteriorly and translucent farther forward, The muzzle from the nostrils forward is dusky. A speci- men from .Oslo, Norway, collected at an earlier date and probably initially preserved in alcoh61, has the uniform 6range brown body traversed by a broad silvery band about as wide as the distance be- tween the inner ba5es of the pectoral fin and extending from the poste- 1958 COHEN: SUBFAMILY ARGENTININAE 113 rior edge of the opercle to the base of the caudal Bn. It has the lateral line as its dorsal border and the next ventral body angle as its ventral border. The fins are all unmarked. The opercle and preopercle are a bright silver, while the rest of the head is a faded white. These two descriptions represent the extremes in coloration. Various intermedi- ate patterns are present in the collection. Schmidt (1906) and Fage (1910) show figures of larvaI A. spligraena in which a row of 7 to 9 dusky blotches are present along the ventral half of the side. DISTRIBUTION.-Argentina sphyraena occurs from the Lofoten Is- lands of Norway south along the Scandinavian coast to the tip of Den- mark in the Skagerrak, in the northern part of the North Sea, south along the western edge of the British Isles through the Bay of Biscay , and down the Atlantic coast of the Iberian Peninsula to Gibraltar. It is taken occasionally in the Irish Sea, and Kotthaus and Krefft (1957) have recorded it off Iceland. In the Mediterranean the species has been taken along the European coast as far east as the Adriatic and is also found off the Balearic Isles and Sicily. Poll's (1953) record of A. sphyraena from the west coast of equa- torial Africa is referred to Glossanodon polli and di5cussed elsewhere in this paper. Gilchrist and von Bonde (1924: 38) record A. sphy- raend from the Indian Ocean off the coast of Natal and Delagoa Bay. Unfortunately, they give neither a description nor a figure of their ma- terial, but merely list synonymy and the station number. Keppel-H. Barnard of the South African Museum writes me that many of the specimens collected by the South African Fisheries Survey, Argentina among them,.have been lost, and that he knows of no others having been collected. J. L. B. Smith of Rhodes University writes me that the 'record of Gilchrist and von Bonde mentioned above is the only one known and that the identification was never verified. He also states that his description of Argentina in his book on South African - fishes (1958) was taken from Smitt (1895), as was the figure (which is actually of A. silus,:not A. sphyraena). Therefore I have refrained from recognizing the Indian Ocean record of A. sphyraena. If an argentinid is discovered in this zoologically surprising area it will prob- ably be a Glossanodon. GEOGRAPHICAL VARIATION.-When certain of the counts and meas- urements summarized above are broken down into thre6 geographical groups, differences in the ranges and average values independent of the size of the fishes become apparent. The geographical areas ar6 the Mediterranean, the Atlantic off Portugal, and the North Atlantic - and Norwegian Sea from Great Britain to the north. The collections 1 114 BULLETIN FLORIDA STATE, MUSEUM V61. 8 TABLE 1 COMPARISONS OF COUNTS AND MEASUREMENTS OF ~HREE GROUPS OF Argentina sphyraena, Group I the North Atlantic and Norwegian Sea from Great Britain to the north, Group II thi Atlantic off Portugal, Group III the Mediterranean. All lengths are expressed as percentages of standard length. Head I II III Prepectoral I II III length length 24.6-.9 1 25.0-.9 2 25.0-.9 6 26.0-.9 1 26.0-.9 10 27.0-.9 10 27.0-.9 6 '2 28.0-.9 679 28.0-.9 665 29.0-.9 793 29.0-.9 4 10 4 80.0-.9 884 80.0-.9 138 81.0..9 4 31.0-.9 1 82.0-.9 1 Average 27.5 29.2 29.4 Average 28.2 29 2 .29.6 Preanal I .II III Predorsal II IIIlength length 78.0-.9 2 41.0-.9 79.0-.9 ' 8 42.0-.9 M C le o ¤ r - e l 80.0-.9 5 2 43.0-.9 281.6-.9 . 16 8 3 44.0-.9 3 582.0-.9 16 5 5 45.0-.9 9 5 83.0-.9 5 8 46.0-.9 5 784.0-.9 8 8 47.0-.9 8 85.0-.9 1 Average, 44.1 45.4 45:9 Average 81.4 82.8 83.2 Eye I Il III Snout I II III lengthdiameter 7.0-.9 6 7.0-.9 4 8.0-3 21 12 5 8.0-.9 22 2 8 9.0-.9 5 6 18 9.0-.9 7 14 14 10.0-.9 2 10.0-.9 3 8 Average 8.8 8.8 9.8 8.7 9.6 9.4Average Maxillary I II III Anal En I II III length ray counts , 5.0-.9 14 11 1 6.0-.9 14 15 11 12 48 7.0-.9 8 3 9 18 10 18 10 14 21 2 Average 6.1 6.7 7.0 15 1 1958 COHEN: SUBFAMILY ARGENTININAE 115 from the most northerly of the three areas show; on the average, a highef anal count, shorter preanal distance, shorter predorsal distance, lesser head length, smaller eye, shorter prepectoral, shorter snout, and shorter maxillary. The Portuguese specimens are intermediate, while those from the Mediterranean show the opposite extremes f6r the characters listed aboye. The data are,summarized in table 1.. Schmidt (1918) investigated geogiaphical variation in the vertebral number of A. sphymena, and records a higher average number of vertebrae for fishes taken north of Scotland than for those taken north of Spain. The variation of A. sphyraena is perhaps another example of the widely known tendency of certain fish species to respond morpholbgically to environmental influences, particularly temperature as influenced by latitude. HABITS.-A. sphyraena has occasionally been taken on a. hclok and line close to the surface, but it is most frequently captured in trawls on or close to the bottom. Tambs-Lyche (1954) states that in the Ber- gen area the species is generally taken at depths between 90 meters and 200 meters and that it is a member of the mud-bottom community. Holt and Calderwood (1895) identified A. sphi/raena in the stomach of Raia oxyrhynchus captured somewhere between 875 and 500 fathoms. In general A. sphyraena seems to pfefer the coastal shelf, although it avoids extreme shoal waters and for this reason perhaps has not been taken in the shallow southern part of the, North Sea or in the English Channel. Although A. sphyraena may wander down the continental slope on occasion, it is certainly not a characteristic member of the deep sea fauna as many authors since before the days·of Linnaeus have labeled it. Schmidt (1906, 1918) describes the eggs and lame as being pelagic in depths to 184 meters. He places the spawning of the Mediterran- ean population in the winter, the more northern populations later in the year: The boxlike armor of scales which encases A. sphyraena would seem to preclude the possibility of rapid movement, yet the fish has adipose eyelids and seale sheaths for the ventral fins, both character- istics of relatively fast-moving Bshes. Valiani (1940) examined the stomach contents of several hundred A. sphyraena ih the Mediter- ranean, and his studies give some clue as to the habits of the species. The greater part of the food consisted of b6ttom-dwelling mollusks and drustaceans. However, the presehce of Clupea pitchardus and Pyrosoma elegans, the latter a pelagic tunicate, indicates that a certain am6unt of time is spent off the bottom and that if necessary the 8sh can swim with the speed necessary to capture Clupea. 116 BULLETIN FLORIDA STATE MUSEUM Vol. 8 NOMENCLATURE.-A. sphyraena in the Mediterranean was described under two names other than A. sphyraena, The first of these, Goni- osoma argentinum, was proposed in 1844 by Costa because of a mis- understanding of tHe mouth-part terminolegy used by Cuvier (1815) in defining the genus of Linnaeus. One is led to believe that a surfeit of chauvinism also played a part in the proposal of the new name if we are to judge from Costa's comment, "1'illustre zoologa francese ne avesse diradate le tenebre, -in reference to Cuvier's treatment of Argentina. Valenciennes (1848) also decided that the species needed a new name, and, prompted by motives exactly the opposite of those which moved Costa, came forth with the name Argentina cuvieri, in honor of his late master. Meanwhile, farther to the north, Risberg had recorded the species from Norway in 1835 (in a paper sbmetimes attributed to Nilsson) and demonstrated his opinion of its affnities by calling it Argentina silus iunior. Yarrell rediscovered the species in the Hebrides in 1838, and called it Osmerus hebredicus. Valenciennes placed Yarrell's spe- cies in the genus Argentina and in so doing decided to honor its dis- coverer by renaming the species Argentina yarrelli. In 1855 Nilsson identified A. silus iunior with Argentina hebredica and also placed A. yarrelli in the synonymy of A. hebredica. In the same work he com- pared A. hebredica with A. sphyraena and maintained the species were distinct. The first author to lump the two forms was Collett in 1875. He was followed by Day (1881) who pointed out that many of the char- acters used to separate the two populations are extremely Variable and to a certain extent dependent on size. Almost all authors- since,Collett and Day have agreed with their treatment of the species. In the above synonymy I have treated Goniosoma argentinum, Argentina cuuieri and Argentina yarrelli as nomina nova rather than new species because in no case was the author describing a form which he considered new to science, but instead was substituting a new name for one which he considered untenable. I mention this somewhat obscure point because Bertin (1940) has listed the paratypes of the latter two, thereby implying he considers them to have been new species. A nomen nouum based on a previous author's work is based - on that author's material. SPECIMENS EXAMINED.- Norwav : Nordaland county, Lofoten, BM 5202 (5); Hordaland county, Herdla, BM 4202 (2) and 4208 (2); Bergen, BM 2248 (1); Rogaland county, Mastra- fjord, BM 3024 (3); Rogaland county, Gannsfjord, BM 8078 (7); Rogaland . county, Baasfjord, BM 3058 (6); Oslo, BM 2254 (5) 1958 COHEN: SUBFAMILY ARGENTININAE 117 Denmark: Near Hirshals, "Michael Sam" st. 388, BM 2417 (2) Scotland: Morayshire, Lossiemouth, CNHM 85488-37 (5); 49' 32' N/ 10' 49' W, BM .8283 (I); 49~ 27' 3/ 8' 86' W, BM 8858 (2) Portugal: 38' 21' N/ 9' 10' W, SU 49267 (20). Spain: Barcelona, SU 49269 (15), Frdnce: Herault, Sete, SU 48872 (7). Italy: Gulf of Genoa, SU 49268 (1); Palermo, SU 1627 (1). Yugostavia: 48' 18:4' 3/ 16' 28.5' E, SU 48878 (12). _ Argentina silus (Ascanius) Salmo silus Ascanius, 1775 : 8, pl. 24 (original description, type 16cality: Bergen, Norway, holotype: unknown, poor figure). Muller, 1776 : 49 (Scandinavian and Icelandic references, common names). Bloch and Schneider, 1801 : 414, pl. 81 (description, pl after Ascanius). Coregonus silus Cuvier, 1817 : 168 (listed, subsequent editions of ·this work use the same name). Nilsson, 1882: 19 (description, synonymy, Norway). Bona- parte, .1846 : 24 (sydonymy, Norway) silus ascanii Reinhardt, 1888 : 11 (nomen nouum pro Salmo silus Ascanius). Argentina situs Risberg, 1885 : 8 (description, synonymy, Bergen, S6ndm6r, Christi- ana, comm611 names). Valenciennes, 1848 : 421 (description, discussion of synonymy, Bergen, common names). Nilsson, 1855 : 469 (description, synony- my, west coast of Norway, food, habits). Gunther, 1866 : 202.(description, synonymy, northwest coast of Norway). Collett, 1875 : 178 (distribution along .Norwegian coast, darly history of common names). Collett, 1879 : 98 (de- scription, distribution along Norwegian coast) Winther, 187.9 : 45 (synonymy, Blaavands Huk, Denmark). Edward, 1881 : 884 (description, Moray Firth, Banffshire, color, food). Gunther, 1887 : 217 (listed, distribution). Lillieborg, 1891 : 679 (description, synonymy, Scandinavian distribution, osteology, hab- its). Smitt, 1895 : 918, fig. 229 (description, synonymy, distribution, habits). Goode and Bean, 1895: 52, fig. 61 (description, synonymy, Sable Island Bank, 200 fms., Belfast, Maine, Biddefoid Pool, Maine, poor figure). Jordan and Evermann, 1896 : 526, fig. 282 (description, distribution, Egure from Goode and Bean, 1895). Holt, 1897 : 841 (50' 20' 3/ 8' 25' W, 74 fms., food) Carman, 1899: 403 (listed). Boulenger, 1900 : 160 (sold in market at Boston, Lincolnshire, England, caught off coast of Norway). Ehrenbaum, 1901 : 138 (synonymy, distribution). Collett, 1903 : 165 (Norwegian records from Christi- anafj6rd to Finmarken, depths to 900 m., length to 475 mm.). Holt and Byrne, 1906 : 5, 22 (west coast of Ireland; south of Porcupine Bank, Irish Sea; 105 and 164 fms.). Schmidt, 1906 : 1, pls. 1-2 (eggs, larval and post- larval development, comparisons with _early stages of A. sphymena, distribu- tion and depths of eggs and young in eastern north Atlantic). Kendali, 1908 : 49 (listed). Ehrenbaum, 1909 : 347, fig. 132 (eggs and larvaes after Schmidt, 1906). Holt and Byrne, 1910 : 8 (listed). Kendall, 1914 : 89 (Maine records). Meek, 1916 »: 143 (distribution, spawning). Schmidt, 1918 : 10, chart 1 (rec- ords of eggs, laryae and postlarvae in eastern north Atlantic). Johnsen, 1919 : 56, 85 (Norwegian records, age, spawning time of Norwegian specimens). Dons, 1921 : 87 (records from northern Norway). Kendall and Crawford, 1922 : 8, fgs: 1-2 (osteology, soft anatomy, North American records, relation- lr8 BULLETIN FLORIDA STATE MUSEUM Vol. 3 ships). Grenholm, 1928:28 (fin muscles). Bigelow and Welsh, 1925 : 147, fig. 59 (description, distribution, Gulf.of Maine records, habits, fig. from Goode and Bean, 1895, and Schmidt, 1906). Jenkins, 1925 : 248 (description, eggs, distribution). Kyle and Ehrenbaum, 1929:XII. F 46, fig. 26 (description, common names, fig. after Smitt). Scheuring, 1929 : 658, fig. 88 (distribution, after Meek, 1916). Jordan, Evermann and Clark, 1980 : 68 (synonymy, distribution). Firth, 1931 :10 (eastern N. Atlantic records). Schnakenbeck, 1931; fig. (description, distribution, common names, fig: from Smitt). Vlady- kov and McKenzie, 1985 : 59, fig. 87 (listed, fig. from Goode and Bean, 1895). Bigelow and Schroeder, 1936 : 828 (off Georges Bank and Cape Cod in deep water). Iversen, 1986 : 99 (65' 86" N/ 81' 15' W, only Greenland record). McKenzie and Homans, 1987 : 277 (43' 05' N/ 62' 85' W off Nova Scotia, 54 fms.). Fridriksson, 1987 : 72, fig. (distribution and depth off Iceland, life history data). McKenzie, 1939 : 15 (W to SW of Emerald Bank, Nova Scotia). Masloy, 1944 : 168 (Barents Sea east of N. Cape, Norway, spawning at Lofo- ten). Lozano Rey, 1947 : 129, Egs. 27-8 (descriptioh, distribution, incorrtctly recorded from Bay of Biscay, figs. from Smitt, 1895 and Schmidt, 1906). Breder, 1948:75, fig. (Amefican distribution, spawning, eggs). Saemundsson, 1949 : 94 (distribution, Icelandic records and references). Bruun and Pfaff, 1950 : 28 (Skagerrak). Bigelow and Schroeder, 1953 : 189, fig. 59 (descrip- tion, eggs and larvae, distribution, records from the Gulf of Maine), Andria- shev, 1954 : 119, fig. 56 (description, synonymy, Arctic records). Anon. 1955 : 35 (1000-pound catch at edge of cohtinental shelf, New England). Schroeder, 1955 : 862' (S Of Nova Scotia, over 200 fms.). Kotthaus and Krefft 1957 : 171 (Iceland-Greenland Ridge, SW, W, and NW coasts of Iceland, Anton Dohrn Bank; 190-500 m.). Acantholepis situs Kr0yer, 1846-49.: 97, 4 (description, synonymy, nomenclature, common names, distribution). Kr0yer, 1847, pl. 17. Argentina surtensium Goode and Bean, 1879 :.261 (original description, type 10- cality: "from the stomach of a hake (Phucis tenuis ?)" in 200 fms. on Sable Island Bank off Nova Scotia coast, holotype: USNM 21624). Jordan and Gilbert, 1883 : 295 (description after Goode and Bean). Jordan, 1887 : 880 (listed). DIAGNOSIS.-An Argentina with spiny scales, 6 branchiostegals, and the swimbladder impregnated with silvery pigment. A. silus dif- fers from ali other species of Argentina in its high' lateral line scale count (64 to 69 in A. silus as against 48 to 56 in all other species), a higher average anal ray count (15 [13 to 17] as against 11 to 15 in all other species of Argentina),and a higher vertebral count (65 to 67 in A. silus as against 49 to 54 in all other species). COUNTS AND MEASUREMENTS.-Based On 59 specimens. D. 12 (11- I3); A. 15 (11-17); pectoral 17 (15:18); ventral 12 (12-13); gill rakers on first arch 6 (5-7) +1+13 (11-15); lateral line scale rows 66 (64-69); standard length 77.4-321.9 mm. (Smitt, 1895, quotes Nilsson as saying that this species may reach.2 Swedish feet (59 cm.); preanal 82.4 (79.8- 85.4); preventral 52.5 (50.0-55.4); predorsal 48.1 (40.7-45.4); prepectoral 1958 COHEN: SUBFAMILY ARGENTININAE 119 27.8 (24.6-80.1); head length 27.5 (25.4-29.6); snout 7.9 (7.0-8.8); eye 9.0 , (6.0-11.1); maxillary length 6.2 (5.2-7.4); maxillary width 1.8 (1.5-2.2); depth at dorsal fin 16.1 (11.3-22.0); width behind head 11.9 (9.3-14.7); head depth 14.8 (12.7-15.5); interorbital 7.5 (6.5-8.8); depth caudaI peduncle 6.5 (6.0-7.6); vertebral count of two western Atlantic speci- mens 67, of an eastern Atlantic specimen 65; pyloric caeca of thfee western Atlantic specimens 21, 22 and 28, two from the eastern At- lantic 15 and 27; branchiostegals 6. DESCRIpTIoN.-Boay deepest at dorsal origin, slender to robust, laterally compressed. Gravid fem-ales with bellies much distended. Greatest depth at or slightly ahead of the dorsal insertion. The dorsal proBle curves gently to the snout from the point of greatest depth and extends posteriorly to the caudal peduncle in a straight line. The dorsal profile 6f the head is interrupted by the dorsal margin of the eye which juts up prominently in larger specimens, less so in smaller ones. The ventral profile of the head runs forward almost in a straight line with only a slight angle at the union of the articular and the quadrate. The general shape of the head reminds me, as it did Smitt, , of a rather blunt, four-sided wedge. The interorbital is a broad flat- bottomed space betwe~en the supraocular- canals, which are raised above the surface of the median portion of the interorbital space. Lateral to these canals the frontals arch upwards to accommodate the dorsal edge of the very large eyes. The posterior adipose eyelid is continuously connected with a layer of clear hyaline tissue Over the' . preopercle. The dorsal edge of the maxillary fts under the lachrymal and extends to the jugal as: well. The posterior edge of the maxillary is closer to the eye than to the tip of the snout. The epidermis cover- ing the ventral edge of the maxillary bears a considerable number of small papillae, which in many larger specimens are visible to the naked eye. These papillae, much reduced in size, extend in a narrow band onto tbe epidermis covering the premaxillary where they meet a similar band from the. other side. The jaws are subequal with tha upper jaw rounded and the lower jaw rather bluntly pointed and bearing a distinct knob at the symphysis. The palatines and the vomer bear a band of teeth from i to,8 teeth wide. There are about 25 teeth on each palatine and about 18 teeth on the head of the vomer. The tongue is armed with 8 to 10 strong recurved teeth. Thickly distributed along the margins of the tongue and interspersdd between the teeth are considerable numbers of minute papillae, which are relatively broad based and send up from 1 to 5 short, twiglike branches. These papillae are similar to. those ~ previously noted on the premaxillary and maxillary. Similar struc- 120 BULLETIN FLORIDA STATE MUSEUM Vol. 8 tures are also found in large numbers on ridges in the floor and roof 6f the pharynx. In the only two larvae of this species I have examined (total lengths 25 mm. and 35 mm.) the dentaries bear a single row of minute cardiform teeth. The next smallest specimen available (71 mm. total length) lacked teeth on the lower jaw, as did all larger specimens. The ceratobranchial of the 5th gill arch bears 5 small teeth. At the anterior end of the 4th suprabranchial are two small patches of teeth with about 10 teeth in the posterior patch and 8 in the anterior. The gill rakers are medium sized, widely spaced, roughly triangular in shape, and compressed. The 4th raker forward from the angle on the lower arm of the 1st arch is equal to or slightly shorter than one-fourth the inte~orbital width. The rakers on the upper arm are fewer in number and slightly smaller than those on the lower arm of the arch. The pectoral fn originates close to a vertical from the posterior edge of the opercle. The posterior end of. the pectoral base forms an angle of 30 to 85 degrees with a plane parallel to the venter. The distance between the innermost rays of each pectoral fin is less than the least depth of the caudal peduncle. A ventral adipose fin, about as high and twice as long as the dorsal adipose is present in frcpnt of the anus of a specimen 54.6 mm. in standard length, while a specimen 68.0 min. in standard length was observed to have a small adiposd flap in front of the anus. No ventral adipose fin was observed in any other . specimen of A. silus examined. The longest dorsal ray is approximate- ly equal to the depth of the body at the origin of the dorsal fin. I have seen no specimens with unbroken anal rays. The tip of the pectoral fin extends to a point about midway between the anterior of the pectoral fin base and the anterior of the ventral fin base. The ventral fn is slightly shorter than the pectoral fin. A distinctive feature of the scales is the presence of small, but strong spines on their posterior exposed portions. These spines may be easily discerned by running a Bnger forward over the side of a. specimen. The spines grow from the posterior edge of the scale as well. as on the suBface, thus giving the scales a distinctly ctenoid ap- pearance. Smitt (1895) Rgured these scales. As in other species of Argentina, the scales mold the body shape, at least in smaller speci- mens, while specimens which lack seales are oval in cross section. The angles of the body in the few specimens available with complete squamation are'less sharply defined than those of A. sphyraena. The outer surface of the peritoneum is coated with a silvery pig- ment, while the ventral and lateral surfaces of the inner face of the peritoneum are so densely punctate they app~ar dark brown or 1958 COHEN: SUBFAMILY ARGENTININAE 121 , blackish. The dorsal surface bears fewer chromatophores and the silvery swimbladder is visible dorsally through the peritoneal mem- bfanes. The Stomach is almost unmarked, with only an occasional chromatophore. Ths Pyloric caeca and the remainder of the gut are lightly peppered with chromatophores. The stomach runs posteriorly to a point slightly anterior to the insertion of the ventral fins.' The outer surface of the posterior cOrve of the stomach bears an extension which Kendall and Crawford (1922) have illustrated and compared to a short caecum. The .intestine is twisted, into a single loop at a point slightly behind a vertical from the posterior tip of the stomach. Distal to the loop the intdstine runs straight back to the anus. I have dissected the intestines of 6 specimens of A. silus, 4 from the western Atlantic and 2 'from the ebstern Atlantic, and in none of these did I observe a "functional spiral valve" as figured and described by Ken- dall and Crawford (1922). In general, the intestine and Ringfalten -are in good agreement with those of A. sphyraena. I counted 23 Rin- gen in the intestine of a spdcimen of A. silus, 300 mm. in standard length. The swimbladder is similar to that described for A. sphyraeim with the exception that it is more elongate and its anterior end lies slightly behind the anterior end of the body cavity. The coloration of alcohol specimens is light on the ventral half of the body, while the dorsal half is more heavily pigmented. Many of the specimens are a rich orange brown on the dofsal half and also along the belly. Closer examination shows this color is due to count- less small oil droplets in the skin beneath the scales. These fish are quite greasy to handle, and slight pressure on the side or belly of such a fish causes a film of oil to exude. Some Norwegian specimens, collected at an earlier date than those described above and probably initially preserved in alcohol, have a broad silvery. band running the length of the body with the dorsal edge slightly above the lateral line and the ventral edge about two-thirds of the distance between the lateral line and the venter. The caudal fin is rather heavily pigmented at its base, less so at the extremities of the rays. The dorsal and anal fin are lightly pigmented and the pectoral and ventral fins have pig- ' ment on their lateral rays only, while the ·median portion of each Bn is uncolored. The opercle is bright silver. The cheeks are about the same color as the ventral half of the fish. The top of the head is trans- parent posteriorly and translucent farther forwards. The muzzle is generally dusky, and the entire underside of the head' is lightly sprinkled with chromatophores. Adolescent specimens have 10 to 12 faint dusky blotches alang' the dofsal half of each side of the body, 122 BULLETIN FLORIDA STATE MUSEUM Vol. 8 DISTRIBUTION.-Argentina silus has a disjunct distribution with populations on each side of the North Atlantic. The southern limit of this species on the American side seems to be the edge of Georges Bank east of Cape Cod. The species is also found in the Gulf of Maine (see Bigelow and Schroeder, 1986, for records) up the southeast coast of Nova SOotia, in the Gulf of St. Lawrence, and according to Squires (1954, in litt.) it is taken "fairly frequently on ihe S. W. edge of the , Grand Banks and on St. Pierre Bank in depths generally over 100 fathoms. Mr. Squires has given me several records from.the Grand Banks, of which 54° 38' N/ 53° 88' W is the most northerly. The European population is more widely·distributed with its south- ernmost occurrence southeast of Ireland at about 49° N (Schmidt, 1918). From this point it ranges along the west coast of the British Isles, also entering the Irish Sea. Murray. and Hjort (1912) record A. silus from south of the Canary Islands, but I examined several of their specimens and found them, as Schmidt (1918) suggested, referable to Argentina ·leioglossa (=Glossanodon leioglossus). A. silus occurs northwest to the Faeroes and is also known from off Iceland. The spe- cies is absent from the English Channel and the shallow southern part of the North Sea; however, it has been taken at Blaavands Huk on the west eoast of Denmark, and eggs and larvae are known from the deep part of the Skagerrak. The species has been recorded many times from the west coast of Norway as far north as West Finmarken. More re- eently Andriashev (1954) has summarized Russian records of this fish at the northern tip of Norway, at Bear Island, and near the southwest . coast of Spitzbergen. It should be emphasized, however, that Murray and Hjort (1912) characterize A. silus as a member of the boreal fauna rather than the cold water arctic fauna. In view of the species' known distribution, the single record from the east coast- 6f Greenland (Iver- sen, 1986), an area-whose fauna Murray and Hjort label as arctic, should probably be considered as a stray rather than any indication of a per- manent population. GEOGRAPHICAL VARIATION.-The samples of A. silus I have exam- ined from each side of the Atlantic have been remarkably constant for the characters inyestigated. I noted only three proportional differ- ences, and these overlap to a considerable extent. Differences are ' listed below as percentages of standard length with the mean (variation in parentheses) of the American specimens followed by the same data for the European specimens. Head length 28.0 (26.8-29.6), 26.8 (25.4- 28.9); eye 9.3 (8.0-11.1), 8.7 (6.0-9.7); interorbital 7.7 (7.1-8.3), 7.2 (6.7- 7.9). These populations have not diverged far enough to warrant sub- specific recognition. 1958 COHEN: SUBFAMILY ARGENTININAE 128 HABITS.-A. silus, like A..sphyraena, has also been taken with hook and line close to shore, but this is a rare event. A. silus is probably the . deepest-living member of the genus. On the American side it has been taken at depths as great as 820 fathoms (R. C. Wilson, in litt.) but is most commonly encountered between 75 and 200. fathoms. It seems to prefer the edge of the continental shelf, but Bigelow and Sehroeder (1958) give records from the deeper parts of the Gulf of Maine. Col- lett (I908) gives a record of 500 fathoms off Norway, but other Nor- wegian and British records are shallower and 50 to 200 fathoms is the general range. Fridriksson (1937) gives the bathymetric range off Ice- land as 120 to 1100 meters: Data and dxperiments noted by Schmidt (I906) show. the eggs and larvae to be bathj,pelagic at depths to 900 meters (the figure 1800 m. which some authors have cited from Schmidt ' is the amount of wire out, not the probable depth of the net) with the greitest concentration of eggs at 800 to 400 meters. Schmidt has taken eggs between May and September in the eastern Atlantic, while Bige- low and Schroeder (1958) note the occurrence of eggs in the Gulf of Maine during April. Although A. silus is not a commercially important species, it is fairly commonly encauntered in commercial catches, particularly those of the otter frawlers. These catches indicate that the fish is'not a solitary - species, but may occur in large aggregations. Firth (1981) records a catch of 15,000 pounds on the edge of Georges Bank. Wilson (in litt.) mentions a catch of 1000 individuals south of Nova Scotia near the 100 fathom line. Boulenger (1900) notes a large catch from the Nor- wegian coast marketed in England. A specimen of A. silus with a nearly complete set of scales seems more fiexible than A. sphyraena. This may perhaps indicate that A. silus is a more powerful swimmer, which might be correlated with its = more extensive bathymetric range and generally more oceanic habitat. , I have examined the stomach contents of several American specimens and found partially digested shrimplike crustaceans. Kotthaus and Krefft (1957) note the occurrence of the bathypelagic fish, Cyclothone braueri, in the gut of A. silus. Fridriksson (1987) studied the grbwth of A. silus off the coast of Iceland. He found that sexual maturity occurred at 8 to 15 years of age, that about 60 percent of his sample consisted of females, and that the time of spawning was probably April and May. He also found growth to be relatively slow after the Brst several years. His oldest but not largest specimen was 24 years old and had a standard lendth of 51.0 cm. NOMENCLATURE.-A. silus was originally described by Ascanius as 124 BULLETIN FLORIDA STATE MUSEUM Vol. 3 a member of the genus Sabno. On the basis of Ascanius' poor figure,· Cuvier (1817) referred the species to the genus Coregonus. Reinhardt (1888) differentiated the fish from both Salmo and Coregonus, chiefly on the basis of jaw structure, and he proposed a new genus and species, Silus ascanii. The first to demons~rate the true relationships of the fish by placing it in the genus Argentina was Risberg (1835). Still another generic name, Acantholepis, was proposed by Kr0yer (1846-49) who recognized the close relationship to A. sphyraena, but separated the two species generically because Cuvier's (1815) description of A. sphyraena stated the latter species lacked scales, whereas Kr0yer noted strongly developed ctenoid scales on his specimen of A. silus. Nilsson (1855) listed Acantholepis silus in the synonymy of Argentina silus with no .comment. He has been followed in synonymizing the two genera by all subsequent authors. A. silus was first recorded from the American coast by Goode and Bean (1879), who described it as a new species, A. syrtensium. In 1895 the same authors referred their nominal species to the synonymy of A. silus where it has remained to this day. Mention should also be made of Miiller's 1776 work (p. 279 in the addendum, which - is absent from some copies) in which he includes Salmo immaculatus, presumably of Linnaeus (1758) in the -synonymy of Sabno silus. He has been followed in this by various authors, the most recent being Ehrenbaum (1901). If S. immaculatus Linnaeus is conspecific with S. silus Ascanius, the name S. immaculatus would take precedence by reason of being proposed 18 years earlier. There are, however, several reasons to question the identity of S. immaculatus Linnaeus with S. silus Ascanius. First of all, Linnaeus places S. im- maculatus in the section characini of his genus Salmo. He defines the characini thus: "Membr. branch. radtis tantum IV." However, A. silus has 6 branchiostegals. Furthermore, the anal, pectoral, and ventral counts given by Linnaeus are one ray below the minima I have found in A.·silus. In the "Systema Naturae" Linnaeus (1758) records S. immacu- latus from America. However, Kr0yer (1846-9) tells us that in the "Museum S.R.M. Adolphi Friderici" (1754) where Linnaeus first de- scribed the species, he listed it as coming from India, far out of the range of Argentina silus. Last of all, Linnaeus states of S. immaculatus, 3. corpore immaculato" which is certainly not true of Argentina silus. It seems safe to conclude, therefore, that Salmo immaculatits Linnaeus is not conspecific with Argentina silits (Ascanius). ~PECIMENS EXAMINED.- Noruiay: Nordaland county, Lofoten, BM 5203 ( 1 ); Sogn og Fjordane county, Ask- 'vold, BM 1849 (1); Sogn dg Fjordane county, Lysterfjord, BM 8819 (1); Roga- 1958 COHEN: SUBFAMILY ARGENTININAE 125 land county, W. Salhaus, BM 4212 (1); Rogaland county, Gannsfjord; BM 8074 (8); Rogaland county, Baasfjord, BM 8052 (18). Denmark: "Michael Sars" st. 828 in Skagerrak, BM 2417 (2) and 2299 (2). Scotland: 58' 46.5' N/ 8' 52' E, SU 49525 (5); 58' 25' N / 5' 00' E, SU 49526 (4). Ireland: 50' 57' N/ 10' 46' W, BM 2516 (1). Canada: Probably Gulf of St. Lawrence, ROMZ 17684 (1), Atlantic coast, ROMZ 17633 (1); 48' 28' 3/ 68' 10' W, ROMZ 10865 (1); 43' 18' N/ 64' 14' W, NFRS, 4965, 4968-9 (8); 43' 39' N/ 59' 08' W, SU 49585 (24); 42' 41' N/ 64' 13' W, MCZ 87765 (11 - United States: Massachusetts Bay, MCZ 82715 (1). Argentina sialis Gilbert Argentina dalia Gilbert, 1890 : 56 (original description, type locality: 29' 54' 80" N/ 118' 01' W in the Gulf of California, -Albatross" st. 8017, 58 fathoms, holotype: in U.S. National Museum). Goode and Bean, 1895.: 510 (type 10« cality incorrectly cited). Jordan and Evermann, 1896 (description after Gil- - bert), Meek, 1916 : 148 (type locality incorrectly listed as California coast). Higgins, 1920 2 86 (800 specimens taken off Huntington Beach, California). Ulrey and Greeley, 1928 : 4 (coast of California, Huntington Beach). Ulrey, 1929 : 8 (listed). Jordan, Evermann and Clark, 1980 : 68 (cdast of. California). Hubbs, 1982 :.105 (redescription of holotype, comparison with original de- scription and Higgins' (1920) specimens). Barnhart, 1936 : 18 (description, coast of southern California). Follett, 1945 : 148, fig. 1 (description, 36' 45' N/ 121 0 58' W, Monterey Bay, California, discussion of type locality and past ' racords). Fitah, 1949 : 48 (6 miles east of Gaviota, California, vicinity of Pt. Conception, California). Clothier, 1950 : 87 (vertebral, dorsal and anal counts; Oceanside, Rincon Pt. near Santa Barbara, Pt. Vicente and Pismo Beach all in California). Fitch, 1950 : 68, fig. 15 (description, summary of all records plus additional records from southern California). DIAGNOSIS.-This species can be separated from A. sphyraena and A. silus by its smooth instead of spiny scales and by its 5 instead of 6 branchiostegals. A. sialis agrees with the above two species and differs from A. strlam and A. elongata in possessing a silvery swimbladder. It can be differentiated from all other species of Argentina except A. silus by its larger number of gill rakers on the lower arm of the 1st arch (14 to 21 in A. sialis, 11 to 15 in A. silus, 5 to 10 in all other species of Argentina. COUNTS AND MEASUREMENTS.-Based on 50 specimens. D. 12 (11- 18); A. 14 (18-15); pectoral 16 (15-18); ventral 11 (10-12); gill rakers 8 (7-9) +1+17 (14-20); lateral line scales 49 (48-51); standard length 91.4-200.5 mm.; preanal 81.6 (77.7-84.2); preventral 51.3 (48.9-55.3); predorsal 45.3 (41.6-47.5); prepectoral. 28.0 (26.0-80.1) head length 28.8 (27.8-81.2); snout 9.7 (8.9-10.5); eye 7.0 (6.1-7.9); maxillary length 6.8 (5.9-7.8); &axillary width 2.0 (1.6,2.8); depth at dorsal Rn 16.9 (12.1- 126 BULLETIN FLORIDA STATE MUSEUM Vol. 3.' 19.0); width behind head 10.6 (9.1-14.9); head depth 18.4 (12.0-15.8); interorbital 6.5 (6.0-7.2); depth of caudal peduncle 6.8 (6.0-7.6); verte- bral count of two specimens, one Monterey Bay, the other Santa Bar- bara, 49 (Clothjer [1950] gives 47 to 49 including the hypural); pyloric caeca 8 and 12 in two specimens from Monterey Bay, 10 in a specimen from Santa Barbara. DESCRIer[ON.-Body elongate but relatively deep and laterally corn- pressed. Greatest depth at the dorsal origin, tapering to the caudal peduncle. A lateral view shows the dorsal profile of the head descend- ing in almost a straight line from the nape and unbroken by the dor- sal edge of the eye. The ventral profile rises to the snout in a gentle curve. The interorbital space between the supraocular canals is broad- ly concave. The, frontals lateral to the canals arch upwards very slight- ly. The head of recently preserved or formaldehyde specimens has clear hyaline tissue covering the top of the head from the interorbital area forward to the level of the nostrils, also on the side of the head over the anterior part of the opercle and pfeopercle, and· continuously connected with the posterior adipose lid of the eye. A narrow band Of this clear tissue runs under the eye and connects with the tissue filling the area bounded by the dorsal profile of the skull and the canals of the jugal and lachrymal. There is a deep fold in the hyaline tissue along a semicircle at the anterior edge of the eye. The tissue' then extends posteriorly as the anterior adipose lid of the eye. The dorsal portion of the maxillary lies tucked under the lachrymal and in some specimens extends to the jugal as well. The posterior edge of the max- illary lies closer to the eye than to the snout. The epidermis covering the ventral edge of the maxillary bears minute papillae which extend onto the upper lip where they meet a similar band from the other side. These structures are visible to the naked eye in some specimens, while in others they are seemingly absent. (These papillae may be what "Hubbs (1932) has called very weak" premaxillary teeth. In examin- ing more than 200 A. sialis, including a number of alizarin specimens, I have encountered no premaxillary teeth, nor indeed have I found them in any other argentinoid fishes. Histological studies on the re- lated A. silus show that maxillary and premaxillary papillae strongly resemble structures on the tongue and in the mouth and pharynx which may be sensory in function.) A small knob is present below the symphysis of the lower jaw. Both the upper and lower jaws are , , broadly rounded. In most specimens the jaws are subequal, but in others the upper or lower may protrude slightly. The palatines and the head of the vomer bear a band frorn 2 to 4 teeth wide composed of small, closely spaced, postariorly directed 1958 COHEN: SUBFAMILY ARGENTININAE 127 teeth. There are 30 to 40 teeth on each palatine and 40 to 50 on the vomer. E. H. Ahlstrom (in litt.) reports the presence of up to 12 pairs of dentary teeth in the larvae. I have not found teeth on the dentaries of adolescents or adults. Very small sens6ry papillae are present along- the lateral edges of the tongue. The ceratobranchial of the 5th gill arch bears 11 to 15 small, conical teeth. Two oval patches of short, conical teeth, 6 to 9 on the anterior patch and 15 to 20 on the posterior patch, appear at the anterior end of the 4th suprabranchial. The gill rakers are elongate and relatively widely spaced and compressed. The 4th raker forward from the angle on the lower arm of the 1st arch is equal in length to between one-third and one-fourth of the inter- orbital distance. The rakers on the upper arm are fewer in number. and much smaller anteriorly. The pectoral fin originates slightly ahead of a vertical through the posterior tip of the operele. The posterior end of the pectoral base forms an angle of 20 to·35 degrees with a plane parallel to the venter. The distance between the bases of the innermost rays is approximately equal to the least depth of the caudal peduncle. The greatest height of the dorsal fin is approximately equal to the depth of the body at the dorsal origin. The greatest height of the anal fin is equal to or slightly less than the depth of the caudal peduncle. The pectoral fins extend more than half the distance between the peeteral .and ventral bases. The ventrals are slightly shorter than the pectorals. The scales lack spines. Specimens with scales show an angular boxlike shape, but the body is more flexible than in A. sphuraena. The peritoneum is silvery on its inner and outer surfaces and is lightly punctate with large chromatophores in a specimen from Mont- erey Bay, while an example ffom Santa Barbara is more densely puric- tate with smaller chromatophores. The gut is unspotted. The outer surface of the posterior curve of the stomach is smooth; the longest pyloric caeca extend to a point between the, anterior and posterior eurves of the stomach. Fr6m the caecal region of the gut the intestine runs to the anus in a straight line. The Ringen are weakly developed , in the intestines of several specimens examined; however, the walls of the intestine are more muscular than in other species of Argentina and not at ail transpareht. The body cavities of all specimens ex- amined contained .considerable amounts of fat. The swimbladder is heavily impregnated with silvery pigment and extends from the level of the anterior curve of the stomach to a point slightly past the 6rigin of the ventral fins. Two types of coloration are present in the fishes I have studied, undoubtedly caused by differences in preservation. Most of the speci- 128 BULLETIN FLORIDA STATE MUSEUM Vol. 8 mens are a uniform light brown with chromatophores scattered over the dorsum and the-upper third of the sides. The muzzle is dusky, as is the nape and'the upper portion of the opercle. The scales are iri- -descent. Other specimens have the body straw colored with a bright silvery band, about equal in width to the pupiI, running along the side of the ·body at the level of the lateral line. The opercles, cheeks; isth- mus, and irises are also bright silver. The body above the silvery band is slightly darker than the body below the band. The muzzle, nape, and dorsaI third of the opercle are dusky. Adolescent specimens bear 8 to 10 dusky blotches along the upper half of the side. DISTRIBUTION.-Argentina sialis has been recorded in California at various localities from Monterey Bay to Huntington Beach. The only other record is the type locality, off the coast of Sonora in the upper part of the Gulf of California (see Follett, 1945 fbr a discussion of the type locality and early records). E. H. Ahlstrom of the U.S. Fish and Wildlife Service has kindly furnished me with unpublished data on the distribution of Argentina sialis larvae during 195.5, when larvae were taken from the Santa Barbara Channel south to Magdalena Bay. Dr. Ahlstrom states that the distributions in 1951 and 1952 were quite similar. During each of the 3 years iarvae were most abundant at a locality slightly north of the north tip of Cedros Island, off the coast of Baja California. The U.S. Fish and Wildlife Service has also col- lected Argentina sialis at three localities inside the Gulf of California, two at the head of the Gulf, the third near the mouth of the Gulf on the west side. HABITS.-The larvae and postlarvae are pelagic, while adults are generally taken closer to the bottom at depths ranging from 11 to 250 meters, the mostcommon occurrence being between 50 and 80 meters. Fitch (1950), in summarizing 19 occurrences of A. sialis off California, concludes that seismic blasting. during offshore oil explorations have shown this species to be more common than was previously thought. One of his records of interest is of a mixed collection of more than 1,000 pounds of A. sialis and the osmerid fish Spirinchus starksi; an- other is his record of three specimens taken from the stomach of a Squalus. I found the stomach of a specimen of A. sialis from Monterey Bay tightly packed with small, stalk-eyed crustacea. SPECIMENS. EXAMINED.~- California: Monterey Bay, SU 86069 (1) and SU 35811 (2); Monterey Co., off Pt. Pinos, SU 49787 (8); San Luis Obispo Co., off Avila, UCLA W49-171 (8); 1-2 mi. below Pt. Conception„SU 15198 (2), Santa Barbara Co., 5-8 mi. east 1958 COHEN: SUBFAMILY ARGENTININAE 129 of Gaviota, SU 15199 (1), Santa Barbara Co., 445 mi. off Santa Barbara, UCLA , + W49-149 (5); vicinity of Santa Barbara, SIO 55-24 (7); 5 mi. southeast of Santa Barbara, SU 16878 (11);, Santa Barbara C6., off Rincon Pt., UCLA W49-178 (7); Santa Barbara Channel off Rincon Pt., SIO 55-23 (58); 442 mi. east of Santa Barbara, SIO 55-27 (38); Santa Barbara Channel, 84' 20' N/ 119' 84' W, SIO 55-26 (87) and SIO 55-25 (12); Santa Monica Bay, SU 35280 (6). Argentina striata Goode and Bean ..4 Argentina striata Goode and. Bean, 1895 : 52, pl. 17, ·fig. 62 (original descrilition, ' type locality: "Albatrdss" st. 2402,28' 86' NJ 850 33' 80" W, Gulf of Mexico south of Cape San Blas, Florida, 111 fms., syntypes: USNM 48858). Jordan and Evermann, 1896 : 526 (descriptidn„ after Goode and Bean, ibid.). Gar- .man, 1899 : 408 (listed). Jordan, Evermann and- Clark, 1980 : 68 (listed). Longley and Hildebrand, 1941 : 14 (Tortugas, 15 ,specimens, largest 190 mm., 160-190 fms,). Schroeder, 1955 : 862 (continental slbpe between Nova Scotia and V.irginia, less than 206 fms.). Springer and Bullis, 1956 : 49 (localities in the Gulf of Mexico). DIAGNOSIS.-This species can be separated from A. silus, A. sphy- raena, and A. sialis by the absence of silvery pigment impregnating its swimbladder. It differs from A. sphi/raena and A. elongata in having a higher pectoral ray count (18 to 20 in, A. striata, 12 to 15 in A. sphy- raena, and 13 to 17 in A. elongata). A. striata also differs from A. elongata in having more vomerine teeth (approximately 45 in A. striata, approximately 25 in A. elongata). COUNTS AND MEASUREMENTS.-Based on 42 specimens. D. 11 (11- 12); A. 13 (12-14* pectoral 19 (18-20); ventral 12 (11-18); gill rakers on Hrst arch 8 (3-4)+1+6; lateral line scale rows 51 (49-52); standard length 98.6-185.6 mm.; preanal 84.1 (81.6-86.5); preventral 55.0 (50.8- 58.8); predersal 47.8 (48.8,51.8); prepectoral 29.2 (26.8-33.0); head length 80.0 (26.1-84.8); snout 9.9 (8.5-11.1); eye 9.4 (7.8-11.6); maxillary length 5.5 (419-6.8); maxillary width 1.7 (1.8-2.0); depth at dorsal fin 12.8 (11.7-15.8); width behind head 11.8 (10.2-12.4); head depth 13.0 (12.0-14.9); interorbital 7.9 (6.9-8.9); depth of caudal peduncle 6.0 (5.5-6.8); vertebral couilt of three specimens 49; pyloric caeca of three specimens 9; branchiostegals 5. DESCRIPTION.-Body elongate, slightly laterally compressed be- tween the head and the dorsal base, more so from that point posteriorly. Greatest' depth between the head and the dorsal origin, tapering to the caudal peduncle. The dorsal profile of the head descends in a gentle curve from the eye to the snout. The dorsal rim of the orbit projects slightly into the profile. The ventral profile of the head rises to the snout in an almost unbroken curve with a very slight angle at the union of the articular and the quadrate. The interorbital is broad 180 BULLETIN F.LORIDA STATE MUSEUM Vol. 8 and Hat between the supraocular canals, which are themselves raised slightly above the flat surface. Lateral to the canals the frontals are still further raised to accommodate the top of the eye. The dorsal edge of the maxillary lies under the lachrymal but does not extend to the jugal. The posterior edge of the maxillary lies slightly closer to the eye than to the snout. Extremely. minute papillae are scattered on the membrane covering the maxillaries and premaxillaries of most , spesimens, but are seemingly absent in several. Both the upper and the lower jaws are rounded and the upper jaw projects slightly. The knob at the symphysis of the lower jaws is reduced or absent. The palatines and the head of the vomer bear a continuous band of small, closely spaced, posteriorly directed teeth. The band varies in width from 1 to 3 teeth on the palatines to 4 to 5 on the vomer. There are about 35 teeth on each palatine and about 45 teeth on the vomer. T'he tongue bears 5 t6 9 strong, recurved teeth. Sensory papillae were not Observed on the tongue; however, many of the Bshes examined had small, hemispherical protuberanees raised from the surface and clus- tered along the tooth-bearing areas of the tongue. The ceratobran- chial of the 5th gill arch bears 18 small, conical teeth. Two patches of short, conical teeth, about 11 teeth on the anterior patch and 17 teeth on the posterior patch, are present at the anterior end of the 4th suprabranchial. The gill rakers on the lower arm of the 1st arch are medium sized, very widely spaced, triangular in shape, and com- pressed. The base of the inner edge of each raker bears a Heshy pad. The 4th raker forward from the angle on the lower arm of the 1st arch is equal to slightly more than one-sixth of the interorbital distance. The 8 or 4 rakers on the upper arm decrease in Size from the angle forward and are markedly smaller than those on the upper arm. The pectoral fin·usually originates in line with a vertical from the posterior tip of the opercles, although the origin may be ahead of or slightly behind this line. The posteri6r end of the pectoral base forms an angle of 40 to 60 degrees with a plane parallel to the venter. The distance between the bases of the innermost pectoral rays is slightly less than the least depth of the caudal peduncle. One specimen was noted which lacked the left ventral, fin. No marks or sears of any sort were present, and dissection showed that the pelvic girdle on the left side was only partially formed. The fish seemed normal in all other respects. The longest dorsal ray is equal to approximately the greatest depth of the body. The longest anal ray equals the distance between the bases of the outermost rays of the ventral fins. The tip of the pectoral fin extends to a point less than half the distance between the 1958 COHEN: SUBFAMILY ARGENTININAE 131 anterior base of the pectoral and the anterior base of the ventral. The ventral fin is approximately equal in length to the pectoral fin. The scales of this species are evidently very deciduous, for I have seen no specimens with complete squamation. Several specimens have a few scattered lateral line scales on the posterior half of the body. The scales are thin and lack spines. The peritoneum is· silvery on its outer surfaee; the dorsal half of the inner surface is lightly colored by large, relatively widely dis- persed chr6mat6phores; these are more closely crowded together on the ventral half of the body cavity and give this area a much darker appearance. The entire gut lacks pigment. The stomach runs for- ward to a point slightly anterior to the posterior tip of the pectorals before swinging downwards. Th'e outer curve is smooth. The pyloric caeca extend to a point betiveen the anterior and posterior curves of the stomach. Distal to the caecal section of the gut the intestine runs straight back to the anus as a narrow, thin-walled, transparent tube. About 12 or 18 weakly developed Ringen are present. The swimbladder is similar to that of A. sphymena, but lacks silvery pigment. In several specimens examined the bladder had an irides- cent sheen. In preserved specimens the ventral half of the body is light brown with a lighter brown band running along the lateral line; while a nar- row dark brown band borders the dorsal edge of the lateral line band. The entire 2orsum of the body is medium brown. Tlie skin of some specimens gives an irideseent sheen and a few examples haye the remnants of what must have been a broad silvery band running the length of the body with the lateral line as a dorsal border. The fins are all lightly flecked with brown except for the median half of the pectorals which are unspotted. The opercles of some specimens are bright silver, as are the cheeks, while in others the silver pigment has dissolved and the Opercle is transparent with a sprinkling of chromato- phores near the bottom and a rather thick concentration of chroma- tophores near the top, The top of the head is transparent posteriorly and translucent farther forward. The muzzle from the nostrils forwaFd is dusky. Several specimens less than 97.0 mm. in standard length bear 11 faint dusky blotches along the dorsal half of each side of the body. I have been fortunate in being able to examine a postlarva of 35 mm. (through the courtesy of Giles W. Mead), and I find it bears a strong resemblance in color pattern to that of the postlarvae of Gloss- anodon leioglossus (see Schmidt, 1918). The specimen I examined bears distinctive transverse stripes along the body, one on the opercle, and one across the caudal fin. None of the stripes is continuous over 182 BULLETIN FLORIDA STATE MUSEUM Vol. 8 the dorsum, but the last 7 body stripes are continuous around the venter. The ventral fins are strongly pigmented, the bases of the pec- torals less so. DIsTRIBUTION.-Until 1955 A. striata had been known from only two localities in the Gulf of Mexico. Recent biological explorations by the U.S. Fish and Wildlife Service in the western Atlantic have added considerably to our knowledge of the distribution of this species. Springer and Bullis (1956) list 26 localities for A. striata, ranging from north of the Yucatan Pehinsula south to Campeche Bay and to the coasts of Texas, Louisiana, Alabama, Florida, and the vicinity of Tor- tugas. Schroeder (1955) captured A. striata at several localities off the Atlantic coast between the northern Gulf of Maine and the Virginia Capes at depths greater than 100 fathoms. I have also examined speci- mens from several places along the east coast of Florida and from the Atlantic coast of Venezuela southeast of Trinidad. HABITs.-A. striata has been captured at depths ranging from 52 fathoms to 250 fathoms, although it has been taken most Often between 100 and 200 fathoms, and most frequently over a mud bottom. All the specimens I examined from the Gulf of Mexico *ere taken in shrimp trawls which, according to Springer and Bullis (1956), selec- tively sample the area above but not directly on the bbttom. Depth probably plays an important part in limiting the distribution of A. striata and perhaps other species of the genus as well. A. striata is conspicuously absent from the catches of the "Oregon" along an ex- tensive stretch of the east coast of Mexico and the southeast coast of Texas and south coast of Louisiana. The station records of the «Ore- gon" given by Springer and Bullis (1956) show that most of the col- lecting in these areas was in depths shallower or much deeper than the known bathymetric range of A. striata. The .stomach: and intestines of several specimens I examined were empty, but one from 168 fathoms at 259 08' N/ 84° 19' W, northwest of Tortugas contained what appeared to be the remains of sponges with siliceous, monaxial spicules. Intermixed with the sponge remains were a considerable number of calcareous protozoan skeletons. Nothing is known of the eggs and larvae of A. striata. SPECIMENS EXAMINED.- Venezu~a: 9' 41' N/ 59' 47' W, USNM 159356 (8); 9' 45' N/ 59' 45' W, USNM 159857 (11) and UF 8026 (6); 9' 58' 3/ 59' 58' W, UF 5287 (6), Gulf of Mexico:.22' 41.9' N/ 86' 41,2' W, CNHM 45728. (1); 19' 37' N/ 92' 40' W, SU·49721,(1); 26' 46' N/ 96' 20' W, UF 1816 (2); 26° 55' N/ 96' 25,5' W, CNHM 46266 (2); 27' 15' 3/ 96' 15' W, CNHM 45050 (1); 29' 15' N/ 88' 18' W, SU 49727 (1); 29" 15.5 N/ 87' 58' W, SU 17441 (1); 29' 20' 3/ 87' 1958 COHEN: SUBFAMILY ARGENTININAE 188 42' W, CNHM 46269 (5) and SU 49722 (4); 29' 27.5' N/ 87' 26.5' W, CNHM 46268 (2) and SU 49728 (8); 28' 80' 3/ 86' 11' W, SU 49724 (2); 28° 19' N/ 86° 04' W, CNHM 45932 (2); 29' 09' N/ 859 59'- W, SU 49725 (4); 28' 48' N/ 85' 40' W, CNHM 46270 (12); 25' 08' 3/ 84«.19' W, CNHM 59886 (21) and SU 49726 (4). Atlantic between Florida and Virginia: 28' 08.5' N/ 79' 55.5' W, SU 49754 (8); 29' 47' 3/ 80' 12' W„. SU 49755 (1); 87' N/ 74' W, MCZ 38305 (1); 87' 88' N/ 74' 15' W, MCZ 88213 (1) Argentina elongata Hutton See subspecies for synonymies and counts and measurements. DIAGNOSIS.-A. elongata differs from A. sphyraena.and A. silus in having only 5 branchiostegals instead of 6 and in having smooth rather than spiny scales. It further differs from the above two species and from A. sialis as well in lacking silvery pigment on the swimbladder. It can also be distinguished from A. sialis by having fewer gill rakers on the lower arm of the 1st arch, (5 to 10 in A. elongata, 14 to 21 in A. dalia). It differs from A. striata in having-a lower pectoral fn counf (18 to 17 in A. elongata, 18 to 20 in A. striata), and iIi having fewer , vomerine teeth (approximately 25 in A. elongata, approximately 45 in A. striata). DESCRIPTION.-Body elongate, adults Alightly laterally compressed to the dorsM fin, more so Dosteriorly; adolescent specimens almost Square in ,.cross-section. A lateral view shows the dorsal profile of the head descending to the snout in a gentle eurve which is not broken by ·the dorsal edge of the orbit. The ventral profile of the head rises more gently in an unbroken line to the snout. Both the upper and 16wer jaws are broadly rounded, with the lower jaw bear- · ing only a slight suggestion of a point below the symphysis. The gill rakers are short, widely spaced, triangular in shape, and compressed/_The bases of the inner edges of the rakers on the lower arm of the Ist arch bear fieshy pads. The pectoral fin usually originates in line with a vertical from the posterior tip 'of the opercle. The posterior end of the pectoral base forms an angle of 85 to 45 degrees with a plane parallel to the venter. The distance between the innermost rays of each pectoral Hn is equal to or greater ·than the least depth of the caudal peduncle. The longest dorsal ray is equal to or greater than the depth of the body at the dorsal origin. The longest anal ray is approximately equal to the dis- > tance between the bases of the outermost rays of the ventral fins. The scales are deciduous, but ,scattered patches on specimens of each of the three subspecies indicate they mold the body into the typi- 184 BULLETIN FLORIDA STATE MUSEUM Vol. 8 cal, boxlike Argentina shape when a full set of scales is present. The scales lack spines. The gut is unpigmented. The swimbladder lacks silvery pigment, but in other respects seems similar to that of A. sphyraena. The anterior end is slightly anterior to the level of the posterior curve of the stomach, while the posterior end is at the level of the ventral Bns. GEOGRAPHICAL VARIATION.-Although series of A. elongata from Australia and New Zealand and A. kagoshimae from Japan are sep- arable by a combination of characters, the three populations do not seem sufficiently divergent morphologically to warrant recognition as full species. Therefore I accord them subspecific rank. Key to the Subspecies of Argentina elongata (See also table 2) la. Gill rakers on lower arm 6f 1st arch 6 (5-7) - 2a. Pectoral rays 16 (15-17); maxillary length 4.2 8.7-4.6) per- cent of standard length ..Argentina elongata kagoshimae p. 141 2b. Pectoral rays 18. (13-14); maxillary length 5,4 (4.9-5.9) per- , cent of standard length . Argentina elongata australiae p. 188 lb. Gill rakers on lower arm of 1st arch 9 (8-IO); pectoral rays 15 (13-16); maxillary length 5.1 (4.2-5.7) percent of standard length . . . . . . . . . . Argentina elongata elongata p. 185 TABLE 2 VARIATION IN NUMBER OF GILL RAKERS ON LOWER ARM OF FIRST ARCH AND IN NUMBER OF PECTORAL RAYS IN THREE SUBSPECIES OF Argentina elongata Hutton. Gill rakers Pectoral rays 5 6 7 8 9 10 18 14 15 16 17 A. e. kagoshimae 2 10 692 A. e. australiae 9 1 8 3 A. e. elongata 5 12 18 1 9 9 1 A. e. australiae is closer to A. e. kagoshimae with regard to gill raker counts, while it agrees with A. e. elongata in having a relatively longer ' maxillary than kagoshimae. A. e. elongata is intermediate between the ' other two forms with regard to pectoral rays. Other less distinctive 1938 COIIEN: SUBFAMILY ARGENTININAE 185 differences between the subspecies are presented in the descriptions which follow. NOMENCLATURE.-Argentina elongata Hutton, 1879 should not be confused with Argentina e/ongata Weiler, 1943, a species described from a fossil otolith from the early Tertiary of southern Rumania. I informed Dr. Weiler that his fossil species is a homonym of A. elongata Hutton; he replied that he intends to emend the name of his fossil spe- cies in one of his next publications. Argentina elongata elongata Hutton Figure 7 Figure 7.-Photograph of a watercolor sketch of the holotype of Argentina decagon by F. E. Clarke. Published with the permission of the Dominion Museum, Wellington, New Zealand. Argentina e/(mgata I Iutton, January 1879 : 58 (original description, type locality: Cape Campbell, Marlborough, South Island, New Zealand, holotype: British Museum), Gunther, 1887 : 218, pl. 55, fig, B (redescription and figure of holotype). Goode and Bean, 1895: 52 (in key) Hutton, 19()4 : 51 (listed). Waite, 1907 : 10 (listed). Waite, 1911 : 161, pl. 24 (description, synonyiny, Otago to Bay of Plenty, Chatham Islands, 16-105 fathoms). Waite, 1928 : 63, fig. (in part, description, distribution), Phillipps, 1927a : 14 (synonymy). Phillipps, 19271, : 11 (listed). Graham, 1989 : 402 (in and around Otago Har- bor, New Zealand), Phillipps, 194() : 45, fig. (description, synonymy), Gra- ham, 195,3 : 121, fig. (description, distribution, habits, food). Munro, 1957 : 15, fig. 199 (in part, description, distribution). Argentina decagon Clarke, May 1879 : 295, pl. 14 (original description, type 10- cality: South Spit, Hokitika, South Island, New Zealand, holotype: unknown). Gill, 1898a : 112 (listed). Waite, 1907 : 10 (listed) Argentina sphyraenu Gunther , 1887 : 218 ( in part , listed). Hutton , 1904 : 51 ( in part, listed). COUNTS AND MEASUREMENTS.-Based on 21 specimens. D. 11 (10- 12); A. 18 (11-14); pectoral 15 (18-16); ventral 11 (11-18),gill rakers on 136 BULLETIN FLORIDA STATE MUSEUM Vol. 3 - first arch 2 (1-8) +1+6 (6«7); lateral line scale rows 54 (51-54); standard length 78.9-216.0 (Graham [1958] states this fish reaches a size of 18 inches [511.2 mm.]); P.reanal 82.8 (79.2-85.2); preventral 51.7 (49.3-59.7); predorsal 45.2 (42.5-47.6); prepectoral 28.0 (25.2-80.0); head length 28.0 (23.2-30,0); snout 9.8 (8.9-10.8); eye, 8.0 (6.8-9.2); maxiJlary length 5.1 (4.2-5.7); maxillary width 1.5 {1.3-1.6); depth at dorsal fin 18.3 (11.9.15.8); width behind head 9.6 (8.4-11.8); head depth 12.1 (10.9-18.6); interorbital 6.6 (6.2-7.2); depth of caudal peduncle 6.3 (5.7- 6.9); vertebral count of two specimens 53 and 54; pyloric caeca in two I specimens 12 and 18; branchiostegals 5. DEsCRIPTION.-Greatest depth at d6rsal origin, tapering to caudal peduncle. The medial area of the interorbital, between the supraorbi- tal canals, is flat. The canals are only slightly raised above the medial area. The supraorbital portions of the frontals lateral to the canals are raised only slightly if at all from the level of the canals. Well- developed adipose eyelids are present. The posterior lid is continu- ous with a clear, hyaline layer covering the cheek, while the anterior lid extends forward f6r a distance about equal to the width of the iris and then doubles back on itself for about a third of the above dis- tance and then runs forwards to the level of the nostrils. The area bounded by the frontals and the canal running fr6m below the eye to a point immediately below the nostrils is covered by the interior exten- sion of the adipose eyelid. The dorsal edge of the maxiilary usually lies 'under the lachrymal only, but in some specimens extends to the joint between the lachrymal and the jugal. The posterior tip Of the maxillary is about midway between the tip of the eye and the snout. The membranes covering the ventral edge of the posterior portion of the maxillary bear papillae only in larger specimens. Much wdaker papillae are present on the epidermal covering of the premaxillary in larger specimens. The upper jaw projects slightly in most specimens but is subequal in several. The palatines and the head of the vomer bear an irregular band of small, closely spaced, posteriorly directed teeth from 2 to 3 teeth wide. There are about 80 teeth on each palatine and about 25 teeth on the vomer. The tongue is armed with 4 to 8 strong, recurved teeth. Strongly developed papillae are present along the lateral edge of the tongue. The ceratobranchial of the 5th gill arch bears 15, smaII con- ical teeth. Two oval-shaped patches of short conical teeth, about 8 teeth on the anterior patch and 17 on the'posterior patch. are present at the anterior end of the 4th suprabranchial. The 4th raker forward from the angle on the lower arm of the first arch is equal in length to 1958 COHEN: SUBFAMILY ARGENTININAE 187 about one-seventh of the interorbital distance. The rakers on the upper arm are much reduced in size. The peritoneum is lightly stippled with chromatophores; these are closer together on the ventral portion and make this area slightly darker. The outer curve of the stomach bears a small protuberance. The thin-walled intestine has about 13 well-developed Ringen in .a specimen 170 Inrn. iii standard length. A considerable amount of ·fat is present in the body cavity. Several examples I examined were heavily parasitized with roundworms. The body color in alcohol is a uniform brownish yellow. The eau- dal and dorsal fins bear a few scattered chromatophores, the other fins are unmarked. The opercle is silvery and the top of the muzzle is darkly pigmented. The top 6f the skull behind the eyes is reddish -brown where the pigment on the inner surfaces of the brain case shows through the transparent cranium. Graham (1953) describes live Bsh - as follows: "... glistening, silvery body, though in medium sized fish the colour is yellowish with pale brown criss-cross bands, but these bands do not extend below the lower half,of the body. There is a bright, polished, glistening silvery band along the middle ofeach side over the lateral line. The fish has blue eyes, with a silvery patch above the pupil." DISTRIBUTION.-The northernmost record for A. e. elongata is the Bay of Plenty on the North Island of New Zealand. It has been taken at various localities along the east coasts of both North and South Islands as far south as Dunedin and has also been recorded from the Chatham Islands. -Hokitika, the type locality of A. decagon on the west coast of South Island, is the only locality recorded for A. e. elon- gata in the Tasman Sea. HABITS.-The known bathymetric range is between the surface and 105 fathoms. Graham (1958) gives considerable data on the habits of this Bsh around Otago Harbor, where he has found them most 6ften in tidal areas. They are taken on hook and line, but mostly in seines and trawls. He also notes seasonal migrations into and out of the har- bor, but his data on this point seem rather confused. He gives further information indicating the populations of this species in the Otago area are subject to considerable fluctuations in abundance. He notes popu- latjon increases for 1916, 1924 to 1925, and 1980 to 1983. A. e..elongata was canned at one time in New Zealand; however, a decline in the catch ended the enterprise. · It is of. interest to note a general agree- ment between those periods of abundance of Argentina and periods of abundance Graham (1958) gives for the New Zealand species of 188 BULLETIN FLORIDA STATE MUSEUM Vol. 8 Sardinops. He states the most common food of Argentina consists of, "crabs (especially Cyclograpsus .lavauxi), Whalefeed, isopods and worms (especially Glycera ouigera)." He also found that red cod (Physiculus bachus) and groper (Polyprion oxygeneios) were predators of A. e...Blongata. Live Argentina invariably died from scale loss When transported for aquarium exhibits. NOMENCLATURE.-This form was first deseribed in January of 1879 by Hutton as Argentina elongata. In May of the same year it was re- described as Argentina decagon by Clarke, who had evidently not seen Hutton's description. Gunther (1887) recognized A. elongata, .but placed A. decagon in the synonymy of A. sphyraena. He was followed in this by Hutton (1904). In 1907; Waite listed both A. elongata ~nd A. decagon with no comment, but in 1911 he pointed out Gunther's erfor and placed A. decagon in the synonymy of A. elongata, an ar- rangement that all later authors have followed. SPECIMENS EXAMINED.- New Zealand: North Island.1-Bay of Plenty, DMW (1); Castlepoint, on east coast, SU 49782 (8). South Island.-Off Cape Campbell, SU 49527 (4), SU 49729 (12), DMW (2). Argentina elongata australiae, new subspecies Argentina elongata McCulloch, 1911 : 18 (description, Oyster Bay, Tasmania; east coast of Flinders Island, Bass Strait, 80 fins.; 15 miles off Norah Heads New South Wales, 45-58 fms.). McCulloch, 1914 : 87 (80 miles west of tile merid- ian of Eucla, Western Australia, 80-120 fms.) McCulloch, 1921 : 29, pl. 6, fig. 61a (key, references, figure of A. e. elongata from Waite, 19I1). Waite, 1921 : 42, fig. 61 (synonymy, record from McCulloch, 1914; figure of A. e. elongata from Waite, 1911). Waite, 1928,: 68, fig. (in part, description, figure of A. e. elongata from Waite, 1911). Lord and Scott, 1924 : 84 (description, Tasmanian records). Waite, 1928 : 4 (listed). McCulloch, 1929 . 44 (listed). Norman, 1987 : 52 (42' 40' S/ 148' 27' 30" .E, off Ta5mania). Fairbridge, 1951 : 122 (food of. Neoplatycephalus macrodon, localities along the southeast coast of Australia between Newcastle, New South Wales and Cabo Island, Victoria; 88-65 fms.). Whitley, 1954 : 29 (seined off· Pt. Phillip, Victoria). Munro, 1957 : 15, fig: 199 (in part, description, distribution, figure of A. e. elongata from Waite, 1911). (OUNTS AND MEASUREMENTS.-Based on 12 specimens. D. 11 (10 12); A. 18 (12-18); pectoral 18 (18-15); ventral 12 (11-18); gill rakers 1 (1-2) + 1 + 6 (6-7); lateral line scale rows 54 (51-54); standard lehgth 61.1-123.0 mm. (McCulloch [1911] records a specimen 125 mm: long); , preanal 88.8 (82.5-85.0); preventral 58.0 (50.6-54.8); predorsal 45.5 (43.9-47.0); prepectoral 28.7 (26.8-80.1); head length 28.8 (22.8-80.4); snout 9.9 (9.4-10.4); eye 8.2 (7.5-9.7); makillary length 5.4 (4.9-5.9); 1958 COHEN: SUBFAMILY ARGENTININAE 189 maxillary width 1.4 (1.2-1.5); depth at dorsal En 10.8 (10.1-11.5); width behind head 10.8 (9.5-11.6); head depth 11.0 (10.5-11.4); interorbital 7.1 (6.6-7.8); least depth of caudal peduncle 5.5 (4.8-6.2); vertebral count of one specimen 50; pyloric caeca in one specimen 6; branchio- stegals 5. DESCRIPTIoN.-Greatest depth between head and dorsal origin, tapering to the caudal peduncle. The medial area 6f the interorbital, between the supraorbital canals, is weakly concave. The canals are only slightly raised above the medial area. The supraorbital portions of the frontals lateral to the eadals are flush with, or below the level of the canals. Remains of adipose eyelids are present in several speci- mens. The dorsal edge of the maxillary lies under the lachrymal only and does not extend to the jugal. Th6 posterior edge of the maxillary lies midway between the snout and the ant*ribr edge of the eye. No papillae are found on the epidermal covering of the maxillary and pre- maxillary, possibly because I have only relatively small specimens. The upper jaw projects slightly. The palatines and the head of the vomer bear an irregular band of small, closely spaced, posteriorly directed teeth from 2 to 4 teeth wide. There are about 25 teeth on each palatine and about 25 teeth on the vomer. The tongue is armed with 5 to 9 str6ng, recurved teeth. Papillae are present on the tongue. About 8 small, conical teeth are present on the ceratebranchial of the Sth gill arch. Two oval-dhaped patches of short, conical teeth, about 5 teeth on the anterior patch and 18 on,the posterior patch, are present at the anterior end of the 4th s-uprabranchial. The 4th gill raker for- ward from the angle on the lower arm of·the first arch is equal in length to abbut one-ninth of the interorbital distance. The rakers on the up- per arm are reduced to small stubs. The tip of the pectoral fin extends to a point less than halfway between the anterior base of the pect6ral and the anteYior base of the ventral. The ventral fin is approximately the same length as the pectoral. The peritoneum has a silvery outer surface and is punctate on its inner surface. The muscular stomach runs posteriorly only for a short distance after entering the body cavity and then swings downward. The outer surface of the curve is smooth. The pyloric caeca extend past the curve of the stomach. The Ringen of an example 86 mm. in standard length are so poorly' differentiated from the Grundnetze that it is impossible to count them with certainty. Many of the paratypes have been preserved in.a manner that has left a considerable amount of silvery pigment on the specimens. The body of the fish below the lateral line is straw colored except for a 140 BULLETIN FLORIDA STATE MUSEUM Vol. 3 silver stripe running from the head to the base of the caudal fin imme- diately below the lateral line. At its anterior·end the stripe is equal in width to the vertical diameter of the pupil; it becomes narrower as it runs posteriorly. The body above the lateral line is slightly darker. Along the upper half of the body are 9 to 11 dark blotches which are not continuous over the dorsum. In some specimens the posterior 5 blotches alternate with smaller blotches placed along the lateral line. The opercle, cheeks, isthmus, and iris are a bright silver. The posterior part of the cranium is transparent, the anterior part translucent, and the upper part of the muzzle is dusky. The scales are iridescent. DISTRIBUTION.-First recorded from Australian waters by McCul- loch (1911) who found the subspecies at Tasmania, Bass Straits and the southern coast of New South Wales. Since that time it has been recorded north to Newcastle along the coast of New South Wales. The westernmost record is McCulloch (1914) near Eucla on the Great Australian Bight. HABITS.-Most of the available information on the habits of this flsh is given incidentally by Fairbridge (1951) in his study of the bi- ology of Neoplatycephalus macrodon, a species taken by the southeast- ern Australia trawl fishery. Argentina was taken at depths ranging from 38 to 65 fathoms, generally over a mud bottom and in company with Apogonops anomalus, Trachichthodes a#inis, and Chlorophthal- mus nigripinnis as well as Neoplatycephalus. Argentina was found to be of some importance as a forage species for Neoplatycephalus, especial- ly in Tasmania. Fairbridge notes that most of the Chlorophthalmus taken were juveniles and that size and abundance were correlated with depth. This may also be true of Argentina in Australia, as all the specimens I have examined from fisheries collections have been juve- niles and the southeastern Australia trawl Bshery is centered in waters generally shallower than the bathymetric center of distribution of oth6r species of Argentina. NOMENCLATURE.-The name australiae refers to Australia, the hab- itat of this subspecies. SPECIMENS EXAMINED.- Australia: Holotype.-New South Wales, off Eden, collected November 1948, received from Ian Munro, SU 49780. Paratypes.-Data as for the holotype, SU 49781 (1); Victoria, within 50 miles of Lakes Entrance (88' S/ 148' E), 20-80 fathoms, collected by 1958 COHEN: SUBFAMILY ARGENTININAE 141 ' . fishermen from the throats of Neoplatvcephalus macrodon from January to June 1955, received from Dudley E. Kurth, SU 49528 (12). Argentina elongata kagoshimae Jordan and Snyder Argentina kagoshimae Jordan and Snyder; 1962 : 590, fig. 5 (original description, holotype SU 6854, poor condition, type locality: Kagoshima, Kyushu, Japan). Jordan, Tanaka, and Snyder, 1918 : 46 (listed). Schmidt, 1981 :.21 (descrip- tion, Nagasaki, Kagoshima). Tanaka, 1931 : 49 (in part, listed, synonymy). Okada, Uchida, and Matsubara, 1985 :'78 (description). Matsubara, 1948 : 64,Egs. 16, 17 (description,.references, comparisons Kith A. semifasciata, Kum- ano-Nada, off Maisaka), Kamohara, 1952 : .13 (description, Prov. Tosa, Shi- koku, Japan). Bahlke, 195.3 . 16 (holotype in Stanford Museum). Matsubara, 1955:216, pl. 14, fig. 54 (description, distribution). COUNTS AND·MEASUREMENTS.-Based on 17 specimens. D. 11 (10- 12); A. 12 (11-13); pectoral 16 (15-17); ventral 12 (11-12); gill rakers 2 (0-3) +1+6 (5-6); iateral line scale rows 54 (51-54); standard length 61.0-200.5 mm. (Kamohara [1952] reports a specimen 220 mm. in stand- ard length); preanal 85.0 (83.1-87.2); preventral 50.7 (48.0-52.5); pre- dorsal 48.1 (41.8-45.2); prepectoral 26.2 (28.8-27.9); head length 26.6 (24.0-27.9); snout 9.1 (8.2-10.0); eye, 7.8 (5.9-9.8); maxillary length 4.2 (8.7-4.,6); maxillary width 1.4 (1.2-1.5); depth at.dorsal fin 12.5 (IO.4- 14.7); width behind head 11.0 (9.4-12.1); head depth 11.1 (10.4-12.6); interorbital 7.1 (6.4-8.2); depth of caudal peduncle 5.8 (5.5-6.4);.verte- bral count 51 in two specimens; pyloric caeta in two specimens 12 (Matsubara [1943] gives 9 to 11); branchiostegals 5. DESCRIPTIoN.-Greatest depth at the dorsal origin„ tapering to the caudal peduncle. The medial area of the interorbital between the supraorbital canals is weakly concave. The canals are only slightly raised above the medial area. Supraorbital portions of the frontals lateral to tjhe canals are raised Only slightly, if at all, from the level of the canals. The dorsal edge of the maxillary,lies under the lachrymal only and does not ext6nd to the jugal. The posterior edge of the max- illary lies closer to the snout than to the eye. The membrane cover- ing the ventral edge of the posterior portion of the maxillary bears minute papillae only in larger examples. Papillae are seemingly ab- sent on the covering of the premaxillary. The upper jaw projects slightly. The palatines and the vomer bear a band from 2 to 4 teeth wide composed of small, closely spaced, posteriofly directed teeth. There are approximately 45 teeth on each palatine and about 25 teeth on the vomer. The tongue is armed with 5 to 9 strong, recurved teeth. Papillae about half as long as the tongue teeth are present on the lat- eral margins 6f the tongue (see Matsubara, 1948, fig. 17). On the ceratobranchial of the Sth gill arch are 8 to 12 small, conical teeth. Two oval-shaped patches of short, conical teeth, about 7 teeth on the -142 BULLETIN FLORIDA STATE MUSEUM Vol. 8 anterior patch and 14 teeth on the pos~erior patch, are present at fhe anterior end of the fourth suprabranchial. The 4th raker from the angle on the lower arm of the 1st arch is equal in length to about one- eighth of the interorbital distance. The rakers on the upper arm, when present, are reduced to small stubs. The tip of the pectoral fin extends to a point about half the dis- tance between the anterior of the pectoral base and the anterior of the ventral base. The ventral fin is approximately the same length as the pectoral fn. The peritoneum is lightly stippled with chromatophores which, be- ing closer together on the ventral portion give this area a slightly darker appearance. The stomach runs posteriorly_ only li short dis- tance before swinging downwards. The outer surface of the curve ·is smooth. The longest caeca extend beyond the curve of the stomach. The gut runs straight back from the caecal portion of the stomach. The thick-Mailed intestine has 10 or 11 weakly developed Ringen in a speciknen 121 min. in standard length. A considerable amount Of fat is present in the body cavity. -The color of preserved specimens is light brown ventrally, slightly darker dorsally. Along each side of the body above the lateral line are 10 or 11 dusky, well-defined blotches in specimens as large as 160 mill. in- standard length. Between each of the well-defined blotches is a lighter-colored smaller blotch. The fihs have only an occasional chromatophore. The muzzle and the dorsal third of the opercle is dusky; the remainder of the head is unpigmented. DISTRIBUTION.-As far as known this subspecies is restricted to Japan, where it is found along the coasts of Kyushu, Shikoku, and the southern half of Hotishu. The northernmost record along the Pacific coast of Japan is given by Matsubara (1943) as "off Maisaka" (roughly 84° 45' 3/ 187° 40' E). A. e. kagoshimae has been taken on Kyushu Island as far south as Kagoshima and also at Nagasaki. There is no record in the literature for the Sea of Japan; however, I have examined a single specimen with localiti' data "probably U6zu or Namerikawar both of which are on Toyama Bay on the Japan Sea. ' HABITS.-Matsubara (1948) has taken this form at depths of ap- proximately 125 fathoms. SPECIMENS EXAMINED.- Japan Honshu: probably Uotsu or Namerikawa, on Toyama Bay, Japan Sea, TU 46255 (l); Mie Prefectilre, off Owashi, SU 49530 (10). : Shikoku: Kochi Prefecture, off mouth of Ukitsu River, TU 39178 (1) 1958 COHEN: SUBFAMILY ARGENTININAE 148 Kvushu: Nagasaki, TU 2689, (1), TU 82251-3 (3); Kagoshima, SU 6854 (1), hulo- type; Kagoshima, Sakurajima, off Kurokami Village, TU 44274 (1); Kagoshima Market, TU 24146 (1), TU 28882 (1). - GENUS GLOSSANODON GUICHENOT Glossanodon Guichenot, 1867 : 7 (type species Argentina leioglossa Valenciennes, , 1848, bx original designati6n). Argentinine fish that have the medial ends of their maxillaries in contact or separated by a narr6w space less than one-fourth the width of the narrow, underlying mesethmoid. The medial portion of the maxillary is enlarged in some species. Distal to the median line the maxillary sends down a prong (not shown in fig. Sa) which articulates with the posterior face of a prong sent up by the premaxillary. The 'vomerine dentition consists of small, conical teeth placed in an irregular row across the head of the vomer and set back from the edge a distance equal to three or more times the length Of a tooth. The palatine dentition varies in the subgenera. The palatine and vomerine teeth are easily distinguished, as an area of unossified carti- lage generally separates the two b6nes. The lingual teeth, when pres- ent, are placed only at the anterior end of the tongue and,are absent along the lateral edges. Anteriorly the body is laterally compressed slightly, posteriorly more so. A slight but distinct knob is usually present below the sym- physis of the dentaries. The posterior tip' of the. maxillary is closer to the eye than to the snout. The distance between the bases of the innermost rays of each pectoral fin is approximately equal to the least depth of the caudal fin. As far as known, the scales lack spines. HISTORY.-Guichenot proposed the genus Glossanodon for Argen- tina leioglossa on the basis of the absence of tongue dentition. The genus never gained recognition. Furthermore, more careful observa- .tions by Schmidt (1918) showed that small teeth are presen-t on the tongues of many specimeits. Chapman (1942) listed Glossanodon as a synonym of Argentina. In this paper I have removed Glossanodon from synonymy and assigned to it the species G. semifasciatus, G. polli, G. Zineatus, and G. pygmaeus, as well as the type species. Key to the Subgenera and Species of Glossanodon la. Distance from v®t to anterior end of anal fin base equal to or less than half the depth of caudal peduncle; lateral line scale rows 48-to 56; pectoral rays 18 to 22; dentary teeth present in most adults. (Subgenus Glossanodon p. 144.) ...... 2 144 ' BULLETIN FLORIDA STATE MUSEUM Vol. 3 2a. Dentary teeth not extending more than half the distance from the angle of the gap to the symphysis of the dentaries 8 Sa. Stomach with dark pigment; medial ends of maxillaries without large knobs . . lossano(ion leioglossus p. 144 Sb. Stomach without dark pigment; medial ends of maxilla- ries with large knobs : Glossanodon semifasciatus p.. 148 2b. Dentary teeth always extending more than half the distance from the angle of the gape to the symphysis of the dentaries, usually to the symphysis ............4 4a. Lateral line scales 48 to 51; branchiostegals 5 .. ........ . . Glossanodon polli p. 151 4b. Lateral line scales 56; branchiostegals 4 . . . . ...... . . . Glossanodon Zineatus p. 154 lb., Distanc6 from vent to anterior end of anal fin. base equal to or ~ greater than the depth of the caudal peduncle; lateral line scales 43 to 46; pectoral rays 12 to 14; dentary teeth absent in adults. (Subgenus hosoarchus p. 155.) . . Glossanodon pygmaeus p. 156 Subgenus Glossanodon Guichenot The. subgenus Glossanodon differs from the subgenus Prosoarchus in .having the anus closer to the anterior end of the anal fin base, a distance equal to one half or less, than .the depth 'of the caudal ped- uncle in Glossanodon, compared to a distance equal to or greater than the depth of the caudal peduncle in Prosoarchus. Furthermore, den- tary teeth are present in most adult specimens of each of the species of Glossanodon, while dentary teeth are absent in adult specimens of Prohoarchus. Glossanodon also has a lateral line count of 49 to 56 compared to 44 to 46, in Prosoarchils, and a pectoral ray count of 18 to 22 compared to 12 tb 14 in Prosoarchus. Glossanodon (Glossanodon) leioglossus (Valenciennes)· Argentina'leioglossa Valenciennes, 1848 : 27, fig. 624 (original description, type locality: harbor of Alger, Algeria; syntypes: in Paris Museum, poor Egure). Guichenot, 1850 : 97 (description of syntypes). Schmidt, 1918 : 1,. figs. 1 (1), 2.5, chart·1 (comparison with A. sphyraena, figures of.postlarvae, Cadiz Bay, southern Mediterranean coast of Spain, Corsica, Sardinia, near Naples, Tu- nisia, Algeria). De Buen, 1926 : 49 (Gulf of. Mdlaga, Spain). Sanzo, 1938: 22, pl. 2, figs. 1-11 (description of developmental stages from egg to 42.4 mm., comparisons with developmental stages of A. sphyraena, Tyrrhenian Sea, Straits of Messina, Ionian Sea). De Buen, 1935 : 47 (references). Bertin, 1940 : 807 (paratypes = syntypes in Paris Museum). Lozano Rey, 1947 : 181, fig. 29 (description, distribution). Karlovac, 1949 : 127, figs. 1-8, 1 map '(Post- larvae in the Adriatic). 1958 COHEN: SUBFAMILY ARGENTININAE +145 2 Argentina lioglossa Gunther, 1866': 204 (emendation of A. leioglossa, description) Gunther, 1887 : 217 (listed). Carus, 1898 : 556 (description). Soljan, 1948 : 185, fig. (description). Glossanodon teloglossum Guichenot, 1867 : 7 (description). Argentina sph!/raena Giglioli, 1880 : 12 (in part, variability in dentition of tongue, lumps A. spht/raena and A. lioglossa) Goode and Bean,- 1895 : 51 (in part, description, synonymy). Meek, 1916 : 148 (in part, distrib®On) Argentina situs (non Ascanius) Mzirray and Hjort, 1912 : 894, fig, 261 (26' 08' N/ 15' 00' W, .figure is A. silus [Ascanius] after Smitt). Fowler, 1936 : 196, fig. 84 (after Murray and Hjort, 1912, .figure is A. silus [Ascanius] after Goode. and Bean). DIAGNOSIS.-This species can be distinguished from G. semifasci- atus in having dark pigment on the stomach, and in lacking knobs 6n the medial ends of the maxillaries. G. leiogloss us differs from G. lineatus in having a lower anal fin ray count (llto 13 in G. leioglossus, 15 in G. lineatus), a lower lateral line count (49 to 50 in G. leioglossus, 56 in G. lineatus), and a higher branchiostegal ray count (5 in G. Zeio- glosms, 4 in G. lineatud). G. leioglossus differs from G. polli and G. lineatus in· having dentary teeth extending only half the distance from the gape to the symphysis, while in the latter two the dentary teeth extend to, or almost to.the symphysis. COUNTS AND MEASUREMENTS.-Based on four specimens. (Also in- cluded are several measurements given by Schmidt (1918) based on six specimens 25.5-88.0 mm. in standard length.) D. 18 (13-14); A. ·12 (11-13); pectoral »21 (20-22); ventral 12; gill rakers on Brst arch 11 + 1 + 24 in a specimen from the Adriatic; lateral line scale rows 49-50; ' standard length 51.0-111.0 (Karlovac [1949] lists a specimen 6f 14 cm.); preanal 82.7 (81.3-85.2); preventral 58.8 (52.7-54.9), Schmidt gives 51.0- 54.8; predorsal 49.5 (49.4-49.7), Schmidt giyes 47.1-50.0; prepectoral 80.8 (29.7-32.8); head length 30.7 (29.9-82.3), Schmidt gives 26.0-30.1, snout 9.5 r9.0-10.8); eye 9.8 (9.0-9.4); maxillary length 8.4 (8.0-8.6); maxillary width 8.0 (2.9-3.4); depth at dorsal fin 12.8 (11.4-18.9); width behind head 10.6 (9.9-11.9); head depth 18.7 (18.5-14.1); interorbital 7.3 (7.2-7.4); depth of caudal peduncle 6.4 (5.5-7.2); vertebral count of a specimen from the Adriatic 49; branchiostegals 5. DESCRIPTION.-LGreatest depth behind the head, tapering to , the caudal peduncle. A lateral View shows the dorsal profile of the head descending in a gentle curve from the nape to the snout. The dorsal profile is interrupted slightly by the eye in the largest specimen ex- amined, not in the others.. The ventral profile rises in a gentle curve from the posterior edge of the preopercle to the union of the articular and the quadrate,,from whence it runs in. a straight line to the tip of 146 BULLETIN FLORIDA STATE MUSEUM Vol. 3 i the lower jaw. The interorbital area between the supraocular canals is broad and flat posteriorly, weakly convex dnteriorly, The canals rise slightly above the level of the median area and the frontals lateral to the canals arch slightly upwards from the canals. The dorsal edge of the maxillary lies under the lachrymal and extends back under the jugal as well. The medial ends of the. max- illaries are not enlarged into knobs, but instead bear articulating sur- faces which receive enlarged knobs sent up by the premaxillaries. The palatines bear 10 to 20 short, widely spacedi conical teeth in an irregular row which is single posteriorly and 2 to 3 teeth wide an- teriorly. The 10 to I5 vomerine teeth are arranged in an irregular band I to 8 teeth wide which resembles the palatine teeth bands. The tongues of all four specimens examined bear teeth, 2 teeth in 2 fishes, 8 teeth in each of the others. Schmidt (1918) found from 3 to 4 small teeth on the tongues of all specimens he examined. Two to eight short, conical teeth are scattered on the inner face of the dentary as -far forward as the midpoint 6f the distance between the angle of the gape and the symphysis. Pharyngeal teeth were not examined. The gill rakers are elongate- and closely spaced. The 4th raker forward from the angle on the lower arm of the 1st arch is equal to from one- third to one-half of the interorbital distance. The posterior ends 6f the pectoral bases form angles of 45 to 60 degrees with a plane parallel to the venter. The tips of the pectorals extend to a point slightly farther than the midpoint of the distance between the pectoral base and the base of the ventrals. The ventral fin is equal in length to the pectoral fin. The inner surface of the peritoneum is colored a light brown ven- trally by very small brown chromatophores; dorsally the peritoneum is transparent. The dusky stomach has ah e16ngate elbow which reaches to the level of the posterior tips of the pectoral fins. Th2 en- tire gut is stippled with brown. The swimbladder, which lacks silvery pigment in a specimen from the Adriatic, originates ,above the posterior curve of the stomach. and terminates at the level of the tips of the ventral Ens.,, A preserved specimen from 'the Adriatic has a light chocolate ground color. A lighter brown band, equal in width to the djameter of the pupil, runs ·down the midline of the body. A narrower and darker brown stripe runs along the dorsal edge of the light-colored band. The muzzle, upper third of the opercle, and posterior portion of the skull including the brain case are pigmented, while the remain- 1958 COHEN: SUBFAMILY ARGENTININAE 147 der of the head is uncolored. The throat and the transparent opercles bear a few remnants of silvery pigment. The dorsal and anal fins are slightly pigmented, the other Bns are not. The three smaller speci- mens are badly faded, but it is possible to make out 8 to 10 dusky blotches along the dorsum. Sanzo (1988) gives color figures of larvae and postlarvae showing 9 dark transverse bands along the side of the body and a broad horizontal silver stripe down the side of a specimen 42.4 mm. long. DISTRIBUfION.-G, leioglossus occurs in. the western part of the Mediterranean from near the straits of Gibraltar to the Ionian Sea. Karlovac (1949) reports the species from a considerable number of localities in the southern two-thirds of the Adriatic. It has been re- corded from two .localities in the eastern Atlantic, one the Bay of Cadiz and the other north of the Canary Islands at 26° 03' 3/ 15° 00' W. The latter record was referred by Murray and Hjort (1912) to A_rgentina silus; I have examined their specimens and find, as Schmidt (1918) suggested, that the specimens are G. leioglossus. HABITs.-The young are pelagic in the Mediterranean at depths of 100 to 200 meters according to Schmidt (1918), while Karlovac (1949) finds postlarvae in the Adriatic between 80 and 860 meters and adults closer to the surface. The Murray and Hjort specimens from the Canary Islands were taken in a trawl from between 267 and 280 meters. Schmidt (1918) states that spa-wning in the Mediterranean takes place in the winter and spring, and Karlovac (1949) finds the Adriatic population in .general agreement with this. Karlovac also notes that adults together with postlarvae are taken only in the southern third of - the Adriatic, while farther north only postlarvae, presumably carried , by cu~rents, are found. NOMENCLATURE.-Aftef the original discovery of G. leioglossus. in 1848; the species Was not taken again until 1880, when Giglioli exam- ined what was undoubtedly a mixed collection and, considering only the dentition of the tongue as a specific character, lumped the species with Argentina sphyraena. He was followed by several workers until 1918, when Schmidt was able to show other ways of separating the two species. SPESIME*S EXAMINED.- Yugoslavia. Adriatic, 48' 18.4' 3/ 16' 25.5' E, SU 49786 (1). Canary Islands: 26' 08' N/ 15' 00' W, "Michael Sars" st.· 89 B, BM 8238 (8). 148 BULLETIN FLORIDA STATE MUSEUM Vol. 8 Glossanodon (Clossanodon) semifasciatus (Kishinouye) Argentina semifasciata Kishinouye, 1904a : 110 (original description, type locality: Toyania Bay, holotype: unknown). Kishinouye, 1904b : 197. Jordan, Tanaka and Snyder; 1913 : 48, (Toyama Bay, Nagasaki), Jordan and Hubbs, 1925 : 152 {Sea of Japan)., Tanaka, 1981 : 49 (in part, listed). Okada, Uchida and Matsubara, 1985 : 78, pl. 21, fig. 1 (description, colored figure).- Matsubara, 1940 : 8I7 (Kambara, near Numazu, Sizuoka Prefecture). Matsubara, 1943.: 66, figs. 18, 19 (description, references, compared with A. kagoshimae, Ku- mano-Nada off Heta). Kuroda, 1951 : 818 (Sagami Bay, Saruga Bay, Ku- mano-Nada, Kii, Shikoku, Kyushu). Kamohara, 1952 : 12, fig. 8 (description, 65 Prov. Tosa, Shikoku). Mori, 1952 : 37 (Fusan and Pohang, Korea). Ochiai, 1952 : 189, figs: 1-4 (analysis of length-weight relationships). Ochiai; 1954 968, figs. 1-8 (studies on the gonads, Maizuru, Kanaiwa, Ogi, Owase, Mimase). Matsubara, 1955 : 216, pl. 15, fig. 5 (description, distribution). Okada, 1955 : 59, fig. 55 (description, distribution, common names, habits). DIAGNOSIS.-This species can be distinguished from G. Zeioglossus - by the absence of pigment on its stomach and by the knobs developed on the medial ends of the maxillaries. G, semifasciakus differs from G. lineatus in having a lower anal fin ray count (11 to. 18 in G. semi- fasciatus, 15 in G. lineatus), a lower lateral line count (50 to 58 in G. semifasciatus, 56 in G. lineatus), and a higher branchiostegal ray count (5 in G. semifasciatus, 4 in G. lineatus). G. semifasciatus differs from G. polli and G. lineatus in having dentary teeth only half the distance from the gape to the symphysis; in polli and lineatus the dentary teeth extend to, or almost to the symphysis. COUNTS AND MEASUREMENTS.-Based on 32 specimens. D. 12 (11- 18); A. 12 (11-13); pectoral 19 (18-21); ventral 12 (10-12); gill rakers on first arch 11 (10-12) +1+ '24 (21-27); lateral line scale rows 51 (50- 53); standard length 98.5-206.0 mm. (Ochiai [1952] records a specimen, 224 mm. long); preanal 88.3 (82.0-85.9); pfeventral 54.7 (52.9-58.5); predorsal 47.8 (46.0-49.8); prepectoral 25.9 (26.6-29.1); head length 28.1 (27.0-29.4); snout 10.3 (9.7-11.8); eye 7.4 (6.6-8.3); maxillary length 6.5 (5.8-7.9); maxillary width 2.8 (1.9-2.8); depth at dorsal fin 12.3 (10.4- 15.8); width behind head 10.6 (9.0-12.3); head depth 18.2 (11.3-14.4); interorbital 6.4 (5.6-7.8); depth of eaudal peduncle 5.5 (4.8-6.6); verte- bral count 48 in two specimens from Tottori Prefecture and one speci- men from Maizuru; pyloric caeca Fll (Matsubara [1943] gives 15-18); branchiostegals 5. DESCRIPTION.-Greatest depth between the head and the dorsal origin, tapering to the caudal peduncle. A lateral view shows the dor- sal profile of the head descending in an almost straight line from the nape to the snout. The dorsal profle is not broken by the ·dorsal rim of .the orbit. The ventral profile rises in a gentle curve from the 1958 COHEN: SUBFAMILY ARGENTININAE 149 posterior edge of the preoper,cle to the union of the articular and the quadrate from whence it runs in a straight line to the tip of the lower jaw. The interorbital space between. the supraocular canals is a broad Bat afea, lower than the supraocular canals. The frontals lat- erally are slightly below the level of the canals. Almost the entire maxillary lies underthe lachrymal, occasionally extending to the jugal. The medial ends of the maxillaries are en- larged into knobs which are connected by a strong ligament to a flat face on the premaxillary below. The epidermis covering the ventral edge bf the maxillary and the premaxillary bears minute papillae iii only a few specimens, while they' seem absent in most. The upper jaw is rounded; the projecting lower jaw is bluntly pointed. The palatines bear a single, irregular, widely spaced row of 20 to 25 short, conical teeth. The vomer carries an irregular band of 5 to 15 similar teeth. The tongue bears 2 to 6 small teeth which are often buried in the mucous membrane. ·The lateral edges of the tongue are serrate. Matsubara's figure (1948, fig. 19c) shows the serrations, but no teeth. The tongue also has very small papillae scattered along its lateral edges. From S to 10 short conical teeth are present in a - single row . on the inner face of each dentary up to a point midway between the angle of the gape and symphysis. Several specimens examined lacked teeth on the dentary. Absence of teeth. was not cor- related with size. On the ceratobranchial of the Sth gill arch are lto 8 small, conical teeth. Two oval-shaped patches of short, conical teeth, about 8 teeth on the anterior patch and 10 to 12 teeth on the posterior patch, are present at the anterior end of the 4th suprabran- chial. The gill rakers are elongate and closely spaced. The 4th raker forward from the angle on the lower arm of the 1st arch is equal in length to from one-third to one-half of the interorbital distance. The posterior ends of the pectoral fin bases form angles of 20 to 85 degrees with a plane parallel to the venter. The tip of the pectoral extends to a point about one-third of the way between the anterion of the pectoral fin base and the anterior of the ventral fin base. The peritoneum is silvery on its outer surface. The inner surface is lightly punctate ventrally, while the chromatophores are smaller ' and closer together dorsally. The entire gut is immaculate. The stomach has a posterior elbow which stretches to a point midway be- tweeh the pectoral and the ventral bases. The anterior curve of the stomach is located at the level of the tip of the pectoral fn (see Matsu- bara [1948] for diagrams of the gut, dention, and gill rakers). The ven- tral section is short, equal to half the length of the posterior el bow, and displaced toward the right side of the body. The pylor ie caeca 150 BULLETIN FLORIDA STATE MUSEUM Vol. 8 extelid only slightly past the anterior curve of the stomach: In a speci- men 145.7' mm. in standard length the intestine. has about 12 strongly developed Ringen which seem to be arranged spirally, The Gru,ul- netze is present as a few strongly developed ridges. The swimbladder is similar to that of Argentina in form. It lacks silvery pigment, but d6es have a faint iridescent ·sheen. The origin of the bladder is above a point midway between the anterior and posterior curves of the stomach and its termination is a point slightly ~ behind the origin of the ventral fills. G. semifasciatus agrees with Argentina in lacking a centralized rete; however, the surface of the bladder carries a considerable number of parallel blood vessels in series of 2 to 6. At the anterior end of the bladder a pair of veins run from the sides of the bladder to the kidneys. In ·the.specimens avail- able it was not possible to observe the arterial supply. The color of specimens ip alcohol varies from light straw to light brown ventrally, while the dorsal half of the body is a darker brown. Almost all specimens bear 7 to 10 dusky blotches along the dorsolateral contour of the· body. Each flap of skin marking a lateral line scale pocket bears a small brown spot. The fins are all lightly flecked with black. The opercle, isthmus, cheek, and iris are silvery, while the muzzle, the most posterior part of the roof of the cranium, -and the upper portion of the opercle are dusky. The portion of the cranium covering the brain is transparent, allowing the brown pigment on the inner surface of the brain case to show through, while the anterior part of the skull is translucent. Okada, et al., (1985) give a color plate of G. semifasciatus which shows the head and body below the lateral line silvery and the body above the lateral line brown with indistinct blotches. DIsTRIBUTION.-Glossanodon semifasciatus is, so far as known, re- stricted to the southern half of Japan and southeastein K6rea. The northernmost record for the species on the Pacific side of Japan is - Sagami Bay (Kuroda, 1951). It has been taken at various localities southward and westward along the coast to Heta near the southern tip of Kyushu. On the western coast of Japan the fish has been taken0 from Nagasaki in the south, north t6 Niigata (approximately 38° N) Mori (1952) gives Pusan and Pohang as the only Korean records. HABITS.-Okada (1955) states: "In Toyama Bay, it is caught in great quantities at the depths of 70 to 240 m., from late February to late March. After March it migrates to deepet bottom, and spawns in the latter part of May and in June." Matsubara (1948) reports speci- mens from 125 fathoms in the Kumano-Nada. 1958 COHEN: SUBFAMILY ARGENTININAE 151 SPECIMENS EXAMINED.- Japan Honshu: Mie Prefecture, Owashi, TU 28786 (1); Wakayama Prefecture, Tanabe, TU 28630-2 (8); Japan Sea, Matsue, TU 31481 ('1); Tottori Prefecture, SU 49738 (12); Kyoto Prefecture, Miyazu, TU 24849-58 (5), TU 24896 (1), Japan Sea, Maizuru, SU 49529 (10); Toyama Bay, Namerikawa, TU 84808-8 (6); Toyama Bay, probably Uotsu or Namerikawa, TU 45188-4 (2), TU 46224 (1), TU 46252«4 (8), TU 46256 (1), TU 46297 (1), TU 46812-4 (3), TU 46716 (1), Niigata, TU 28429 (1), TU 24897 (1), SU 23441 (5). Shikoku: Mimase, near Kochi, TU 24268 (1); off Kochi'Prefecture, TU 16780-1 (2). No data: TU 39178 (1), TU 26869-10 (2). Glossanodon (Glossanodon) polli, new species Figure 8 ZI k. J . Figure 8.-Glossanodon poUi. Adult and adolescent sbecimens from off West Africa (from Poll, .1958). Argentina sphymena (non Linnaeus) Poll, 1953 : 58, fig. 20 (description, habits, 80' 05' 5/ 9' 25' E, 9' 81' 30" N/ 169 28' W) DIAGNOSIS.-G. pom differs from G. leioglossus and G. semifasci- atus in having dentary teeth· extending more than half the distance from the angle of the gape to the symphysis. In most specimens the teeth extend to the symphysis. G. polli differs from G. lineatus in having 18 to 14 anal rays and 48 to.·51 lateral line scales, while lineatus has 15 anal rays and 56 lateral line scales. G. polli has 5 branchi- ostegals, G lineatus has 4. COUNTS AND MEASUREMENTS.-Based on six specimens,from the eastirn Atlantic,· D. 12 (12-14); A. 14; pectoral 21 (19-22); ventral 12 (12-18); gill rakers on first arch 10 + 1 + 24 (21-25); lateral line sbale rows 49 (48-50); standard length 64.0-184.0 mm.; preanal 83.8 (82.8- 85.2); preventral 58.4 (51.9-54.1); predorsal 48.6 (47.8-49.5); prepectoral 28.9 (28.4-29.7); head length 28.7 (26.9-29.8); snout 8.9 (8.0-9.4); eye 152 BULLETIN FLORIDA STATE MUSEUM Vol. 8 8.6 (8.1-9.4); maxillary length 6.8 (6.3-7.8); maxillary width 2.3 (2.2- 2.5); depth at dorsal fin 11.8 (9.4-12.3); width behind head 11.0 (10.2- 11.7); head depth 12.2 (11.7-12.6); interorbital 7.3 (6.7-7.7); depth of caudal peduncle 5.7 (5.2-6.8); branchiostegals 5, The following counts and measurements are based on two speci- mens from the western Atlantic: D. 12,12; A. 13,13; pectoral 21, 22; ventral 12„ 11; gill rakers 11+1+22, 10+1+20; lateral line scale, rows 51, S standard length 75.0 mm., 100.5 mm.; preanal 84.7, 82.1; preventral 58.8, 50.8; predorsal 48.7, 46.8; prepecforal 29.9, -; head length 30.0, 26.9; snout 9.1, 8.9; eye 9.8, 7.4; maxillary length 6,7,6.6; maxillary width 2.6,2.0; depth 9.6, 10.9; width behind head 9.4,9.9; head depth 12.0, 10.9; interorbital 8.1, -; depth of caudal peduncle 6.0, ·5.5; branchiostegals 5 (?), 5 (?). DESCRIPTION.-Greatest depth behind the head, tapering to the eau- dal peduncle. A lateral view shows the dorsal profile of the head de- scending in a gentle curve to the snout. The dorsal profile is not. broken by. the dorsal rim of the eye. The ventral profile rises in a gentle curve from the posterior edge of the preopercle to the union of the articular and the quadrate, from whence it runs in a straight line to the tip of the lower jaw. The interorbital space between the supra- ocular canals is a broad flat area, lower -than the· supraocular canals. The frontals lateral to the canals slope downwards over the orbit. The dorsal edge of the maxillary lies under the lachrymal and ex- tends to the ioint between the lachrymal and the iugal. The epidBrmis covering the ventral edge of the maxillary and the premaxillary bears minute papillae. Both the upper jaw and the slightly projecting lower jaw are rounded. - Each palatine bears 20 to 35 short, conical teeth which are re- stricted to the lateral edge of the bone in several specimens, but may be scattered about the surface or in a row along the medial edge of the bone in others. The vomer bears a patch of 20 to 25 similar teeth across its head. Lingual teeth are apparently absent. From 10 to 20 Short, conical teeth are. present in a single -row on the inner face of each dentary extending from the angle of the gape past the midpoint of the dentary and ilsually to the symphysis. The branchiaL arches were not examined for teeth. The gill rakers are elongate and c16sely spaced., The 4th raker forward from the angle on the lower arm of the firit arch is equal in length to from one-third to one-half of the interorbital distance. The posterior end of the pectoral fin base forms an angle of 45 to 60 degrees with a plane parallel to the venter. The tip of the pectoral extends to a,point about two-thirds of the way between the anterior 1958 , COHEN: SUBFAMILY ARGENTININAE 153 of the pect6ral fin base and the anterior 6f the ventral Bn base. The ventral fin is slightly shorter than the. pectoral fin. Tbe peritoneum is silvery on its outer surface. Tbe inner surface ' is densely punctate. with dark brown chromatophores at the anterior end of the body cavity and more. lightly colored posteriorly. The stomach appears to lack pigment. The swimbladder has 'no silvery Digment. The body of a preserved specimen is light tan. A dark brown streak, equal in width to the diameter of the pupil, runs along -the dorsolateral contour of the body. Eight or nine more densely pig- mented areas are present along the band and are more pronounced in smaller specimens. An iridescent band, below and equal in width to the dark brown stripe, runs along the lateral line. Some of the scale pockets along the dorsum are edged with brown. The muzzle is dusky; the inner surface of the braincase shows dark brown through the transparent frontals and parietals. The isthmus, opercles, cheeks, and irises are bright silver. The postlarval color pattern c6nsists of 9 dark lateral streaks along the side of the body (fig. 8). DISTRIBUTION.-This species has been collected at three localities by the recent Belgian Oceanographic Expedition along the coasts of West Africa. One of the localities is, unfortunately, unknown. The other two are off French Equatorial Africa and off the border of Sierra Leone and French Guinea. In addition, I have examined specimens, apparently referable to G. polli, collected off the Brazilian coast close to the equator. The western Atlantic population may prove to be separable when more material is available. HABITS.-The known bathymetric range in the eastern Atlantic is 150 to 200 meters. One collection was made 6ver a sand b6ttom and the pther over a sand and rock bottom. Poll (1958) states that a stomach examined contained crustaeeans and that larger specimens from the northern locality (collected in June) were sexually mature or almost s6. Western Atlantic specimens were taken at depths of 175 to 225 fathoms with a 40-foot shrimp trawl. SPECIMENS EXAMINED.- , Eastern Atlantic off Equatorial Africa: Holotype.-9' 81' 30" N/ 16' 23' W, 150-200 meters, rock and sand bottom, tempdrature at 145 meters 15.38' C., salinity 36.68 ppm, oxygen 1.79 grams per liter, trawl hauled for li hours, collected by the Expedition Oceanographique Belge dan les'Eaux 'Cotidres Afficaines de l'Atlantique - Sud (1948,1949), st. 214, 6-7 June, 1949, RCB 95.005. Paratypes.-Data as for the holotype, RCB 95.006-9 44); 3' 5' S/ 9' 25' E, . 154 BULLETIN FLORIDA STATE MUSEUM Vol. 8 150-200 meters, sand bottom, temperature at 145 meters 14.58° C., salinity ' 83.6 ppm, oxygen 2.27 gm, per liter, trawl hauled for one hour, same expedition, st. 162, 25-26 March 1949, RCB 90.004 (1). O# Brazil: 1' 49' N/ 46' 48" W, 225 fms., 17 November ,1957, "Oregon" st. * 2088, USNM 159355 (1), 1' 52' N/ 46' 54' W, 175 fms., 17 November 1957, "Oregon" st. 2081, USNM 159852' (1) Glossanodon (Clossanodon) lineatus (Matsubara) Leuroglossus Zineatus Matsubara, 1948 : 70, figs. 20, 21 (original description, type - locality: Kumano-Nada, holotype: no. 4247 Matsubara fish collection; emenda- tion of Microstomidae to receive Leuroglossus), Kamohara, 1952 : 18 (Mi- mase, Shikoku). Matsubara, 1955 : 217, fig. 57 (listed in family Bathylagidae). DIAGNOSIS.-G. lineatus can be separated from the other species in the subgenus Glossanodon by its higher lateral line count (56 in G. Zineatus, 48 to 53 in the other species), a higher anal fin ray count (I5 in G. lineatus, 11 to 14 in the other species), and fewer bran6hiostegal rays (4 in G. lineatus, 5 in other species). G. lineatus resembles G. polli and differs from G. teioglossus and G. semifasciatus in having dentary teeth extending t6, or almost to, the symphysis. COUNTS AND MEASUREMENTS.-Based on two specimens and given for each. Measurements given in ,millimeters. D, 12, 18; A. -, 15; pectoral.21, 18; ventral 12,.11; gill rakers on first arch 9+ 1 + 16 (?), -, (Matsubara [1943] gives 8-9 + ,19-20); lateral line scale rows -, 56; standard length 89.9,97.4 (Kam6hara [1952] notes a specimen which is damaged but "about 120 mm. long"); preanal 84.0, 88.2; preventral 51.4, 52.4; pred6rsal 48.3,48.2; prepebtoral 25.8,25.9; head length 26.4, 27.5; snout 8.2,8.8, eye 8.2,7.8; maxillary length 6.9,6.2; maxillary width 2.3,2.6; depth at dorsal fin 11.1, 12.1; width behind head 10.9, 9.7; head depth 11.7, 11.8, interorbital 6.7, 6.l; depth of caudal pe- duncle 5.7,5.6; branchiostegals apparently 4 (I have not examine"d an alizarin preparation of this .species). DESCRIPTION.-Body slender with the greatest depth between the dorsal fin and the head, tapering to the caudal peduncle, The dgrsal profile of the head slopes gently to the snout and is unbroken by the orbit. The ventral profile of the head rises gently from the posterior margin of the opercle to the union of the articular and the quadrate, from whence it rises more steeply.' The interorbital space between the supraocular canals is a broadly rounded, slightly convex area, lower than the supraocular canals. The frontals lateral to the canals are on a leyel with the cani{ls. The dorsal edge of the maxillary lies under the lachrymal and the jugal. Both the upper and the projecting lower jaw are. rounded, the upper jaw more broadly so. 1958 COHEN: SUBFAMILY ARGENTININAE I55 The conical palatine teeth are arranged in two separate rows, one along the lateral edge of the bone,. and the other along the medial edge of the bone. Each palatine bears 20 to 25 teeth. The vomerine teeth number about 15. Each dentary has about 15 small pointed . teeth in an irregular row extending from the angle of the gape to the symphysis of the dentaries. Teeth are apparently absent from the tongue. The gill rakers are elongate and closely spaced (Matsu- bara, 1948, fig. 21). The pectoral base forms an angle of about 45 degrees with a plane parallel to the venter. The peritoneum is silvery on the outside, while the inner surface is densely punctate with dark brown pigment. The body of an alcohol specimen is straw colored and has a dark brown stripe along its dorso- lateral contour. About 6 brown blotches are distributed along the body on the stripe. Matsubara (1943) mentions' the presence of a silvery band below the brown stripe. The muzzle and upper part of . ' the opercle are dusky. DISTRIBUT[ON.-Recorded from only two localities on the Pacific coast of southern Japan, one the type locality, Kumano-Nada off Hon- shu, and the other off Mimase, Shikoku. HABITS.-Taken at about 125 fathoms. SPECIMENS EXAMINED.- _Japan: HonAhu.-Kumano-Nada, CKM 4245 (1), paratype; Kumano-Nada, off Owashi, CKM 18867 (1). Subgenus Prosoarchus, new Type species Glossanodon (Prosoarchus) pygmaeus, new. The subgenus Prosoarchus differs from the subgenus Glossanodon in having the anus more advanced. In Prbsoarchus the anus lies at a distalice equal to or greater than the depth of the caudal peduncle in front of the anterior end of the anal fin base, while in Glossanodon, the anus lies at a,distance equal to half or less, than the depth of the caudal ped-uncle in- front of the anterior end of the anal fin base. Prosoarchus also differs from Glossanodon in having a lower pec- toral ray count (12 to 14 in Prosoarchus, 18 to 22 in Glossanodon), and a lower lateral line count (43 to 46 in Prosoarchus, 48 to 56 in Glossano- don). The name prosoarchus is from the Greek proso, forward, and archos, anus, in reference to the forward position of the anus. 156 BULLETIN FLORIDA STATE MUSEUM V61. 8 Glossanodon (Prosoarchus) pygmaeus, new species Figure 9 ~OUNTS AND MEASUREMENTS.-~OUntS based on 18 specimens, measurements based on 14 specimens. D. 11 (10-12); A. 12 (11-13); pectoral 18 (12-14); ·ventral 10 (10-12); gill rakers on first arch in four - specimens 10 (9-11) +1+22 (21-28); lateral line scale rows 46 (43-46); standard length 52.5-98.5 mm.; preanal 81.5 (80.1-84.8); preventral 54.2 (51.6-58.7); predorsal 47.6 (45.8-49.6); prepectoral 28.8 (26.7-30.0); head length 28.7 (27.2-30.1); snout 9.2 (8.0-10.2); eye 8.9 (8.1-9.7); maxillary length 8.0 (6.7-8.9); maxillary width 2.8 (2.4-3.8); depth at dorsal fin 12.1 (10.1-14.0); width behind head -10.4 (8.5-11.8); head depth 18.0 (12.0-14.8); interorbital 7.2 (6.1-8.8); depth of caudal peduncle 6.0 (5.8- 6.8); vertebral count of a single specimen 43; pyloric caeca 8 in two specimen5,9 in another, branchiostegals 5 (the first ray on each side much reduced). DESCRIPTION,-A small species with a slender body, even in gravid females. · Greatest depth behind the head, tapering gradually to the caudal peduncle. A lateral view shows the upper profile of the head descending to the snout in a gentle curve which is unbroken by the dorsal margin of the eye in smaller specimenl. but in »larger ones the eye projects somewhat. The ventral proBle rises more gently from the posterior edge of the preopercle to the union of the articular and the quadrate, from whence it rises more rapidly. The interorbital - space between the suprao¥bital canals is flat. The canals are raised slightly above the central area, and the frontals lateral to the canals arch upward very slightly. The dorsal edge of the maxillary lies tucked under the lachrymal and the jugal. Minute papillae are distributed On the skin along the ventral edge of the premaxillary. The ventral edge of the maxillary, however, seems to lack papillae. In most specimens the lower jaw projects slightly, though in several the jaws are subequal. The tooth-bearing areas of the palatines are poorly ossifted. The anteriormost portion of the bone is a thin, tooth-bearing plate which is present in the.same body of cartilage as the main body of the pala- tine, but is not connected with it by bone. The main body of the palatine bears an irregular row of 10 to 25 conical teeth. The row is - usually single on the posterior part of the bone and 2 or 8 teeth wide on the anterior part of the palatine proper. The anterior plate of the palatine bears 10 to 17 teeth in a band 2 to 4 teeth wide, which repre- sents a continuation of the tooth band on the anterior part of the pala- tine proper. The vomer also bears a continuation of this band con- 1958 CO HEN: SU BFAM ILY AR G EN TIN IN AE Figure 9.-Holotype of Glossatotion (Prosoarchus) pygmaeus. CNHM 64845. 158 BULLETIN FLORIDA STATE MUSEUM Vol. 8 si5ting of 11 to 18 widely scattered conical teeth. The vomer is also poorly ossified, especially in a hemispherical area at the tip. The tongue is armed with 2 to 6 teeth. The ceratobranchial of the 5th gill arch bears 4 small conical teeth. Two ovaI-shaped patches of short conical teeth, about 2 teeth on the anterior patch and 7 on the posterior, are present at the anterior end of the 4th suprabranchial. The gill rakers are elongate and closely spaced. The 4th raker from the angle on the lower arm of the 1st arch is equal in length to about oile=half of the inteforbital distance. The posterior end of the pectoral base forms an angle of 45 to 55 degrees with a plane parallel to the venter. The longest dorsal ray is slightly longer than the greatest depth of the body at the origin of the dorsal fin. The tip of the pectoral fin extends mere than half of the distance between the anterior end of the pectoral fin base and the anterior, end of the ventral fin base. The ventral fin is shorter than the pectoral fin. The inner surface of the peritoneum is densely punctate ventrally, appearing almost solid black, and considerably lighter dorsally. The gut is unpigmented. Tlie smoothly rounded posterior curve of the stomach extends almost fo the level ,.of -the posterior tip of the pec- toral En. The intestine of a mature individual 77.2 mm. in standard 16ngth is transparent for most of its length and has the Ringen so pborly differentiated from the Grundnetze that it is impossible to make a Count. The swimbladder lacks silvery pigment. Alcohol specimens have a uniform light tan ground color. In some specimens a dark brown stripe, equal in width to half the inter- orbital, runs along the dorsolateral margin of the body and continues onto the upper edge of the opercle; in others the stripe is reduced to a lightly pigmented band. A faint iridescent sheen alc,ng the lateral line suggests a silvery band may have be,en present. Large' brown chromatophores sprinkled along the lateral line form a poorly defined band venttal to the one mentioned above. A similar poorly defined band is present on the ventrolateral contour of the body above the area between the anus and the caudal base. The dorsal and ventral areas of pigmentation meet at the caudal base, coloring the entire area a light brown. In many specimens the scale pockets on the dor- sum and the venter are neatly outlined by dark brown chromato- phores. The muzzle is dusky; the rami of the lower jaw·and the ventral third of the opercle and preopercle are lightly washed with brown. The isthmus is colored a darker brdwn by large brown chromato- 0 , 1958 COHEN: SUBFAMILY ARGENTININAE 159 phores. The posterior portion of, the cranium is lightly dotted with very small brown chromatophores through which the outlines of the brain are visible. The dorsal fin rays are colored by evenly distributed brown chromatophores. The caudal fin is also pigmented, but less evenly, while the pectoral, ventral, and anal fins. are pigmented at their bases but not on their distaI portions. Several smaller specimens have faded considerably and show only faint remnants of the above pattern. In one of the smaller specimens the dorsolateral stripe is broken into 8 or 9 faint dusky blotches. DISTRIBUTION.,-G. pygmaeus has been taken off the coast of South Carolina, in the northern half of the Gulf of Mexico off the coasts of Texas and Florida and northwest of Tottugas Island, off the coast of Venezuela southeast of Trinidad, and off the c6ast of Brazil near the equator. HABITs.-The known bathymetric range of G. pygmaeus is 52 to I68 fathoms. Three collections are from over a mud bottom, while a third is from a sand and gravel bottom. G. pygmaeus was taken sev- eral times in company with Argentina striata which it strongly re- sembles in color pattern. The gut of one specimen contained several copepods along with a considerable amount of unidentifiable organic matter. G. P!/ginaeus is evidently a small species, or at least reaches sexual maturity at a small size. A specimen 77.2 mm. in standard length had ripe eggs in the ovary, while another specimen 86.6 mm. in standard length had well-developed testes. NOMENCLATURE.-The name pygmaeus is Latin for dwarf and re- fers to the small size of this species. SPECIMENS EXAMINED.- Gulf of Mexico o# Tortugas: Holotype.-25' 08' 3/ 84' 19' W, 168 fathoms,» sand and gravel bottom, bottdni temperature 10' C., 40-foot shrimp trawl, 2115-2866 hours, "Ore- gon" st. 1026, 19 April 1954, CNHAA 64845. Paratypes.-r Gulf of Mexico o# Tortugas: Data as for h616type, CNHM 64346 (6), SU 49785 (1). Gulf of Mexico, southeast of Pensacola Bay, Ftorida: 80' 05' N/ 86' 55' W, 52 fms. blue-mud bottom, bottom' temperature 15' C., 40-foot shrimp trawl, 1480-1500 hours, "Oregon" st. 858, 29 Oct. 1958, CNHM 45983 (1) Gulf of Mexico, southeast of Port Aransas, Texas: 27' 22' 3/ 96' 08' W, 108 fms., mud bottom, bott6m temperature 15' C., 100-foot 5hrimp trawl, 1550-1700 hours, "Oregon" st, 156, 27 Sept. 1950, CNHM 46267 (1); 26' 55' N/ 96' 25' W, 125 fins., gray-mud and shale bottom, bottom temperature 19' C., 40-foot shrimp trawl, 1535-1600 hours, "Oregon" st. 550. 18 April. 1952,. SU 49734 (1). 160 BULLETIN FLORIDA STATE MUSEUM - Vol. 8 N. Atlantic o# South Carolina: 32' 58' N/ 77' 52' W, 32' 56' N/ 77' 56' W, 74 fms., "Albatross III" cruise 818, st. 8, tow 1, USNM 152104 (2). N. Atlantic o# Venezuela: 9' 45' 3/ 59' 45' W, 40-foot shrimp trawl, "Oregon" st. 1989, USNM 159351 (8); 9' 41' N/ 59' 47' W, 150 fms,, 40-foot shrilnp trawl, "Ordgon" st. 1985, 3 Nov. 1957, UF 8027 (2). N. Atlantic o# Brazil: 2' 41' N/ 47' 48' W, "Oregon" st. 2067, USNM 159850 (1). LrrERATURE CITED Andriashev, Anatoly P. 1954. Ryby severny m6ryea SSSR. Opredeliteli po Fauna SSSR, Izdavaenmia Zoologicheskii Institutom Akademia Nauk SSSR, no. 58, 566 pp. Anon. 1955. New England exploratory fishery program. Commercial Fish. Rev., vol. 17, nb. 7, p. 35. Artedi, Peter 1788. Ichthyologia, sive opera omnia de piscibus silicet: pam iii. Genera piscium, etc.; iv, 88 pp, pars. v. Descriptiones specierum piscium. Lei- den, iv, 102 pp. Ascanius, Peder 1775. Icones rerum naturalium, ou figures enlumindes d'histoire naturelle du Nord. Troisieme cayer. Copenhagen, 6 pp., tables 21-80. Barnhart, Percy Spencer 1986. Marine fishes of southern California. Berkeley: Univ. Calif6rnia Press, 209 pp. Berg, Leo Semenovitch 1940. Classification of fishes both recent and fossil. Travaux de l'Institut , Zoologique de l'Acad6mie des Sciences de l'URSS, vol. 5, pp. 87-517. Bertin, I~on 1940. Catalogues des types de poissons du Museum National d'Histoire Natu- relle. Pt. 2. Dipneustes, Chondrost6ens, Hglost6ens, Isospond)des. , Bulletin Museum d'Histoire Naturelle, ser. 2, vol.,12, pp. 244-822. Bigelow, Henry Bryant, and William Schroeder 1986. Supplemental notes on fishes of·the Gulf of Maine. Bull. U.S. Bur. Fish., vol. 48, pp. 819-343. 1958: Fishes of the Gulf of Maine. Fishery Bull. 74, U.S. Fish and Wildlife Service, vol. 58, vili + 577 pp. Bigelow, H. B,„ and William W. Welsh 1925. Fishes of the .Gulf of Maine. Bull. U.S. Bur. Fish., vol. 40, pt. ' 1, 567 pp. Bloch, Marc Elieser, and Johann Gottlob Schneider 1801. Systema ichthyologiae iconibus ex illustratum. Berlin: Saxo, 1x + 584 pp., 110 pls. , B6hlke, James Erwin . , 1953. A c~alogue of the type specimens of recent fishes in the Natural History , Museum of Stanford University. Stanf6rd Ichthyol. Bull., vol. 5, 168 pp. 1958 COHEN: SUBFAMILY ARGENTININAE 161 Bonaparte, Charles Lucien Jules Laurent I846. Catalogo metodico die pesci Europei. Naples, 97 pp. Boulenger, George Albert 1900. Abundance of the greater sill-smelt (Argentina silus, Asean.) on the mar- ket of Boston, Lincolnshire. Ann. Mag. Nat. Hist., ser. 7, vol. 6, p. 160. Breder, Charles M. 1948. Field book of marine fishes of the Atlantic coast from Labrador to Texas. Revised edition. New York [etc.1: G. P. Putnam and Sons, xxxvii + 882 *D., 16 pls. Brito Capello, Felix De 1881. Catalogo dos peixes de Portugal. Memorias Acad. Sci. Lisbon, vol. 46 (new ser., vol. 6, pt. 11), 78 pp Brunnelli, Gustavo, and Giorgio Bini 1934. Ricerche comparative sulle pesche profonde di diversi mari Italiani. Bollettino di P.esca, Piscicoltura e Iilrobjologia, vol. 10, pp. 783-744, 1 pl. Bruun,' Anton Fr., and J. R. Pfaff 1950. List of Danish vertebrates. Fishes. Copenhagen, pp..20-66. Brunnich, Morten Thrane 1768. ' Ichthyologia .MassiliensiS, sistens piscium descriptiones eorumque apud incolas nomina. Accedunt spolia maris Adriatica. Copenhagen' and Leipzig, xvi + 110 pp. Canestrini, Giovanni. 1870. Fauna d'Italie, parte terza. Pesci. Milan, 208 pp. Carus, Julius Victor - 1893. Prodromus faunae Mediterraneae. Vol. 2. Brachiostoma, Mollusca, Tuni- cata, Vertebrata. Stuttgart, 854 pp. Chapman, Wilbert McLeod 1942. The osteology and relationships of the Argentinidae, a family of oceanic fishes. Jour. Washington Acad. Sci., vol. 82, pp. 104-117. Clarke, F. E. 1879. ·On a new fish found at Hokitika. Trans. and Proc. New Zealand Inst. for 1878, vol. 11, pp. 295-297, pl. 14. - Clothier; Charles R. 1950. A key to some southern California fishes based on vertebral characters. California Dept. Fish and Game, Fish Bull. 79, 88 pp. Collett, Robert 1875. Norges fiske, med bemaerkninger_ om deres udbredelse. Oslo, 240 pp., 2 pls., map. 1879. Meddelelser om Norges fiske i aarene 1875-78. F6rhandlinger i Videns- kabsselskabet i Kristiania, no. 1, 107 pp. , 1903. Meddelelser om Norges fiske i aarene 1884-1901 (8die hovedsupplement til -Norges Fiske"). II. Ibid., rio. 9, 175 pp. 162 BULLETIN FLORIDA STATE MUSEUM Vol. 8 Costa, Oronzo Gabriele 1836-1844. Fauna del regno Napoli ossia enumerazi6ne di tutti gli aniniali. che abitano le diverse regione di questo regno. Vol. 8. Pesci. Naples, 227 pp. (not numbered consecutively), pls. Couch, Jonathan 1865. A history of the fishes of the British Islands, vol. 4. [Edit. 1.] London, vii + 489 pp., pls. 180-248. Cuvier, Geprges L6opold Chr6tien Fr6deric Dagobert 1815. Observationes et'recherches critiques sur diff6rents poissons,de la Mdditer- rande, et a leur occasion sur des poissons d'autres mers, plus ou moins li6s avec eaux. M6moirs Museum d'Histoire Naturelle, vol. 1, Paris, PP. 226-241. 1817. Le regne·animal distribue d'apres son organization, vol. 2. Paris, xviii + 532 pp. Day, Francis 1880-1884. The fishes of Great Britain and Ireland, vol. 2. London, 888 pp., pls. 94-174. 1881. On the Hebredal Argentine. Tour. Linnean· Soc. Zool., vol. 15, pp. 78-84.- De Buen, Fernando 1919. Las costas sur de Espaaa y su fauna ictiologica marina. Boletih de Pes- cas. Madrid, vol. 4,.nos. 87-38, pp. 249-320. 1926. Catilogo ictiol6gico del Mediterraneo Espafiol y de Marruecos. Res. Cainp. Realizado por Acuerdos Internadionales, no. 2, 221 pp 1985. Fauna ictiol6gica. Cattilogo- de los peces Ib6ricos, pt. 1. Inst. Espafi61 de Oceanografia, Notas y Resumenes, ser. 2, no. 88, 89 pp. Doderlein, Pietro 1878. Prospetto metodico d6lle varie specie di pesci riscontatre nelle acque marine e fluviale della Sicilia annesso al prodromo della fauna ittiologica. Atti dell' Accademia di Scienze, Lettre ed Art. Palenno, new ser. 6, pp. 25-64. Dons, Carl 1921. Endel s jeldnere, fisk i Nord-Norge. Troms0 Museums Arshefter, 48 (1920)1 no. 6, 40 pp., 2 pls. Edward, Thomas 1881. On the occurrence of the Norwegian -Argentina situs on the shore of Moray Firth, Banffshire. Jour. Linnean Soc,, vol, 15, pp, 834-886, Ehrenbaum, Ernst 1901. Fauna Arctica. Fische. Vol. 2, no. 1, pp. 67-168. 1909. Eier und Larven von Fischen, pt. 2. Nordisches Plankton. Vol. 10, pp. 217-418. Faber, George L. 1883: The fisheries of the Adriatic.and the flsh thereof. London, xxvi + 292 pp. Fage, ·Louis 1907. Essai sur la faune des 'Iles Baldares et description de quelquas esp6ces nouvelles. Archives de Zoologie Experimentale et Generale, ser. 4, vol. 7, pp. 69-98, pl. 6. 1958 COHEN: SUBFAMILY ARGENTININAE 163 1910. Recherches sur les stades p61agiquei de quelques T616ost6ens de la mer de Nice (Parages de Monaco) et du Golfe du Lion. Annales de l'Institut Oc6anographique, vol. 1, fasc. 7,58 pp., 1 pl. Fairbridge, W. S. 1951: The New South Walei flathead, Neoptat!/cephalus macrodon (Ogilby). ~ Australian Jour. Marine and Freshwater Research, vol. 2, pp. 118-178, ' 4 pls. Firth, Frank E. 1981. Some fishes collected recently in New England waters. Bull. Boston Soc. Nat. Hist., no. 61, pp: 8-14 Fitch, John E. 1949. Some unusual occurrences of fish on the Pacific coast. California Fish , and Game, vol. 35, pp. 41-69.. 1950. Notes on some Pacific fishes. Ibid., vol. 36, pp. 68-70. Follett, W. I. 1945. Notes on Argehtina sialls Gilbert, an isospondylous fish of western North America. Copeia, 1945, pp. 148-145, Fowler, Henry Weed 1912. Notes on salmonoid and' related fishes. Proc. Acad. Nat. Sci. Philadel- phia, vol. 63,'pp 551-571. 1986. The marine fishes of West Africa, pt. 1. Bull. Amer. Mus. :Nat, Hist., vol. 7.0, 605 pp. Fridriksson, Ami 1937. Um Gullax (Argentina silus Ast.) Natturufraed, Reykjavik, vol. 7, pp. 72-78. Frost, G. Allan, 1983. Otoliths 01 fishes from the iower Tertiary formations of southern Eng- land. I. Isospondyli, Apodes, Berycomorphi. Ann. Mag. Nat. Hist., ser. 10, vol. 12, pp. 887-896, pl. 12. Carman, Samuel 1899. Reports on an exploration off -the west coast of Mexico, Central - and South America, and off the Galapagos Islands in, charge of Alexander Agassiz, by the U. S. Fish Commission steamer "Albatross" during 1891, Lieut. Commander Z. L. Tanner, U.S.N., commanding. 26. The Fishes. Mom. Mus. Comp. Zool. Harvard, vol. 24, 481 pp,·pls. 1-85 and A-N. Gesner, Conrad 1558. Historiae animalium liber IIII, qui est de piscium & aquatilium animan- tium natura. Zurich, 1297 pp Gigholi, Enrico Hillyer ' 1880. Elenco dei Mammiferi, degli Uccelli e dei Rettili ittiofagi appartenenti alla fauna Itabca e Catalogo degli Anfibi e dei Pesci Italiani. Florence, 55 pp. Gilbert, Charles Henry 1890. A preliminary report on the fishes collected by the steamer "Albatross" on the Pacific Coa5t of Norih America during the year 1889, with de- 164 BULLETIN FLORIDA STATE MUSEUM. Vol. 8 scriptions of twelve new genera and ninety-two new species. Proc:. U.S. Natl, Mus., vol. 18, pp 49-126. Gilchrist, John Dow Fisher, and Cecil von Bonde 1924. Deep-sea fishes procured by the S. S. "Pickle" (Part II). Report no. 8 for the year 1922. Union of South Africa, Fisheries and Marine Survey, spec. rept. 7, 24 pp., 6 pls. Gill, Theodore Nicholas 1868. On the subfamily of Argentininae. Proc. Acad. Nat, Sci: Philadelphia, vol. 14, pp. 14-15. 18988. A comparison of antipodal faunas. 'Natl. Acad. Sci., vol, 6, mem. 5, pp. 91-124. . 1893b. Families and subfamities of fishes. Ibid., vol. 6, mem. 6, PA 127-138. Gmelin, Johann Frederick , 1778. Caroli a Linnd.'. . Systema naturae per regna tria naturae, secundum classes, ordines, genera, ,species, cum charactefibus, differentiis, synony- . mis, locis, vol, 1, pt. 3. Editio decinia tertia, aucta, reformata. Leipzig, pp. 1033«1516. Goode, George Brown, and TarIeton Hoffman Bean 1879. Description of Argentina surtensium, a new deep-sea fish from Sable Island bank. Proc. U.S. Natl. Mus., vol. 1, pp. 261-268. 1895. Oceanic ichthyology. U.S. Natl. Mus., Spdcial Bull. 2, xxxv + 529 pp., 128 pls. Goodrich, Edwin Stephen , 1909. Vertebrata, craniata (cylostoines and fishes). In Lankester, E. R. Treatise on zoology, vol. 9. London, pp. xvi + 518. Graham, David H. 1989. Fishes of Otago Harbour and adjacent seas with additions to previous records. Trans. and Proc. Roy. Soc. New Zealand, vol. 68, pp. 899-419. 1958. A treasury of New Zealand fishes. Wellington, 400 pp. Grenholm, Ake 1928. Studien iiber die flossenmuskulatur der Teleostier. Upsala, 296 DP. Gronovius, Laurentius Theodorus 1754. Museum Ichthyologicurn,... vol. 1. Lugduni [? Lyons] ,70 pp., tables '1-4. 1756. [Same title]_, vol. 2. ' Lugduni [? Lyons]; 46 pp,, tables 5-25. Guichenot, Alphonse 1850. Exploration scientifique de l'Alghie. Vol. 3. Zoologie. Histoire Natural des Reptiles et des Poissons. Paris, iv + 144 pp., 12 pls. ' 1867. Ichthyologie. III. L'Argentine leiglosse, noveau genre de salmonides. Annales de la Soci6t6 Linnd D6partment Maine-et-Loire, vol. 9, pp. 7-9. Gunther, Albert Carl Lud*ig Gotthilf 1866. Catalogue of the Bshes in the British Museum, vol. 6. London, xv + 868 pp 1887: Report on the deep-sea fishes collected by H. M. S. Challenger during the years 1878-1876. Rept. Sci. Res. Voyage H. M. S. Challenger. Zoology. Vol. 22. 1xv + 385 pp., 78 pls. 1958 COHEN: SUBFAMILY ARGENTININAE 165 1888. Report on the fishes obtained by Mr. J. Murray in deep water on the northwest coast · of Scotland, between April, 1887 and March, 1888. Proc. Roy. Soc. Edinburgh, vol. 15, pp. 205-220, pls. 3-4. 1899. Address delivered at the anniversary meeting of the Linnean Society of London on the 24th of May, 1899. (Complete catalog of' Linne's pri- vate collection of fishes, now ih possegsion of the Linnean Society.) Proc. Linnean Soc, London, Session 1898-99, pp. 15-38. Higgins, Elmer 1920. Scientific work of the "Albacore" in December. California Fish and Game, vol. 6, pp. 86-88. - Holt, Ernest W. L. 1897. The great silver smelt, Aigentina silus, Nilss. An addition to the list of British fishes. Jour. Marine Biol. Assoc., new ser. vol. 5, pp. 841-342. Holt, E. W. L., and Lticius Widdrington Byrne 1906. First report on the fishes of the Irish, Atlantic slope. Dept. of Agri. and Tech. Instruction for Ireland, Fisheries Branch, Sci. Invest. 1905, no. 2, 28 pp., 1 pl. 1910. Fourth report on the fishes of the Irish Atlantic slope. List of recorded species with references. ,Ibid: 1908, no. 5,7 pp. Holt, E. W. L., and W. L. Calderwood 1895. Survey· of fishing grounds, west 'coast of Ireland, 1890-1891. Report on the rarer fishes. Sci. Trans. Roy. Dublin Soc., vol. 5, ser. 2, pA 861-512, pls. 89-44. Hubbs, Carl Leavitt 1982. The Californian species of the fish genus Argentina. Copeia, 1982, p. 105, Hutton, Frederick Wollaston 1879. Descriptions of two new fishes from New Zealand. Ann. Mag. Nat. Hist., ser. 5, vol. 3, p. 58. 1904. Index fauhae Novae Zealandiae. Philosophical Institure of Canterbury, 372 ppi ' Iversen, Thor 1936. Syd0stgr0nland. Report on Norwegian Fishery and Marine Investiga- - tions, vol. 5, no. 1, 100 pp. Jacobshagen, E. 1987. Mittel und Enddarm (rumpfdarm). In L. Bulk, et aL Handbuch clef vergleichenden Anatomie der Wirbeltiere. Band 8. Berlin und Vienna, pp:. 568-724. Jenkins, J. Travis , , 1925. The fishes of the British Isles. London, 876 pp., 148 pls. Johnsen, Sigurd 1919. Ichthyologiske notiser 1. Bergens Museum Aarbok 1918-1919, Naturvid. Raekke, no. 6, 95 pp, 1 pl Jones, F. R. Harden, and Norman Bertram ,Marshall . f953. The structure and function of the teleostean swimbladder. Biol. Rev., vol. 28, pp. 16-88. 166 BULLETIN FLORIDA STATE MUSEUM Vol. 8 Jordan, David, Starr 1887. A catalogue of the fishes known to inhabit the waters of North America, north of the Tropic of Cancer, with notes on the species discovered in 1888 and 1884. ,Rept. U.S. Comm. Fish and Fisheries.for 1885, pp. 789-973. 1928. A classification of fishes including families and gel*ra as far as known. Stanford Univ. Publ., University Series, Biological Sciences, vol. 3, pA 79-248 + x. Jordan, D. S., and Barton Warren Evermann 1896. The fishes of North and Middle America. Bull U,S. Natl, Mus,, no. 47, pt. 1, 1x + 1240 pp. Jordan, D. S.; B. W. Evermann; and Howard Walton Clark 1930. A check-list of the fishes and fish-like vertebrates of North and Middle America north of the northern boundary of Xenezuela and Colombia. Rept U.S. Comm. Fish. for 1928, append. x + 670 pp. Jordan, D. S., and C. H. Gilbert 1888. Synopsis of tlie fishes of North America. Bull. U.S. Natl. Mus., no. 16, 1v + 1018 pp. JordBn, D. S., and C. L, Hubbs 1925. Record of fishes obtained by David Starr Jordan in Japan, 1922. Mem. Carnegie Mus., vol. 10, pp. 93-846, pls. 5-12. Jordan, D. S., and John Otterbein Snyder 1902. A review of the salmonoid Bshes of Japan. Proc. U.S. Natl. Mus., vol. 24, pp. 567-593. Jordan, D. S.; Shigheo Tanaka~ and J, 0, Snyder 1913. A catalogue of the Eshes of Japan. Jour. Coll. Sci. Imperial Univ. Tokyo, vol. 38, art. 1, 497 pp Kamohara, Toshiji 1952. Revised descriptions of the offshore bottom-fishes of Prov. Tosa, Shikoku, Japan. Repts. Kochi Univ. Nat. Sci.,no. 3,1'22'pp. Khrlovac, Jozica 1949. Premidres d6couvertes de postlarves de la famille Argentinidae, en Adri- atique. Acta Adriatica, vol. 4, no. 4, 16 pp; 2 charts. Kendall, William Converse - 1908. Fauna of New Efigland. 8. List of.the Pisces. Occ. Pap. Boston. Soc. Nat.. Hist., no. 7, 152 pp. 1914. An annotated catalogue of the fishes of Maine. Proc. Portland Sac. Nat. Hist., vol, 8, pt. 1, 198 pA Kendall, W. C., and Donald R. Crawford 1922. Notice of a spiral valve in the teleostean fish Argentina sdus, with a dis- cussion of some other characters. Jour. Washington Acad. Sci., vol. 12, PP. 9-19. Kishinouye, Kamakichi 1904a. On nigisu, Argentina semifasciata, n. sp. Zoological Magazine, T6kyo, vol. 16, p. 110 [not seen]. 1 1 1958 COHEN: SUBFAMILY ARGENTININAE 167 1904b.The nigisu again (Argentina semifasciata Kish.). Ibid., vol. 16, pp. 197- 198. KIauber, Laurence Monroe 1948. Some misapplications of the Linnaean names applied to American snakes. Copeia, 1948, pp. 1-14. Kotthaus, Adolph, and Gerhard Krefft 1957. Fischfaunenliste der Fahrten mit F, F. S. "Anton Dohrn" nach Island- Grilnland. Ber. Deutsch Wissen. Komm. fur Meeresforschung,'new ser. vol. 14, no. 8, pp. 169-191. Kr0yer, Henrik Nikolaj ' 1846-1849. Danmarks fiske, vol 3, pt. 1. Copenhagen, 704 pp. - 1847. Poissons de la mer du Nord, regus, pendent le voyage scientifique en Sc/ndinavie, en Laponie, etc., ex6cut6 sous la direction de P. Gaimard, Paris. (Atlas of 20 plates, unaccompanied by explanatory text.) Kuroda, Nagamichi -\ , 1951. A nominal list with distribution of the fishes of Suruga Bay, inclusive of the freshwater species found near the coast. Japanese Jour. Ichthyol., vol. 1, pp. 814-888. Kyle,· Ifarry M„ and E. Ehrenbaum - 1929. Teleostei Physostomi. 1. Clupeiformes. In Die Fische der Nord und Ostsee. Leipzig, [no regular pagination]. Lac6pdde, Bernhard Germaifi Etienne 1808, Histoire naturelle des 'poissons, vol: 5, Paris, brviii + 892 pA, 18 215. Lilljeborg, Wilhelin 1891. Sveriges och Norges fiskar, vol. 3. Upsala, 788 pp. Linnaeus, Caroli 1744. Systema naturae. Editio quarta ab auct6re emendata and aucta. Paris, 108 + xxvi pp. [not seSn]. 1746. Museum Adolpho-Fridericianum. Stockholm,·48 pp., 2 pls. [not seen]. 1748. Systemae naturae. Editio sexta. Stockholm, pp. .ii + 224,8 pls. [not seen]. 1749. Museum Principis, Ainoenitates Academiae, vol. 1, pp. 277-826,2 tables. 1754. Pisces. In Museum S. R. M. Adolphi Fiederici Regis Suecorum. Stock- holm, 183 pp., 32 pls. [not seen]. 1756. Systema naturae. Editio ... auctior and emendatior. Leiden, pp. viii + 227, 8 pls. [not seen]. 1758. Systenia naturae. Editio decinia, vol. 1. Stockholm, 824 pp Longley; William H., and Samuel Hildebrand - 1941. Systematic catalogue of the fishes of Tortugas, Florida. Papers Tortugas Lab., vol. 84, Carnegie Inst., Washington, publ. 585, xiii + 381 pA, 34 pls. Lunnbdg, Einar 1896. Linnean type-specimens of birds, reptiles, batrachians and Ishes in the Zoological Museum of the R. University in Upsala. Bihang Till Svenska Vetenskapsakademiens Handlingar, vol. 22, afd. 4, no. 1, 45 pp. 168 BULLETIN FLORIDA STATE MUSEUM Vol. 8 Lord, Clive E,, and H. H. Scott 1924. A synopsis of the vertebrate animals of Tasmania. Hobart, 840 pp., pls. Lozano Rey, Luis 1947. Peces Ganoideos y Fisostomos. Mem. Real Academia de Ciencas Ex- actas, Fisicas y Naturales, Madrid, voI. 11, pp. xv + 885, 20 pls. McCulloch, Allan Riverstone 1911. Zoological results of the fishing experiments carried out by the F. I. S. "Endeavor" 1909-10, pt. 1. Rept. on the fishes, pt. 1. 87 pp., 16 pls. 1914. Zoological results of the fishing experiments carried out by the F. I. S. "Endeavor" 1909-1914, vol. 2, pt. 8. Rep'ort on some fishes obtained by the F. I. S. "Endeavor" on the coasts of Queensland, New South Wales, Victoria, Tasmania, South and South-western Australia, pt. 2, pp..77.165, pls. 13-24. 1921. Check-list of the fish and fish-like animals of New South Wales, pt. 2. Australian Zool., vol. 2, pt. 7, pp. 24-68, pls. 4-24. 1929.'A check-list of the fishes of Australia. Mem. Australian Mus., no. 5, - x + 554 pp, McKenzie, R. A. 1989. Some marine fish and.salp records, Proc. Nova Scotian Inst, Scir, vol. 20, pp. 15-20. McKenzie, R. S., and R. E. S. Homans 1987. Rare and interesting fishes ahd salps in the Bay of Fundy and off Nova Scotia. Ibid., vol. 19., pp 277.-281. - Maslov, N. 8. j 1944. The bottom food-fishes of the Barents Sea. Traris. Knipovich Polar Sci. Inst. of Sea-Fisheries and Ocean6gr., Murmansk, vol. 8, 186 pp., 48 tables. Matsubara, Kiyomatsu 1940. Studies on the deep-sea fishes of Japkn. XIII. On Prof: Nakazaw's col- lection .of fishes referable to Isospondyli„ Iniomi and Allotriognathi (1) Suisan Kenkiu-Shi, vol. 85, no. 12, pp. 815-319. 1948. Ichthyological annotations from the depth of the sea of Japan. I-VIII. Jour. Sigenkagaku,Kenkyusyo, vol. 1, pp. 37-81, 1 pl. 1955. Fish morphology and hierarchy, vol. 1. Tokyo, 788 pp. Meek, Alexander , - 1916. The migratiohs of fish. London, xvili + 427 pp., 11 pls, Moreau, Emile, 1881. Histoire naturelle des poissons de la France, vol. 8, Paris, 697 pp, Mori, Tamezo 1952. Check list of the fishes of Korea.- Mem. Hyogo Univ., Agric., vol. 1, no. 8 (biol. ser. no. 1), 228 pp., map. Muller, Otto Frederik 1776. Zoologiae Danicae prodromus, seu animalium Daniae et Norvegiae indi- genarum characteres nomina, et synonyma imprimis popularium. Hau- niae [- ?], xxxii + 282 pp. 1958 C0HEN: SUBFAMILY ARGENTININAE 169 Munro, Ian S. R. 1957. Handbook of Australian fishes (pp„ 29-82). Fisheries Newsletter, Can. berra,a:01. 16, pp. 15-18. Murray, John, and Johan Hjort 19f2. The depth5 of the ocean. London, xx + 821 pp., 9 pls., 4 maps. Nilsson, Sven - 1882. Prodromus ichthyoiogicae. Lundae [?'Lund, :Swedenl, iv + 124 pp 1855. Skandinavisk fauna. Fjerde delen: Fiskarna. Lund, xxxiv + 768 pp. Nirmi, Emilio 1912. Ca~logo di pesci del Mare Adriatico. Venice, 271 pp. , Norman, John Roxborough 1987. B. A. N. Z. Antatctic research expedition 1929-1981. Fishes. Reports- Series B (Zoology and .86tany) Adelaide, vol. 1, pp. 49-88. Ochiai;, Akira 1952. Ec616gical studies on a. clupeoid fish Argentina semifasciata Kishinouye. I. 'The relation between the body length and the body weight. Bull. Japanese Soc, Sci. Fish., vol, 18, pp. 139-146. 1954. [Same title. ] III. Gonad., vol. 19, PR. 968-972. Okada, Yaichir'o 1955. Fishes of Japan. Takyo, 484 'pp. Okada, Y., Uchida; and K. Matsubara , 1935. Iconography .of Japanese fishes: Tokyo, 425 pp., 166 pls. Oken, Lorenz 1886. Allgemeine naturgeschichte fiir alle stlinde, vol. 6. Stuttgart, 698 pin P.alombi, Arturo, and Mario' Santarelli 1958. Gli animali commestibili dei maYi d'Italia. Milan, 349 PA - Poll,·Max 1958. Poissons. III. T616osuens , Maladoptdrygiens. Exp6d. Oc6anogr. Belge dans les eaux c6ti@res~ Afficaines de.1'Atlantique Sud (1948-1949). R6- sultats Scientifique, vol. 4, fasc. 2,285 pp., 8 pls. Phillipps, William John , 1927a. Bibliography Of Ndw Zealand F.ishes. Wellington: New Zealand Marine ~ Dept., Fish. Bull. no. 1, 68 pp 1927b. A check«list of the Rshes o,f New Zealand. Jour. Pan-Pacific Res. Inst., vol. 2, pp. 9-15. 1940. The fishes of. New Zealand, vol. 1. New Plymouth, New Zealand, vii + 87 pp. Rafinesque-Schmaltz, Constantine Samuel . 1810. Indice d'ittiologia Sicilana. Messida, 70 pp., 2 pls. Ray, John 1710.- Syrtopsis methodica avium & piscium. London [not seen]. Regan, Charles Tate 1918. The Ar;tardic fishes Of the Scottish Nati6nal Antarctic Expedition. Trans. Roy. Soc. Edinburgh, vol. 49, pt. 2, pp. 229-292, 11 pIs. 170 BULLETIN FLORIDA STATE MUSEUM, Vol. 8 Reguis, J. B. Marius 1882. Essai sur l'histoire naturelle des vertdbr65 de la Provence et des d6Darte- ments circonvoisins. Vertdbrds anallantoidiques (poissons et batraciens). Marseille, 429 pp. - Reinhardt, Johannes Christopher Hagemann , 1883. Bemtirkningen til den Skandinaviske ichthyologie. S7 pp. Risberg, Gustave 1885. Observationes ichthyologicae. Lundae [? Lund, Sweden], [not seen]. , Risso, Antoine ' 1826. Histoire naturelle des principales productiones de l'Europe Meridionale et particulierment de celles environs de Nice et des Alpes Maritimes, vol. 8. Paris, xvi + 480 pp., 16 pls. ' Rondelet, Cillielmus 1554. Libri de piscibus marinus, in quibus verae piscium effigies expressae sunt. Lugduni [? Lyons], 14, 588, 28 pp. Saemundsson, Bjarni 1949. Marine Pisces. The, zoology of. Iceland, vol, 4, pt. 72. Copenhagen, ' 150 pp Sanzo, Luigi 1933. Salinonoidei. Uova, larvae e stadi giovadili di Teleostei. Fauna e Flora del Golfo di' Napoli, Monogr. 88, PA 21-42; pls: 2-8. Scheuying, Ludwig , 1929. Die Wanderungen def Fische, pt. 1. , Ergebnisse der Biologie, vol 5, 691 pp Schmidt, 'Ernst Johannes 1906. On the larval and pMt-larval development of· the argentines (Argentina - situs (Asean.) and Argentina sphyraena Linne) with some notes,on Mal- lotus uillosus '(O. F.- Muller). Meddelelser K6mmissionen f6r Havunders- 0gelser: Ser. Fiskeri, vol. 2, no. 4, 29 pp., 2 pls. , 1918. Argentinidae, Microstomidae, Opisthoproctidae, Mediterranean Odonto- stomidae. Rept. Danish Oceanogr. Exped., 1908-1910 to the Mediterran- - ean and adjacent seas, vol. 2,.no. 4, art. 5,40 pp., 4 charts. Schmidt, Peter J. 1981. Fishes of Japan collected in 1901. Trans.. Pacific Comm, Acad. Sci., . USSR, *01.2, 176 pp Schnakenbeck, W. 1981. Argehtind situs and Argentina sphuraena, In Joubin, L, (ed.) Faune , , Ichthyologique de l'Atlantique Nord, no. 8. Conseil Permanent· Internat- ional Pour EExploration de la Men [No pagination.] Schroeder; William 1955. Reports' on the results of ex]Dloratory otter-trawling along the continental shelf and slope between. Nova Scotia and Virginia during 'the summers of '1952 and 1958. Pap. Marine Biol. and Ocean6gr., Deep-sea Research„ suppl. to vol. 8. pp. 358-372. Smith, J. L. B. 1958. The ·sea fishes of southern Africa. R]evised editjon. Capetown, xvi + 564 pp., 107 pls. - 1958 COHEN: SUBFAMILY ARGENTININAE 171 Smith, W. C. 1928. Occurrence of Argentina sphyraena. ' Proc. Liverpool Bibl. Soc., vol. 42, p. 88. Smitt, Frederik Adam 1895. A history of Scandinavian fiShes. By B. Fries„ C. U. Ekstrom, and C. Sundevall, pt. 2,2nd edit., revised and completed by F. A. Smitt. Stock- holm, pp. 567-1240: - Soljan, Tonko , 1948. Ribe Jadrina. Zagreb, 487 pp. Sonnini de Maboncourt, Charles Nicolas Sigisbert 1804. Histoire naturelle generale et particuliere des poissons. Ouvrage faisant suite a l'histoire naturelle. . . par Leclerc Buffon, vol. 12, 482 pA, pls. , 62-72. Springer, Stewart, and Haryey R. Bullis 1956. Collettions by the Oregon in the Gulf of Mexico. U.S. Fish and Wild- , life. Serv. Special Sci. Rept. Fisheries, no. 196, 184 pp. Steindachner, Franz 1868. Ichthyologischer Bericht iiber eine nach Spanien und Portugal unte,r- nommene Reise. Uebersicht der Meeresfische an den Kusten Spaniens und Portugals (II).. Sitzb. Kaiserlichen Akademie Wissenshaften in Wien. Math.-Naturwiss. classe, vol. 57, pp. 667-789, 6 pls. , Swainson, William 1889. The natural history of fishes, amphibians, & reptiles or monocardian ani- mals, vol. 2. London, 488 pp. Tambs-Lyche, Hans 1954. The fish fauna of the fjords between Vaksdal and Romarheim in the. Bergen District. Univ. Bergen Aarb6k, 1954, Naturvitenskapelig rekke ha. 5, 17 pp. Tanaka, Shigeho 1981. On the distribution of fishes in Japanese waters. Jour. Fae. Sci. Imperial Univ. Tokyo, sect. 4, zool,, vol, 8, pt. 1, 90 pp., 8 pls. Tortonese, Enrico, and Leopoldo Trotti 1949. Catalogo dei pesci del Mar Ligure. . Centro Talassografico Tirren6, Dubl. 1. Atti dell Academia Ligure di Scienze ed· Lettere, vol. 6, fasc. 1, 118 'pp. Ulrey, Albert B. 1929. A check-list of the fishes of southern California and Lower California, Jour. Pan-Pacifc Res. Inst., vol. 4, no. 4, pp. 2-11. Ulrey, A. B., and Paul O. Greeley 1928. A list of the marine fishes (Teleostei) of southern California with their distribution. Bull. South. California Acad. Sci., vol. 28,· no. 1, 58 pp. Valenciennes, Achille 1848. Hist6ire naturelle des poissons, vol. 21. Paris, xiv + 513 pp. Valiani, S. 1940. Contributo alla biometria e allo studio dell'alimentazione dei pesci (Ar- 172 BULLETIN FLORIDA STATE MUSEUM Vol. 8 gentina sphuraena L.). Bollettino di Pesca, Piscicoltura e Idrobiologia, voI. 16, pp. 125-129. ' Vladykov, Vadim D., and R. A. McKenzie 1985. The marine fishes of Nova Scotia, Proc, Nova Scotian Inst. Sci., vol. 19, pp. 1-8. Waite, Edgar Ravenswood 1907. A basic list of the fishes of New Zealand. Rec. Canterbury Mus., vol. 1, no. 1, 89 pp. 1911. Scientific results of the New Zealand Government trawling expedition, 1907, Pisces, pt. 2. Ibid., vol, 1, no, 3, pp 157-272, pls. 24«57. 1921. Catalogue of the fishes of South Australia. Rec, South Australian Mus., vol. 2, no. 1, 208 pp., 1 pl. 1928. The fishes of South Australia. Handbooks of the Floraand Fauna of South Australia. British Science Guild (South Australia Branch),243 pp. 1928. Check list of the marine fishes of South Australia. Jour. Pan-Pacific Res. Inst., vol. 3, pp. 8-10. Walbaum, Johann Julius 1792. Ichthyologiae. Pars. III. Petri Artedi sueci genera piscium. Grypes- . waldiae [= ?], 728 pp., 3 pls. 1793. [Same title.] Pam IV. Petri Artedi Angermannia-Sueci synonymia nominum Discium fere omnium. Crypeswaldiae [- ?], 140 pp Weiler, Wilhelm 1943. Die Otolithen aus dem JungtertiUr Sud-Rumdniens. 1. Buglow und Sarmat. Senckenbergiana, 'vol. 26, pp. 87.-115, 1 pl 1950. Die Otolithen aus dem Jung-TertiUr Sud-Rumiiniens. 2. Mittel-Mioziin, Torton, Buglow und Sarmat. Ibid., vol. 31, pp- 209-258, 12 pls. Went, A. E. J. 1951. A specimen of-an Argentine, Argentina sphyraena L. from the Irish Sea. Irish Nat. Jour., vol. 10, p. 197. - Whitley, Gilbert P. 1954. New locality records for some Australian fishes. Proc. Roy. Soc. New South Wales, 1952-58, pp. 23-30. Willughby, Francis 1686. Historia piscium libri quator. Oxford, 348, 80 pp., 168 pls. Winther, Georg 1879. Prodromus Ichthyologiae Danicae Marinae: Fortegnelse over di i de Dansk farvande hidtil fund fiske. Naturhistorisk Tidsskrift, ser. 8, vol. 12, pp. 1-68. ¥arrell, William 1888. On a new species of smelt from the Isle of Bute (Osmerus hebredicus). - Rept. Bfitish Assoc. Adv. Sci. for 1888, no. 8, pt. 2, pp. 108-109. 1859. A history of British fishes, Edit. 3, vol. 1. Edited by Sir John- Richard- son. ' Londod, pp. xxxviii + 675. 1860. Supplement to 2nd volume of the history of British fisbes, editions 1 and 2. Edited by Sir John Richardson. London, 72 pp. - - Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM may be in any field of biology. Manuscripts dealing with natural history or systematic problems involving the southeastern United States or the Caribbean area are solicited especially. Manuscripts should be of medium length-12 to 200 printed pages. Examination for suitability is made by an Editorial Board. The BULLETIN is distributed worldwide through institutibnal subscriptions and exchanges only. It is considered the responsibility of the author to distribute his paper to all interested individuals: To aid in this, fifty copies are furnished Khe author without cost. PREPARATION OF MANUSCRIPT Highly recommended as a guide is the "Style sheet for the scientific serial publica- tions..of' the American Museum of. Natural. History," second edition, revised, 1958. Manuscripts should be typewritten with double spacing, with ample maigins, and on only one side of the paper, The authof should.keep a copy; the copy submitted must be the original. Tables, legends of figures, and all footnotes should be as- sembled separate from the text. Several legends or footnotes may be placed on a single sheet. Illustrations, including maps and photographs, should be referred to as "figures" wherever possible. All illustr-ations are reduced to a maximum of 444 by 748 inches. The scales, wherever it is necessary, should be incorporated into the figure. All references to literature should conform with the bibliographic style used in recent numbers of the BULLETIN. Spell out in full the titles of non-English serials. Footnote" material should be kept t6 a minir*um. However, provide copy for a footnote detailing the title, affiliations, and address of the author (see recent numboks of the BULLETIN). Manuscripts must be accompanied by a synopsis-a brief and factual summary, (not a mere description) of tbe contents and conclusi6ns, which points out the piesence. of any new information and indicates its relevance. In it list all new organisms described and give their ranges; indicate all taxonomic changes pro- posed. The synopsis, written in full sentences, sl\ould bd concise, but completely intelligible in itself without reference to the paper, thereby enabling. the busy reader to decide more surely than he can from the title alone whether the paper merits his reading. The synopsis will be published with the paper, hence it does not replace the usual conclusions 6r summary sections: It will also serve as copy for the abstracting services. Manuscripts and all editorial matters should be addressed to: Editor of the BULLETIN Flint Hall University of Florida Gainesville, Florida / ..'"/'Ililillillillillillillillillillillillillil