(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Fishes of the western North Atlantic. Editorial board: editor-in-chief John Tee-Van [and others]"

Fishes of the Western North Atlantic 



MEMOIR 

SEARS FOUNDATION FOR MARINE RESEARCH 

Number I 

Fishes of the 
Western North Atlantic 





PART THREE 

Soft-rayed Bony Fishes 

Class Osteichthyes 

Order Acipenseroidei, Order Lepisostei 

Order Isospondyli, Suborder Elopoidea 

Suborder Clupeoidea^ Suborder Salmonoidea 

NEW HAVEN ig63 
SEARS FOUNDATION FOR MARINE RESEARCH, YALE UNIVERSITY 



PRINTED IN DENMARK FOR 

SEARS FOUNDATION FOR MARINE RESEARCH 

BINGHAM OCEANOGRAPHIC LABORATORY 

YALE UNIVERSITY 

Yngve H. Olsen, Editor 

Bianco Luno's Printing, Copenhagen, Denmaric 



Fishes of the 
Western North Atlantic 




Authors 



HENRY B. BIGELOW 

Museum of Comparative Zoology 



MARGARET G. BRADBURY 

Hopkins Marine Station 

JOHN R. DYMOND 

University of Toronto 

JOHN R. GREELEY 

New York Conservation Department 

SAMUEL F. HILDEBRAND 

U. S. Fish and Wildlife Service 

GILES W. MEAD 

Museum of Comparative Zoology 



ROBERT R.MILLER 

The University of Michigan 

LUIS R. RIVAS 

University of Miami 

WILLIAM C. SCHROEDER 

Woods Hole Oceanographic Institution 

ROYAL D. SUTTKUS 

Tulane University 

VADIM D. VLADYKOV 

University of Ottawa 



NEW HAVEN, 1963 



SEARS FOUNDATION FOR MARINE RESEARCH, YALE UNIVERSITY 



Editorial Board 



Editor-in-Chief HENRY B.BIGELOW Mmeum of 

Comparative Zoology, Harvard University 



CHARLES M. BREDER 

American Museum of Natural History 
New Tork 



YNGVE H. OLSEN 

Bingham Oceanographic Laboratory 
Tale University 



DANIEL M. COHEN 

U. S. Fish and Wildlife Service 
Washington, D. C. 



WILLIAM C. SCHROEDER 

Museum of Comparative Zoology 
Harvard University 



GILES W. MEAD 

Museum of Comparative Zoology 
Harvard University 



LEONARD P. SCHULTZ 

United States National Museum 
Washington, D. C. 



DANIEL MERRIMAN 

Bingham Oceanographic Laboratory 
Tale University 



JOHN TEE-VAN 

New Tork Zoological Society 
Bronx, N. T. 



Table of Contents 



Preface xiii 

Introduction xvii 

Maps 

BONY FISHES 

Superclass, Class, Subclasses, and Orders, by H. B. Bigelow i 

Superclass Pisces 2 

Class Osteichthyes 2 

Key to Orders i6 

Text and Footnote References i8 

Names of Bones, by G. W. Mead and Margaret G. Bradbury 20 

Order Acipenseroidei, by V. D. Vladykov and J. R. Greeley 24 

Family Acipenseridae 26 

Genus Acipenser 28 

Acipenser brevirostris 36 

Acipenser fulvescens 41 

Acipenser oxyrhynchus oxyrhynchus 46 

Acipenser oxyrhynchus desotoi 5^ 

Text and Footnote References 59 

Order Lepisostei, by R. D. Suttkus 61 

Family Lepisosteidae 68 

Genus Lepisosteus 69 

Lepisosteus oculatus "Ji 

Lepisosteus osseus 75 

Lepisosteus platyrhincus 80 

Lepisosteus spatula 83 

Text and Footnote References 87 

Order Isospondyli, Composite Authorship 

Characters and Keys to Suborders and Families, by H. B. Bigelow 89 

Key to Suborders 95 

Key to Families of Isospondyli, Iniomi, and 

Giganturoidei 97 

Text and Footnote References 105 

Suborder Elopoidea, Composite Authorship 

Characters and Key to Families, by H. B. Bigelow 107 

Text and Footnote References 109 



Editorial Board 



Editor-in-Chief HENRY B.BIGELOW Mmeum of 

Comparative Zoology, Harvard University 



CHARLES M. BREDER 

American Museum of Natural History 
New York 

DANIEL M. COHEN 

U. S. Fish and Wildlife Service 
Washington, D. C. 

GILES W. MEAD 

Museum of Comparative Zoology 
Harvard University 

DANIEL MERRIMAN 

Bingham Oceanographic Laboratory 
Tale University 



YNGVE H. OLSEN 

Bingham Oceanographic Laboratory 
Tale University 

WILLIAM C. SCHROEDER 

Museum of Comparative Zoology 
Harvard University 

LEONARD P. SCHULTZ 

United States National Museum 
Washington, B.C. 

JOHN TEE-VAN 

New Tori Zoological Society 
Bronx, N.T. 



Table of Contents 



Preface 

Introduction 

Maps 



XIII 

xvii 



BONY FISHES 

Superclass, Class, Subclasses, and Orders, by H. B. Bigelow i 

Superclass Pisces 2 

Class Osteichthyes 2 

Key to Orders i6 

Text and Footnote References 1 8 

Names of Bones, by G. W. Mead and Margaret G. Bradbury 20 

Order Acipenseroidei, by V. D. Vladykov and J. R. Greeley 24 

Family Acipenseridae 26 

Genus Acipenser 28 

Acipenser brevirostris 36 

Acipenser fulvescens 41 

Acipenser oxyrhynchus oxyrhynchus 46 

Acipenser oxyrhynchus desotoi 56 

Text and Footnote References 59 

Order Lepisostei, by R. D. Suttkus 6 1 

Family Lepisosteidae 68 

Genus Lepisosteus 69 

Lepisosteus oculatus 7 1 

Lepisosteus osseus 7^ 

Lepisosteus platyrhincus 8 o 

Lepisosteus spatula 83 

Text and Footnote References 87 

Order Isospondyli, Composite Authorship 

Characters and Keys to Suborders and Families, by H. B. Bigelow 89 

Key to Suborders 95 

Key to Families of Isospondyli, Iniomi, and 

Giganturoidei 97 

Text and Footnote References 105 

Suborder Elopoidea, Composite Authorship 

Characters and Key to Families, by H. B. Bigelow 107 

Text and Footnote References 109 



viii Table of Contents 

Editorial Comments on the Hildebrand Manuscripts iio 

Family Elopidae, by S. F. Hildebrand 1 1 1 

Genus Tarpon 1 1 2 

Tarpon atlantkus 113 

Genus Elops 123 

Elops saurus 124 

Family Albulidae 132 

Genus Albula 133 

Albula vulpes 134 

Genus Dixonina 143 

Dixonina nemoptera 143 

Text and Footnote References 146 

Suborder Clupeoldea, Composite Authorship 

Characters and Key to Families, by H. B. Bigelow 148 

Text and Footnote References 151 

Family Engraulidae, by S. F. Hildebrand 152 

Genus Anchovia 155 

Anchovia clupeoides 155 

Anchovia nigra 158 

Genus Anchoa 162 

Anchoa spinifer 167 

Anchoa argenteus 169 

Anchoa duodecim 1 7° 

Anchoa trinitatis 171 

Anchoa cay or urn 173 

Anchoa pectoralis 1 74 

Anchoa mitchiUi 176 

Anchoa parva 181 

Anchoa januaria 1 8 3 

Anchoa choerostoma 185 

Anchoa tricolor 186 

Anchoa cubana 188 

Anchoa ginsburgi 1 90 

Anchoa I ampro taenia 192 

Anchoa hepsetus 194 

Anchoa lyolepis 200 

Anchoa filifera 202 

Genus Anchoviella 204 

Anchoviella eury stole 208 

Anchoviella estauquae 211 

Anchoviella cayennensis 212 

Anchoviella perfasciata 213 



Table of Contents ix 



Anchuviella brevirostris 


219 


Anchoviella guianensis 


'2.1'] 


Anchoviella alleni 


219 


Anchoviella nattereri 


220 


Anchoviella pallida 


221 


Anchoviella elongata 


223 


Anchoviella lepidentostolc 


224 


Anchoviella hlackburni 


226 


Anchoviella (Amplnva) jamesi 


227 


Genus Pterengraulis 


228 


Pterengraulis atherinoides 


229 


Cienus Hildebrandichthys 


230 


Hildebrandiclithys setiger 


231 


Genus Lycengraulis 


233 


Lycengraulis grossidens 


234 


Lycengraulis limnichthys 


237 


Lycengraulis batesii 


240 


Lycengraulis abbotti 


242 


Lycengraulis barbouri 


243 


Genus CetengrauUs 


245 


Cetengraulis edentulus 


245 


CetengrauUs juruensis 


247 


Text and Footnote References 


249 


Family Alepocephalidae (Interim Account), by H. B. Bigelow 


250 


Text and Footnote References 


253 


Family Searsiidae (Interim Account), by H. B. Bigelow 


254 


Text and Footnote References 


256 


Family Clupeidae, by S. F. Hildebrand; except 




Genus Harengula by L. R. Rivas, and 




Genus Dorosoma by R. R. Miller 


257 


Genus Etrumeus 


262 


Etrumeus sadina 


263 


Genus Jenkinsia 


267 


Jenkinsia lamprotaenia 


268 


'Jenkinsia viridis 


272 


Genus Clupea 


274 


Clupea harengus 


275 


Genus Alosa 


293 


Alosa sapidissima 


295 


Alosa alabamae 


308 


Genus Pomolobus 


312 


Pomolobus chrysochloris 


z^i 



Table of Contents 

Pomolobus mediocris 3 1 9 

Pomolobus aestivalis 324 

Pomolobus pseudoharengus 332 

Genus Brevoortia 342 

Brevoortia tyr annus 346 

Brevoortia brevicaudata 363 

Brevoortia patronus 365 

Brevoortia smithi 372 

Brevoortia gunteri 376 

Genus Opisthonema 380 

Opisthonema oglinum 3 8 i 

Genus Harengula, by L. R. Rivas 386 

Harengula humeralis 387 

Harengula clupeola 390 

Harengula pensacolae 393 

Genus Sardinella 397 

Sardinella pinnula 399 

Sardinella anchovia 401 

Sardinella brasiliensis 407 

Genus Rhinosardinia 411 

Rhinosardinia amazonica 412 

Rhinosardinia serrata 413 

Genus Ilisha 4 1 5 

Ilisha altamazonica 4 1 7 

Ilisha castelnaeana 419 

Ilisha amazonica 42 i 

///j/^<z harroweri 423 

Ilisha narragansetae 426 

Genus Pristigaster 427 

Pristigaster cay ana 428 

Genus Odontognathus 430 

Odontognathus mucronatus 431 

Odontognathus compress us 433 

Genus Neoopisthopterus 435 

Neoopisthopterus cub anus 436 

Genus Chirocentrodon 438 

Chirocentrodon bleekerianus 439 

Genus Dorosoma, by R. R. Miller 443 

Dorosoma cepedianum 444 

Dorosoma petenense 448 

Text and Footnote References 452 



Table of Contents xi 

Suborder Salmonoidea, Composite Authorship 

Characters and Key to Families, by H. B. Bigelow 455 

Family Salmonidae, by J. R. Dymond 457 

Genus Salmo, by J. R. Dymond 459 

Salmo salar 460 

Salmo trutta 498 

Salmo gairdneri 499 

Text and Footnote References 500 

Genus Salvelinus, by H. B. Bigelow 503 

Salvelinus alpinus 507 

Salvelinus fontinalis 5^5 

Genus Cristivomer, by H. B. Bigelow 542 

Cristivotner namaycush 543 

Text and Footnote References 545 

Family Coregonidae, by H. B. Bigelow 547 

Genus Coregonus 549 

Coregonus clupeaformis 549 

Text and Footnote References 552 

Family Osmeridae, by H. B. Bigelow and W. C. Schroeder 553 

Genus Osmerus 554 

Osmerus eperlanus mordax 559 

Genus Mallotus 573 

Mallotus villosus villosus 576 

Text and Footnote References 595 

Index of Common Names 599 

Index of Scientific Names 606 



Preface 



THE SYSTEMATIST has divided the Superclass Pisces into two major groups: 
the elasmobranchs with an internal supporting skeleton of cartilage only, and bony 
fishes with a skeleton consisting of true bone in greater or lesser amount, or some deri- 
vation of it. Parts i and 2 of this series have already treated the sharks, skates, and rays, 
which comprise the elasmobranchs. With publication of Parts 3—5, this series enters the 
realm of bony fishes, by far the larger and more diverse of the two groups. The fishes 
treated here, commonly referred to as "soft-rayed bony fishes," include the anadromous 
sturgeons and gars, the wide-ranging and numerous groups of herring-like isospondylids, 
the little-known giganturoids, the broad assemblage of iniomes, and finally the strange- 
looking lyomerids. The Introduction and Table of Contents for each volume provide 
detailed information as to the subjects treated by the various authors. 

The diverse shapes and characters of the forms detailed in these volumes, well 
known to the ichthyologist and many students of oceanography, constitute a notable 
display of the devious developments and modifications that have occurred in the course 
of evolutionary changes throughout time. This is also quite apparent to the average 
reader from a mere glance at the illustrations. Certainly none among nature's multi- 
plicity of creatures is more bizarre than the oddly shaped hatchet fish with its tubular 
eyes and luminous organs, or the pelican fish with its long eel-like body and an enormous 
mouth that allows it the unusual ability to devour prey larger than itself. Indeed, so 
unusual and diverse are many of the groups treated here that even the trained system- 
atist is put to it even to approximate their correct position on the ichthyological tree. 
Parallel with their diverse development in form, we also find pronounced modifications 
in their mode of life and habitat, as in the anadromous sturgeons and gars who have 
assumed the capacity to live in both fresh and salt water, and the deep-sea creatures 
who have adjusted to life at great depths under tremendous pressure where no sunlight 
penetrates. 

In Parts 3—5, as in the two previous volumes, our objectives have remained the 
same — to provide for both professional and nonprofessional readers "critical reviews and 
revisions of each group rather than perfunctory compilations or mere reprintings of 
previously published works." The extent to which these objectives are fulfilled in each 
section will be determined by the usefulness of the account to both groups of readers. 



xiv Preface 

With multiauthorship, instead ot dual authorship as in the case of Parts i and 2, and 
with wide variation in both subject matter and available material, treatment and em- 
phasis vary extensively according to authorship. All of the papers in these three Parts 
have been edited by me to a greater or lesser extent, except the Family Sternoptychidae 
in Part 4, which is published as it was submitted by the Editors-in-Chief. 

All of these volumes should prove to be of considerable value in one way or an- 
other to the ichthyologists and to other students of the sea. To the average reader, and 
to sport and commercial fishermen in particular. Part 3 will be of special interest, for 
here is contained a large accumulation of facts about the tarpon, ladyfish, bonefish, shad, 
salmon, and trout, all of which are sportsmen's favorites, and about the sturgeon, an- 
chovy, herring, pilchard, sardine, smelt, menhaden, and shad, many of which are im- 
portant to man either directly as table food or indirectly as food for the larger in- 
habitants of the sea that are of economic value to man. 

As in the case of Parts i and 2, the geographic area for these studies has been 
arbitrarily defined as that extending from the vicinity of Hudson Bay, just below the 
fringe of the Arctic, southward to the tropical latitudes of the Amazon, and seaward 
more or less to the mid-Atlantic, including Bermuda and all of the Caribbean archipel- 
ago. Although most of the accounts in these volumes treat fishes that spend at least 
part of their life-span in waters ranging from brackish to maximum salinities, a few, 
known at present from fresh water only, have been included when, in the author's 
opinion, they are likely to be taken in brackish or saline situations at some future date. 

The format, general organization, and arrangement of the included material have 
been modified from Parts i and 2 only insofar as the subject matter and the demands 
of economy dictated. In Parts i and 2 there was excessive footnoting, and this has been 
eliminated largely by the use of numbered references. Long museum and institutional 
names have been replaced by the abbreviations listed in the Introduction for each Part. 

It is regrettable that a number of deep-sea groups — particularly the isospondylous 
Alepocephalidae and Searsiidae and the iniomous Myctophidae and Neoscopelidae — 
are represented in their respective positions by only interim accounts, inasmuch as the 
detailed studies could not be completed for these volumes. The genus Cyclothone, though 
treated in somewhat more detail than the groups just noted, also awaits further study. 
It is hoped that these missing sections can be incorporated in the following Parts. 
While these volumes were in preparation, a comprehensive study of the Searsiidae by 
A. E. Parr has appeared in Dana Report No. 51, i960, to which the reader is referred 
in lieu of a detailed account here. 

It is of interest to note that the contents of these three volumes were originally 
intended for inclusion in Part 3 only, a commentary indicating quite clearly the impos- 
sibility of predicting the ultimate total number of volumes for this series. Because of 
the redistribution of the subject matter into three volumes, the sections on Bony Fishes 
and Class IsospondyH in Part 3, prepared originally when only one volume was assumed, 
now obtains for all three volumes. Of particular importance in these sections are the 
Keys — to Orders in Bony Fishes, and to Suborders and Families in Class IsospondyH. 



Preface xv 

The Keys and other material in these two sections have not been repeated in Parts 4 
and 5. 

Despite the ever-increasing progress of ichthyology, and of oceanography in gen- 
eral, and despite the impressiveness of the contents in these volumes, the ichthyologist, 
and anyone else for that matter who becomes acquainted with the contents here, cannot 
avoid being impressed with the great gaps in our knowledge of fishes, particularly with 
regard to the deep-sea groups in Parts 4 and 5. For all the study, observation, and ex- 
ploration throughout the centuries, our present knowledge is indeed meager when com- 
pared to what is unknown. Taxonomic descriptions aside, our knowledge of the deep- 
sea fishes is limited primarily to speculation — to what may be surmised from their shape 
and color, stomach contents, or location of capture. Of many shallow-water forms we 
know little more. And even of those that have been subjects of extensive investigation, 
such as the salmon and some of the commercially important herring-like groups, much 
of what is known appears to be contradictory or debatable. 

Many species are as yet unknown, and of those now known, many are still re- 
presented by only one or a few specimens, these often damaged. Not until larger and 
more extensive collections are assembled will it be possible to clarify the inter-relation- 
ships of many species and groups, to obtain a more complete knowledge of distribu- 
tion in area and depth, and much other important information. For the great majority 
of deep-sea fishes, the records of depth distribution are sparse or crude at best, a 
lack that stems largely from the fact that most of the deep-sea collections have been 
taken with open nets rather than closing nets. With the former device, one knows only 
that a collected specimen frequented some level of the water column between the sur- 
face and the estimated fishing depth of the net. Even with the closing net, exact depth 
is questionable because of uncertainties in determining the wire angle. Add to this the 
fact that the scientific expeditions that have explored the depths have been few and 
widely dispersed. Considerable areas of all oceans, more so in some than in others, still 
remain unexplored. 

Even as the present volumes constitute a tribute to the expeditions and countless 
investigators who have contributed to our present knowledge of fishes, so do they re- 
emphasize, by what is not known, the need for earnest and continued expansion of 
marine exploration and investigation, particularly at a time when exploration of the 
"sea" beyond our atmosphere holds the stage. Especially notable in Parts 4 and 5, 
dealing mainly with the deep-sea isospondylids and iniomes, are the contributions of the 
Oregon, Silver Bay, and Combat, whose surveys, sponsored by the U.S. Fish and 
Wildlife Service, have provided a wealth of study material. Similarly, the contributions 
of specimens obtained from stomach contents during investigations of the tuna and 
other commercially important fish have enhanced many accounts; in fact, a number of 
species, presumably fast-swimming forms that are consistently successful in evading 
the slow-moving nets, are known only from such a source. Hand in hand with more 
extensive collecting must go technical development of gear such as closing nets, devices 
for accurate measurement of wire angle, underwater luminescent and photographic 



xvi Preface 

equipment, and a long list of other essentials. For experimental work at sea, fully 
equipped floating laboratories are needed to study live specimens at the source. And 
perhaps most important of all is the need for an increasing number of well-trained 
marine biologists and technicians. 

The Sears Foundation extends its gratitude to Henry Bigelow and Giles Mead for 
their various contributions in bringing these three volumes to fruition, particularly for 
harvesting the manuscripts. To those authors who have given their generous cooper- 
ation in achieving uniformity in the presentation of the material and in supplying ad- 
ditional information, I extend special thanks. Many of the authors have acknowledged 
the help and cooperation of individuals and institutions; to them, and particularly to 
the many who have been omitted, I add the gratitude of the Sears Foundation. It is also 
a special privilege to note the continued interest and confidence of Henry Sears, who 
has generously aided in the financing of these volumes, and to acknowledge financial 
support of various author investigations by the National Science Foundation. In the past 
two or three years, the Editorial Board has been expanded by the addition of Daniel 
Cohen, Giles Mead, and Daniel Merriman to its roster; to them and to the older mem- 
bers of the Editorial Board, the Sears Foundation is grateful for many and varied 
services. 

Finally, it is with no small regret that I note the discontinuance of Albert Parr's 
active participation in the functions of the Sears Foundation and in the publication of 
these Memoirs and the Journal of Marine Research, both of which were conceived and 
brought into being as a result of his sincere and deep interest in all things marine and of 
his keen perception of the need for vehicles to distribute the thoughts and scientific 
observations of an expanding oceanographic community. The results seen in the publi- 
cations themselves speak for his contribution far more significantly than mere words 

Yngve H. Olsen 
May 1963 Editor 



Introduction 



PART 3 OF THIS Series deals with the more primitive groups of soft-rayed bony 
fishes. The volume commences with a characterization of the class Osteichthyes (bony 
fishes) and comparison of this group with the class Chondrichthyes (elasmobranchs), 
followed by a characterization of the subclass Actinopterygii and a key to the various 
orders dealt with here and in Parts 4 and 5 : Acipenseroidei, Lepisostei, Isospondyli, 
Giganturoidei, Iniomi, and Lyomeri, all of which were originally to be included in this 
volume. The characterization and discussion of the Isospondyli concludes with a short 
key to the suborders and an extensive key to the external characters of all the families 
of the orders noted above; this, then, also applies to Parts 4 and 5 as well as to this 
volume. Because of the large number of groups in the Isospondyli, only part of the 
families, mostly shallow-water, are included in this volume; the rest, mainly deep-sea, 
are relegated to Part 4. Thus, in addition to the sturgeons and gars, this volume includes 
accounts of the tarpon, ladyfish, bonefish, the numerous anchovies, the herring, men- 
haden, shad, sardine, pilchard, gizzard shad, round herring, salmon, trout, and smelt. 

The geographic area covered extends from Hudson Bay southward to the Amazon 
and seaward more or less to the mid-Atlantic, with Bermuda and the Caribbean areas 
included. For the most part, only fishes living in brackish water or salt water have 
been included. 

Unfortunately, the accounts of the Alepocephalidae and Searsiidae could not be 
completed for inclusion in this volume. These are therefore represented by brief interim 
accounts that contain enough information for general identification of specimens in this 
category. 

In these volumes, museum and institutional names have been abbreviated ac- 
cording to the following list: 

AM — Amsterdam Museum, Holland 

AMNH — American Museum of Natural History 

ANSP — Academy of Natural Sciences of Philadelphia 

BLBG — Biological Laboratory, U. S. Fish and Wildlife Service, Bureau 

of Commercial Fisheries, Brunswick, Georgia 

BMNH — British Museum (Natural History), London 

BNM — Bergens Museum, Norway 



XVlll 



Introduction 



BOC — Bingham Oceanographic Collection, Yale University 

BU — Boston University 

BrU — Brown University 

CAS — California Academy of Sciences 

CF — Carlsberg Foundation, Copenhagen 

CFG — California Division of Fish and Game 

CM — Carnegie Museum 

CNHM — Chicago Natural History Museum 

CU — Cornell University 

FSM — Florida State Museum 

I J — Institute of Jamaica, B. W. I. 

IRSNB — Institut Royal des Sciences Naturelles de Beige, Belgium 

LMNH — Leiden Museum of Natural History, Holland 

MCZ — Museum of Comparative Zoology, Harvard University 

MMF — Museu Municipal do Funchal, Madeira Islands 

MHNBA — Museo de Historia Natural de Buenos Aires, Argentina 

MNHN — Museum National d'Histoire Naturelle, Paris 

MOM — Musde Oceanographique, Monaco 

MRAC — Musee Royal d'Afrique Central, Tervuren 

MSNF — Museo di Storia Naturale, Firenze 

NHR — Naturhistoriska Riksmuseum, Stockholm 

NYZS — New York Zoological Society 

POFI — Pacific Oceanic Fishery Investigation, U.S. Fish and Wildlife 
Service (Bureau of Commercial Fisheries), Honolulu, Hawaii 

ROMZ — Royal Ontario Museum of Zoology, Canada 

SIO — Scripps Institution of Oceanography 

SU — Natural History Museum, Stanford University 

TU — Tulane University 

UCLA — University of California at Los Angeles 

UF — University of Florida 

UI — University of Indiana 

UL — University of Louisville 

UMIM — University of Miami Ichthyological Museum 

UMML — University of Miami Marine Laboratory 

UMMZ — University of Michigan Museum of Zoology 

USNM — United States National Museum 

UT — University of Texas 

UW — University of Washington 

VOM — Vanderbilt Oceanographic Museum 

WHOI — Woods Hole Oceanographic Institution 

ZMA — Zoological Museum, Amsterdam 

ZMC — Zoological Museum, Copenhagen 



Introduction xix 

Of the other abbreviations used in this volume, the following require explanation 
here: 

SL — standard length 

FL — fork length 

hi — head length 

m.w. or m.w.o. — meters of wire or meters of wire out, used in lieu of depth of 
capture when discussing the catches of nets fished at indeter- 
minate depths and not equipped with closing devices. 

Finally, acknowledgment with gratitude is extended to all of the authors who 
contributed to this volume and to the institutions that supported their work; to the 
Sears Foundation for Marine Research, Yale University, and to Mr. Henry Sears for 
his assistance in financing publication; to the National Science Foundation for funds 
to finance part of the cost of research (through grant NSF G 7123); and to the numer- 
ous ichthyologists and many others who have contributed information for this volume. 

Henry B. Bigelow 
March 1963 Editor-in-Chief 



lOCTTS* 50" 25" 




North America 




South America 



Bony Fishes 

Superclass, Class, Subclasses, 
and Orders 

HENRY B. BIGELOW 

Museum of Comparative Zoology 
Harvard University 

ACKNOWLEDGMENTS 

With grateful acknowledgment for many helpful criticisms and suggestions from 
William W. Anderson, U. S. Fish and Wildlife Service ; Frederick H. Berry, U. S. 
Fish and Wildlife Service ; Rolf L. Bolin, Hopkins Marine Station ; Daniel M. 
Cohen, U. S. Fish and Wildlife Service ; Myvanwy M. Dick, Museum of Compara- 
tive Zoology, Harvard ; Robert H. Gibbs, Boston University ; Joseph Gregory, 
Peabody Museum of Natural History, Tale ; N. B. Marshall, British Museum 
[Natural History); Giles W. Mead, Museum of Comparative Zoology, Harvard; 
James E. Morrow, University of Alaska; George S. Myers, Natural History Mu- 
seum, Stanford University ; Tngve H. Olsen, Bingham Oceanographic Laboratory, 
Tale ; Alfred S. Romer, Museum of Comparative Zoology, Harvard; Bobb Schaeffer, 
American Museum of Natural History ; William C. Schroeder, Woods Hole Oceano- 
graphic Institution ; Leonard P. Schultz, U. S. National Museum ; Royal D. Sutt- 
kus, Tulane University ; John Tee-Van, New Tork Zoological Society ; Ethelwynn 
Trewavas, British Museum {Natural History) ; Vladimir Walters, University of 
California ; and Norman J. Wilimovsky, University of British Columbia. 



2 Memoir Sears Foundation for Marine Research 

Superclass PISCES 

Both the cartilaginous fishes (S) and the bony fishes have been grouped together 
under the name Pisces by Goodrich (^5) as a Grade intermediate between Branch and 
Subgrade, by Berg (4) as a Series intermediate between Superclass and Class, and by 
Bertin and Arambourg (jj) as a Superclass. This last arrangement, with Pisces ranked 
as a Superclass, is accepted here to embrace both cartilaginous fishes (Chondrichthyes) 
and bony fishes (Osteichthyes). The cyclostomes, not regarded here as fishes in the 
usual sense, are considered a separate Branch (Agnatha) of the vertebrate Subphylum 
Craniata. 

Bertin and Arambourg (jJ: 1956) have expanded the Superclass Pisces to include 
the fossil Placodermi (including the Acanthodii and the Athrodiri); this ancient group, 
however, was only remotely related to either the living cartilaginous fishes or the 
living bony fishes. See Romer (75: 38—59) for an account of the placoderms and 
their evolutionary pattern. 

Class OSTEICHTHYES 

Characters of Living Members. The internal skeleton contains true bone in greater 
or lesser amount, especially in the regions of the skull, the jaws, and the pectoral arch. In 
most the jaws are well developed, but in a few they are greatly reduced. The palato- 
pterygoid part of the upper jaw-complex is fused solidly with the lower surface of the 
skull in the Dipnoi (lungfishes) but not in the others, with one exception (75: 273). 
The teeth are usually embedded in the bone (but see p. 6 for exceptions). In a few there 
is a spiracular opening between the jaws and the hyoid arch (polypteroids and some 
acipenseroids). The external opening of each nostril is typically double but is single in 
a few (p. 6); the nasal sac ends blindly in most but opens internally into the mouth in 
some. There is only one external gill opening on each side, and the two openings are 
often united across the throat; the common chamber into which the internal gill clefts 
open is roofed over by a dermal opercular flap supported in most cases by a series of 
opercular bones. A typical bony fish has four or five pairs of gill arches (the fifth the 
smallest) and four to six pairs of internal gill clefts. ^ 

The majority has two sets of paired fins — pelvic and pectoral. The pectoral girdle 
is well developed and is attached at its upper end to the temporal region of the skull 
in the great majority (see p. 6 for exceptions). The endoskeletal support of the pectoral 
and pelvic fins consists ot either one or two series of short basal elements;^ in the typical 
pectoral fin these elements are in parallel or fan-like arrangement (Actinopterygii in 
general, p. 11), with the outer part of the fin supported by fin rays; but in some there 
is a jointed midrib with either a series of short side branches on each side, as in the 
Dipnoi, or a terminal fan of jointed rays (Fig. 2), as in the living coelacanth Latimeria 

1. According to Tchernavin {S4: 284), there are six gill clefts in Eurypharynx (gulper eel). 

2. The basalia of the pectorals are fused in Dallia to form a single plate; in Lophius there are only two; and in 
general they are much reduced in the pelvics. 



Fishes of the Western North Atlantic 3 

(jjt: 2567, 2570, fig. 1840). The marginal parts of both the single and paired fins 
are supported, at least in part, by jointed (segmented) and usually branched rays which 
are either bony or fibrous in composition; in some the rays are apparently single,^ 
but in the great majority each ray consists of a right-hand and left-hand half, the halves 
closely pressed together (Fig. i). In a great majority the dorsal, caudal, and anal fins 
are separated, one from the next, but in some these fins are continuous around the tip 
of the tail. 

In a great majority the scales are wholly of mesodermic origin and are embedded 
in the skin below the epidermis. The outer surface of the scales consists of a simple 
calcified layer in most but of an enamel-like substance (ganoine) in a few. In most the 
scales persist throughout life and grow as the fish grows, generally without increase 
in number or modification except as they may be replaced when accidentally lost. How- 
ever, in some of the goby-like fishes, additional scales are normally interpolated with 
growth* in the region of the caudal peduncle. 

The anal, genital, and urinary tracts open separately to the exterior in most but 
into a common cloacal pouch in a few. Some of the members of most of the major 
groups have a swim bladder developed as a diverticulum from the dorsal side of either 
the oesophagus or the stomach; a few (Dipnoi, polypteroids) have a functional lung (or 
lungs) developed as an outpocketing from the ventral side of the pharynx.^ 

The notochord is persistent and unconstricted from end to end in a few, but in 
most it is sharply constricted in its passage through the vertebral centra, often to thread- 
like proportions, if it is not obliterated altogether. 

Nomenclature.^ The name Osteichthyes, proposed by Howes (j^: 78, 87), has been 
adopted by Goodrich (^5: 210) and by numerous subsequent students, including Berg 
(j: xv), Goodrich,' Romer (7J: 75), Schultz and Stern {jy. 227), Matsubara (^7: 
78, 154), and Bertin and Arambourg (jj: 1978, 2068). It is equivalent to the Class 
Pisces of Linnaeus (46), except that Linnaeus excluded the genera Acipenser (sturgeons) 
and Lophius (anglerfishes) ; it is also equivalent to the Subclasses Lophobranchii plus 
Potomabranchii of Bonaparte {10 : add. 1 3), with the addition of the lungfishes, which 
were unknown to Bonaparte;* it also equals: the combined Subclasses Dipnoi, Teleostei, 
and Ganoidei of Muller (5J: 201—204); the Class Pisces as restricted by Jordan 
(59: 108) and by Regan {ji [1929]: 305); the Teleostomi of Stensio;^ the Classes 

3. The situation in this respect is not known for either the coelacanths or the Dipnoi. 

4. Steindachner and Dbderlein {8i: 269) have reported this for Cepola; Koumans {43: 267-279) has described in detail 
and pictured clearly a similar situation for Oxyeleotris. Furthermore, J. R. Dymond (personal communication) 
states that he has found in salmon and trout "... .what appear to be rows or partial rows [of scales] that appear 
to have been developed later than the rows above and below"; and Hubbs (j6: 82) has reported an increase, 
caused by parasites, in the number of scales in the genus Platygobius. 

5. Millot and Anthony (ji: 2584) have reported a vestigial lung ("poumon degenere") of large size in the living 
coelacanth Latimeria. For a general survey of the relation between swim bladder and lung of living fishes, see Romer 

{74- 344-345' 347-35°)- 

6. See Myers {$4: 31-40) for nomenclature of terms that have been used for the higher categories of fishes. 

7. Ref. (jo: xvii); considered a Subgrade of Class Pisces. 

8. Ref. {10: 16); name Dipnoa used for Amphibia. 

9. Ref. {82: III); taxonomic rank not stated. 



4 Memoir Sears Foundatio?i for Marine Research 

Dipnoi and Teleostomii, combined, of Berg (^); the Teleostea of Fowler {2y: 53); 
and the Class Osteichthyes of Bertin and Arambourg (jr: 1978). 

The Class Osteichthyes as defined above includes all of the living groups of fishes 
apart from the elasmobranchs (sharks and batoids) and chimaeroids. The cyclostomes 
are not regarded here as "fishes" in the usual sense of the word. 

Bony Fishes {Osteichthyes) Compared with Cartilaginous Fishes {Chondrichthyes). All 
living bony fishes differ fundamentally from all living cartilaginous fishes in that the 
internal skeleton of the former consists of true bone in greater or lesser amount, or of 
some derivation of true bone, whereas the skeleton of living elasmobranchs and chimae- 
roids contains no true bone. Bony fishes differ to an even greater extent from the cy- 
clostomes in that the former have well-developed jaws and (typically) two olfactory 
organs. The bone of the Osteichthyes represents in part the ossification of the primary 
skeletal cartilages by the action of bone-forming cells from without (for details, see 
Goodrich, 4^: 65-67); but in part, especially in the regions of the skull and pectoral 
arch, it consists of so-called membrane bone, which is not preformed in cartilage. In 
neither case is it derived from calcifications of the kind that stiffen the cartilaginous 
skeletons of elasmobranchs and chimaeroids. 

Due to the presence of bone, the skull of all living bony fishes is marked by su- 
tures,!" which is not so in the cranium or brain case of living cartilaginous fishes. How- 
ever, this seems an appropriate place to note that in fossil placoderms, which are only 
remotely related to the bony fishes, the internal skeleton was more or less bony; 
many of them were more or less completely armored with bony plates, and their skulls 
showed sutures corresponding to those in Osteichthyes. Also, many of the fossil Ag- 
natha, of which the cyclostomes of today are commonly regarded as degenerate descen- 
dants, were variously encased in bony armor, and at least in some cases their skulls 
were marked with sutures." 

Hardly less diagnostic for most of the living bony fishes, including the living 
coelacanths,i2 is the presence (typically) in some part of their fins of segmentally jointed 
rays" (mostly branched) which are either bony or fibrous. These are the soft rays known 
to every student of fishes. Even among the living Dipnoi (lungfishes), some of the 
fin rays (hair-fine in this group and of doubtful homology) are segmented, though most 
of them continue unsegmented throughout life, thus resembling superficially the horny 
rays (ceratotrichia) of sharks (for details, see Goodrich, 28: 480-482). In the fossil 
Dipnoi, however, all of the fin rays were segmented. 

No living elasmobranch or chimaeroid has in its fins any structures that correspond 
structurally to the segmented fin rays of bony fishes. Thus the horny fin rays of sharks 
are not jointed, and they grow inward from the outer margin of the fin, not outward 

10. The bony neurocranium of the extinct palaeoniscoids, a primitive group of bony fishes, is not marked with sutures 
(Stensio, 83; Rayner, 6s- 287). 

11. For a readable account of these ancient groups, see Romer (7J: 25, 38). 

12. For the fin rays of Latimeria, see Millot and Anthony {31: 2567, 2570, fig. 1840). 

13. The fin rays are not jointed in the lyomerids, the trachypteroids, the stylophoroids, and perhaps in other 
groups as well. 



Fishes of the IVestern North Atlantic 




from its base as the fin rays grow in actinopterygian fishes,'* and presumably in coela- 
canths. Although the radial cartilages that support the wing-like pectoral and pelvic 
fins of skates and rays recall in appearance the jointed fin rays of bony fishes, they are 
a part of the internal skeleton ; and they are single whereas the soft 
rays of bony fishes are double (p. 3; Fig. i). 

Although the fin rays of the Dipnoi, which are intermediate 
between elasmobranchs and actinopterygian bony fishes in this 
respect, are jointed, they resemble the horny fin rays of elasmo- 
branchs and chimaeroids in their growth inward from the outer edge 
of the fin instead of outward from the base, in their arrangement in 
two series, and in their seemingly single nature.'* 

The contrast between bony and cartilaginous fishes is further 
blunted by the presence of hair-thin unjointed horny rays like those 
of the elasmobranchs in the so-called adipose dorsal fin of certain 
bony fishes (salmons and catfishes)'* and in the free edges of the 
rayed fins of other bony fishes." 

All of the other differences that have been proposed as alter- 
native between bony and cartilaginous fishes are beset with excep- 
tions. For example, it has been cited repeatedly that the dermal 
fold, which roofs the common branchial cavity (into which the in- 
ternal gill clefts open), is supported in bony fishes by a series of 
opercular bones, whereas in the only living cartilaginous fishes that 
have a corresponding dermal flap (chimaeroids) it is supported by a 
series of cartilaginous rays only. The polyodontids (paddlefishes), 
however, have only vestiges of the opercular bones; in the gigantu- 
roids, in some stomiatoids, and in some eels they are ossified but 
little; and in the lyomerids (an eel-like group of deep-sea fishes with 
enormously distensible mouths) and in the lophobranchs they are lacking altogether." 

Below the gill covers, the great majority of bony fishes has a series of parallel 
branchiostegal rays, but these rays are lacking in the elasmobranchs, chimaeroids. 
Dipnoi, acipenseroids, and deep-sea Lyomeri. 

The scales of bony fishes are deeply implanted in the skin, are wholly of mesodermal 
origin, and they usually persist throughout the life of the fish (for exceptions, see ftn. 4); 

14. For the development of soft rays (leptotrichia) in the actinopterygians, see especially Pennant (6j: 190-195) and 
Bertin {31: 735). 

15. For the structure, growth, and homologies of the fin rays of the different major groups of living fishes, see espe- 
cially Goodrich {28: 480-491), Eaton {2^: 195-200), and Bertin {31: 731-736). 

16. According to Dr. George S. Myers (personal communication), what appear to be ossified spines are developed 
in the adipose fin of some African catfishes, and with age in some characids of the genus Serrasalmus ; also the lori- 
cariid and calichthyid catfishes have the adipose fin supported by an initial bony spine. 

17. For a review of early literature on this subject with accounts and illustrations of these rays in the adipose fin of 
salmon and trout, see Valette St. George [88: 187-192, pi. 15); for a more recent review, see Goodrich {28: 
473-477)- 

18. For a discussion of the evolution of the operculum in different groups of fishes, fossil and living, see Eaton 
{24: 42-46). 



Figure i. Soft dor- 
sal fin ray of Esox 
lucius (pike) with 
its skeletal support; 
LEFT-lateral view; 
RiGHT-front view. 
R ray; p basal seg- 
ment, and M median 
segment of support- 
ing radial. After 
Goodrich, emended 
somewhat. 



6 Memoir Sears Foundation for Marine Research 

in elasniobranchs and chimaeroids the so-called placoid scales are tooth-like in structure, 
have an extensive pulp cavity, and are tipped with an enamel-like substance of ectodermic 
origin. Furthermore, the scales of a shark are not only short-lived but increase in number 
as it grows. Although the presence of bony-fish scales is positively diagnostic for those 
fishes that have them, their absence is not negatively diagnostic, for it is common 
knowledge that scales are wholly lacking in some bony fishes. Furthermore, certain 
groups of the bony fishes have placoid denticles of the elasmobranch type on their 
bony scales, e. g. the polypteroids, the lepisosteids (gars), and some ostariophysids (cat- 
fishes); and some of this last group have placoid denticles on their fin spines as well.^^ 
Denticles, believed to be placoid in nature (an interpretation calling for verification), 
have also been described for the snouts of young Xiphias (swordfish) and young Istio- 
phorus (sailfish) (jj: 321, pis. 1,2); and Dr. Ethelvyynn Trewavas informs us that they 
are present on the sucking disc of remoras as well (personal communication). 

It has been stated repeatedly that, whereas the teeth of elasmobranchs are simply 
implanted in the gums, those of most bony fishes are set in the bone. The polyodontids 
(paddlefishes), however, and certain Salariinae (blennies) parallel the elasmobranchs in 
this respect, their teeth having no special connection with the jaw bones; and the teeth 
in Plecoglossus (related to the salmonoids) are attached to the jaws by connective tissue 
only (j^: 437). 

The nasal openings are single on either side among the elasmobranchs in general 
and among the chimaeroids, but among bony fishes they are double (typically) on each 
side. However, some bony fishes, e. g. pomocentrids and cichlids, have only one nasal 
opening on each side. On the other hand, it has been discovered recently that each 
nasal opening in one genus of torpedo rays (Diplobatis) is divided into two (55: 482, 
fig. 2; 7: 562, 563, fig. 7). 

Various texts have emphasized the mode of suspension of the pectoral girdle as a 
diagnostic feature: it is attached to the skull in typical bony fishes; it is not connected 
at all to the axial skeleton in living sharks and chimaeroids; and it is attached to the 
anterior part of the vertebral column in batoids. However, among bony fishes, the 
pectoral girdle has lost its attachment to the skull in the deep-sea giganturoids and sac- 
copharyngoids, in the amphipnoids (a group that is eel-like in appearance but differs 
widely from the true eels in anatomy), and in the Apodes (eels) as a whole; and in some 
Apodes (Muraenidae) the pectoral girdle is lacking altogether. On the other hand, the 
girdle is attached by a ligament to the third or fourth vertebra in the Opisthomi.^" 

The presence of either a swim bladder or a functional lung is diagnostic for such 
of the bony fishes as have either, unless a small diverticulum from the dorsal side of 
the oesophagus, observed in embryos of certain sharks by Miclucho-Maclay (.^9: 448, 
pi. 5), represents the rudiment of a swim bladder. But the absence or presence of a 

19. For illustrations conveniently available of placoid denticles among bony fishes, see especially Goodrich, in Lan- 
kester {4$: 231, fig. 201; 289, fig. 262; 291-292, figs. 264, 265). For a recent account, with references and excellent 
illustrations of these denticles among living catfishes, see Orberg (56: 487, figs. 4 C, 40). 

20. An Order of eel-shaped freshwater fishes of doubtful affinity, known from tropical Africa, the Euphrates, and 
southern Asia north to Peking {68: 219). 



Fishes of the Western North Atlantic 7 

swim bladder is not a feature of much direct service to the systematist, for groups 
that differ widely in other respects may agree in lacking the swim bladder, and 
vice versa.^i 

In some families, even where most of the genera have a swim bladder, others do 
not. Familiar examples of such families are the Blenniidae (blennies), the Polynemidae 
(threadfins), and the Scombridae (mackerels); in the last group the chub mackerel 
{Fneumatophorus) has a swim bladder while the common mackerel of the North Atlantic 
{Scomber) has none (Starks, 92: 223). 

It has long been emphasized that, while the gill folds of elasmobranchs and chi- 
maeroids are attached to the interbranchial septa outward nearly to their tips, those of 
the higher groups of bony fishes are free from the septa for most of their lengths. These 
extremes, however, are bridged on the one hand by the Dipnoi, where the gills reach 
only slightly beyond the septa, and on the other hand by the polypteroids, acipenserids 
(sturgeons), and lepisosteids (gars), where they extend only a little farther outward. 
Furthermore, most of the bony fishes lack functional respiratory folds on the anterior 
(hyoidean) wall of the first gill pouch, such as are characteristic of elasmobranchs and 
chimaeroids; but the acipenserids and lepisosteids are exceptions to this rule, as is the 
living coelacanth Latimeria (Millot and Anthony, jr: 2584). 

The ventral or pleural ribs between the adjoining surfaces of the peritoneum and 
body wall are diagnostic for such of those bony fishes as have them, for the ribs of 
the living cartilaginous fishes^- that have any lie in the horizontal septum that separates 
the musculature of the trunk into an upper and lower division.^^ Many bony fishes, 
however, lack ribs of any sort: the lophobranchs (26: 28-30), for example, the gi- 
ganturoids {jo: 57), the stylophorids {80: 21), and some trachypteroids (genus 
Agrostichthyes) (information from Vladimir Walters). The tetraodontids (swellfishes), 
the diodontids (porcupinefishes), and the ostraciodontids (trunkfishes), among the 
plectognaths, lack ribs {j6: 8, 10; 41: 325, pis. 15, 16), while their relatives, 
the Balistidae (triggerfishes) and Aleuteridae (filefishes), have ventral ribs.-* It is also 
evident from a glance at the skeleton of Lophius (anglerfish) that it is ribless, as are 
the Ogcocephalidae (batfishes), by personal observation, and Rhynchoceratias (60: 13), 
among pediculate fishes. 

So-called dorsal ribs are not peculiar to the elasmobranchs alone, for the po- 
lypteroids, salmonids (salmons), some clupeids (herrings) (26: 209), Esocoidea (pikes) 
(4: 242, 429), and balistids (triggerfishes) among the plectognaths (41: 304) have 
both ventral ribs and ribs that occupy the dorsal position. However, it appears that 

21. N.B.Marshall contributes the information (personal communication) that the major groups of bathypelagic 
fishes, such as stomiatoids, deep-sea salmonoids, and Miripinnati, can be characterized by the structure of the 
swim bladder. 

22. The chimaeroids and some of the rays are ribless. 

23. But these ribs are considered by Emelianov (26: 244) to be homologous in their development with the pleural 
ribs of bony fishes. 

24. Regan {66: 285, 286, fig. 56) interpreted the rib-like bones of the Balistidae and Triacanthidae as intermuscular 
in nature and "not bordering the abdominal cavity." But it seems clear from Kashkaroff's account {41: 303, 
325, pis. 12-14) and from our own dissections that they are true ventral ribs and do embrace the body cavity. 



8 Memoir Sears Foundation for Marine Research 

these dorsal ribs may not be homologous, embryologically, with those of the elasmo- 
branchs (26: 215-217). 

A spiral valve in the intestine, which is universal in elasmobranchs and chimae- 
roids, is well developed in the living coelacanths {^0\ 426, 2582, Millot and Anthony), 
in the Dipnoi, and in the polypteroids alone among bony fishes; but it is represented in 
vestigial form in the acipenserids (sturgeons) and polyodontids (paddlefishes). A struc- 
ture that more or less suggests the elasmobranch spiral valve has been reported also 
as occurring in Chirocentrus, an Indo-Pacific genus falling among the clupeoids {18: 
160, pi. 565; ^5: 116, fig. 77 A, Goodrich), in Argentina si/us (the common herring- 
smelt of the North Atlantic),-^ in Salmo gairdneri (rainbow trout), Thymallus (grayling), 
and the Coregonidae (whitefishes) among the salmonids (information from Daniel Co- 
hen), in Macropinna among the Argentinoidea {15: 282), in the young of the characid 
genus Curinata (information from G. S. Myers), and in Alepocephalus (18: 172, 176). 
But the so-called spiral valve of these isospondylous fishes probably is not homologous 
with the spiral valve of elasmobranchs and chimaeroids.-* 

On the males of living cartilaginous fishes among which internal fertilization in 
the female is universal, a copulatory organ commonly called a clasper is developed in 
connection with each pelvic fin. But the male copulatory organ of the few bony fishes 
whose eggs are fertilized internally is single (typically) and has no connection with the 
pelvic fins, being developed either from the genital papilla on the urogenital orifice, or 
in connection with the anal fin, or as a special structure on the chest. However, the 
phallostethids (Indo-Australian freshwater fishes of the Order Percesoces) are an ex- 
ception, for their "much modified pelvic fins and girdle" join in the formation of the 
copulatory organ (2: 478). 

Finally, such of the bony fishes as pass their lives in salt water maintain their body 
fluids in osmotic balance with the surroundings in one way, elasmobranchs in another. 
In each case it is a problem of maintaining the concentration of salts in the blood at a 
level lower than that of the surrounding sea water; the osmotic pressure in the body 
fluids of marine teleosts is only about 40^/0 of that of sea water." Elasmobranchs 
solve this problem by retaining in their blood and tissues much of their nitrogenous 
wastes in the form of urea; thus they raise the total concentration of materials in solution 
to a level that is nearly as high as, or even higher than, that of the surrounding sea water 
without increasing the salt concentration (7^: 393). As a result of this mechanism, the 
marine sharks and rays have no special need of conserving water. According to Black, 
"They rarely drink salt water and obtain free water in their food" (J2: 187). Marine 
bony fishes, never having chanced on this device, prevent an accumulation of salt-ions 

25. Kendall and Crawford {42: 10, fig. lA) pictured Argentina as having a complete spiral valve, but Cohen writes 
that while he has noted spiral thickenings in the intestine of various Argentinidae, including six specimens of Ar- 
gentina silus {16: 121), in none of these last did he "observe a functional spiral valve as figured by Kendall and 
Crawford." 

26. According to Jacobshagen (9: 611), the structure that has been called by this name among the Isospondyli actu- 
ally represents a series of simple circular ridge-like thickenings of the inner surface of the intestine. For further 
discussion, see Cohen {16: 96, 98). 

27. Estimated by Denton and Marshall {21: 754) from data assembled by Krogh {44: 130 ff). 



Fishes of the Western North Atlantic 9 

in their body fluids by a complicated excretory process. Thus, via their gill membranes, 
they excrete most of the salt from the sea water that is consumed, and at the same time 
they reduce their excretion of water to the minimum that will provide the necessary 
amount of urine. In some groups this reduction is effected by atrophy or loss of the 
glomeruli,-* but in other saltwater groups that still retain the glomeruli, including the 
living coelacanths (jJ: 2588, Millot and Anthony), this goal appears to be reached by 
constriction of the glomerular arterioles.-* 

We have yet to learn how bony fishes that spend part of their life in fresh water 
and part in salt water or that suffer no apparent harm from rapid changes in the 
salinity of the water manage to maintain themselves in osmotic balance with their sur- 
roundings. This applies equally to such of the cyclostomes as spawn and pass through 
their early stages in fresh water but make their growth in salt water. Those that pass 
their entire life in salt water are isotonic to their surroundings, or very nearly so (see 
especially McFarland and Munz, 4.8: 348). 

Subclasses. Students of phylogeny have disagreed widely concerning the relation- 
ships of the various groups that make up the ichthyological tree. In the first instance, 
a decision on the best arrangement of the various groups of the Class Osteichthyes 
into Subclasses depends on the relationship of the Dipnoi (lungfishes) to the other 
bony fishes. Johannes Miiller (5j), writing in 1844, rated the Dipnoi as a separate 
Subclass of the Class Pisces, and this scheme was later adopted by Smith Woodward 
{gi: 234), by Goodrich (.^5: 230), by Bridge, and by Jordan (jp). Berg's (4) later 
characterization of them as a class of the Series Pisces embodies this same point of view, 
as does Bertin and Arambourg's treatment of them as the Subclass Dipneusti of the 
Class Osteichthyes (jJ: 2522). It is likely, indeed, that this scheme would still be 
adopted generally were our knowledge of the Dipnoi limited to those of today, for these 
are set apart from all other living groups of bony fishes by the solid fusion of the palato- 
pterygoid-quadrate elements of the upper jaw-complex with the lower surface of the 
skull, by the general morphology of their brains and hearts, and by their total lack of 
both premaxillary and maxillary bones (a significant difference); in addition, the living 
Dipnoi are set apart further from all other bony fishes except the coelacanths by the 
presence of a jointed axial skeleton in their paired fins (Fig. 2g). None of the other 
various features that have been cited as alternative between Dipnoi and other bony fishes 
is strictlv so. Palaeontological evidence, furthermore, is to the effect that the oldest 
known fossil Dipnoi (Devonian) resemble in many characters the Rhipidistia (Middle 
Devonian), whose nearest relatives among present-day fishes are the coelacanths 

{73- 114)- 

Accordingly, the Dipnoi and the coelacanths among living fishes have been 
united by several recent writers {55: 376; yi [1937]: 327; 75: 589; yy. 227; 4y: 

28. Examples cited by Smith [jg: 104) are Lophitts (anglerfishes), Opsanus (toadfishes), Hippocampus (seahorses), and 
Syngnathus (pipefishes). 

29. For more extended discussions of this general subject, from which the foregoing summary has been drawn, see 
Prosser et al. (6^: 48-52), Smith (79: 62-64, i02-t05), Romer (74: 392-393), Denison (20: 429-430, 439-441), 
and Black {12: 182-199, ^^^^- 199-206). 



lO 



Memoir Sears Foundation for Marine Research 



1 54) to form one of the Subclasses of the Class Osteichthyes, and this same course is 
followed here. 

In 1 86 1 Huxley proposed for this Subclass the name Crossopterygidae (subse- 
quently changed to Crossopterygii) as a Subordo of his Ordo Ganoidei (j^: 23, 25). Other 
names proposed later are: Amphiboidei by Hubbs in 1919 (J5: 589); Choanichthyes 
by Romer in 1937 (72: 56), with Crossopterygii retained as an Order to include fossil 
as well as living Choanichthyes (/j: 589); and Sarcopterygii by Romer in 1955 to 
replace his earlier name Choanichthyes {y^: 126). 




Figure 2. Skeleton and basal part of pectoral fin rays of: top, Latimeria, after Millot and Anthony, somewhat 
emended; lower left, Neoceratodus (lungfish), after Giinther, somewhat emended; lower right, Salmo trutta 
(European trout), after Parker and Haswell, somewhat emended, c pectoral girdle; b basal segment, and d distal 
segment of radialia; r fin rays. 



None of these names seems wholly appropriate. While Crossopterygii, in a restric- 
ted sense, has been widely employed in scientific literature,^" its continued use seems 
likely to lead to confusion because it has long been associated with the polypteroids, 
which were the basis for Huxley's name but which have been excluded from the Cross- 
opterygii, Goodrich {2g: 91) having shown that the affinities of the polypteroids lie 
with the actinopterygian fishes. 

Amphiboidei does not seem a happy choice, for while the fishes concerned may 
be "related to amphibian ancestry," the living coelacanths are about "as unamphibian 
as a fish can be," as Romer has aptly expressed it {j^: 125, 126). 

Choanichthyes is a misnomer if the coelacanths are included, for internal nares 
neither were nor are present among fossil {y^: 126) or living coelacanths, regardless 
of the situation in their hypothetical ancestors; and it is doubtful whether the internal 

30. For a recent defense of this course, see Trewavas, et al. [8y: 126-127). 



Fishes of the Western North Atlantic 



1 1 



nares of the Dipnoi are homologous with those of the tetrapod vertebrates. At any rate, 
they are not used in respiration. 

Although objection may be raised to the use of Sarcopterygii when older names 
are available, this name does not carry with it any evolutionary implication that may 
not be warranted; and furthermore, it has never been employed in more than one sense, 
hence it is accepted here. This Subclass need not concern us further, for the Dipnoi are 
strictly confined to fresh water while living representatives of the coelacanths have been 
taken only off the southeastern coast of Africa and near Madagascar. 

Actinopterygii has been chosen here as a Subclass name to represent the bony 
fishes that remain after subtracting the Sarcopterygii, a course which accords with the 
weight of contemporary opinion. An alternative scheme, proposed by Regan {6g: 458; 
'JI [1937]: 312) and adopted by Norman (55: 376-377), would distribute them in two 
Subclasses: Palaeopterygii for acipenserids, polyodontids, and polypteroids; and Neop- 
terygii for the remaining groups. But this seems an unnatural arrangement; although 
the polypteroids and acipenserids are similar in that the number of fin rays is greater 
than the number of their basal skeletal supports and because there is an unmistakable 
spiral valve in the intestine, the polypteroids differ widely from the acipenserids in the 
nature of both caudal fin^i and dorsal fin, in the segmentally constricted notochord, 
and in the presence of a pair of bony gular plates between the branches of the lower 
jaw. Indeed, the polypteroids differ so widely from all other living bony fishes that 
Bertin and Arambourg's separate Subclass for them, Brachyopterygii (jJ: 198 i, 2500), 
may represent a step forward in attempts to present more accurately the affinities of 
the various groups of bony fishes. 



Subclass ACTING PTERTGIP-' 
Ray-finned Fishes 

Characters of Living Members. The skull in its early stages is a primitive cartilaginous 
cranium that is replaced during development by bone in varying degrees in different 
groups; in most of the living bony fishes it is completely ossified in adults. Its outer 
surface is invested with a covering sheath in an intricate pattern of so-called dermal or 
membrane bones that develop as new structures "in the membranes of certain regions 
of the skull" and that appear to represent such modified scales; thus the entire skull 
is "so welded together to form a compact whole that in the adult fish it is often im- 
possible to decide as to which category a particular element belongs" {55: 161). 

Generally the upper jaw includes paired premaxillary and maxillary bones ;'^ the 
premaxillaries, supplemented in many cases by the maxillaries, form the upper region 

31. Symmetrical in the polypteroids, with the posterior part of the vertebral column continuing the axis of the body; 
strongly asymmetrical in the acipenserids and polyodontids, with the posterior part of the vertebral column bent 
upward (heterocercal). 

32. Including the polypteroids. 

33. Premaxillaries are lacking among acipenserids (sturgeons) and polyodontids (paddlefishes) {32: fig. 20); in Bathy- 
saurus of the Iniomi and in Gigantura maxillaries are lacking; both premaxillaries and maxillaries are lacking 



1 2 Memoir Sears Foundation for Marine Research 

of the mouth; the premaxillaries, and often the maxillaries, are attached to the skull 
with varying degrees of firmness in different groups. The palatopterygoid-quadrate 
elements of the upper jaw-complex are articulated anteriorly with the ethmoid (nasal) 
region of the skull and are suspended posteriorly (usually via a symplectic bone) from 
the hyomandibular arch ; but they are not fused solidly anywhere with the lower surface 
of the skull, except in one known case.'* A bony gular plate is present in the chin region 
between the two branches of the lower jaw in the Amiidae and Elopidae, and there is 
a pair of such plates in the polypteroids. The cheek region of the skull is without 
squamosal bone. 

In most of the bony fishes, the jaw teeth, if present, are separate from each other, 
and their substance is separate from that of the jaw except for those in which it is 
cemented basally "by bony substance which is resorbed when the tooth is shed" (45: 
272 — Goodrich). 

A persistent spiracular opening is present in only a few (polypteroids, some aci- 
penserids, and polyodontids). The nasal cavities end blindly in the great majority; 
however, in Jstroscopus {22: 993; 23'. 348-365; J: 371) and in Uranoscopus (original 
observation) of the Uranoscopidae they open both inwardly to the mouth and out- 
wardly to the exterior, as they do in various members of the eel tribe as well (j: 371); 
this is suggestive of similar conditions among the Dipnoi (lungfishes), but the two 
are probably not homologous. Gill rakers, present in the great majority,^^ vary widely 
in number. Branchiostegal rays, though present in the great majority, are lacking in a 
few (sturgeon and paddlefish tribes, polypteroids, and Lyomeri). 

The endoskeletal support of the pectoral fin consists typically of one or two series 
of short basal radialia, either in parallel or in more or less fan-like arrangement; these 
radial ia articulate with the pectoral girdle directly in most Actinopterygii but indirectly 
in a few. The endoskeletal support of the pelvic fin is essentially similar, but here the 
radialia are greatly reduced. The dorsal and anal fins are supported basally by a single 
series of more or less rod-like radialia; in most, the rays of both dorsal and anal fins 
are equal in number to the basal radialia, but in both acipenserids (sturgeons) and 
polyodontids (paddlefishes) the rays of both fins are much more numerous than the 
radialia; among the polypteroids, while the dorsal rays more or less equal the number 
of radialia, the anal rays are much more numerous. 

The outer surface of the scales (if any) consists of a simple calcified layer in most, 
but is enamel-like in appearance (ganoine) in a few. 

or are very small in some members of the Argentinoidea [86: 605-614, pi. 11; 62: 30-31, figs. 9-11); and the 
Monognathidae (an aberrant family of lyomerids) have no upper jaw-complex at all (5: 533-540, figs. 1-4; 6: i, 
5, figs. 2, 3). ^ 

34. In the argentinoid genus Macropinna, according to Chapman {15: 273), the palatoquadrate complex is "immov- 
ably synchronized" with the occipital portion of the skull. Among the Ophiocephalidae and Symbranchii, the 
metapterygoid bone (part of upper jaw-complex but not forming part of gape) is in contact with the sphenotic 
bone, which is part of the posterior rim of the orbital capsule of the skull. For references, see Berg {4: 306, 471). 
For an explanation of the mode of suspension of the upper jaw that is typical of ray-finned fishes, see especially 
Parker and Haswell {$g: 243, fig. 207; 244). 

35. Lacking in some members of the Cetomimidae {61: 20-32), in the Istiophoridae, and perhaps in some others. 



Fishes of the Western North Atlantic i 3 

Usually each of the genital, urinary, and intestinal tracts opens separately to the 
exterior, but among the syngnathids (jj: xxvii, ftn. i) and lophotids (information 
from Vladimir Walters) they empty into a common cloacal pouch. In many, a swim 
bladder is developed as a single outpocketing from the dorsoanterior part of the di- 
gestive tract, and an open pneumatic duct persists throughout life in some but not in 
others; in the polypteroids a functional paired lung is developed as a ventral out- 
pocketing from the pharynx.'" 

The brain, consisting of a large cerebellum, large optic lobes, and a feebly de- 
veloped cerebral region, is roofed (typically) in its anterior part with simple nonnervous 
epithelial tissue. The notochord is more or less constricted segmentally except in the 
aclpenserids, polyodontids, and deep-sea halosaurids. The jugal sensory canal (forward 
extension of the lateral line) is represented in most by a horizontal line of sensory pits 
that cross the cheek region {8g: 378). The pericardium Is thick-walled posteriorly; the 
heart chambers show no indication of subdivision into arterial and venous channels." 

Nomenclature. The choice of a name for this Subclass lies between Actinopterygll 
and Teleostomi. 

In 1839 Bonaparte proposed the name Teleostomi'* as a sectio to include his Ga- 
noidei and certain other groups of bony fishes, but excluding the aclpenserids (sturgeons) 
and some others (jO: 15). Many years later Owen expanded Bonaparte's Teleostomi 
to Include the bony fishes as a whole, excepting the Dipnoi, which were unknown to 
Bonaparte {^y. 7). Since 1868 Teleostomi has been used In this sense by numerous 
authors. 

The Subclass name ActInopterI,'» changed subsequently to Actinopterygia and 
Actinopterygll, was coined by Cope to Include the aclpenserids, lepisosteolds, amioids, 
and all higher groups of bony fishes, except the polypteroids {ly : 449). The Subclass 
was employed In this same sense by Traquair {8^\ 505, 507), Goodrich (45), Jordan 
(jp: III), Lehman, Bertin, and Bertin and Arambourg (jj: 2070-2500). In 1928 
it was expanded by Goodrich to Include the polypteroids {2g: 91), and It has been 
employed with this modification by Stensio,*" Berg (4: 159, 392), Romer (75: 105, 
579), and Schultz and Stern {^y. 228). 

Superorders. Several students have distributed the Orders of Actinopterygll among 
three Superorders, namely: Chondrostei (or Chondrosteoldea) for the aclpenserids (stur- 
geons) and polyodontids (paddlefishes) ; Holostel for the Lepisosteoldea (gars) and 
Amioldea (bowfin); and Teleostei for the remaining bony fishes. It does not seem pos- 
sible to draw a line of distinction between these Superorders if the fossil record be 
taken into account, hence they are not recognized here." 

36. For discussion of the relationship of swim bladder to lungs, see especially Romer {74: 343-35°> figs. 225, 227). 

37. Among the living Dipnoi, the pericardium is thin-walled, and there is an "incipient division of the heart into 
a venous and an arterial channel" [45: 249 — Goodrich). 

38. From the Greek words teUos (meaning perfected) and stoma (meaning mouth). 

39. Based on the Greek words aitis, meaning a ray, and pteron, wing or fin. The spelling was later changed to 
Actinopterygia by Cope and to Actinopterygii by subsequent writers. 

40. In a series of papers (1932-1936). 

41. For a discussion of this matter, see especially Berg {4: 160-163, or 392-395). 



1 4. Memoir Sears Foundation for Marine Research 

The name Teleostei*^ has proved far too useful to be banished from nontechnical 
literature, whatever its fate may be in technical writings; we urge that it be trans- 
literated henceforth to fit the language in which an author may be writing; for example, 
teleosts in English, teleosteens in French, Teleostier in German, etc.*^ However, the 
sooner "Ganoidea" and "ganoids" be relegated to oblivion the better, for the groups 
of living or extinct fishes** that were united under this name by Miiller (55: 203) 
and that have been grouped similarly by subsequent authors do not represent a 
natural assemblage. 

Orders of the Subclass Actinopterygii. The various classifications that have been pro- 
posed for the living ray-finned fishes since the appearance in 1844 of Muller's classic 
"system" (55: 201—204**) have been reviewed by Lonnberg (xj: 2-62), by Berg (4), 
and more recently by Matsubara (47: 1-54). Hence it seems sufficient to remark here 
that the number of Orders that have been recognized within the Subclass Actinopter- 
ygii during the past 50 years or so has ranged from nine by Goodrich (jo: x— xvii) 
and 12 by Romer (75: 579-585) to 24 by Matsubara (^7), 32 by Regan {^T. 76-82), 
36 by Schultz and Stern {yj: 220-247), 43 by Jordan (59), 44 by Berg (4), and 35 
by Bertin and Arambourg (jj: 1 978-1 981). This clearly illustrates the diversity of 
opinions and conclusions at which eminent students have arrived from consideration of 
essentially the same facts. 

Part 3 of this series of volumes deals with the Acipenseroidei (sturgeons) of the 
western North Atlantic, with such of the Lepisostei (gars) as enter brackish or salt water, 
and with part of the Isospondyli, namely the Elopoidea (Elopidae, Albulidae), Clupeoidea 
(Engraulidae and Clupeidae, with only interim accounts of Alepocephalidae and Searsii- 
dae), and Salmonoidea (Salmonidae, Coregonidae, and Osmeridae). Parts 4 and 5 deal 
with the remainder of the Isospondyli, the Iniomi, the Giganturoidei, and the Lyomeri. 

Order of Presentation. Since it is not possible to represent the true interrelation- 
ships of different large groups of animals on the printed page, any sequence of pres- 
entation that may be adopted (other than an alphabetical one) must necessarily be 
artificial. This would be true even if the taxonomic units in question could be traced 
back through the ages to their earliest known fossil ancestors and even if we had a 
uniform yardstick by which we could measure the relative extent to which the dif- 
ferent groups have diverged during their evolutionary histories. In any general account 
of a group as large and as varied as the bony fishes, the most one can hope for is that 
the sequence of treatment — Orders within Subclasses, and Families within Orders — 
shall be consistent with animal affinities "insofar as is practical" {j8: 13). The sequence 
followed here represents a compromise between antiquity of ancestry, degree of special- 
ization, and accepted precedent. 

42. From the Greek teleas (meaning perfected or completed) and osteon (meaning bone). 

43. Ichthyostes, as used by Moreau (57: 2; 52: 89), is an equivalent. 

44. Among living fishes — the sturgeon family, the paddlefish family, the polypteroids, and the gars. For a list (per- 
haps only partial) of the various extinct groups that have been joined together recently as "Ganoidei," see 
Jordan (jp: 111-116). 

45. For other places of publication, see Dean (19: 174). 



Fishes of the Western North Atlantic 



15 




Figure 3. Caudal fin of: top, Acipenser oxyrhynchus (sturgeon), showing especially the extended scaly axis and 
fulcral scales, drawn by E. N. Fischer; lower left, Sahelinus alpinus (Arctic charr), from eastern Greenland, 
drawn by Jessie H. Sawyer, lower right, Lepisosteus (longnose gar), after KoUiker, combined with drawing 
of L. osseus by Jessie H. Saw}er, with fulcral scale from upper margin, about x 2. 

The sturgeons head the list, with the gars in second place, for while the known 
fossil record does not reach back beyond the upper Cretaceous for the living repre- 
sentatives of either of these groups,*' fishes that seemingly were ancestral to both of 
them are known from as far back as the Upper Triassic. The Isospondyli and the 
Iniomi come next in recorded antiquity; the Elopidae (tarpon and ladyfishes), Clupei- 
dae (herrings), and Chanidae (milkfishes) are known from the Lower Cretaceous, with 
the Albulidae (bonefishes) and even the Myctophidae (or some closely allied genera) 
and Aulopidae known from the Upper Cretaceous. There is no known fossil record 



46. Wilimovsky {go: 1205—1208, pi. 132) has described and figured an undoubted sturgeon from the Upper Creta- 
ceous of Montana. 



1 6 Memoir Sears Foundation for Marine Research 

for either the Bathylaconoidea or the Lyomeri, curiously modified groups that are in- 
cluded in Parts 4 and 5 because they appear to stem back either to isospondylic or to 
eel-like ancestry; for further details, see Romer (75: 579-584). 

Identification. The sturgeons, the gars, the lyomerids, and the giganturoids are set 
apart by external features so obvious that no one at all acquainted with fishes would 
be likely to mistake any one of them for any other kind of fish that is known from the 
western North Atlantic. Although the Isospondyli, and probably the Iniomi as well, 
represent several phyletic lines of descent, with their members being correspondingly 
varied in appearance, they are easily separated as a group from the other Orders that 
are dealt with in Parts '^-i,. 

The following Key to the Orders described in these three volumes is offered solely 
as an aid to identification. The reader is referred to the sections dealing with the several 
Orders for information as to their internal features. 



Key to External Characters of Orders Represented in 
the Western North Atlantic 
and Described in Parts 3-5 

Separate rayed caudal fin clearly marked off from anal fin, from dorsal fin, or from 
both anal and dorsal. 

2 a. Caudal fin with conspicuous fleshy axis (marking rearward extension of verte- 
bral column) bending sharply upward at base of caudal and continuing out- 
ward, close to upper margin of fin, nearly to its tip; upper margin of fin much 
longer than lower margin (Fig. 3). Acipenseroidei, Part 3, p. 24. 

2 b. Caudal fin without conspicuous fleshy axis; upper margin of fin little if any 
longer than lower margin. 

3 a. Trunk, rearward from gill openings, completely clothed below and above 
with thick rhomboid scales in mosaic pattern, interlocking but hardly 
overlapping, forming a flexible armor; rear boundary of fleshy base of 
caudal fin sloping obliquely rearward-upward (Fig. 3); anterior part of 
upper margin of caudal and nearly entire length of lower margin 
edged with a series of large "fulcral" scales, shaped as in Fig. 3. 

Lepisostei, Part 3, p. 61. 

3 b. Trunk, rearward from gill openings, not clothed in a continuous armor 

of interlocking scales ; rear boundary of fleshy base of caudal either nearly 

vertical or symmetrically rounded; margins of caudal fin without fulcral 

scales. 

4a. Point of origin of pectoral fins higher on sides than upper end of 
gill openings; fin rays not branched. Giganturoidel, Part 4. 



Fishes of the Western North Atlantic 1 7 

4 b. Point of origin of pectoral fins at least no higher than upper end 
of gill openings, and usually much lower; at least most of fin rays 
branched. 

5 a. Upper edge of mouth formed by the maxillary bones as well 
as by the premaxillaries. Isospondyli, Part 3, p. 89, 

and Part 4. 

5 b. Upper edge of mouth formed by the premaxillary bones 

alone. Iniomi, Part 5. 

I b. No distinct rayed caudal fin; extreme tip of tail either like a whiplash or narrow 

band; its membrane, if any, without rays. Lyomeri, Part 5. 



TEXT AND FOOTNOTE REFERENCES 



1. Atz, Quart. Rev. Biol., 27, 1952. 

2. Bailey, J. Morph., ^g, 1936. 

J. Berg, Freshwater Fish. Russia, 1923 (Russ.). 

4. Berg, Class. Fish., Trav. Inst. zool. Acad. Sci. 

URSS, 5 (2), 1940; also reprinting by J. W. 
Edwards, Ann Arbor, Mich., 1947 (Russ. 
and Eng.). 

5. Bertin, Bull. Soc. zool. Fr., 61 (7), 1937. 

6. Bertin, Dana Rep., 75, 1938. 

7. Bigelow and Schroeder, J. Mar. Res., 7, 1948. 

8. Bigelow, Schroeder, and Farfante; Bigelow and 

Schroeder, Mem. Sears Found. Mar. Res., I 
(i), 1948; 1(2), 1953. 

9. Bohlk, et ah, Handb. Vergl. Anat. Wirtbelt., 

J. 1937- 

10. Bonaparte, Mem. Soc. Neuchatel Sci. Nat., 2, 

1839. 

11. Bronn's Klassen, 6(1), 1901. 

12. Brown, et al., Physiol. Fish., I, 1957. 

JJ. CHARTER, Proc. zool. Soc. Lond., 1919. 

14. Chapman, J. Morph., 68, 1941. 

J5. Chapman, Ann. Mag. nat. Hist., (i i) 9, 1942. 

16. Cohen, Bull. Fla. St. Mus., j, 1958. 

ly. Cope, Trans. Amer. philos. Soc, J^, 1871. 

18. Cuvier and Valenciennes, Hist. Nat. Poiss., 29, 



24^. xlyATON, Copeia, 1939. 

25. Eaton, J. Morph., 'j6, 1945. 

26. Emilianov, Zool. Jb., (Anat.) 60, 1935. 



2^. rowLER, Fish. Red Sea, J, 1957. 



28. vjooDRicH, Quart. J. micr. Sci., 4."], 1904. 

2g. Goodrich, Palaeobiologica, Jena, I, 1928. 

JO. Goodrich, Stud. Struct. Devel. Vert., 1930. 

JJ. Grasse, et al.. Trait. Zool., JJ, 1958. 

22. Gregory, Trans. Amer. philos. Soc, 2j(ii), 

1933- 

22- Guitel, Arch. Zool. exp. gen., 4(2), 1908. 



24- xdowEs, Sci. Progr., j, 1894. 

25- Hubbs, Science, N. S., 4g, 1919. 
56. Hubbs, J. Parasit, 14, 1937. 

J7. Huot, Ann. Sci. nat. Zool., {8)4, 1902. 

28. Huxley, Mem. Geol. Surv. U.K., Dec 10, 



2g. Jordan, Stanf. Univ. Publ. Biol., J (2), 1923. 
40. Jordan, Evermann, and Clark, Rep. U. S. 
Comm. Fish. (1928), 2, 1930. 



ig. 
20. 
21. 



23- 



L)ean, Bibliog. Fish., 2, 1917. 
Denison, Fieldiana, Geol., II (8), 1956. 
Denton and Marshall, J. Mar. biol. Ass. U.K., 

J7(3). 1958- 
Dahlgren, Science, N. S. 2y, 1908. 
Dahlgren, Sci. Mon., Lond., 24, 1927. 



41. K.ASHKAR0FF, BuU. Soc imp. Nat. Moscow, 

NS 2y (1913), 1914- 

42. Kendall and Crawford, J. Wash. Acad. Sci., 12, 

1922. 
42. Koumans, Temminkia, I, 1936. 
44. Krogh, Osmot. Reg. Aquat. Anim., 1939. 



Fishes of the Western North Atlantic 



19 



^5. LvANKESTER, Treat. ZooL, 9, 1909. 
^6. Linnaeus, Syst. Nat., 1758. 

4^. JVIatsubara, Fish. Morph. Hierarch., Pt. i, 

1955 (Japanese). 
48. McFarland and Munz, Biol. Bull. Woods Hole, 

II4(S), 1958- 
4.g. Miclucho-Maclay, Jena Z. naturw., J, 1867. 
^O. Millot, Nature, London, 17./, 1954.. 
57. Moreau, Hist. Nat. Poiss., 2, 1881. 
52. Moreau, Man. Ichthyol. Fr., 1892. 
5J. MuUer, Abh. Akad. Wiss. Berl., Phys. Mat. KL, 

1844. 
^4. Myers, Stanf. Ichthyol. Bull., 7 (3), 1958. 

$$. JNoRMAN, Hist. Fish., 1931, and subsequent 
editions. 



^6. v:9rberg. Ark. ZooL, 10 (12), 1957. 
^■j. Owen, Vert. Anat., I, 1868. 



6g. Regan, Proc. zool. Soc. Lond., 1923. 

"JO. Regan, Ann. Mag. nat. Hist., (9) jj, 1925. 

7/. Regan, Encyc. Britt., 9, 1929; 9 (4th ed.), 

1937- 
72. Romer, Bull. Mus. comp. Zool. Harv., 82, 

1937- 
7J. Romer, Vert. Paleont., 2nd ed., 194;. 
7./. Romer, Vert. Body, 1955. 
J^. Romer, Nature, London, IjG, 1955. 
'j6. Rosen, Ark. Zool., 10 (8), 1916. 

Jj. OCHULTZ and Stern, Ways of Fish., Pt. i, 1948. 
75. Simpson, Bull. Amer. Mus. nat. Hist., 55, 

1945. 
jg. Smith, Fish to Philospher, 1953. 

80. Starks, Bull. Mus. comp. Zool. Harv., ^2, 

1908. 
See also g2 below. 

81. Steindachner and Doderlein, Denkschr. Akad. 

Wiss. Wien, Math-Nat. Kl. 5J, 1887. 

82. Stensio, Triass. Fish. Spitzbergen, Pt. i, 192 1. 
5j. Stensio, Medd. Gronland, 5j (3), 1932. 



84. IcHERNAviN, J. linn. Soc. (ZooL), ^0(279), 

1947. 
(§5. Traquair, Nature, London, 62, 1900. 
86. Trewavas, Proc. zool. Soc. Lond., 1933. 
(S7. Trewavas, et al.. Nature, London, lyS, 19^5. 



5(S. Jr^ALMER, Ann. Mag. nat. Hist., (12) j, 1950. 
^g. Parker and Haswell, Textb. Zool., 2 (6th ed.), 
1940; and earlier editions. 

60. Parr, Occ. Pap. Bingham oceanogr. Coll., 3, 

1930. 

61. Parr, Bull. Bingham oceanogr. Coll., 4 (6), 

I93+- 

62. Parr, Bull. Bingham oceanogr. Coll., j (7), 88. Valette St. George, Arch. micr. Anat., 77, 

1937. 1880. 

6j. Pennant, Bull. Soc. zool. Fr., 62, 1937. 
64. Prosser, et al., Comp. Anim. Physiol., 1950. 



6^. rVAVNER, Philos. Trans., 2JJ B (601), 1948. 

66. Regan, Proc. zool. Soc. Lond., 2, 1902. 

6^. Regan, Ann. Mag. nat. Hist., (8) J, 1909. 

68. Regan, .\nn. Mag. nat. Hist., (8) 9, 1912. 



8g. Westall, J. Anat., Lond., JI, 1937. 
go. Wilimovsky, J. Paleont., JO, 1956. 
gi. Woodward (Smith Woodward), Cat. Foss. 
Fish.^rit. Mus., 2, 1891. 

g2. Starks, Science, N. S. 5./, igzi. 



Names of Bones 



GILES W. MEADi 

Museum of Comparative Zoology, Harvard University 

and 
MARGARET G. BRADBURY 

Hopkins Marine Station, Stanford University 

The bones of the skull and pectoral girdle of the hickory shad, Pomolobus mediocris 
(Mitchill), are depicted in Figs. 1-3. These figures do not represent an attempt at a 
critical osteological study, and the names used were chosen with regard to those most 
frequently used in the descriptive accounts that follow, as well as to questions of 
homology. The species illustrated may serve as an adequate guide to the bones of the 
more common and conservative of the Isospondyli, but it will be of limited value in 
the study of grotesque forms such as the stomiatoids and the Lyomeri. Notes on the 
osteology of these, or references to such studies, are included in the accounts of the 
various species. 

The fishes used were caught in Chesapeake Bay north of the town of Solomons, 
Maryland, in November 1958. All were between 280 and 310 mm SL. One dry 
skeleton was prepared and two were cleared in potassium hydroxide and stained 
with alizarine. These preparations and two additional whole specimens are now in the 
United States National Museym (USNM 186078, 186090, and 1 86091). 



I. This work was done while both authors were with the Ichthyological Laboratory, U.S. Fish and Wildlife Service. 

20 



Fishes of the Western North Atlantic 



21 



dermosphenotic 
dorsal postorbital 
ventral postorbital \ 
sclerotics. 



tabular 



supraorbitals 
nasal 



premaxilla 

maxilla 
dentory 
lacrimal 
supramaxillae 



articular 

ectopterygoid 



postlemporal 




opercle 
subopercle 
preopercle 
interopercle 
suborbital 



angular \quadrate 



/hyomandibular 

metapterygoid 

endppterygoid (mesopterygoid) 
ectopterygoid 

^palatine 




opercle^ \y 
subopercle^ zss. 



dentary 
articular 
quadrate 



angular 
symplectic 



preopercle' 

interopercle' 

interhyai 

Figure 4. Superficial face bones and suspensorium of Pomolobus mediocris. a lateral view of left side of head; 
underlying bones of the neurocranium not shown, b inner view of left suspensorium, lower jaw, and opercular 
series. Drawn by Margaret G. Bradbury. 



22 



Memoir Sears Foundation for Marine Research 



vomer (prevomer) 
mesethmoid 



prefrontal - 
(lateral ethmoid) 

frontal 
sphenotic 
pterotic 
parietal 
supraoccipital 

opisthotic 
epiotic 
exoccipital 
parasphenoid 

basioccipital 





mesethmoid 
vomer (prevomer) 

prefrontal 
(lateral ethmoid) 
frontal 
orbitosphenoid 

pterosphenoid 
alisphenoid) 

prootic 
sphenotic 
parasphenoid 
opisthotic 

^^_____ -pterotic 

epiotic 

exoccipital 

basioccipital 



supraoccipital 

epiotic 
exoccipital 
pterotic 




.frontal 



opisthotic 
basioccipital 
parasphenoid 



prefrontal 
(lateral ethmoid) 



mesethmoid 
vomer (prevomer) 

orbitosphenoid 
pterosphenoid (alisphenoid) 



supraoccipital 

parietal 
.epiotic 
pterotic 



exoccipital 
opisthotic 
basioccipital 

prootic 
parasphenoid 

basisphenoid 



Figure 5. Neurocranium of Pomolobus mediocris. a dorsal view; b ventral view; c posterior view; d lateral 
view. Drawn by Margaret G. Bradbury. 



Fishes of the IVestern North Atlantic 



23 



first 
arch 



glossohyal 
hypohyals 
ceratohyal 




pharyngobranchial 
epibranchial 
ceratobranchial — ^o"^ 
hypobranchial 
basibronchiol 



interhyal 
epihyal 



branchiostegals 
urohyal 



first (or suspensory) phoryngobranchiais 

pharyngobranchial 



basibranchiais 



hypobranchials 

ceratobranchials 




epibranchials 



B 




posttemporal 

supracleithrum 
postcleith 
scapula 
cleithrum 
coracoid 



posttemporal 
supracleithrum 
cleithrum 
postcleithra 
mesocoracoid 
scapula 
coracoid 



Figure 6. Hyoid arch, branchial basket, and pectoral girdle of Pomolobus mediocris. a lateral view of left hyoid 
arch with associated branchiostegal rays and first branchial arch; b three-quarter view of branchial arches; 
c outer view of left pectoral girdle; d inner view of left pectoral girdle. Drawn by Margaret G. Bradbury. 



Order Acipenseroidei' 



VADIM D. VLADYK0V2 

and 

JOHN R. GREELEY3 



Acknowledgments. We are indebted to the late Leon Bertin and to Miss Rolande 
Esteve for their help and for their permission to examine Dumerirs types of Sturgeons 
in the Paris Museum National d'Histoire Naturelle. We are also grateful to Henry 
W. Fowler and Leonard P. Schultz for making available for study the Sturgeon 
collections in the Academy of Natural Sciences at Philadelphia and of the United 
States National Museum ; to Lor en P. Woods for specimens from the Chicago 
Natural History Museum; to Carl E. Gut he and Ralph S. Palmer for their kind 
offer of a rich collection of Acipenser brevirostris from the New York State 
Museum; to the late Alexander D. Bajkov, R. E. Dimick, Ivan f. Donaldson, and 
Charles E. Warren for collecting Sturgeons in Oregon ; and to Henry B. Bigelow 
and William C. Schroeder for their careful editing of this manuscript and for making 
available their unpublished notes. 

The illustrations were drawn by Paul I. Voevodine, artist of the Department 
of Fisheries, Quebec. 



Scope of Study. In the following text we give the characters of the Order Aci- 
penseroidei and of the family Acipenseridae; also descriptions, life histories, and geo- 
graphical distributions, as well as synonyms and references, for the following Stur- 
geons : Acifenser brevirostris, A. fulvescens, A. oxyrhynchus oxyrhynchus, and A. oxyrhynchus 

1. Contribution No. 53, Department of Fisheries, Quebec, P. Q., Canada. 

2. Present address: Department of Biology, University of Ottawa, Ontario, Canada. 

3. Present adress: Division of Fish and Game, Conservation Department, Albany, New York, U.S.A. 

24 



Fishes of the Western North Atlantic 2 5 

desotoi\ all of these occur in the western North Atlantic or Gult of Mexico in fresh, 
brackish, or salt water. Since A. sturio of the eastern Atlantic and Mediterranean has 
sometimes been considered synonymous with A. oxyrhynchus, the relationship of these 
two is discussed on p. 57. 

Most of the material for our study consisted of collections made in New York 
(J5: 45-103) and Quebec (83: 143-204; 84: 129-154). Supplementary data were 
obtained by examination of the material at the Paris Museum National d'Histoire 
Naturelle, the United States National Museum, and the Academy of Natural Sciences 
of Philadelphia. 

The Descriptions are based on the Study Material listed for each species. An 
explanation of terms, measurements, and counts is given on p. 33. The accounts of the 
habits and geographical distributions are based upon published records and personal 
observations. 

Many of the earlier authors who dealt with the taxonomy of Sturgeons based their 
conclusions on differences associated with the age of the specimens, with the conse- 
quences that they recognized several nonexistent species. Hence, among the refer- 
ences, we use only those that we consider to be the most important from a taxonomic 
point of view, and particularly those dealing with the life histories of the species. 

Characters. Body subcylindrical. Snout greatly extended, blade-like and flexible 
in some, in others only slightly extended, if at all, and rigid. Mouth on lower sur- 
face of head, protractile or not. Chin barbels prominent in some but lacking in others. 
Dorsal fin situated far rearward; only one. Caudal fin more or less deeply forked, 
with fleshy axis (enclosing notochord) extending rearward to tip of upper lobe, bending 
upward beginning at base of fin; heterocercal, wider below than above the fleshy axis. 
Ventral fins well developed, abdominal in position. Fin rays much more numerous 
than their basal skeletal supports. Pectorals with the first ray transformed in some 
into an ossified ray, the fins otherwise without spines. Skin of trunk rearward from 
eyes seemingly naked in some but actually with minute scattered Platelets; others 
with five longitudinal rows of Scutes (large bony shields), the skin roughened else- 
where with small scattered Dermal ossifications.* A continuous series of rhombic to 
lanceolate Plates, closely set, on axis of caudal; the upper margin of caudal with a 
dorsal fringe of large v-shaped Fulcra (spine-like scales), each representing the fusion 
of a pair of opposing plates. Teeth minute, if present. Branch iostegal rays absent. 

Skeleton cartilaginous for the most part, the primitive cartilaginous cranium 
(chondrocranium) persisting and growing throughout life; ossification chiefly repre- 
senting development of membrane bones (p. 4) in regions of skull, jaws, and pectoral 
girdle. Upper jaw not articulating with skull either in the ethmoid region or in the 
sphenoid. Premaxillary bone fused with maxillary. Pectoral girdle fused with clavicle. 

4. There is no definite terminology for the plates or granules found in the skin between the regular rows of bony 
shields; they have been variously called: stellate plates, stellate roughnesses, stellate ossifications, minute rough 
spinules, small prickle-like plates [43: 104-106), dermal plates, and small spinous asperities (Ryder, The Sturgeons, 
with an account of experiments bearing upon Sturgeon culture, 6$: 23i-3z8). In the present manuscript they are 
called "dermal ossifications." 



2 6 Memoir Sears Foundation for Marine Research 

Myodome absent. Interopercular bone absent. Radials in dorsal and anal fins cartilagi- 
nous. Vertebral centra absent, the notochord entirely unconstricted throughout its 
length. Air bladder well developed, oval, with smooth inner surface, opening into 
oesophagus through a short, wide duct. Intestine with well-developed spiral valve (see 
Berg, J-o: 387). 

Habitat. Fresh water, or, if marine, entering fresh water to breed. 

Families. Of the two living families, the Acipenseridae with four genera is much 
more important throughout the northern hemisphere than Polyodontidae, which is rep- 
resented now by only two freshwater genera, each with a single species; Polyodon 
spathula (Walbaum), with a nonprotractile mouth, is found in eastern North America,^ 
whereas Psefhurus gladius (Martens), with a protractile mouth, inhabits China; one 
family, Chondrosteidae, is extinct, being known only from fossils of the Lower Lias 
to Lower Cretaceous periods. 

Key to Modern Families 

I a. Snout in front of eyes little if at all longer than remainder of head, subconical, 
rigid, more or less flattened in some but not blade-like in shape; a transverse 
series of four fleshy and conspicuous barbels close in front of mouth; mouth on 
lower surface of head, protrusible ; jaws toothless in adult ; five longitudinal rows 
of large bony shields (scutes) on body, the lateral row extending onto base of 
caudal, the skin elsewhere between them rough with small dermal ossifications; 
bony plates separated by evident sutures covering head; anterior edge of pectoral 
fins with a stout ossified ray; upper lobe of caudal much longer than lower. 

Acipenseridae, below. 
I b. Snout in front of eyes much longer than remainder of head, blade-like In form 
and flexible in American species; no barbels in front of mouth; mouth lateral in 
position, protrusible (in P. gladius) or nonprotrusible (in P. spathula) ; jaws with 
minute teeth; no bony shields on body, the skin everywhere on trunk smooth to 
the touch; head not covered with bony plates; no ossified ray in pectoral fins; 
lower lobe of caudal at least as long as upper lobe in American species. 

Polyodontidae. Paddlefishes. 
Fresh water of eastern North America 
and China. 

Family ACIPENSERIDAE 
Sturgeons 

Characters. Body elongate and fusiform. Scutes or bony shields in five rows: 
one dorsal, two lateral, and two ventral ; all scutes very sharp and strongly developed 

5. According to Dr. R. D. Suttkus (personal communication, July 9, 1957), a specimen of P. spathula, 481 mm FL 
and weighing 354 g, was taken in the brackish water of Vermilion Bay at Cypremont Point, Louisiana, on May 15, 
1954, by Antoine Carrere while trawling for penaeid shrimp. 



Fishes of the Western A^orth Atlantic 2 7 

in young individuals, but becoming progressively blunter with age or even disappearing 
through absorption. Skin between scutes with small ossifications. Snout protruding. 
Mouth inferior, protractile. Teeth absent in adults. Barbels 4, in a crossrow in 
front of mouth. Gills 4, and an accessory opercular gill. Branchiostegals absent. 
Gill rakers fewer than 50. Opercle absent. Head covered by bony plates separated 
by sutures; particularly visible in younger specimens. Dermal skeleton without 
ganoine. Caudal fin with typical fulcra. Dorsal and anal fins behind ventrals. Pec- 
toral fin with first ray enlarged and ossified. Tail heterocercal. Air bladder large, 
simple. Stomach with numerous pyloric appendages, forming a compact and rather 
large gland. Rectum with spiral valve. 

Anadromous and freshwater fishes of the northern hemisphere; Upper Cretaceous 
to Recent. 

Remarks. The cartilaginous skull has an opening under the frontal and parietal 
plates, usually completely roofed over by these membrane bones. Hence there is a cav- 
ity, evident in sagittal section of the head, for which we propose the nzme frontal sinus. 

In young specimens of A. oxyrhynchus of America and A. sturio of Europe, there 
is a membranous, elongated, nonossified area on the top of the head which Ryder^ called 
the "fenestra" or "fontanelle." In specimens less than 40 inches long, this can usually 
be found by probing with a needle between the frontal and parietal plates; thus a probe 
can be passed downward into xht frontal sinus. Adults, and in rare instances young, have 
the plates fully joined, closing the fontanelle. 

The protractility of the mouth is made possible by movements of the jaws and 
palatoquadrate bone {10: 409). 

The pseudobranchs are rudimentary In Scaphirhynchus (jj: 26) and completely 
lacking in Pseudoscaphirhynchus {li: 104). The large element of the gill cover repre- 
sents the subopercle, not the opercle (jo: 48—50). 

The appearance of the Sturgeon changes considerably with age; the snout 
becomes shorter and blunter and the shields smoother while the lower caudal 
lobe, not fully formed' in very small specimens, becomes longer. Apparently the 
shields may be partially or completely absorbed in connection with ripening of the 
gonads. 

Genera. The family Acipenseridae is divided into two subfamilies. Acipenserini 
(true Sturgeons), with spiracles present, is represented by two genera: Huso, Lower 
Pliocene to Recent, with two species; and Acipenser, Upper Cretaceous to Recent, 
with about 16 species. Scaphirhynchini (Shovelnose Sturgeons), without spiracles, 
also includes two genera: Scaphirhynchus, with two species;* and Pseudoscaphirhynchus, 
with three species. 

6. Ryder was probably the first American author to describe the presence of a fontanelle in A. oxyrhynchus {65: 233); 
for that in A. sturio, see Stannius {68: 26) and Antoniu {4: 113, 114). 

7. Grote, et at. {36: 215, fig. 158) and Roule {62: pi. 4, fig. 7) have presented drawings of young A. sturio in which 
the lower lobe is not well grown as yet. 

8. Although one of the American species has been described as a separate genus, Parascaphirhynchus [31: 37-44), 
recent studies fail to substantiate this (S: 174). 



2 8 Memoir Sears Foundation for Marine Research 

Key to Genera' 

I a. Spiracles present; snout subconical; tail not depressed or completely mailed; gill 
rakers lanceolate. 
2a. Gill membranes joined to isthmus. Acipenser YJvxWi-atw's, 1758, below. 

2 b. Gill membranes free from isthmus and broadly connected between themselves. 

Huso Brandt 1869. 
Basins of the Adriatic, Black, 
and Caspian seas, and the Far 
East (Amur River). 
lb. Spiracles absent; snout shovel-shaped, depressed above; tail depressed and mailed; 
gill rakers fan-shaped. 

3 a. Caudal peduncle completely mailed, and long, with tip of anal fin remote 

from caudal base; barbels fringed. Scaphirhynchus Heckel 1835. 

Mississippi River Basin. i" 
3 b. Caudal peduncle incompletely mailed, and short, with tip of anal fin reaching 
almost to caudal base; barbels smooth. Pseudoscaphirhynchus Nikolski 1900. 

Turkistan (Basin of the Aral Sea). 



Genus Acipenser Linnaeus 1758 

Acifenser Linnaeus, Syst. Nat., ed. 10, 1758: 237; type species, A.stiirio Linnaeus. Atlantic Ocean, along 
European coast. 

Generic Synonyms: 

Ichthocolla GeofFroy, Descr. de 719 Plantes, etc., 1767: 399 (presumably A. huso L.). 

Sterleta Giildenstadt, Nov. Com. Acad. Petropol., 16, 1772: 533 {A. rutkenus L.). 

5turio Rafinesque, Indice Ittiol. Sicil., 1810: 41 (Sturio vulgaris Rafinesque = A.sttirio L.). 

Sterletus Rafinesque, Ichthyol. Ohiensis, 1820: 80 {A . serotinus Rafinesque = A.fuhescens Rafinesque). 

Helops Brandt and Ratzeburg, Medizinische Zool., 2, 1833: 3, 349 {A.stellatus Pallas = A.helofs Pallas). 

Sterletus Brandt and Ratzeburg, Medizinische Zool., 2, 1833: 349 {A. rutkenus L.). 

Sturio MiJller, Vergleich. Anat. Myxinoiden, 1835: 77 [A. sturio L.). 

Antaceus Heckel and Fitzinger, Ann. Wien. Mus., J, 1836: 293 {A. schypa Eichwald). 

Lioniscus Heckel and Fitzinger, Ann. Wien. Mus., I, 1836: 370 {A. glabra Fitzinger). 

Sterletus Bonaparte, Cat. Metod., 1845: 21 {A.vultharius L.). 

Ellops, Gistel, Naturg. Thierreichs, 9, 1848 {A.helofs Pallas). 

Schifa Brandt, Bull. Acad. Sci. Petersb., 7, 1850, 113 {A. schypa Eichwald). 

Characters. Gill membranes joined to isthmus. Spiracles present. Snout sub- 
conical. Tail depressed and completely mailed. Gill rakers lanceolate. Characters 
otherwise those of the family. 

9. Since Sturgeons change with age in such features as head proportions and characteristics of scutes, the preparation 
of keys applicable to both young and adult stages has involved selection of those characters that are most constant 
throughout hfe. 
10. According to Dr. R. D. Suttkus (personal communication, July 9, 1957), a specimen of S. album, 532 mm PL 
and weighing 553 g, was taken in the brackish water of Vermilion Bay at Cypremont Point, Louisiana, on May 
10, 1954, by Lester Boudreaux while trawHng for penaeid shrimp. 



Fishes of the Western Nortli Atlantic 2 9 

Subgenera. There have been several attempts to subdivide the genus Acipenser as 
defined by Linnaeus in 1758. The preferences of earlier authors may be judged from 
the Generic Synonyms. Among contemporary authors, Berg (jl : 66), for instance, favors 
three subgenera: Lioniscus Heckel and Fitzinger for A. nudiventris, Gladostomus Holly 
for A. stellatus, and Acipenser s. str. for the rest. On the other hand, Antoniu (5: 317, 
3 1 8) has grouped Black-Sea Sturgeon in two subgenera : Euacipenser Antornu, comprising 
A. ruthenus and A. nudiventris, and Acipenser Linnaeus, embracing A. sturio, A. giilden- 
stddti, and A. stellatus. We believe that in the future any subdivision of the genus 
Acipenser should be based on meristic and anatomical characters; this is not possible 
yet because the anatomy of the Pacific species is completely unknown and that of 
Atlantic species has not been sufficiently studied. 

Size. Among Acipenser species, the range in size is wide. The largest species 
is A. transmontanus, with occasional individuals weighing up to 1,800 pounds {21: 
74), as in the case of one taken about 60 years ago at Mission, British Columbia; 
the smallest is no doubt A. brevirostris, about nine pounds (p. 38). The species of 
Huso are much larger. Berg (jj: 61) has mentioned the capture of a female H. huso 
in the Volga in 1924 that weighed 2,707 pounds (1,228 kg). A still larger specimen, 
weighing 6,614 pounds (3,000 kg), was taken in 1827. Females grow much larger 
than males. 

Span of Life. The species of Acipenser and Huso may have a longer life span than 
most fishes. For instance, through study of cross sections of the pectoral bony ray, 
an A. sturio 230 cm long was found to be 25 years old {20: 61). The age of a Huso 
dauricus, 500 cm long and weighing 1,600 pounds (S^^ kg), was estimated to be be- 
tween 50 and 55 years {60: 199)- Further information is given In the accounts of the 
western North Atlantic species. 

Age in Sturgeons has been studied by two methods, one employing cross sections 
of the pectoral ossified ray, the other otoliths." 

Breeding Habits and Sex. Sturgeons breed for the first time at a much more ad- 
vanced age than most other fishes, and apparently the small species are more precocious 
than the large. For instance, the relatively small A. ruthenus from the Volga may mature 
at four years (jj: 75) and A. brevirostris from the Hudson River at five years (jJ: 
82-83). O" the other hand, such a large species as A. oxyrhynchus reaches maturity 
in 10-12 years. According to Harkness," ^./«/wJC^«i from Lake Nipissing, Ontario, 
spawns for the first time at 22 years of age.^^ 

Apart from the period when they are sexually mature, there is practically no 
exterior distinction between the sexes. The cow Sturgeons, when full of ripe eggs, 
have a greatly swollen abdomen with rather thin walls, and the mature males have a 
more elongated appearance, with the walls of the abdominal cavity much thicker. 

11. For schematic drawings of otoliths of A . oxyrhynchus and brevirostris, see Greeley {35: 81); and ior ful'v-e 
scens, see Harkness {38: 16-19). Harkness (3S: 13-42) and Greeley {35) used otoliths for determination of age 
while Classen {20) and other European investigators based their findings on cross sections of the pectoral 
bony ray. 

12. For a review of the rate of growth and of the age at spawning of different species, see Classen (20). 



3 o Memoir Sears Foundation for Marine Research 

Among the large species, accurate determination of the sex in specimens smaller than 
30 inches is difficult and often requires the use of a magnifier. 

All of the species are oviparous and spawn in fresh water, and there is no adap- 
tation for the care of eggs or young. The eggs are large, at least 2.5 mm in diameter, 
demersal and adhesive (prior to water-hardening), and their number varies with the 
species and with the size of fish; in large specimens they exceed several million, as 
in the case of the 2,707-pound female reported by Berg (above), which yielded an 
estimated 7.7 million eggs weighing 542 pounds. More details are given in the ac- 
counts of the respective species. 

In all Acipenserldae" the ripening gonads^* have a characteristic appearance, 
being thick and elongate and extending along each side of the air bladder near the 
dorsal surface of the body cavity. The immature gonads in both sexes are covered on 
the underside by a large mass of yellow fat which apparently plays a role of energy 
storage for the ripening gonads; in those close to spawning it is almost entirely absorbed. 
The fully mature ovaries are very large and make up 10-20 "/o of the total weight of 
the fish. As the ripe eggs become dark,!* ^^g whole ovary darkens. Mature testes 
are pale yellow to whitish. 

It is a common observation that, even in the spring during the spawning season, 
large individuals with immature ovaries are found among fully mature females. This 
may be explained by the fact that the fish, after the first spawning, may spawn only 
at intervals of two or even three years. 

Studies of gonad maturation in J. nudiventris from the Aral Sea found that 
there are two distinct "waves" in the ripening process (75: 581-584). Fish ascend 
the Syr Darya River for spawning around the end of July, when they have rather 
well developed gonads. Soon the advanced sex cells are absorbed. After about two 
weeks, during the second "wave," the maturation of gonads starts again and is nearly 
complete by mid-October. During the winter the maturing fish hibernate in the river 
and spawn the following April or May. However, the material at Trusov's disposal 
was rather limited (160 gonads) and was collected during four different years (1941, 
1946-1948). 

Habits and Food. Sturgeons are sluggish but strong. During the summer they often 
leap from four to six feet above the water. There are many known cases of Sturgeons 
jumping into a passing boat. About 70 years ago Ryder mentioned such a case for 
A. oxyrhynchus near Delaware City, and on August 14, 1951, the Quebec newspaper 
Le Soleil (p. 20) reported that a 35-pound A.fulvescens jumped into a small rowboat 
on the Ottawa River, near Pembroke, Ontario. 

The small protractile mouth preceded by four sensitive barbels in the middle of 
the lower surface of the head may be considered an adaptation for bottom feeding. 

13. The same phenomenon has been reported for Polyodon as well (^6: 116-124). 

14. For further details on the ripening of eastern American Sturgeons, see Ryder {65: 268-273). 

15. Nikitin (56: 76) gives practical suggestions for the recognition of eggs from different species of Sturgeon made 
into caviar. The eggs may be distinguished by size, color, and particularly by the tint of their germinal disc 
("eye" of the egg). 



Fishes of tJie IVestern North Atlantic 3 i 

However, it has been observed that Sturgeon may rise to the surface to seize swimming 
objects {^6^: 265). 

The main Sturgeon food consists of bottom organisms such as oligochaete worms, 
insects, mollusks, and crustaceans; occasionally plant material is found in their stomachs 
as well. Sturgeons may feed on their own eggs, at least in the case oi A. fulvescem from 
Lake St. Peter, Quebec (unpublished data from Messrs. L. Philippe and J.-P. Cuerrier), 
and on different species of fish and birds. The gigantic Huso huso of the Caspian Sea 
may even feed on newborn seals {^^J: 435)- 

Habitat. While the majority lives in both fresh and salt water and makes most of 
the growth in the sea, at least two, fulvescens and ruthenus, are found typically in 
fresh water, although the former has been reported as occurring in brackish water. 

Range. The species of Acipenser are spread throughout the whole northern 
hemisphere, being particularly numerous in the Caspian and Black Sea basins. 

Commercial Importance. The fisheries for Sturgeon in North America have declined 
greatly in the last hundred years, and their maintenance on a sustained-yield basis 
constitutes a challenge to conservationists. Toward the end of the 1 9th century, catches 
for certain years were over 5,000,000 pounds for all species combined, but in 1956 
only 392,000 pounds were taken in Canada and 719,000 pounds in the United States 
(combined value about $386,000).'^ The Lake Sturgeon {A. fulvescens) in Canada and 
the White Sturgeon {A. transmontanus) in the western United States yield the greater 
part of the commercial catch. 

Most of the catch is consumed in the form of smoked Sturgeon and as caviar 
made by curing the eggs with salt. Over the years, production of caviar has dropped" 
so that in 1951 only 2,000 pounds, made from eggs o? A. fulvescens, were produced 
in Canada, worth about $10,000 to the fishermen. 

In Europe, particularly in Russia, Sturgeon fisheries are extensive. The fish are 
valuable not only for the flesh and eggs but for several other products; the commercial 
product from the spinal cord is called "vyaziga," and the inner lining of the swim 
bladder is used for the preparation of gelatine and glue of the best quality.^* The 
head of a Sturgeon, rich in gelatinous substances, is a valuable food product; in Russia 
it is sold separately from the flesh under the name "golovizna." In the Province 
of Quebec, older people use oil made from Sturgeon liver as an ointment against 
rheumatism. 

Although angling for Sturgeon by hook-and-line is not common, it is practiced 
in some places, as around Montreal, Canada; there also is an important spear fishery 
for sport for A. fulvescens in several northern waters of the United States." The 

16. A comprehensive summary of former catches of Sturgeon in North America is given by Rostlund (6j: 10-12). 
The most recent available data for the whole continent are in "Fishery Statistics of the United States, 1956" 
(74) and in "Fisheries Statistics of Canada, 1956" (26). 

17. According to Ryder, in 1888 fishermen from the Delaware River produced for the German market about 50 
tons of caviar from eggs of A. oxyrhynchus {fi^: 278). 

18. For the preparation of isinglass, see Ryder {^5: 278) and Classen {ig: 426-428). 

19. Some information on Sturgeon spearing, for instance in Lake Winnebago, has been reported (S6: 29). 



3 2 Memoir Sears Foundation for Marine Research 

occurrence of Sturgeon scutes in materials excavated from prehistoric American Indian 
villages indicates that these large fish were utilized in aboriginal times {pi: 10—12). 

Hybridization. Hybrids are found in nature quite frequently, not only between 
different species of Acipenser but between Acipenser and Huso. Antipa described at 
least eight different types of Sturgeon hybrids (j: 270-273), and recently Antoniu 
added two new findings among Roumanian Sturgeon (6: 308—313). In some areas, for 
instance the estuary of the Lena River, Siberia, the hybrid A. baeri x A. ruthenus is 
much more common than either of the parent species {14.: 141). 

Species. At present at least 16 species are recognized, five of which are found in 
North America. In eastern North America there are three: A. oxyrhynchus Mitchill 
1 8 14, A. brevirostris LeSueur 1 8 1 8, and A. fulvescens Rafinesque 1 8 1 7. The two found 
in the Pacific coastal waters of America are A. medirostris Ayres 1854 and A. trans- 
montanus Richardson 1836 (p. 36). 

In Europe there are five species oi Acipenser: A. ruthenus Linnaeus 1758, a fresh- 
water species inhabiting the rivers of the Black Sea and Caspian regions and the Arctic 
basin from the Ob to the Kolyma; A. nudiventris Lavetsky 1828, of the Black, Caspian, 
and Aral seas and their tributaries; A. guldenstadti Brandt 1833 and A. stellatus Pallas 
1 77 1, found principally in the Caspian and Black sea regions, but occasionally entering 
the Mediterranean; and A. sturio Linnaeus 1758, inhabiting both the Atlantic and 
Mediterranean areas. Furthermore, in his list of Acipenser species, Magnin recognized 
A. nacarii Bonaparte from the Adriatic Sea as a distinct species {48: 280). Chalikov has 
described a new species from the Caspian Sea under the name of A. primigenius 
(77: 47—50), but Berg considered it a hybrid A. ruthenus x A. giildenstadti (ll: 77). 

According to Professor Tamezo Mori (personal communication, December 20, 
1954), seven species of Acipenser are represented in the waters of eastern Asia: 
A. schrencki Brandt 1869 is limited to the Amur River and Okhotsk Sea; A. kikuchii 
Jordan and Snyder 1901 and A. multiscutatus Tanaka 1908 are restricted to Japan; 
A. dabryanus Dumeril 1868 has been reported from only the Yellow Sea basin in 
China and western Korea; and A. sinensis Gray 1834 is known from China, Korea, 
and Japan ; the sixth, A. medirostris Ayres, which is found on our Pacific coast, also 
frequents the western Pacific coast; the seventh Asiatic species, A. baeri Brandt 1869, 
inhabits the Arctic Ocean basin along the coast of Siberia, from the River Ob to 
Kolyma. Some information on the taxonomic characters and distribution of Asiatic 
species has been given (5J: 6-10; 5^: 1-228; 18: 69-71; $5: 15-17; Ji: 65-68). 

Relationship between Species. The total number of gill rakers increases in American 
species in the following order: medirostris (18-20), oxyrhynchus (17-27), brevirostris 
(22-29), ftilvescens (25-40), and transmontanus (34-36). The American species may 
be classified in order of increasing development of their bony shields in all the rows 
(using specimens of corresponding sizes) as follows : brevirostris., fulvescens, transmon- 
tanus., oxyrhynchus, and medirostris. Several small plates, with or without a weak crest, 
are present immediately behind the dorsal fin and in front of the anal; these plates are 
typically paired in the case of oxyrhynchus but are single in the case of the two other 



Fishes of the JVestern North Atlantic 3 3 

western North Atlantic species. In American species, the development of the modified 
predorsal shield increases in the following order: transmontanus (not only rudimentary, 
but sometimes absent), brevirostris, fulvescens, oxyrhynchus, and medirostris. In the case 
of the enlarged fulcrum along the lower lobe of the caudal and of the dermal ossifica- 
tions, the order is brevirostris, fuhescens, transmontanus, oxyrhynchus, and medirostris. 

Explanation of Terms, Measurements, and Counts. In order to facilitate the use of 
the Key and Descriptions, a brief explanation of certain terms and methods used in 
measuring and counting follows. 

Regular rows of bony shields: The dorsal row and the lateral row on each side 
are the most important from a taxonomic standpoint. The first shield in the dorsal row 
corresponds to the occipital plate or nuchale, and the last is just in front of the modified 
shield (see below) at the beginning of the dorsal fin. The first shield of the lateral 
row, called the supracleithrum, borders the posterior margin of the branchial cavity; 
although its size may vary, it is typically larger than the next posterior shield; its 
spine is usually less developed than the spines on the shields posterior to it. The last 
shield of the lateral row (a small one) is situated at the posterior extremity of the 
caudal peduncle, just anterior to the shields that continue on from the lateral row 
along the upper lobe of the caudal. 

The elongated fulcrum at the base of the lower caudal lobe (difficult to see on 
large fish) is of considerable taxonomic importance when compared to the length of 
the anal fin base. 

There are also modified shields at the beginning of both dorsal and anal fins; 
these, because of their respective positions, can be called predorsal and preanal. The 
one in front of the dorsal fin is more important from the taxonomic point of view; 
this modified shield is oval in shape, has no median crest or spine, and its posterior end 
is somewhat bent upwards to embrace the first dorsal ray; the anterior end is not forked 
but does have a slight indentation. There are two or more preanal shields; the nearest 
one to the fin is very similar in shape and structure to the predorsal shield ; those farther 
in front, according to the species, are present in pairs or in a single file; the disposition 
of these preanal shields is the same as that for postdorsal shields (see below). 

Dermal ossifications: These small bony structures (ftn. 4), which vary in 
size and shape with species and with age of fish, lie in the skin between the regular rows 
of bony shields; they are best developed on the sides between the dorsal and lateral 
rows. The degree of development, according to species, corresponds closely to that 
of the regular bony shields. 

Total length (TL) is taken along the median line from the tip of the snout to 
the posterior of the upper lobe of the tail when the fish is lying in a natural position. 
Fork length (FL) is measured from the tip of the snout to the posterior end of the 
median caudal rays — that is, to the fork of the tail. [In specimens less than five inches 
(125 mm) FL, the lower caudal rays are still not fully developed and are therefore 
short. The absence of the lower caudal lobe is not only typical of young oxyrhynchus 
but of other Sturgeons as well. We observed this m fulvescens; and Grote, et al. (j6: 



34 



Memoir Sears Foundation for Marine Research 




'mm. 



Fishes of the Western North Atlantic 3 5 

215, fig. 158) and Roule {^2: pi. 4, fig. 7) have presented drawings of young A. sturio 
in which the lower lobe is not well grown.] 

Head length is the distance from the anterior tip of the snout to the rearmost 
point of the opercle, opercular membrane excluded. Snout length Is from the ante- 
rior tip of the snout to the anterior margin of the orbit, the membranous rim included. 
Mouth width is the greatest transverse distance across the mouth slit, with the thick 
lips excluded and the mouth closed. Postorbital distance is that from the posterior 
edge of the orbit, membranous rim included, to the rearmost point of the opercle, 
membrane excluded. Interorbital width is the minimum distance across the top of 
the head between the bony edges of the orbits. Gill rakers are counted on the outer 
surface of the first branchial arch, including the rudimentary ones; they are usually 
counted on the left side of the fish; in specimens smaller than eight inches (20 cm), 
some rakers on the upper limb are not yet developed ; therefore the counts for young 
individuals are not given (see also 84: 149-154). 

Base of anal fin is the length from the structural base of the first ray to the 
point where the membrane behind the last ray contacts the body. Fin ray counts 
for the dorsal and anal are sometimes given, but it is difficult to count these accurately, 
even after staining with alizarin and clearing in glycerin {83: 3—9); the fin membrane 
is very thick, and the rays are numerous, close together, and show little bifurcation ; 
and in young individuals several rays are detached from the basal elements (radials). 
Since the counting of rays is not only tedious but uncertain, we omit the number of 
fin rays from the taxonomic characters. 

Remarks on Identification. Within the same species, various characters change con- 
siderably with growth; young have much longer snouts than adults, and their scutes 
(shields) are sharper and closer together. Because of these natural changes it is difficult 
to prepare a universal key for the specific identification of young specimens as well as 
adults. There are, it is true, several fairly constant interspecific differences in anatomical 
features, such as the intestinal tract, gill arches, grinding ridges within buccal cavity, 
and otoliths, to mention a few. But to present these in abbreviated form seems likely 
to prove confusing rather than helpful, hence they are not included in the following 
Key to Species. 

Key to American Atlantic and Pacific Species of Acipenser 

la. Species of the Atlantic Seaboard. 

2a. Mouth width less than 55''/o of interorbital; average difference between TL 
and FL i4''/o of FL; gill rakers 17-27 (av. 21.6); postdorsal and preanal 
shields in pairs; viscera pale, unpigmented. 

3a. In young specimens 50—70 cm long, head length 26—28 % of FL; bony 
shields of dorsal row oval in shape, their longitudinal length being greater 
than their transverse width; carina on dorsal shields low, without a 
pronounced hook; dermal ossifications between dorsal and lateral rows 

3* 



3 6 Memoir Sears Foundation for Marine Research 

of shields only weakly developed; spleen short, not reaching farther back 
than middle loop of small intestine. 

oxyrhynchus oxyrhynchus Mitchill 1814, p. 46. 
3 b. In young specimens 50—70 cm long, head length 30—3470 of FL; bony 
shields of dorsal rows rather square in shape, their longitudinal length 
much shorter than their transverse width; carina on dorsal shields high, 
typically with two strong hooks; dermal ossifications between dorsal and 
lateral rows of shields strongly developed; spleen long, reaching much 
farther back than middle loop of small intestine. 

oxyrhynchus desotoi Vladykov 1955, p. 56. 
2 b. Mouth width over 62 "/o of interorbital; average difference between TL and 
FL less than i2*'/o of FL; gill rakers 22—40; postdorsal and preanal shields 
in a single row; viscera blackish, heavily pigmented. 

4a. Gill rakers 22—29 i?-^- 2 5-4); interorbital width 34— 40''/o (av. 37"/o) 
of hi; dorsal shields 8—13 (av. 10); lateral shields 25—32 (av. 28.3); 
dorsal and lateral shields pale, contrasting with dark background; 
length, so far as known, not over 100 cm. 

brevirostris LeSueur 18 18, below. 

4b. Gill rakers 25-40 (av. 33.1); interorbital width i<)-2S''lo (av. 32<'/o) 

of hi; dorsal shields 9—17 (av. 13.4); lateral shields 29—42 (av. 35.4); 

dorsal and lateral shields brownish, of same color as background; 

length commonly over 100 cm. fulvescens Rafinesque 18 17, p. 41. 

I b. Species of the Pacific Seaboard. 

5a. Caudal peduncle short; tip of anal fin reaching or even extending beyond base 
of lower lobe of caudal ; modified shield at beginning of dorsal fin rudimentary 
or absent; 11 — 14 dorsal shields; 38—48 lateral shields; gill rakers 34—36; 
viscera blackish, heavily pigmented. transmontanus Richardson 1836. 

Southern Alaska to California; enters salt water. 

5 b. Caudal peduncle long; tip of anal fin not reaching base of lower lobe of caudal ; 

a large modified shield at beginning of dorsal; 8 — 1 1 dorsal shields; 23—31 

lateral shields; gill rakers 18—20; viscera pale, practically without pigment. 

medirostris Ayves 1854. 
Southern Alaska to California; enters salt water. 

Acipenser brevirostris LeSueur 1 8 i 8 

Shortnose Sturgeon 

Figures 7, 8 

Study Material. A total of 109 specimens: 95, 450— 900 mm TL, from the Hudson 
River, New York;^* i, presumably LeSueur's type, ANSP 16953;-! i, from Dela- 

20. For age and localities of these specimens, see Greeley (jj: 89, 90). 

21. This specimen, originally No. 84, was described by Ryder (65: 236) and later by Fowler (jj: 604, pi. 38, fig. i). 
On June 23, 1949, the specimen (stuffed) measured 746 mm FL. 



Fishes of the Western Nort/i Atlantic 



37 



ware River at Bay-at-Green Creek, Cape May County, 
New Jersey, in salt water, May 5, 1 907, ANSP 2547 1 ; 
4, from Delaware River at Torresdale, Philadelphia 
County, Pennsylvania, in fresh water, November 
191 1, ANSP 44298-44301; 3, 470-580 mm FL, 
from Delaware River, March 29-31, 1913, USNM 
I258i6-i258i8;2^ I young, 185mm FL, Salmon 
Creek, North Carolina, USNM 64330; i of head, 
fins, and dried strips of skin with plates, from Poto- 
mac River, USNM 26273; 3 stuffed specimens in 
Paris Museum. 2^ 

Distinctive Characters. A. brevirostris is distin- 
guished from A. oxyrhynchus by blackish viscera, a wider 
mouth (see Key), the absence of a fontanelle, almost 
complete absence of the postdorsal shields, and by 
preanal shields arranged in a single row. Also, indi- 
viduals of A. brevirostris assume adult proportions 
when they are only about two feet TL; A. oxyrhynchus 
retains juvenile characteristics even up to four feet. The 
most conspicuous difference between brevirostris and 
fulvescens is in the lateral scutes, which are much 
paler than the background in the former but of the 
same shade as the background in the latter. 

Description. Scutes in all five main rows not 
closely set and rather weakly developed in adults in 
comparison with other eastern American species, but 
sharp and set close together in a young specimen 
7.75 inches FL (USNM 64330); in 12 specimens, 
8-13 (av. 10) dorsal shields, 25-32 (av. 28) lateral 
shields, and 7— 1 1 (av. 8) ventral shields; shields be- 
hind dorsal fin either in a single row or paired; elon- 
gated fulcrum at base of lower caudal lobe shorter 
than base of anal fin. 

Head averaging 22''/(, of FL in adult, 28 "/o in 
a young specimen 7.75 inches FL. Fontanelle ab- 

22. According to the tag, these specimens were submitted as "young 
Sturgeon" (presumably as A. oxyrhynchus) for identification. 

23. Details on these specimens are given by Berlin {i2: 252). Two of 
these fish were sent to the Paris Museum by LeSueur, apparently 
as paratypes. 



Figure 8. Acipenser brevirostris. Lateral view of spawning female, 
580 mm long, from the Hudson River near Kingston, New York. 




3 8 Memoir Sears Foundation for Marine Research 

sent on top of head. Snout in 22 grown specimens (390-700 mm FL), mucii shorter 
than postorbital distance (av. 70 "/o of the latter), but in a young specimen, longer than 
postorbital distance (i83"/o). Postorbital distance in adults 51-61 "/q (av. 55'*/o) of 
hi, but 33''/o in a young specimen. Interorbital width 34— 40*'/o (av. 37%) of hi, and 
Mouth width (excluding lips) 69-8 i^/q (av. 74''/o) of interorbital width, these two 
characters not differing with age. 

Gill rakers rather long, triangular, in 13 specimens 22—29 (^v- '^S-S) °" ^'^'^^ 
branchial arch; in number of gill rakers, A. brevirostris occupies an intermediate 
position between A. oxyrhynchus with 17-27 (av. 21.5) and A.fulvescens with 25—40 
(av. 33)- 

Caudal fin with short upper but rather long lower lobe; average difference be- 
tween TL and FL, i I'/o of FL; no notch at tip of upper caudal lobe. 

Color. In preserved specimens, general color similar to that of A. oxyrhynchus 
(p. 46). Top of head and back dark, becoming paler on sides ; entire lower surface 
white. Central part of dorsal shields whitish, contrasting with darkly pigmented back 
and black skin between shields. Lateral shields pale, clearly distinguished from sur- 
rounding skin. Ventral shields of the same whitish color as lower surface of fish. Intes- 
tinal tract, air bladder, and fat surrounding gonads darkly pigmented, nearly black; 
the peritoneum only slightly pigmented. 

In an excellent photograph of a live brevirostris, the similarity in color to oxyrhyn- 
chus is particularly striking (^^: 14, 15): in addition to the details observed on preserved 
specimens, the white edging on the paired fins is similar; but the anal fin in brevirostris 
is pigmented while that in A. oxyrhynchus remains whitish. Small live specimens from 
the Hudson River show black pigment blotches similar to those of young A.fulvescens. 

Size: This is the smallest species of all the Sturgeons. Apparently it never grows 
to more than about three feet; the largest specimen recorded thus far is a female 
39.5 inches TL (899 mm FL), from the Connecticut River. The measurements sub- 
mitted by Professor Andrews and two specimens sent to us (p. 40) confirm his deter- 
minations. The largest sizes encountered during the present studies were: a female 
34.8 inches TL, weighing eight pounds 15 ounces, in its 13th year as determined by 
■its otoliths; and a male 25.5 inches, weighing four pounds five ounces; both were taken 
from the Hudson River (j5: 68-70). 

Development and Growth. The early life history is unknown. Few small specimens 
have been recorded; probably the smallest fish is one of 7.3 inches (about 185 mm), 
from North Carolina (USNM 64330). The smallest specimens taken in the Hudson 
River were two females; one was 17 inches (about 433 mm) and weighed only 15 
ounces; the other was a little less than 18 inches (about 600 mm) and weighed 19 
ounces. The sizes of five specimens from the Delaware River observed by Ryder ranged 
between 18-23 inches. Age determinations, based on otolith readings (J5: 68-70), 
have shown that A. brevirostris is a very slow-growing species. Specimens of brevirostris 
from the Hudson River that measured 17-35 inches (about 430-890 mm) TL were 
4-15 years old; on the other hand, A. oxyrhynchus 74-100 long were 1 1-13 years of 



Fishes of the JVestern Nort/i Atlantic 3 9 

age. The growth rate of the two species, however, is very similar in early life, which 
suggests that the subsequent divergence is due to more rapid growth by oxyrhynchus 
while at sea. 

Spawning. Males may mature when they are only about 20 inches TL, and most of 
them do so by the time they pass 21 inches; most of the females mature at about 24 
inches. The ripe eggs are dark brown. Their number is not known. Spawning takes place 
in rivers early in the spring. For Hudson River fish, Greeley (J5: 90) stated that: 

The spawning season evidently included late April during 1936. Several fish approaching spawning 
condition were examined from Rhinecliff during early April. A female, weighing five pounds, from Highland 
April 21 contained eggs which appeared ripe, while a three pound male taken at the same time was spent. Roe 
fish from Highland, May i, were spent (2 pounds, 13V2 ounces; 5 pounds, 8 ounces). Several others, also 
taken during early May had spawned. 

General Habits. On account of its small size, A. brevirostris has attracted little atten- 
tion except when taken in nets in fresh, brackish, or salt water. It is found most often 
in tidal rivers. But the place of capture of Gulf of Maine specimens shows that some 
certainly go out into the open sea and wander for some distance from the parent 
stream. 

Food. Studies of stomach contents from Hudson River specimens showed that A. 
brevirostris feeds upon the bottom, eating small animals and plants intermingled with 
mud (72: 141 — 144, table 4). The organisms consumed were sludgeworms, chironomid 
larvae, small crustaceans, etc., such as are found in the stomachs of young Atlantic 
Sturgeon (p. 52). 

Abundance and Relation to Man. Because of its scarcity and small size, A. brevirostris 
is of slight interest to commercial fishermen, hence there is little information about 
its abundance. 

Ryder {6^: 232), writing of the fishery for oxyrhynchus in the Delaware River, 
stated, "It was my good fortune to secure no less than five specimens of the A. brevirostris 
of LeSueur, which has, so far as I can learn, not been certainly recognized since that 
Naturalist's time," and our own Study Material (p. 36) marks it as at least not uncom- 
mon in the Hudson today. But while LeSueur in 1 8 1 1 stated that the Shortnose 
Sturgeon is "more sought after and commands a higher price than the large common 
species" in the Delaware River (y^: 383—394), it is of virtually no economic importance 
at present due to its small size and apparent scarcity. Its flesh, however, is of good 
quality, and its eggs are suitable for caviar, but they are not very numerous. A few 
fishermen prior to 1937 set gill nets for brevirostris in the Hudson River just before 
the yearly run of American shad {Alosa sapidissima). Some have also been taken in the 
Hudson River with hook-and-line by anglers. 

Range. Atlantic seaboard of North America, from New Brunswick, Canada {^y : 
504, and later comments), to Florida, mostly in rivers, including' the Connecticut, Hud- 
son, Delaware, and Potomac. This range is based upon records for a period of years, 
and the species may now occupy a more restricted area. 

Until recently the southernmost definite record was for Charleston, South Carolina 



40 Mejnoir Sears Foundation for Marine Research 

{43: 106). Smith stated, "while it doubtless ascends all suitable streams in North 
Carolina, actual records of its occurrence are rare" {6^ : 58). However, thanks to Dr. 
Daniel M. Cohen, we had an opportunity to examine a female A. brevirostris (UF 5714) 
735 mm TL taken on May 11, 1949, in Big Lake George of the St. Johns River 
drainage, Putnam County, Florida. This is a proven record of the occurrence of the 
Shortnose Sturgeon in Florida. Thus the geographic range of A. brevirostris is almost 
as extensive as that of A. oxyrhynchus. 

It has been reported as being taken as far north as the St. Lawrence River (j2 : 
70; ^g: 226; 57: 44), but these records seem to be based chiefly on A.fulvescens, 
in part on A. oxyrhynchus. The most northerly proven record of its occurrence is for 
the St. John River, New Brunswick (47 : 504). We also obtained some specimens from 
this same area, thanks to the kindness of the late Dr. Leim. Collections taken at Gage- 
town include three heads of adults caught in 1959 (collected by J. C. Medcof, Biolog- 
ical Station, St. Andrews, New Brunswick) ; also, seven adult females (700—8 70 mm 
TL) and three adult males (740-800 mm TL) were taken in May and early June i960 
(obtained by Fishery Officer J. O. Jenkins). One female and one male of the latter 
sample were almost ready to spawn. In the Connecticut River, near Hadley, Massa- 
chusetts, four adults were caught in October 1951 and September 1952 (obtained by 
Professor T. J. Andrews, University of Massachusetts). 

The breeding range of A. brevirostris is not clearly defined, but it is known to 
include the Hudson River, where the spawning areas appear to be very restricted. 
The Delaware River may still maintain a small local population, and it seems likely 
that the St. John River, N. B., has a spawning population, judging by the near- 
spawning condition of the above-noted male and female taken at Gagetown. If, 
through increased pollution or habitat changes, the population Is no longer able to 
persist in these northern rivers, the species may become dangerously reduced. 

The Shortnose Sturgeon has no close relative in Europe. 

Synonyms and References:^* 

Acifenser brevirostris LeSueur, Trans. Amer. philos. Soc, I, 181 8: 390 (descr., Delaware R.); DeKay, Zool. 
N.Y., 4, 1842: 345 (descr., ident. from Virginia not certain); Ryder, Bull. U.S. Fish Comm. (1888), 
8, 1890: 231 (descr., Delaware R.); Jordan and Evermann,* Bull. U.S. nat. Mus., 47 (i), i8g6: 106 
(descr., synon., Cape Cod to Florida, specimen from Charleston, S. Carolina); Smith,* N.C. geol. econ. 
Surv., 2, 1907 : 57 (genl. acct., doubtless occur. N. Carolina); Kendall,* Occ. Pap. Boston Soc. nat. Hist., 
7 (8), 1908: 17 (New England); Fowler, Proc. Acad. nat. Sci. Philad., 1910: 604; Halkett,* Check List 
Fish. Canad., 1913: 44 (genl. range of distr.); Nichols and Breder,* Zoologica, N.Y.,9, 1927: 31 (genl. 
acct.); Hildebrand and Schroeder,* Bull. U.S. Bur. Fish., .^J, 1928: 76 (Provincetown, Massachusetts); 
Breder,* Field Bk. Mar. Fish. Atl. Cst., 1929: 43 (genl. acct.); Truitt, Bean, and Fowler, Bull. Md. 
Conserv. Dep., 3, 1929: 33 (genl. acct.); Jordan,* Manual Vert. Anim. NE U.S., 1929: 33 (distr., key, 
and brief characteristics); Jordan, Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 2, 1930: 34 
(in checklist); Jordan and Evermann, Amer. Food Game Fish., 1937: 12 (genl. acct., key); Greeley, 
N.Y. Biol. Surv. Lower Hudson Watershed (1936), 1 1, 1937: 90 (size ranges, age determ., Hudson R.); 
Schrenkeisen, Field Bk. Freshw. Fish. N. Amer., 1938: 14 (genl. acct.); Vladykov and Beauheu, Nat. 

24. We follow the lead of Jordan, et al. {45: 34) in calling this species brevirostris. The authors who used the spelling 
bre^irostrum are marked by an asterisk (*) following their names. 



Fishes of the Western North Atlantic 41 

Canad., 7J, 1946: 43 (detail, acct. of shields and gill rakers, size); Nat. Canad., 76', 195 1 : 199 (detail. 

acct. of gill rakers for three w. Atlantic species). 
Acifenser {Huso) ksueurii Valenciennes (Ms.) Dumeril, Hist. Nat. Poiss. 2, 1870: 166 (descr., New York).-^ 
Adpenser (Huso) rostellum Dumeril, Hist. Nat. Poiss., 2, 1870: 173 (descr.. New York?).^^ 
Acipenser {Huso) sinus Valenciennes (ms.) Dumeril, Hist. Nat. Poiss., 2, 1870: 175 (descr., New York?).^^ 



Acipenser fulvescens Rafinesque 18 17 
Lake Sturgeon,^^ Smoothback 
Figures 7, 9 

Study Material. Over 800 specimens, 6-229 ^"^ or 2.5-90 inches TL (5.7— 
21 1 cm or 2.3—83 in. FL), from different sections of the St. Lawrence River, Quebec; 
also odd specimens from Lake Ontario on both the Canadian and New York sides. 
Specimens from a large number of Interior waters not included. For details, see Vla- 
dykov and Beaulieu (83: 18). 

Distinctive Characters. A.fulvescens, like J. brevirostris, is distinguishable from 
A. oxyrhynchus by: the arrangement of its preanal shields in a single row, a mouth 
not less than 66" j^ as wide as the interorbital width, the absence of a soft area 
(fontanelle, p. 27) on the top of the head, and black viscera. From brevirostris it 
differs in that its lateral shields are of the same color as the background. 

Description. Scutes in young specimens (up to 12 in. FL), in all rows, developed 
even more strongly than in A. oxyrhynchus of corresponding sizes; progressively ab- 
sorbed with age, beginning with the ventral rows, until they may be completely so, near 
spawning time, especially in the ventral rows (hence the commercial name Smooth- 
back); in 516 Quebec specimens, 9-17 (av. 13.4) dorsal shields, and in 242 specimens, 
29—42 (av. 35.4) lateral shields; shields behind dorsal fin in a single row; elongated 
fulcrum at base of lower caudal lobe never longer than base of anal fin. Dermal 
OSSIFICATIONS minute, but the skin much tougher than in A. oxyrhynchus. 

Head longer in young than in larger individuals, on the average 24''/o of FL in 
specimens 40-70 cm FL. Snout in specimens less than 50 cm FL longer than post- 
orbital distance, but shorter in older individuals. Interorbital width rather large, 
varying; in 79 young specimens 8-36 cm FL, 2 i. 6-31. 80/0 (av. 28''/o) of hi; in 143 
older specimens 40-211 cm FL, 2^— 4.0" jo (av. 34''/o). Mouth large, its width (ex- 
cluding lips) 66-93 "/o (^v. 77 '/o) of interorbital width. 

Gill rakers rather short and blunt, but numerous; 350 Quebec specimens above 
8 inches FL with 25—40 (av. 33.1) rakers. 

25. The Dumeril holotypes of these three species are kept mounted in good condition at the Paris Museum National 
d'Histoire Naturelle. They were examined in 1950 and identified as A. bre'virostris by Vladykov. A similar identi- 
fication was made by Benin {12: 252). Jordan, et al. {45: 34) quoted two more names given by Dumeril 
{28: 164-168), based on locality only: microrhynchus (New York) and dekayi (New York). 

26. In Quebec, the French-speaking fishermen call the adults Esturgeon jaune or Camus, that is, short and flat 
snout; the latter name was already in use at the end of the last century (52: 188). The young specimens are usually 
named Escargot, or in certain sections (Nicolet) Charbonnier, on account of the black blotches (83: 32). On the 
New York market, this species is often called the Smoothback. 



42 Memoir Sears Foundation for Marine Research 

Caudal fin with rather short upper but long lower 
lobe; the average difference between TL and FL 9"/q of 
FL; no notch at tip of upper lobe. 

Color. There is a pronounced difference between the 
young and the adult. In young below 300 mm (12 in.) FL, 
two pairs of large black blotches present on anterior half 
of upper surface of snout and on back between dorsal and 
\. , lateral rows of shields, the anterior blotch the larger and 

wk\ immediately behind the base of pectoral fin, the smaller 

I posterior one below the dorsal fin;-' in addition, several 

small, irregular, black spots on top of head, on back and 
^' sides, and on lower surface of caudal peduncle. Intestinal 

1 tract darkly pigmented, nearly black. Air bladder and fat 

, surrounding gonads, dark gray. Peritoneum pale or only 

' 7 slightly pigmented. In general, the dark pigmentation of the 

/ intestinal tract is similar to that of A. brevirostris but very dis- 

I tinct from that of A. oxyrhynchus. In half-grown specimens 

' up to about 24 inches or 610 mm FL, the large blotches 

t I have disappeared, but black spots may persist. Older in- 

I dividuals above 40 inches FL are uniformly dark brownish 

or grayish on top of head and on back and sides. In fish of 
all ages, dorsal and lateral shields of the same color as back- 
ground; shields of ventral rows sometimes pigmented on 
upper part but always whitish on lower part. Lower surface 
of young and adults uniformly whitish. In young below 24 
inches FL or about 610 mm, preserved in 4— 5''/o formalin, 
lower surface and sides of head and body clear green. 
; At all ages, fins dark brown or gray, typically without 
J white edging. 

Size and Growth. Among western North Atlantic spe- 
cies, A.fulvescens reaches a size exceeded only by A. oxyrhyn- 
chus. The largest recorded is a female Lake Sturgeon, of 
approximately 275 pounds, taken in Lake Winnipeg, Mani- 
toba, in the summer of 1941 {40 '. 17). The largest male, 
caught in Lake Erie in 1948, weighed 220 pounds {66: 7). 
The specimens from Quebec listed in Study Material, 2.5—90 
inches TL, weighed 0.032 ounces-212 pounds (0.9 g— 96 

27. For a good illustration of these black blotches in young A. fulijescens, see 
Forbes and Richardson {31: 36); for color illustrations of an adult, see Vla- 
dykov {82: 6). 

Figure 9. Acifenser fuhescens. Lateral view of immature female, 655 mm 
long, from the St. Lawrence River near Neuville, Quebec, P. Q., Canada. 




Fishes of the Western North Atlantic 43 

kilo).-' Data for the smallest 1 1 specimens from the St. Lawrence River, taken from 
August to November in pSches located near Quebec City, are tabulated in Table i. 
Judging by their long snouts, dorsal scutes in a continuous row without separation, 
general coloration, and small size, we consider all of these specimens to be less than 
one year old. During the first year, juveniles of A.fulvescens (Table i) and of A. oxy- 
rhynchus grow at about the same rate. 

Table I. Growth Rate of A.fulvescens. 



Date 




Total 


Length 


Fork 


Length 




Weight 


Place 






(mm) 


(inches) 


(mm) 


(inches) 


(g) 


(oz.) 




Sept. 19, 


1952... 


63 


2.48 


57 


2.24 


0.9 


0.032 


St. Nicolas 


Oct. 19, 


1953... 


84 


3-31 


82 


3-23 


2.0 


0.071 


St. Nicolas 


Aug. 8, 


1947... 


87 


343 


79 


3-II 


2.5 


0.088 


Neuville 


Nov. 19, 


1947... 


99 


3.90 


90 


3-54 


3-2 


O.I 1 3 


Neuville 


Oct. 31, 


1952... 


106 


4-17 


98 


3.86 


4-4 


0.155 


Neuville 


Sept. 12, 


1944... 


log 


4.29 


96 


3-78 


4-3 


0.152 


Neuville 


.-^ug. 20, 


1952... 


117 


4.61 


107 


4.21 


4.8 


0.169 


Neuville 


Sept. 30, 


1947... 


121 


4.76 


109 


4.29 


4-7 


0.166 


Neuville 


Aug. 29, 


1947... 


121 


4.76 


III 


4-37 


7-7 


0.271 


St. VaUier 


Aug. 25, 


1952... 


123 


4.84 


112 


4.41 


6.5 


0.229 


Neuville 


Sept. 30, 


1947... 


123 


4.84 


112 


4.41 


6.2 


0.219 


Neuville 


Average 




104.8 


4-13 


95.8 


3-77 


4.29 


0.151 





Although the ages of Lake Sturgeon in our material from the St. Lawrence River 
have not as yet been determined, probably their range is up to 30 years, or even more. 
According to Cuerrier and Roussow, who studied the growth oi fulvescens from the 
upper St. Lawrence River (Lake St. Francis), fish 3-22 years old averaged 16.5-37.5 
inches FL {24: 8—14). The oldest fish in their material, a female 23 years of age, 
was 50.5 inches FL and weighed 42 pounds. 

Spawning and Reproduction.^^ The principal spawning season for A. fulvescens in 
Quebec extends from about the beginning of May to the end of June. The smallest 
male ready to breed, taken at St. Vallier on May 5, 1944, was 38 inches TL (35 in. FL) 
and weighed 1 1 pounds 4 ounces. The smallest mature female, taken at the same place 
on June 13, 1949, was 45 inches TL (41 in. FL) and weighed 17 pounds. The Lake 
Sturgeon from Lake Nipigon, Ontario, does not become sexually mature until it is 
about 22 years of age, when it is approximately three feet TL (j5: 15-25). 

Roussow, in studying cross sections of the first pectoral ray of A. fulvescens^ 
found variations in the broad and narrow zones on this bone (6^: 553—572). He con- 
cluded that "the growth retardations caused by the ripening of the gonads and the 

28. For the length-weight relationship of St. Lawrence specimens from Quebec, see Vladykov {y8: ^7). Roussow 
has given some information on the rate of growth and movements of A . ftd'uescens from the St. Lawrence 
and Ottawa rivers (6j: 1-124). Results of extensive tagging at Quebec of 1,507 A.fulvescens and 2,995 ■^•"xy- 
rhynchus have been summarized by Magnin and Beaulieu [^o: 237-252). 

29. Additional information on spawning habits and experimental hatching of Lake Sturgeon has been given by 
Stone (69: 1 18-123) ''od Carter {16: 60-63). 



44 Memoir Sears Foundation for Marine Research 

losses at the consecutive spawnings are visible in the ray sections in the form of 'belts' 
of 2 to 7 narrower zones (or annuli). These indicate that the interval between spawning 
periods can vary from 4 to 7 years and more." It is highly desirable to corroborate these 
findings by observing the frequency of spawning by tagged specimens. 

The principal spawning rivers in Quebec are the St. Francis for Lake St. Peter, 
and the Batiscan and Chaudiere for the respective sections of the St. Lawrence from 
Three Rivers to Montmagny. 

The diameter of mature but unfertilized eggs preserved in 4-5% formalin varied 
between 2.7—3.1 mm. Thus the ripe eggs of A. fulvescens are somewhat larger than 
those of A. oxyrhynchus. Table 11 gives data on ovaries and eggs from mature females 
of A. fulvescens from the St. Lawrence River near Quebec. 

Table IL Data on Ovaries and Eggs of A. fulvescens. 

TL 

Locality Date ('^■1^) 

St.Vallier June 5, 1944 136 

St.Vallier May 15, 1950 144 

St.Vallier May 15, 1950 142 

St.Vallier June 17, 1946 145 

St.Vallier May 19, 1946 175 

St.Nicolas May 8, 1949 170 

* Approximate. 

Food. In the St. Lawrence, A. fulvescens subsists on about the same type of food 
as A. oxyrhynchus (p. 52). In 29 stomachs oi fulvescens we found that larvae of the bur- 
rowing May fly [Hexagenid) were present in 76 "/o of the samples, amphipods in 73 "/o, 
gastropod mollusks in 48 "/o, bivalve mollusks in 35%? and fish in 2^"/^. Indeed, 
A. fulvescens of different sizes are easily taken by hook-and-line, and the best bait is 
young shad, or, as second choice, banded killifish (Fundulus diaphanus). Harkness^" 
found that A. fulvescens in Lake Nipigon, Ontario, fed principally on May fly and 
chironomid larvae and on mollusks. 

Abundance and Commercial Importance. According to official statistics for 1951, the 
catches of Lake Sturgeon were (in pounds): 9,000 for Saskatchewan, 164,000 for 
Ontario, 194,000 for Quebec, and 13,000 for the U.S. Great Lakes; total, 380,000 
pounds, with a combined monetary value of $ 300,000. Late nineteenth-century yields 
were several times greater. From Lake Michigan alone, in 1880, the catch was 3,839,600 
pounds (jj: 26). Lake of the Woods (Minnesota and Canada) was formerly called 
"the greatest sturgeon pond in the world"; the catch there in 1893 was 1.6 million 
pounds (29: 121-136). 

In the St. Lawrence River, from the Ontario border to Cap St. Ignace at Quebec, 

30. Harkness (jS: 28, 29) has given references to other authors who studied food requirements oi A.ful'vescens from 
various localities. 



Length - 
FL 


FL 


Weigh 


t 


Wgt. 


f Ovaries 


Diam. 
of Eggs 


(cm) 


(in.) 


(kg) 


(lb.) 


(kg) 


(lb.) 


(mm) 


127 


50.0 


17. 1 


37.6 


3-1 


6.9 


2.7 


136 


53-5 


26.3 


58 


47 


10.4 


2.8 


133 


52.4 


19.1 


42 


2.9 


64 


2.9 


135 


53-1 


22.2 


49 


— 


— 


3.0 


170 


66.9 


50-3* 


100* 


— 


— 


3-0 


163 


64.2 


37-0 


81 


II. 2 


24.6 


3-1 



Fishes of the Western North Atlantic 45 

there is a regular fishery for this species; at Quebec, although present catches are less 
abundant than formerly, A.fulvescens is still taken in fair quantity; and it is reported 
as being still very plentiful in the Hudson Bay area. Large A.fulvescens are taken early 
in the spring or late in the fall. During the summer months, in the vicinity of Quebec 
City from St. Nicolas to Montmagny, A. oxyrhynchus is more plentiful t\\2in fulvescens\ 
however, both species are taken together quite often at the beginning and end of the 
fishing season. The fish are caught with either gill nets (8—10 in. stretched mesh) or 
weirs, and a few are taken on set lines baited with fish, or in New York on unbaited 
set lines ("trap lines"). 

Relation to Man. Considering its entire range, the Lake Sturgeon is at present of 
much greater commercial value than both the American Atlantic Sturgeon and the 
Shortnose Sturgeon combined. The principal product is the meat for "smoked 
Sturgeon." In 1951, in Canada, 2,000 pounds of caviar were made from the eggs of 
this species. 

Range and Occurrence in the Western North Atlantic. Larger rivers and lakes, from 
Hudson Bay to the Mississippi Valley; according to Hubbs and Lagler (^j: 30), 
"from the Red River of the North, the Saskatchewan River in Alberta, and the Hudson 
Bay, St. Lawrence and Lake Champlain drainages of Canada (and possibly from Labra- 
dor), southward, west of the Appalachian Mountains, to the Tennessee River of Ala- 
bama, to Missouri and to eastern Nebraska." On several occasions it has been reported 
as being taken on the shores of Hudson Bay (j^: 30 [42]). ^^ In the St. Lawrence River, 
where a general line of demarcation between fresh and brackish water can be drawn 
from Cap Brule on the north shore to St. Jean-Port- Joli on the south shore, it is taken 
in commercial quantities from Lauzon to Cap St. Ignace, together with oxyrhynchus. 
Occasionally Lake Sturgeon venture even farther into brackish water. For instance, on 
October 10, 1954, an A.fulvescens weighing five pounds was taken in a peche at St. Roch- 
des-Aulnaies; information from Mr. Alfred Pelletier, proprietor. For further details 
see Vladykov {yg: 54-57). It should be emphasized, however, that a sea-resident 
stage, characterized by a rapid growth rate, has not been demonstrated for this species. 

Synonyms and References: 

Acipenser fulvescens Rafinesque, Amer. Mon. Mag., I, August, 1817: 288 (descr.. Great Lakes); Hubbs, 

Copeia, 1917: 48 (synon.); Hubbs, Misc. Publ. Mus. Zool. Univ. Mich., J5, 1926: 8 (in checklist); 

Dymond, Univ. Toronto Stud. Biol., 2y, 1926: 33 (descr.. Lake Nipigon, Ontario); Dymond, Hart, 

and Pritchard, Univ. Toronto Stud. Biol., JJ, 1929: 13 (statist, of catches); Jordan, Manual Vert. 

Anim. NE U.S., 1929: 33 (distr., key, and brief characteristics); Jordan, Evermann, and Clark, Rept. 

U.S. Comm. Fish. (1928), 2, 1930: 35 (in checklist); Greeley, N.Y. Biol. Surv. Champlain Watershed 

(1929), 4, 1930: 73 (Lake Champlain); Bajkov, Trans. Amer. Fish. Soc. (1930), 60, 1931: 11 (statist. 

of catches, Manitoba); Greeley, N.Y. Biol. Surv. St. Lawrence Watershed (1930), 5, 1931: 81 (St. 

Lawrence R., New York); Bajkov, Trans. Amer. Fish. Soc. (1930), 60, 1931: 11 (Manitoba, statist. 

of catches); Gowanloch, Bull. La. Conserv. Dep., 23, 1933: 412 (genl. acct.); Vladykov, Contr. Canad. 

Biol., 8 (29), 1933: 30 (42) (Hudson Bay region); Schrenkeisen, Field Bk. Freshw. Fish. N. Amer., 

1938: 13 (genl. acct.); Kuhne, Guide Fish. Tennessee, 1939: 17 (genl. acct.); Dymond, Contr. R. 

31. A young specimen of A.ful'vescens about 107 mm FL (USNM 27784) was taken by Walton Hayden at Moose 
Factory, James Bay (a southern extension of Hudson Bay), no doubt in brackish water. 



46 Memoir Sears Foundation for Marine Research 

Ontario Mus. ZooL, 25, 1939: 8 (statist, of catches, Ottawa R.); Hinks, Fish. Manitoba, 1943: 15 
(Manitoba); Eddy and Surber, North. Fish. Upper Mississippi Valley, 1943: 57 (size, artif. propag., 
Mississippi R.); Vladykov and Beaulieu, Nat. Canad., 7J, 1946: 27 (detail acct. of shields and gill rakers, 
size, Quebec); Hubbs and Lagler, BuU. Cranbrook Inst. Sci., 26, 1947: 30 (distr.); Dymond, Misc. 
Publ. R. Ontario Mus. Zool., I, 1947: 4 (genl. distr., range); Vladykov and Beaulieu, Nat. Canad., jS, 
195 1 : 129 (detail, acct. of gill rakers for three w. Atlantic species); Scott, Freshw. Fish. East. Canada, 
1954:7. 

Acipenser ruhicundus LeSueur, Trans. Amer. philos. Soc, J, 1818: 38 (descr., Lakes Ontario, Erie and all 
upper lakes); DeKay, Zool. N.Y., 4, 1842: 344 (descr.); Jordan and Evermann, Bull. U.S. nat. Mus., 
47 (i), 1896: 106 (descr., synon., distr.); Carter, Trans. Amer. Fish. Soc, 34, 1904: 60 (artif. propag. 
Vermont); Evermann and Goldsborough, Proc. biol. Soc. Wash., 20, 1907 (Canada); Kendall, Occ. Pap. 
Boston Soc. nat. Hist., 7(8), igo8: 16 (Lake Champlain, Vermont); Nash, Vertebr. Ontario, 1908: 
14 (genl. acct.); Kendall, Proc. Portland Soc. nat. Hist., 2, 1909: 219 (checklist, Labrador); Halkett, 
Check List Fish. Canad., 1913: 44 (Canada); Forbes and Richardson, Fish. Illinois, 1920: 24 (descr., 
statist, of catches); Harkness, Univ. Toronto Stud. Biol., 24., 1923: 13 (age, food); Kendall, Contr. 
Canad. Biol. (1923), 23, 1924: 21 (439) (between Lake Winnipeg and Hudson Bay); Breder, Field 
Bk. Mar. Fish. Atl. Cst., 1929: 43 (genl. acct.); Jordan and Evermann, Amer. Food Game Fish., 1937 : 
10 (genl. acct., key). 

Acipenser maculosus LeSueur, Trans Amer. philos. Soc, I, 1818: 393 (descr., young from Ohio R.). 

Acipenser serotinus Rafinesque, Ichthyol. Ohiensis, 1820: 80 (Ohio R.). 

Acipenser ohiensis Rafinesque, Ichthyol. Ohiensis, 1820: 81 (Ohio R.). 

Acipenser macrostomus Rafinesque, Ichthyol. Ohiensis, 1820: 81 (Ohio R.). 

Dinectus truncatus Rafinesque, Ichthyol. Ohiensis, 1820: 82 (erroneous, based on drawing by Audubon). 

Acipenser rupertianus Richardson, Fauna Boreal. Amer., 1836: 311 (specimen from Albany R., Rupert 
Land). 

Acipenser laevis Ag&ii\z, Lake Superior, 1850: 267 (Lake Superior). 

Acipenser carbonarius Agassiz, Lake Superior, 1850: 271 (Lake Superior). 

Acipenser rhynchaeus Agassiz, Lake Superior, 1850: 276 (Lake Superior). 

Acipenser Hopeltis Giinther, Cat. Fish. Brit. Mus., 8, 1870: 341 (Mississippi). 

Nominal specific names, given by Dumeril (jj: 105, 220) to specimens oi A.fuhescens, according to localit)' 
only: copei (supposedly upper Missouri), rauchi (Osage R., Missouri), richardsoni (upper Mississippi), 
anasimos (Missouri R.), paranasimos (Huntsville, Alabama), anthracinus (Lake Erie), megalaspis (Lake 
Champlain), lamarii (supposedly Lake Erie), atelaspis (Saskatchewan R.), rajinesquii (Ohio R.), rosarium 
(supposedly Lake Erie), platyrhinus (supposedly upper Mississippi), kirtlandi (Lake Erie), nertinianus 
(Michigan), Cincinnati (Ohio R.), and buffalo (Lake Erie). 



Acipenser oxyrhynchus oxyrhynchus Mitchill 18 14 

Atlantic Sturgeon, Sea Sturgeon, Common Sturgeon*^ 

Figures 3, 7, 10 

Study Material. Over 3,000 specimens: i, ca. 485 mm (19 in.) FL, New Brit- 
ain, Florida, Jan. 6, 1900, CNHM 35376 (received through kindness of Loren P. 
Woods); 42, 200-2,490 mm TL, Hudson River, New York ^^j the remainder, 6c^- 

32. According to the List of Common and Scientific Names (j: 7), oxyrhynchus should be called the Atlantic Stur- 
geon, but the name Sea Sturgeon is more generally used. In the Hudson River adults are known as the Big 
Sturgeon and small ones as Sharpnose Sturgeon or Pelican (55: 89). In Quebec, French-speaking fishermen 
commonly call the larger ones Esturgeon noir, or Esturgeon d'eau salee, while the young are known as Escargots 
or Ecailles (Sj: 43). 

33. For age and size of these specimens, see Greeley (55: 68, 139). 



Fishes oftJie JVestern North Atlantic 



47 



2,670 mm TL (60-2,500 mm FL), St. Lawrence River, 
Quebec, between Three Rivers and Riviere-aux-Outardes.** 

Distinctive Characters. A. oxyrhynchus oxyrhynchus is 
readily distinguishable from both brevirostris (Shortnose 
Sturgeon) and fulvescens (Lake Sturgeon) by the arrange- 
ment of its preanal shields in a double row, by its pale un- 
pigmented viscera, by a mouth less than 55 "/o as wide as 
the space between the orbits, and by the presence of a soft 
area (fontanelle) on the top of its head. It differs from A. oxy- 
rhynchus desotoi (p. (,€) in that its bony shields are oval and 
longer than broad, and that the carina on its dorsal shields 
do not have a conspicuous hook. Among western Atlantic 
species, furthermore, A. oxyrhynchus has much the longest 
upper caudal lobe but a rather short lower one. In A. oxy- 
rhynchus the average difference between TL and FL is i4''/o 
of FL, in A. brevirostris 11 "jo, and in A. fulvescens <)''/o- 
For a discussion of the differences between oxyrhynchus and 
sturio of the eastern Atlantic, see p. 57. 

Description.^^ Scutes strongly developed in all longitu- 
dinal rows; in 1,814 Quebec specimens, 7—13 (av. 9.8) dorsal 
shields, and in 692 specimens, 24—35 (^^- 2^-7) lateral 
shields ; shields behind dorsal fin in pairs ; elongated fulcrum 
at base of lower caudal lobe decidedly longer than base of 
anal fin. Dermal ossifications also well developed on skin 
between dorsal and lateral rows. 

Head elongate. A membranous Fontanelle between 
frontal and parietal plates in young individuals, and a sinus 
at all ages. Snout longer than postorbital distance in indi- 
viduals up to 95 cm, but in older specimens shorter than 
postorbital distance. Interorbital width rather narrow, var- 
ying; in 239 young specimens (6—49 cm), 22.4-28.4"/o 
(av. 26" jo) of hi, in 439 older specimens (50—231 cm), 25— 
2 5.6° jo (av. 2g°jo). Mouth narrow, its width (excluding 
lips) less than 55*'/o of interorbital width. 

Gill rakers among specimens from the St. Lawrence 
River, Quebec (studied by Vladykov and Beaulieu, 83: 43- 

34. For origin, size, and method of fishing, see Vladykov and Beaulieu {83 : 43-47). 

35. In part from Vladykov and Beaulieu {83: 43-47), but mainly from an unpub- 
lished manuscript by the same authors on the morphometric studies on Quebec 
Sturgeons. 

Figure 10. Acipenser oxyrhynchus. Lateral view of immature male, 581 mm 
long, from the St. Lawrence River near St. Vallier, Quebec, P. Q., Canada. 




V: 



H 



Q 



J/ 






,//' / I v^ 




48 Memoir Sears Foundation for Marine Research 

47): total number on first branchial arch in 70 young below eight inches, 15—24 (av. 
20.4); in 800 large individuals 8—91 inches, 16—27 (av. 21.5). This difference was 
due to the fact that in young specimens not all rakers were developed as yet on the 
upper limb. Therefore, for purposes of comparison of gill raker counts, only individuals 
above eight inches FL should be used. In 70 large specimens oi A. oxyrhynchus, 9-14 
(av. 1 1.5) rakers on lower limb and 7—13 (av. 9.8) on upper limb. 

Caudal fin with average difference between TL and FL, i4''/o of FL; a notch 
at tip of upper caudal lobe in undamaged specimens. 

Color. There is no significant difference in coloration between young specimens 
and adults. ^^ Back and top of the head bluish-black, becoming progressively paler 
down the sides ; entire lower surface whitish. Median carina and spine on dorsal shields 
(in dorsal row) whitish, contrasting strongly with darkly pigmented back, particularly 
with the jet-black skin covering the extremities of each shield, the dorsal row thus 
appearing as a series of whitish spots on a dark background. Upper half of lateral 
shields of same color as background, but central and lower part of these shields whitish, 
contrasting well with pigmented sides of fish, but not as sharply as in the case of the 
dorsal shields. Sides of head and body, including ventral row of shields, white. Iris 
pale golden. Anterior edge of pectorals, ventrals, and lower caudal lobe white, clearly 
defined against remaining dark grey part of these fins; anal fin entirely white; dorsal 
fin and upper caudal lobe dark grey or blue-black in the central portion, their broad 
posterior margins whitish. Intestinal tract and peritoneum nearly white, never darkly 
pigmented. 

The coloration of the three species that occur on the Atlantic coast of America, 
when preserved in 4-5 "/o formalin, is characteristic. In A. oxyrhyrtchus, the ventral sur- 
face and sides of the fish, even after a long period in perservative, are whitish or faintly 
pinkish. The skin of the two other species {brevirostris and fulvescens), however, soon 
acquires and permanently retains a green color that is particularly pronounced on the 
unpigmented parts of the body, such as the ventral surface, sides of the head, etc. 

Size. Ryder found that females averaged about 8 feet TL, and sometimes 
reached 10 feet, whereas males ranged between 6-7 feet {65: 268). Weights and total 
lengths of two specimens taken in the Hudson River in 1936 and recorded by the 
New York State Conservation Department were as follows: unspent male 7 feet 4 
inches, 176 pounds; a spent male 8 feet 4 inches, 199 pounds 15 ounces. For the Gulf 
of Maine, Bigelow and Schroeder (jj: 82) stated that "about 12 feet is perhaps the 
greatest length to be expected today. But i 8 feet, reported for New England many 

years ago, may not have been an exaggeration The heaviest Gulf of Maine Sturgeon 

reliably reported (to our knowledge), was one of 600 pounds, landed in Portland by 
the steam trawler Fabia from Georges Bank, December 21, 1932." A 14-foot female, 
weighing 8 1 1 pounds, was caught at Middle Island Maguerville, about 64 miles off 
the estuary of the St. John River, New Brunswick, in July 1924.^' 

36. For color illustrations of an adult Atlantic Sturgeon, see Vladykov [82: 7). 

37. Caught by Mr. John Neilson of Moncton. This information was kindly furnished by Mr. J. Raymond Tippett, 
Moncton, New Brunswick, a Sturgeon fisherman himself. 



Fishes of the Western North Atlantic 49 

The St. Lawrence River specimens noted in the Study Material^ 65-2,670 mm 
TL, weighed 0.0015-352 pounds (0.7-160 kg). Every year in the St. Lawrence River, 
Quebec, some Atlantic Sturgeons weighing between 200 and 300 pounds are taken 
by local fishermen. The two largest specimens at our disposal were taken in the 
brackish section of the St. Lawrence River at Ile-aux-Patins in a fascine weir.^^ A spent 
female^* 8 feet 3 inches TL (7 ft. 7 in. FL) and weighing 203 pounds was taken 
July 5, 1946. Another female, almost ready to spawn, 8 feet 9 inches TL (8 ft. 2.5 
in. FL) and weighing 352 pounds, was caught June 23, 1954. 

Development and Growth. Eggs hatch in about one week at a temperature of i7.8°C 
(64° F). In one of the New York hatcheries where some fry were kept until they were 
about an inch long, the yolksac apparently lasted for about six days." Unfortunately, the 
temperatures were not recorded. The newly hatched fry are about 1 1 mm (0.4 in.) 
long. The later growth of young oxyrhynchus has not been followed, but in Europe 
A. sturio reach a length of 4—5.5 inches in two months. 

Little is known about the early stages of oxyrhynchus in nature, but it appears 
that yearlings grow rather rapidly, for in freshwater sections of the St. Lawrence River 
in Quebec we collected some young specimens ranging between 60-96 mm FL 
that we consider to be less than one year of age (see Table iii). The smallest of 

Table IIL Data for the Five Smallest Specimens from Fresh Water, 
St. Lawrence River. 
Date Place TL 

(mm) 

Aug. 8, 1945 Berthier-en-Bas, fresh water 65 

Aug. 24, 1950 St. Vallier, fresh water 96 

Sept. 26, 1944 Berthier-en-Bas, fresh water 103 

Sept. 26, 1944 Berthier-en-Bas, fresh water 102 

Sept. 22, 1944 Riviere Quelle 1 10 

these was tangled in one corner of a net with 8-inch stretched mesh, set in 60 feet of 
water. The others were taken in p^ches, that is, weirs made from chicken wire or 
fascines. Berthier-en-Bas and St. Vallier are on the freshwater section of the St. Law- 
rence River. Riviere Quelle Is situated at the estuary of the river of the same name, 
which empties Into the brackish section of the St. Lawrence. Additional Information 
about the growth of young fish that we consider to be yearlings Is given In Table iv. 

Determinations of the apparent growth rates of A. oxyrhynchus tagged in the 
St. Lawrence are made difficult because of damage to their tails." However, Instances 

38. Taken by Mr. Maurice Ouellet, Kamouraska, Quebec, a Sturgeon fisherman. 

39. The ovaries of this fish weighed 30 pounds. 

40. Eighth Annu. Rep. Comm. Fish., New York (1875), 1876: 12-14. 

41. In Quebec, fishermen who catch Sturgeon in a peche buiit in salt water at St. Andre de Kamouraska carry them 
alive in a horse-drawn cart ("tombereau" locally) for about 30 minutes and dump them in a freshwater brook. 
But if one of them is a tagged fish, it is measured and weighed and returned to the sea by horse-cart after a 
sojourn of a few hours to one or two days in the brook. On a hot summer day, fish in the cart thrash so 



FL 


Weight 


(mm) 


(g) 


60 


0.7 


87 


34 


89 


4-3 


91 


4.0 


96 


4.2 



5 o Memoir Sears Foundation for Marine Research 

Table IV. Sizes and Weights of Young Taken during 1 944 in a Peche at 
Ste. Famille He d'Orleans, near Quebec City. 

No. of Fork Length Weight 

Date Specimens (mm) (g) 

Range Average Range Average 

August 31 25 130-144. 122.8 6.8-17.4 107 

Sept. 6-28 43 100-205 150.6 5.6-35.1 21. 1 

Oct. 11-26 40 125-201 168.5 9.9-47.7 28.0 

of good growth are afforded in four specimens, 27.8—33.3 inches FL and weighing 
4.8—8.8 pounds; the indicated annual increases were 6.3— i4.4''/o in length, corre- 
sponding to 28. 8-47. o^ in weight (for further details, see Vladykov, yS). 

Sturgeon 11—34 inches long from the Hudson River were 2-8 winters old, as 
indicated by their otoliths (J5: 68, tab. 10). Otolith studies, though based on few 
specimens, also indicate a greatly accelerated growth following the seaward migration of 
the immature Sturgeon, if estimated ages of 1 1 years for a 75-inch fish and of 12 years 
for two others of 88 and 100 inches are approximately correct. The maximum age, as 
indicated by the winter marks on the otoliths, is at least 18 winters in the case of a 
female 8 feet 8.5 inches and weighing about 225 pounds, taken from the Hudson 
River, Maiden, N. Y., April 30, 1937.*^ Vladykov (76') has given a graph for the 
length-weight relationship for a series of 1,592 Atlantic Sturgeon from the lower St. 
Lawrence River, Quebec. 

Spawning and Migrations. The Atlantic Sturgeon is an anadromous species, in- 
variably spawning in fresh water but making its growth in salt water. Adults migrate 
from the sea to fresh water in advance of the spawning season. The spawning migra- 
tion begins during February in the St. Marys River, Georgia, typically during April 
in Chesapeake Bay (jp: 73-76), at the end of April and in May in the Hudson River, 
and during May and June in the Gulf of Maine (jj: 82). The exact movements of 
mature Sturgeon in the St. Lawrence River have not been reported, but they prob- 
ably begin to ascend the current in May (fishing during this month is prohibited) 
and continue to run upstream through June and sometimes in early July. 

In the Delaware River in 1925, the spawning migration started on April 23; 
a few fish were taken during the last of April and the first part of May, with the 
largest catches between May 12 and 22 inclusive (J5: 184). During the last week of 
May and the first ten days of June the catch was scarcely one fish per day, and on 
June 10 the Sturgeon fishing ceased. The gill nets (13— in. stretched mesh) were set 
in a narrow arm of the river between Pea-Patch Island and the New Jersey shore 
opposite Delaware City, where the channel is about 2.5 miles long, 6—7 fms. deep, 
and the current strong. The Sturgeon gathered for spawning in the upper part of the 
river near the eastern shore, where the bottom is hard clay. Spawning took place 

hard with their tails that the posterior ends of the caudal rays are broken, and only by slow regeneration do 
they regain their original length. Thus some fish, after repeated recaptures within the same season, seem to have 
grown somewhat shorter. 
42. Age determination by Greeley, not previously published. 



Fishes of the JVe stern North Atlantic 5 i 

at water temperatures ranging between 56°— 64°F. Among 63 adult females, three 
were "pasters"; i. e. with the roe in the form of a slimy, grayish white paste consisting 
of squeezed eggs, white yolks, and ruptured membranes. Between the loth and 22nd 
of May, seven females with running roe and three males with ripe milt were taken 
during the act of spawning. 

Little is known about the breeding grounds or spawning behavior along the 
rest of the Atlantic coast. Apparently the sexually mature individuals ascend the St. 
Lawrence River through deep channels that are kept free of nets because of navigation, 
and it is probable that spawning takes place in pools below waterfalls of certain St. 
Lawrence tributaries, for instance the Batiscan River along the southern shore. 
Moreover, there are several rivers along the northern shore (Riviere-aux-Outardes, for 
instance) where A. oxyrhynchus spawn. It is probable that the spent fish return gradu- 
ally to salt water. 

The eggs when laid are light to dark brown. The outside membrane of ripe eggs 
readily imbibes water and becomes glutinous and sticky. Being demersal and adhesive, 
they become attached to weeds, stones, and so forth, and it is believed that the eggs 
are scattered over a wide area. There is no evidence of prenatal care, such as prepara- 
tion of a nest area {6^'. 267-268; jp: 73). 

The diameter of ripe eggs is 2.5 or 2.6 mm; using this size as a basis for esti- 
mates, there would then be 168,000 eggs per gallon (U.S.) and 800,000—2,400,000 
eggs per fish (65: 268; 15: 186). 

The largest ripe female examined by us, caught on June 23, 1954, in the St. 
Lawrence River at Ile-aux-Patins, Quebec, was 8 feet 9 inches TL (8 ft. 2.5 in. FL), 
weighed 352 pounds, and would have spawned in about a week; its ovary weighed 91 
pounds and contained an estimated total of 3,755,745 eggs. No female A. oxyrhynchus 
have been reported as being ready to spawn before reaching at least 1 50 pounds and an 
age of about 10 years. The testes of the smallest ripe male encountered during this 
study, a fish of 69 inches TL (60 in. FL) and 70 pounds, taken in the St. Lawrence 
River at St. Andre de Kamouraska on June 2, 1949, weighed 3 pounds 13 ounces. 

General Habits. Sturgeon are bottom fish and are seldom seen except when taken 
in nets or when jumping. It is of interest that this relatively sluggish species is capable 
of making powerful jumps (p. 30). In the opinion of fishermen, at least in Quebec, 
Sturgeon jump to rid themselves of attached lampreys. 

Very little appears to be known about the behavior of the Sturgeon in salt water. 
It is astonishing how easily these fish can adapt to a sudden change from salt to fresh 
water, or vice versa. Some tagged specimens were forced to abruptly change habitats 
at least twice during the same season, apparently without harmful results, because 
they were recaptured again alive. 

Thus far the actual movements of A. oxyrhynchus have been studied by means of 
tagging in Quebec only.*' During five years, 1945— 1949, a total of 1,948 was liberated 

43. For details, including a photograph of a tagged Sturgeon, see Vladykov [So: 65-69; 76: 47). Recently in Oregon 
several White Sturgeon (A. transmontanus) were marked with Peterson modified tags {58: 4-5). 



5 2 Memoir Sears Foundation for Marine Research 

at different localities in the St. Lawrence River, in both fresh and salt water. Up to 
December 31, 1952, a total of 47 fish was recovered. Several of them were recaptured 
as many as four and five times each and were subsequently reliberated. The majority of 
the recaptures showed definite mass movements toward fresh water in spring (May- 
June) and back to salt water in the fall (Sept. -Nov.). There were four recaptures of 
tagged Atlantic Sturgeon of unusual interest: three liberated at Kamouraska and one 
at He aux Coudres, Quebec. After periods varying between 307-705 days, three of 
them were recaptured not far from Halifax, Nova Scotia, having traveled a minimum 
of 900 miles. The fourth fish was retaken near the Strait of Canso. The weights of these 
fish when recaptured, according to the fishermen, ranged between 6-24 pounds. 

Food and Feeding. The large Sturgeon feeds on mollusks and other bottom organ- 
isms. The fish roots in the sand or mud with its snout, like a pig (the barbels serving as 
organs of touch), as it noses up the worms and mollusks on which it feeds and which 
it sucks into its toothless mouth with considerable amounts of mud (65' '^^S)- The 
Sturgeon also eats small fishes, particularly launce {Ammodytes) (jj: 83). The mature 
Sturgeon, like the salmon, eats little or nothing while it travels up the river to spawn. 

The digestive tracts of 26 young oxyrhynchus weighing 1-7 pounds, from the 
Hudson River, contained bottom mud along with plant and animal matter, including 
sludgeworms {Limnodrilus), chironomid larvae, isopods, amphipods, and small bivalve 
mollusks (Pisidium) (72: 141 -144, tab. 5). The food of A. oxyrhynchus varies with the 
type of habitat, as in the St. Lawrence River, Quebec. In 27 half-grown Sturgeon 
taken in salt water, polychaete worms {Nereis virens) were found — 265 on the average; 
the maximum number in a single stomach was 1,221. In addition, the Sturgeon fed 
on marine gastropods, shrimps (Cragd), amphipods, and isopods, in that order. In fresh 
water, the bulk of the food consisted of aquatic insects, amphipods, and oligochaete 
worms; in 88 "/o of 178 Sturgeon examined, larvae of the burrowing May fly (Hexagenid) 
were present.** 

Abundance and Commercial Importance. The present Sturgeon fishery along the 
western Atlantic coast is very small. In 1956, according to statistics (in pounds), the 
following catches ol A. oxyrhynchus were made: Quebec 57,000, New Brunswick 800, 
Nova Scotia 1,000, New England 9,000, Middle Atlantic, U.S., 15,000, Chesapeake 
Bay 23,000, South Atlantic, U.S., 129,000, and Gulf of Mexico 15,000; the total was 
249,800 pounds. 

Former catches were manyfold greater: the catch in North Carolina in 1880 
was 437,000 pounds {(ty: 56), in Chesapeake Bay in 1890 over 900,000 pounds 
(59: 75), and in the Delaware River that year, 5,000,000 pounds {22: 372). Seven 
years later, however, the Delaware catch was only 2,428,616 pounds, about half as 
large as in 1880. By 1920 the catch had declined to 22,886 pounds in Chesapeake 
Bay, by 1950 to about 1,200 pounds in and at the mouth of Delaware Bay, and to 
18,900 pounds in Chesapeake Bay (see also Details of Occurrence, p. 54)." 

44. For further details on the food of A. oxyrhynchus in Quebec, see Vladykov [yy: 53-55)- 

45. For the abundance of Sturgeon in New England waters in Colonial days, see Bigelow and Schroeder {13: 83). 



Fishes of the Western North Atlantic 5 3 

At the turn of the century, when Sturgeon were caught in very large numbers 
and when mature individuals were the mainstay, fishing was so intensive that very 
few fish were able to reach the upper waters to spawn. This was probably one of 
the reasons for their great decrease in subsequent years. Also, the building of dams in 
many important rivers (e. g. the Susquehanna, Maryland) deprived Atlantic Sturgeon 
of their favorite spawning areas, and pollution by wastes of all kinds from the factories 
and by municipal sewage from the towns and cities along the Atlantic shore aided 
in reducing still further the local populations. 

Laws prohibiting the capture of Atlantic Sturgeon below a minimum size or 
during certain periods of fishing (jp: 76) were helpful in protecting and maintaining 
the species. However, even at its lowest ebb the Atlantic Sturgeon in North America 
never was close to complete extermination, as some authors** were inclined to believe 
(in 1950). There are indications, along the Atlantic seaboard, that this Sturgeon is 
coming back. The small number of large fish taken during recent years is partially 
due to a reduction in the quantity of appropriate fishing gear. In New York, for 
instance, the lo-inch mesh nets, legally specified for taking this species, are not used 
in the Hudson River because fishermen do not consider it likely that enough fish 
could be taken to justify the expense of gear and license (55: 139). 

In Quebec, the Atlantic Sturgeon is perhaps as plentiful now as it used to be 
in bygone years. For instance, at our request a fisherman of St. Vallier counted all the 
young A. oxyrhynchus that he liberated alive from his peche; during the night of Oc- 
tober 2, 1944, he freed 1,000 young 6—15 inches long. 

The principal Quebec catches are made with weirs situated along the southern 
shore of the St. Lawrence River, from Riviere Quelle to St. Nicolas. Some fish 
are also taken regularly with gill nets in the same area. Along the northern shore of 
the St. Lawrence, some large fish are caught in either weirs or in salmon nets, par- 
ticularly in the estuaries of Riviere-aux-Outardes and Riviere Manicouagan, and 
occasionally large fish are taken at Sept lies. 

Relation to Man. The Atlantic Sturgeon is a good quality foodfish, but less so 
than the Lake Sturgeon {fulvescens), which reaches maturity at around 20 pounds and 
is more generally appreciated for "smoking" purposes. The flesh of the young or half- 
grown Atlantic Sturgeon, which make up the great bulk of the catch of this species, 
is coarse, contains many ligaments, and has only a small amount of fat; it is only the 
large adult that furnishes a tender and rich flesh. Some years ago the price paid to Quebec 
fishermen for a fish weighing 30 pounds when dressed was 35 cents a pound for an 
Atlantic Sturgeon as compared with 80 cents or even one dollar per pound for a Lake 
Sturgeon of the same dressed weight; the male fish brings a better price than the female. 
Moreover, the Atlantic Sturgeon carries heavier scutes, thus there is a greater loss of 
weight during preparation of the meat. Dressing for the United States market consists 

46. Darlet and Prioux {25: 6), in their review of the Atlantic Sturgeon supply in different countries, said that there 
were no more left in America ("aux Ameriques il n'y en a plus"). 



5 4 Memoir Sears Foundation for Marine Research 

of removing the head, collar bone (clelthrum), tail, all fins, and the viscera.*' Thus the 
weight of prepared meat, which varies with size of fish, sex, stage of maturity, and 
condition of fatness, is only about 50 "/o or even ^o°Jq of the entire Sturgeon. The roe 
from Atlantic Sturgeon is as good for making caviar as that from Lake Sturgeon, and 
being mature at a larger size, the Atlantic Sturgeon furnishes a larger quantity of eggs. 
In early Colonial days, around 1709 in North Carolina, the rough Atlantic Sturgeon 
scutes were considered "good nutmeg graters" {6y: 56). 

The value of the Atlantic Sturgeon for angling purposes is negligible. Neverthe- 
less, occasional captures of large ones with hook-and-line stir public interest. It is worth- 
while to mention the capture of a 6-foot Sturgeon, foul-hooked, by a skilled woman 
angler at Wasque Point, Martha's Vineyard, Nantucket Sound (jj: 84). 

General Range. The regular range of A. oxyrhynchus is limited to the Atlantic 
coast of North America and the Gulf of Mexico. The northern subspecies, J. oxy- 
rhynchus oxyrhynchus, occurs from Hamilton Inlet on the Atlantic coast of Labrador 
(7: 290) and the Gulf of St. Lawrence to eastern Florida. In the Gulf of Mexico is 
found the southern subspecies A. oxyrhynchus desotoi (p. 56) (j^: 408, 409, 41 1). Oc- 
casional captures of Atlantic Sturgeon have been reported for Bermuda*^ and even for 
French Guiana,*^ but these southern stragglers probably are of the subspecies desotoi. 
And a large Sturgeon, taken near the mouth of the Mississippi River, the photograph 
of which has been published (j^), can be identified as desotoi by its broad and strongly 
developed scutes. 

There are early reports of the presence of A. sturio (equals A. oxyrhynchus^ in 
Hudson Bay, but from the descriptions it is evident that the specimens in question 
were A. fulvescens (see also J^: 30 [42]). 

Details of Occurrence. Hamilton Inlet on the Atlantic coast of Labrador is the 
most northerly point where A. oxyrhynchus has been reported (7: 290). Blanc Sablon, 
on the Quebec side of the Strait of Belle Isle, is the next most northerly point (some 
specimens from this locality are in our collections). Atlantic Sturgeon are found reg- 
ularly throughout the Gulf of St. Lawrence and in the St. Lawrence River up to Three 
Rivers, and odd specimens are taken even in Lake St. Peter, near Sorel, Quebec (25 : 
19). They are found also in small numbers on the Newfoundland side of the Gulf of 
St. Lawrence, are well known in Nova Scotian waters, especially near estuaries {8 5: 
58), and are caught regularly in the St. John River, New Brunswick, as well as at the 
head of the Bay of Fundy {42: [10] 58). 

To the south, Atlantic Sturgeon are (or were) well known in the Penobscot, 
Kennebec, and Merrimack rivers ; indeed, they entered nearly every stream of any size 

47. Ryder {6$: 276) and Hildebrand and Sckroeder {jq: 75) have given details on the methods of preparing Stur- 
geon for the market. 

48. According to Beebe and Tee-Van, two records of A. oxyrhynchus were mentioned in early Bermuda lists, one 
for 1876 and the other for 1887 (9: 32). 

49. In 1867 Dum^ril [z-j: 161) described a new Sturgeon, as A. cayennensis, from French Guiana "de la riviere 
Oyapock dite la Cayenne." Bertin {12: 248-249), after examining the holotype (No. 3493, 740 mm, stuffed, 
in good condition) in the Paris Museum, considered it as a possible hybrid A. sturio x A. bre'uirostris , but Vladykov, 
who re-examined it in 1950, identified it as A. oxyrhynchus. 



Fishes of tJie Western North Atlantic 5 5 

that empties into the Gulf of Maine. There are also definite records of Sturgeon taken 
off the open coast from the Bay of Fundy southward to Cape Cod (jj: 83), and along 
the coasts of southern New York. Along the Middle and South Atlantic seaboard 
of the United States there are several rivers that formerly maintained important 
fisheries: the Hudson, Delaware, Susquehanna, Potomac, York, James, St. Marys 
(Georgia), and St. Johns (Florida). =" 

At Quebec, it seems that young A. oxyrhynchus pass their entire growth in the 
fresh and brackish water of the St. Lawrence River, for Sturgeons of all sizes are taken 
there — from a few inches long up to nine feet or longer {y8'. 53—56). Why it was that 
some half-grown tagged fish were recaptured several hundred miles from the place of 
release is not known (pp. 51, 52). But the small number of these stragglers does not ap- 
preciably change the impression that the majority of the Atlantic Sturgeon produced in 
the St. Lawrence River does not venture far from the estuary, for they are taken there 
in some numbers throughout the entire fishing season. 

In the Hudson River, where gill nets used for shad and other species accidently 
capture Sturgeon 11—34 inches long and 2—8 winters old (age determinations based 
on otoliths), there is abundant evidence from both winter and summer catches that 
these immature Sturgeon inhabit the freshwater areas of the river throughout the 
year. But there is a wide gap in the Hudson River collections between these and the 
adults; since immature Sturgeon are frequently encountered in salt water in the 
general vicinity, it seems likely that all of them go to sea sooner or later to complete 
their growth. This rule seems to apply also to the rivers of the Gulf of Maine. However, 
some of the Delaware River Sturgeon seeking salt water may remain within the lower 
reaches of Delaware Bay while others go to sea, as proven by catch records of fish 
taken in the pound nets near Ocean City, Maryland. In Chesapeake Bay, fish of all 
sizes are taken in both the upper and lower parts of the Bay throughout that portion 
of the year when the pound nets are in operation (59: 73). But it is not known how 
many Sturgeon stray beyond the outer headlands, from Chesapeake Bay, the sounds 
of North Carolina, or the rivers farther southward. 

Published reports of Sturgeon taken at sea are mostly for waters close to shore, 
but landings of 3,700 pounds in 1950 and 3,900 pounds in 1952 in New York 
and New Jersey by otter trawlers show that some wander out on the Continental 
Shelf; and to the east, some even stray to the offshore fishing banks. Thirty 
Sturgeon, for example, were landed in Boston and Portland from Brown Bank, 
Georges Bank, the South Channel, and Nantucket Shoals by otter trawlers during 
the 'years 1927— 1934; and landings of 5,300 pounds in 1940 and 6,600 pounds 
in 1 947 in New England ports from these offshore grounds correspond to about 
50 and 70 fish, respectively, if an average weight no greater than 100 pounds is 
assumed. ^^ 

50. We believe that the specimens from South Carolina and Florida, referred to as A. breiArostris by Jordan and Ever- 
mann {43: 106) and by several subsequent authors, were oxyrhynchus. 

51. From data collected by the late Walter H. Rich, U.S. Bureau of Fisheries, and from notices in the daily press. 



5 6 Memoir Sears Foundation for Marine Research 

Synonyms and American References : 5- 

Aclpenser oxyrhynchus Mitchill, Trans. Lit. philos. Soc. N.Y., I, 1814: 462 (descr., New York); DeKay, Zool. 
N.Y., 4, 1842: 346 (descr.); Provancher, Nat. Canad., 8, 1876: 226 (genl. acct.) ; Smith, N.C. geol. 
econ. Surv., 2, 1907: 57 (Albemarle Sd.); Hildebrand and Schroeder, Bull. U.S. Bur. Fish., 43, 1928: 
72 (Chesapeake Bay); Breder, Field Bk. Mar. Fish. Atl. Cst., 1929: 42 (genl. acct.); Jordan, Manual 
Vert. Anim. NE U.S., 1929: 32 (distr., key, brief descr.) ; Jordan, Evermann, and Clark, Rep. U.S. 
Comm. Fish. (1928), 2, 1930: 34 (in checklist); Gowanloch, Fish. Fishing Louisiana, Bull. La. Conserv. 
Dept., 23, I933"- 411 (mouth of Mississippi R.); Schrenkeisen, Field Bk. Freshw. Fish. N. Amer., 
1938: 12 (genl. acct.); Vladykov and Beaulieu, Nat. Canad., 7J, 1946: 43 (detail, acct. of shields and gill 
rakers, size, Quebec); Vladykov and Beaulieu, Nat. Canad., y8, 195 1". 129 (detail, acct. of gill rakers 
for three w. Atlantic species); Rostlund, Univ. Calif Publ. Geogr., 9, 1952: 248 (import, to native In- 
dians). 

Acifenser slurio Linnaeus,^' Syst. Nat., ed. 10, 1758, 237 (descr., European seas); Ryder, Bull. U.S. Fish 
Comm. (1888), 8, 1890: 232 (descr., Atlant. est.); Jordan and Evermann, Bull. U.S. nat. Mus., 47 (i), 
1896:105 (descr., Atlant. cst.) ; Kendall, Occ. Pap. Boston Soc. nat. Hist., 7 (8), 1 908 : 1 6 (New England) ; 
Kendall, Proc. Portland Soc. nat. Hist., J (i), 1914: 13 (Maine); Huntsman, Contr. Canad. Biol. (192 1), 
J, 1922: 10 (58) (St. John R., head of Bay of Fundy); Bigelow and Welsh, Bull. U.S. Bur. Fish., 40 
(i), 1925: 74 (descr., habits. Gulf of Maine); Borodin, Trans. Amer. Fish. Soc, 55, 1925: 184 (biol. 
observ.); Nichols and Breder, Zoologica, N.Y., 9, 1927: 30 (size, distr. New England, New York); 
Truitt, Bean, and Fowler, Bull. Md. Conserv. Dep., 3, 1929: 33 (Chesapeake Bay); Beebe and Tee- 
Van, Field Bk. Shore Fish. Bermuda, 1933: 32 (2 recs. for Bermuda cited); Vladykov and McKen- 
zie, Proc. N.S. Inst. Sci., ^9(1), 1935: 53 (genl. acct.); Jordan and Evermann, Amer. Food Game 
Fish., 1937: 8 (genl. acct., key); Bigelow and Schroeder, Fish. Bull. (74) U.S. Fish Wildl. Serv., 53, 
1953: 81 (descr., habits, Gulf of Maine). 

Acipenser sturio oxyrhynchus. Smith, Bull. U.S. Fish Comm. (1891), 1893: 190 (N.Carolina rivers: Pasquo- 
tank, Edenton, and Roanoke); Halkett, Check List Fish. Canad., 191 3: 44 (distr.); Greeley, N.Y. Biol. 
Surv. Lower Hudson Watershed (1936), 11, 1937: 89 (age determ., size ranges, Hudson R.); Bailey, 
Biol. Surv. Merrimack Watershed, New Hampshire Fish Game Dept., 1938: 155 (former abund. 
Merrimack R.). 

Acipenser cayennensis, Dumeril, Nouv. Arch. Mus. Hist. nat. Paris, j, 1867: 161 (descr.. River Oyapock 
known as Cayenne, French Guiana; see ftn. 49). 

To the above Synonyms should be added several nominal specific names, which were given by Dumeril (27: 
161-177; 28: 116-228) to specimens of ^. oxyrhynchus according to locality only: mitchilli (New 
York), kennicotti (James R.), girardi (Maryland), macrohinus (New York), milberti (New York), bairdi 
(Maryland), storeri (Boston), holbrooki (Charleston), and lecontei (New York). 



Acipenser oxyrhynchus desotoi Vladykov 1955 

Gulf of Mexico Sturgeon, Common Sturgeon 

Figure 3 

Study Material. Two specimens, 50 and 60 cm FL, obtained by T. Dawson be- 
tween Twin and Rabbit islands at the mouth of Singing River, off Gautier, Missis- 
sippi, November 30, 1953, CNHM 59803, 59804);^* 3 specimens (seen by V. D.V.), 
from western coast of Florida (two from Cedar Keys, the other from Suwannee River), 
UF collections. 

52. Among the references, we have omitted some early authors whose incomplete or erroneous descriptions of the 
species are of little scientific value. 

53. European authors in general, such as d'Ancona (z), Berg (jj: 94), and others, consider the American form identi- 
cal to the European A. sturio. 

54. Kindness of Loren P.Woods. These specimens served as the basis for the original description (S/: 2, pis. i, 3, 5, 7). 



Fishes of the Western North Atlantic 



S7 



Distinctive Characters and Description. The subspecies desotoi Is distinguishable 
from typical oxyrhynchus of comparable size by the following differences, the most 
striking and characteristic difference being the length of the spleen. Head, pectoral, 
and spleen given in per cent of fork length. For other details, see Vladykov {81). 

Character desotoi oxyrhynchus 

Scutes — dorsal row . . . squarish, with length much oval, with length greater than 

shorter than width; carina width;* carina (keel) low, 

(keel) typically with two without a pronounced hook 
strong hooks 

Head 30-9-33-6''/o 26.5-27.6«/o 

Pectoral fins 1 5.5-1 6.3 "/o 1 1.5-15.1 "/o 

Spleen length le-igo/o 3-9% 

* Specimens from eastern Florida and Quebec. f Two specimens. 

Habits. The subspecies desotoi is anadromous, as is the northern subspecies oxy- 
rhynchus. Further than this there is no detailed information on its habits, except that 
Louisiana fishermen occasionally capture large Sturgeon around the mouth of the Mis- 
sissippi River and in the Sound, where they have been reported as "Common Stur- 
geon" (5^: 41 1-412). Adults probably visit several rivers on the northern and eastern 
shore of the Gulf of Mexico for spawning purposes. 

Range. Apparently its occurrence is limited to the Gulf of Mexico, the northern 
coast of South America, and Bermuda (p. 54). 

Synonyms and References: 

Acifenser oxyrhynchus de sotoi Vladykov, J. Fish. Res. Bd. Canada, 12 (5), 1955: 2 (orig. descr.). 
Acifenser oxyrhynchus Gowanloch, Bull. La. Conserv. Dep., 1933: 411 (mouth of Mississippi R.). 
Acipenser sturio Rivas, Bull. U.S. Fish Wildl. Serv., ^^ (89), 1954: 504 ("• and e. shore. Gulf of Mexico). 
Acipenser {oxyrhynchus) de sotoi Springer and BuUis, Spec. Sci. Rep., U.S. Bur. Comm. Fish., J96, 1956: 46 
(Mississippi Sd.). 



Relationship of A. oxyrhynchus to A. sturio 

Some authors (see Synonyms and References, p. 56), following Ryder {6^: 234— 
238), have considered the Atlantic Sturgeon of America to be the same species as 
the Common Sturgeon of Europe, 

A. sturio Linnaeus; the two species Table V. Comparative Counts of A. sturio and 
resemble each other very closely A. oxyrhynchus. 

in certain features, as is shown in Species A. sturio A. oxyrhynchus 

Table v. Authority Berg Vladykov and Beauheu 

D'Ancona, in a summary of . (-^^ ■ 93-95) (*J = 44-57) 

, , r ^ ■ Dorsal shields .. . <5-i^ 7-i3 

the characters of A. sturto, presum- Lateral shields . . 24-35 H-3S 

ably from the Mediterranean Sea, Gill rakers 18-25 17-27 



5 8 Memoir Sears Foundation for Marine Research 

has quoted somewhat higher figures than those given in Table v for the shields: 10—14 
in the dorsal row (excluding the nuchale) and 25—36 in the lateral row (2). Antoniu, 
for A. sturio from the Black Sea, mentioned the presence of pairs of postdorsal shields 
and of a frontal sinus in the cartilaginous skull ; both of these characters are found 
in A. oxyrhynchus (^: 109— 116). 

Of two specimens obtained by Cope nearly a century ago in Adige (Verona), Italy, 
and labeled A. sturio, one is probably A. guldenstddti (ANSP 646) and the other is 
definitely an A. sturio (ANSP 645), 15 inches long (380 mm) FL. The latter has a 
fontanelle, 10 shields (including nuchale) in the dorsal row, and 30 shields (including 
supracleithrum) in the lateral rows. On the first branchial arch we counted 20 gill 
rakers. The body proportions are also very similar to those of A. oxyrhynchus of cor- 
responding size, and its color, or what remains of it, corresponds very closely to that 
of American specimens. 

In a drawing of A. sturio by d'Ancona, the coloration is also very similar to 
that of the western Atlantic Sturgeon (2: fig. i); on the other hand, the mouth of 
d'Ancona's A. sturio is larger than that of A. oxyrhynchus (2 : fig. 2), the shape of its 
lower lip is very different, and there are other differences as well. According to Berg, 
the adult of A. sturio has several (up to 10—12) dense, oblique series of rhombic plates 
(dermal ossifications) between the dorsal and lateral rows of shields (jJ: 97); in western 
Atlantic specimens these ossifications are much less developed and are rather irregular 
in shape. Moreover, Classen {20: 88—102) has mentioned several ripe females of 
sturio from the Gulf of Cadiz, Spain, d^ inches TL (145 cm FL) and weighing as 
little as 62 pounds (28 kg); oxyrhynchus apparently attains sexual maturity at a larger 
size than sturio, the males of oxyrhynchus being approximately six feet TL and 70 
pounds in weight, the females not less than 150 pounds (p. 51). 

In Europe, probably because of a general scarcity of A. sturio {2^: 5—13), little 
work has been done to elucidate the problem of local races. Nevertheless, there are 
observations by Marti that A. sturio from the Black Sea are similar to those from the 
Mediterranean region but differ from those of the Baltic (5J: 435—442, Russ.); 
the Black Sea specimens have on the average higher numbers of dorsal and lateral 
scutes and dorsal rays. Magnin, by studying cross sections of the pectoral ray, found 
that the rate of growth of the European Sturgeon varies according to geographical 
region {^g\ 152-159). The growth observed was fastest in Italy (Tiber and Po rivers), 
intermediate In Spain (Guadalquivir River), and slowest in France (Gironde River). 

In conclusion, until an adequate comparison of anatomical and meristic characters 
of these two species is made, it is preferable to retain separate specific rank for A. 
sturio and oxyrhynchus. 



TEXT AND FOOTNOTE REFERENCES 



1. American Fisheries Society, List of Common 

and Scientific Names . . . Fish . . . U.S. and 
Canada, Publ. I, 1948. 

2. d'Ancona in Faune Ichthyol. d'Atl. nord, 2, 

1929. 
J. Antipa, Fauna Ichthyol. Romaniei (Buch.), 
1909. 

4. Antoniu, Ann. Sci. Univ. Jassy, 2J (2), 1937. 

5. Antoniu, Ann. Sci. Univ. Jassy, 28 {z), 1942. 

6. Antoniu, Bull. Acad. Roumaine, 29 (5), 1946. 

7. IJACKus, Copeia, 195 1. 

8. Bailey, Pap. Mich. Acad. Sci., jg, 1954. 

9. Beebe and Tee-Van, Field Bk. Shore Fish. 

Bermuda, 1933. 

10. Berg, Class. Fish., Trav. Inst. zool. Acad. Sci. 

URSS, 5 (2), 1940; reprinted by J. W. Ed- 
wards, Ann Arbor, Mich., 1947. 

11. Berg, Poiss. des eaux douces . . . U.R.S.S., I, 

1948. 

12. Bertin, Bull. Mus. Hist. nat. Paris, (2) 12 (6), 

1940. 

13. Bigelow and Schroeder, Fish. Bull. (74) U.S. 

Fish Wild. Serv., 5J, 1953. 

14. Borisov, Trav. Comm. Rep. Jakoute, 9, 1928. 

15. Borodin, Trans. Amer. Fish. Soc, 55, 1925. 



16. V_iARTER, Trans. Amer. Fish. Soc. (1904), 

1905. 
ly. Chalikov, Zool. J., 23, 1944. 
18. Cheng, Fukien Univ. Sci. J., I, 1938. 
ig. Classen, Tech. Fish. Prod., 1925 (Russ.). 

20. Classen, Trab. Inst. esp. Oceanogr., ig, 1944. 

21. Clemens and Wilby, Bull. Fish. Res. Bd. 

Canad., 48, 1946. 



22. Cobb, Rep. U.S. Comm. Fish. (1899), 1900. 

23. Cuerrier, Fry, and Prefontaine, Nat. canad., 7J, 

1946. 

24. Cuerrier and Roussow, Canad. Fish Cult., 10, 

1951. 

25. JL/arlet and Prioux, Bull, franc. Piscic, 1^8, 

1950. 

26. Dominion Bureau of Statistics, 1956. 

2y. Dumeril, Nouv. Arch. Mus. Hist, nat., Paris, 
3, 1867. 

28. Dumeril, Hist. Nat. Poiss., 2, 1870. 

29. JivERMANN and Latimer, Proc. U.S. nat. Mus., 

J9, 1911. 

30. Jtorbes and Richardson, Bull. 111. Lab. nat. 

Hist., 7, 1905. 

31. Forbes and Richardson, BuU. lU. nat. Hist. 

Surv., ed. 2, 3 (3), 1920. 

32. Fortin, Ann. Rep. Fish, (i 863), Append., 1 864. 

33. Fowler, Proc. Acad. nat. Sci. Philad., 1910. 

34. Ljowanloch, Bull. La. Conserv. Dep., 23, 

1933- 

35. Greeley, 26th Rep. N.Y. St. Conserv. Dep., 

Suppl., 1937. 

36. Grote, Vogt, and Hofer, Siisswasserfische von 

Mittel-Europa, 1 (text), 11 (Atlas), 1909, 



3y. JtdlALKETT, Check List Fish. Canad., 191 3. 
38. Harkness, Publ. Ont. Fish. Res. Lab., Biol., 
24(18), 1923. 



59 



6o 



Memoir Sears Foundation for Marine Research 



jg. Hildebrand and Schroeder, Bull. U.S. Bur. 
Fish., 4J, 1928. 

40. Hinks, Fish. Manitoba, 1943. 

41. Hubbs and Lagler, Bull. Cranbrook Inst. Sci., 

26, 1947. 

42. Huntsman, Contr. Canad. Biol. (192 1), J, 

1922. 

43- Jordan and Evermann, Bull. U.S. nat. Mus., 

^7(0. 1896. 
f^ . Jordan and Evermann, Amer. Food Game 

Fish., 2nd ed., 1937. 
4^. Jordan, Evermann, and Clark, Rep. U.S. Comm. 

Fish. (1928), 2, 1930. 



46. J_/ARiM0RE, Copeia, 2, 1950. 
4^^. Leim and Day, J. Fish. Res. Bd. Canad., 16 
(4). 1959- 

48. JVIagnin, Rev. Trav. Inst. Peches marit., 2J, 

1959. 

4g. Magnin, Bull, franc. Piscic, 193, 1959. 

^O. Magnin and Beaulieu, Nat. canad., ^7, i960. 

51. Marti, Zool. J., 18, 1939. 

^2. Montpetit, Poiss. d'eau douce Canad., 1897. 

5J. Mori, J. Chosen nat. Hist. Soc, 16, 1933. 

34- Mori, Mem. Hyogo Univ. Agr., J (3), 1952- 



55. Nichols, Nat. Hist. Centr. Asia, 9, 1943. 

56. Nikitin, Fish Products, 1949 (Russ.). 

5y. Nikolski, Rapt. Amph., Fish., 1906 (Russ.). 

55. Oregon State Game Commission, Bull., 1950. 

Sg. Ir^ROVANCHER, Nat. Canad., 8, 1876. 

60. Putchkov, Phys. Fish., 1941 (Russ.). 

61. JKosTLUND, Univ. Calif. Publ. Geogr., No. 9, 

1952. 



62. Roule, Poiss. des eaux douces Fr., 1925. 

63. Roussow, Off. Biol. Min., Chasse Peches, 

Quebec, 1955. 

64. Roussow, J. Fish. Res. Bd. Canada, 14, 1957. 

65. Ryder, Bull. U.S. Fish Comra. (1888), 5, 1890. 

66. ocoTT, Freshw. Fish. east. Canad., 1954. 
6y. Smith, N. C. geol. econ. Surv., 2, 1907. 
68. Stannius, Handb. Zool. Fische, ed. 2, 1854. 
6g. Stone, Trans. Amer. Fish. Soc, 29, 1900. 

yo. 1 ATARKo, Trav. Mus. zool. Acad. Sci. Ukr., 

10, 1936. 
yi. Ting, Copeia, 1949. 

^2. Townes, 26th Rep. N.Y. St. Conserv. Dep., 
Suppl., 1937. 

73. Trusov, Doklady Acad. Sci. USSR, 67 (3), 

1949. 

74. U. S. Fish Wildl. Serv., Statist. Digest., 1956 



V LADYKOv, Contr. Canad. Biol., 8 (29), 1933. 
Vladykov, Rapp. Gen. Dep. Pech. Quebec 

(1945-1946), 1946. 
Vladykov, Rapp. Gen. Dep. Pech. Quebec 

(1947-1948), 1948. 
Vladykov, Rapp. Gen. Dep. Pech. Quebec 

(1948-1949), 1950. 
Vladykov, Rapp. Gen. Dep. Pech. Quebec 

(1950-1951), 1951. 
Vladykov, Rapp. Gen. Dep. Pech. Quebec 

(1952-1953), 1953. 
Vladykov, J. Fish. Res. Bd. Can., 12 (5), 1955. 
Vladykov, Fish. Quebec, Alb. 5, 1955. 
Vladykov and Beaulieu, Nat. canad., 73, 

1946. 
Vladykov and Beaulieu, Nat. canad., 78, 

1951. 
Vladykov and McKenzie, Proc. N.S. Inst. Sci., 

-f9(0. 1935- 



86. Wisconsin Conserv. BuU., August 1954. 



Order Lepisostei 



ROYAL D. SUTTKUS 

Tulane University 

Acknowledgments. My thanks are due the following persons for the loan of specimens 
or for help in making specimens available for study : Leonard P. Schultz, Ernest A. 
Lachner, William Ralph Taylor, Robert Kanazawa, and James Tyler of the United 
States National Museum; fames E, Bohlke of the Academy of Natural Sciences at 
Philadelphia ; Henry B. Bigelow and Mrs. Myvanwy Dick of the Museum of 
Comparative Zoology at Harvard University ; Reeve M. Bailey, Robert R. Miller, 
and Carter Gilbert of the Museum of Zoology, University of Michigan ; Clark 
Hubbs of the University of Texas ; Charles M. Breder and Vladimir Walters of 
the American Museum of Natural History ; Luis Rivas of the University of Miami ; 
William M. Clay and Louis A. Krumholz of the University of Louisville ; Loren P. 
Woods, Robert F. Inger, and Pearl Sonoda of the Chicago Natural History Museum, 
and Shelby D. Gerking of the University of Indiana. Thanks are also due Edward 
C. Raney and Henry B. Bigelow for reading the manuscript and for making helpful 
suggestions for improvement ; to Bill Komodore for delineation of Figs, ii, 14, 75, 16 ; 
to Rudolph y. Miller for Figs. 12, 1 3, 18 ; and to Miss Patricia Hale for Fig. ly. 
Preparation of the illustrations and examination of many specimens at various 
museums and institutions were made possible by financial aid from the Sears Founda- 
tion through a grant-in-aid from the National Science Foundation (N. S. F. No. 
G J 1 23), and from a National Science Foundation grant to the author {N. S. F. 
No. G go26). 

Scope of Study. This account gives detailed descriptions of tlie Order Lepisostei 
and family Lepisosteidae, of the genus Lepisosteus, and of four species — oculatus, 
osseus, platyrhincus, and spatula. The "Key to Species of the Western North Atlantic" 
encompasses subgenera as well as species and gives the alternate characters that are 

61 



62 Memoir Sears Foundation for Marine Research 

readily recognizable. The Study Material is listed for each species; in most cases this 
includes specimens from the many institutions listed. For easy and quick identification, 
the Distinctive Characters, as in the Key, compare the easily recognizable features of 
one species with the others. Many specimens were examined in preparing the taxonomic 
accounts, and the methods of measuring and counting the characters are detailed on 
p. 68. 

The Descriptions are broken down into three major categories: (i) Proportional 
measurements in per cent of length, (2) Proportional measurements in per cent of head 
length, and (3) Meristic counts. Thus measurements for most of the head characters 
(snout, lower jaw, etc.) are given in per cent of head length as well as in per cent of 
length of fish. Following submission of this account for publication, a small specimen 
of spatula, 49-5 m^n (57.3 mm TL), came to hand; the measurements for this individual 
have been incorporated in the Description with figures enclosed in brackets. 

What is known of the life history of Gars in general and of the species in particular 
has been culled from the literature and gained by direct observation and personal com- 
munication with field workers. The Synonyms and References at the end of each species 
account include references to only those taken in salt water; these are not numerous, 
because Gars for the most part frequent fresh water, and less commonly brackish or 
salt water. 

Characters. Lepisostei are slender ray-finned fishes with well-ossified skeletons, 
as in the teleosteans, but with ganoid scales (see below), an archaic character shared by 
the polypteroids alone among living fishes. The arterial cone of the heart has eight trans- 
verse rows of 4—8 valves each,^ a number of rows greater than in any other living ray- 
finned fish; this character also puts the Gars apart from the polypteroids. The caudal 
fin has no prolonged fleshy axis, but all of the hypural bones that bear the fin rays are 
supported by the upturned end of the vertebral column (caudal "abbreviate hetero- 
cercal," Fig. 1 1), and the rear boundary of the fleshy caudal peduncle slopes obliquely 
dorsoposteriorly, the latter character shared by Amia alone among living fishes. 

The vertebrae are completely ossified and opisthocoelus, i.e. posterior face con- 
cave and anterior face convex, a conformation unique among living fishes though paral- 
leled in some tailed Amphibia and in the thoracic region of penguins, gulls, and plovers 
among birds. In the adult the pleural ribs extend from the vertebrae to the skin (Balfour 
and Parker, 7: 387, pi. 28 fig. 72; Emelianov, 20: 176-180, figs. 8-10). 

The elongated snout, with nasal openings and olfactory sacs at its end, is the result 
of a lengthening of the ethmoid region, and olfactory nerves course through long canals 
in the ethmoid cartilage (Regan, 58: 447, fig. 3). The preorbital (lacrimal or maxillary 
of some authors) is subdivided into a row of 6—8 bones that bear small, medium, 
and large teeth (Holmgren and Stensio, 10: 474, fig. 363; Hammarberg, 2g: figs. 41, 
43; Berg, 8: 211, 414); the larger teeth are radially grooved at the base. The 
infraorbital sensory canal is a prominent surface feature of the preorbitals in the early 

I. For an excellent illustration and for a table giving the numbers of valves in various other fishes, see Berlin 
{25: 1402, fig. 1005B, 1404). 



Fishes of the IVestern North Atlantic 



63 



developmental stages (Collinge, 25: 265, pi. 4, fig. 2 ; 14: 511, 512; Landacre and 
Conger, 42'. 593, 594; Hammarberg, 29: 309, figs. 39-43). A small sliver-shaped 
bone is present, at least in the young, at the angle of the mouth ; this bone is continuous 
with the preorbitals in its dentition but is not traversed by the infraorbital canal ; thus 
it is considered to represent the maxillary. The vomer is paired. There is no opisthotic 




Figure ii. Lepisosteus spatula. Caudal fin skeleton; original drawing (natural size) of a specimen from Lake 
Pontchartrain, Louisiana. 



(Mayhew, 47: 327), supraoccipital, gular plate, or myodone in the Garfishes. The lower 
jaw, a complex of six dermal bones — dentary, articular, angular, preangular, prearticu- 
lar, and coronoid (Arambourg and Bertin, 2^: 2185, fig. 1546) — articulates in front 
of the eye. The sacculus and lagena form a common sac, the largest otolith being in 
the former. Three pairs of branchiostegals support the gill membrane. The opercular 
gills are especially well developed in the Garfishes, an upper horizontal "arm" and a 
bisegmented lower one being noticeable (Wright, y6: 483). 

The dorsal fin is situated posteriorly, above the anal fin. The fins are without 
spines, and the rays of the dorsal and anal fins are equal in number to the supporting 
radialia. There is no adipose fin. 

The body of the adult is completely encased in an armor of rhombic ganoid scales, 
which may be denticulated and sculptured on the exposed surface.^ Most of the scales 

2. The degree of sculpturing is of some taxonomic value when specimens of comparable size are used. 



64 Memoir Sears Foundation for Marine Research 

are hinged by a peg- and groove-type articulation. Although the bony scales form nearly 
an impenetrable covering, the articulations, the fibrous connections, and the arrangement 
in diagonal rows allow some flexibility of the body. The lepisosteid ganoid scale is com- 
posed of two layers, ganoine above and isopedine below, both of which are irregularly 
penetrated by vascular canals.* In addition, both layers, especially the latter, are per- 
forated by a system of small tubes; these tubules penetrate the scale from all sides and 
are dendritic at their inner terminations. A similar layer of ganoine is present on the 
dermal bones about the head. Both margins of the caudal fin as well as the leading 
margins of the other median and paired fins are fortified with a biserial row of fulcra. 
The cheeks are covered by numerous irregular plates. 

The swim bladder,* cellular and highly vascularized, is connected to the pharynx 
by the enlarged pneumatic duct (Wiedersheim, yo: i— 16, pis. 1—3). The gonads are 
staggered (Miiller, 49: pi. 6, fig. 2), i.e. the right one is mostly anterior to the pelvic 
fins and the left one mostly posterior to them. The ovaries are closed, and the right 
and left oviducts join the respective urinary ducts. A short distance posterior to their 
junction, a common chamber is formed by the union of right and left parts; thus 
the products from both systems exit from the body through a single urinogenital 
sinus. No vestigial Miillerian ducts exist in the male (Pfeiffer, ^2: 465). The conus 
arteriosus is a contractile vessel with several rows of valves (^9: pi. 5, fig. 2). 
Remnants of the spiral valve occur in the posterior part of the gut. There is no lumi- 
nescent organ. 

Taxonomic Rank. In 1844, Johannes Miiller combined his family Lepidosteini 
(evidently to include the amiids) with his family Polypterini to constitute his Order 
Holostei (^9: 201—204), and they have been similarly ranked as one of the constituent 
divisions of a more widely inclusive Order by some subsequent authors. Thus they were 
associated: with the Amiidae and Polypteridae by Giinther {28: 328) and with the Amii- 
dae (among living fishes) by Bridge {l2\ 495, 502) as the Order Holostei; with the 
Amiidae (among living fishes) by Arambourg and Bertin {2^: 2 181— 2194) in 1958 as 
the Order Amiiformes. 

They have been regarded, however, as the sole living representatives of a separate 
Order by the majority of recent authors: Ginglymodi by Cope {16: 452, 453), Rhom- 
boganoidea by Jordan and Evermann (57: 108), Lepidosteoidei by Goodrich (2J: 340), 
Holostei by Jordan (j6 : 115), Ginglymodi by Regan {^^g : 313), Holostei by Jordan 
et al. (j8: 2^)y Lepidosteiformes by Berg (8: 211, 414), Semionotoidea by Romer 
(62: 580), Lepisosteida by Matsubara (^6: 170), and Ginglymodi by Norman (5X: 
61). Here also they are regarded as representing a separate Order. 

Name of the Order. The choice here lies between Ginglymodi and some derivative 

3. For accounts'of the~scales, see especially L. Agassiz {2: 74, 75, 77, tab. G, figs. 8-io), Williamson (75: 435-447, 
470, pi. 40, figs. I, 2; 74: 651, 658, 687, 699), Reissner (67: 254-268, pi. 5, figs. :-6), Nickerson {50: 115-139, 
4 pis., 31 figs.), Scupin (6j: 166, 167, pi. 10, fig. i), Goodrich (2J, 758, 759, fig. 199; 43: 21S, 219, fig. 192), 
Kerr {^40: 63-66, figs. 3, 4). 

4. Apparently the swim bladder functions as a breathing organ (Potter, $4: 63) in addition to the gills. When 
removed from water, the Garfishes regularly breathe air into and out of the swim bladder. 



Fishes of the Western North Atlantic 6 5 

of Lepisosteus.^ I.episostei is chosen here because of the connotation carried by this 
name. 

Relationships. It is customary in general works on living fishes to place the Lepi- 
sostei next to the amiids, with which they agree : in the abbreviate-heterocercal nature 
of their caudal fin skeleton (made evident externally by the dorsoposterior slope of the 
rear boundary of the fleshy caudal peduncle); in their flattened, overlapping pelvic 
bones; in their chambered swim bladder that serves as an accessory respiratory organ; 
and in the presence of a rudimentary spiral valve in their intestine. However, they dif- 
fer widely from the amiids in many respects. The body scales, for example, which are 
thick and interlocking but not overlapping in the lepisosteids, are thin and widely over- 
lapping in the amiids, but the fulcral scales that arm the margins of the fins in the 
former are rudimentary in the latter. The long snout, the segmented "maxillary" bones, 
a lower jaw articulating anterior to the eye, and the three rod-like branchiostegals of 
the lepisosteids are replaced in the amiids by a short snout, unsegmented maxillaries, 
a lower jaw articulating posterior to the eyes, and by 10—12 broad plate-like branchio- 
stegals ; and the free margin of the branchiostegal membrane, which crosses the isthmus 
in a continuous arc of long radius in the lepisosteids, is deeply incised there in Amia, 
one side overlapping the other. In lepisosteids the preopercular bone does not extend 
as far forward as that in the amiids, the cheek region is covered by irregular plates 
in lepisosteids but not in amiids, there is no bony gular plate in the chin region be- 
tween the branches of the lower jaw as in Amia, and the dorsal is much shorter in 
lepisosteids than in amiids. The following internal differences deserve mention: the 
vertebral centra, which are concave posteriorly but convex anteriorly in the lepisosteids, 
are concave anteriorly as well as posteriorly in Amia\ the vertebral column in the tail 
region is diplospondylous; and the arterial cone of the heart, which has eight trans- 
verse rows of valves in the lepisosteids, has only two rows in the amiids. 

Among fossil groups, lepisosteids are usually associated with the Semionotidae, 
some of which are known from as far back as the Jurassic. 

Spawning and Development. Spawning takes place entirely in fresh water during a 
rather brief period beginning about mid-May and lasting until mid-June, in the latitude 
of New York (j [1879]: 6^, 67). When Garfish move into shallows to spawn, each 
female is accompanied by one to four males and there is a great amount of thrashing 
during the breeding activities. Fertilization is external. The large numbers of individ- 
uals that concentrate in the shoal areas suitable for spawning disperse rather suddenly 
to other areas afterwards. 

The eggs are adhesive and adhere to the substratum. No parental care is given to 
the eggs or young. On the ventral surface at the end of the snout the larvae have peculiar 
adhesive suckers that are used for attachment to objects on or above the bottom. The 
larvae are relatively inactive until the yolk mass is absorbed, but then they become very 
active predators. 

A characteristic caudal filament that develops at the termination of the upturned 

5. Many authors have preferred the spelling Lepidosleus. 

S 



66 Memoir Sears Foundation for Mari?ie Research 

vertebral column is generally kept in constant rapid motion, as are the fringed, fleshy 
pectoral fins. The caudal filament atrophies at different times in the various species. 
Fin rays develop much sooner in the pelvic than in the pectoral fins; hence the pelvics 
are actually well-defined fins while the pectorals are still delicately fringed, fleshy stumps.* 

The lateral-line sensory system appears early in life, before any scales are formed. 
The lateral-line scales develop in a posterior-to-anterior direction, as do the other body 
scales. Usually by the time the anterior lateral-line scales are developed, lateral scales 
have formed on the caudal peduncle; on the lateral areas, scales develop ahead of those 
on the mid-dorsal and midventral regions. Moreover, those of the mid-dorsal area de- 
velop ahead of the ones on the midventral section, so the breast Is the last to become 
scaled. The most posterior fringe of scales that Is superimposed on the bases of the 
caudal rays does not appear until after the scales on the whole anterior part of the lateral 
line and on the body up to the region of the pelvic fins have developed. This terminal 
fringe of scales usually contributes one to three scales to the lateral-line count. Fulcral 
scales appear after the fins are otherwise well developed. The caudal fin Is somewhat 
of an exception In that one to three rays form In the superior region following the caudal 
filament. The last of these upper rays may not be fully developed until the early juvenile 
stage Is reached. 

Differential growth of structures is pronounced in the Garfishes; although it be- 
comes less marked with age, it does not terminate with maturity. The diameter of the 
orbit continually changes in proportion to other parts of the body — particularly to 
snout length, Interorbital distance, and postorbltal distance. Primarily because workers 
have failed to consider such changes properly, the systematlcs of these fishes has been 
in a chaotic condition for many years.' Males not only mature at a smaller size but 
grow to a smaller size than females, and sexual dimorphism exists In some structures, 
e. g. length of snout In the Spotted Gars. 

Habits and Food. The Garfishes are sluggish creatures most of the time. During 
the summer they frequent the surface waters and appear more or less motionless for 
minutes at a time. In streams below barriers, such as dams or falls, frequently large 
numbers of Longnose Gar can be seen near the surface on bright sunny days. In this 
quiet state, the caudal fin is usually undulated slowly while they are at rest or moving 
slowly forward or backward. But they are capable of swimming rapidly for short dis- 
tances by rapid undulations of the body. When disturbed, they immediately "sound." 
In the winter they are found in the deeper waters; the shrimp trawlers frequently 
catch Alligator and Longnose Gars In their trawl nets from the deep holes of estuaries 
and bayous in the salt marshes of Louisiana. 

During the summer months the periodic surfacing of Garfishes Is a common sight, 
and there Is some debate about their successfulness in gulping air during these surface 

6. A young specimen of L. osseus without scales and with the peculiar fleshy pectoral fins induced Rafinesque to 
describe the form as a new genus, Sarchirus {^6: 418). 

7. L. Agassiz {4: 360) recognized the remarkable changes that take place in the ontogeny of Garfishes, and 
possibly this was the reason he never followed through with the description of 22 supposed new species, most 
of which were discovered during a journey in the eastern United States (L. Agassiz, j: 136). 



Fishes of the IV es tern Nortli Atlantic 6 7 

excursions. However, considering their anatomy, results of experimentation, observa- 
tions on individuals in their natural habitat, and the regular breathing actions of in- 
dividuals removed from water, there is little doubt that the swim bladder and associated 
structures leading to it serve as a supplementary respiratory mechanism. 

Gars feed primarily on other fishes, most of which are forage species (Bonham, 11: 
358—360). There are very few published accounts of cannibalism, but Gars with por- 
tions of their tails missing are frequently captured, indicating that they have been 
snapped at by other individuals; this has been observed in streams and canals of penin- 
sular Florida, where the Florida Gar [L. -platyrhtncus) is very abundant. In estuarine 
waters of Louisiana, Alligator Gar frequently consume the blue crab, Callinectes sapidus^ 
as well as garbage (Weed, 6g: 9, 10; Gudger, 26: 120, 121) where refuse is regularly 
thrown into the water. 

Relation to Man. Although Alligator Gar, L. spatula, attain a large size, no authen- 
ticated records of an attack on man are available {6g: 9, 10; 26: 120, 121). The usual 
response by Gar when disturbed by a bather or fisherman is a dash for deeper water. 
While Gars are considered to be a nuisance and are detrimental to game fishes, sport 
fishing for Alligator Gars has become popular in recent years in Louisiana, Arkansas, 
Mississippi, and other states in the Mississippi Valley. 

Alligator Gar are sold in the French Market in New Orleans at the present time, 
and small numbers are consumed elsewhere in the United States. Dr. Robert R. Miller 
contributes the information that L. tropicus is of considerable importance as a food item 
on the Pacific side of southern Mexico and Guatemala. And personal conversation with 
Nicaraguans reveals that L. tropicus is used for food in the Lago de Nicaragua area 
and lower Rio San Juan. The North and Central American Indians used the ganoid 
scales and bones for arrow points, ritual instruments, and ornaments. The late Mr. 
Percy Viosca, Jr., of New Orleans, at one time made an array of ornaments and jewelry 
from their scales. 

Habitat and Range. Their range extends from Quebec in the northeast, west to 
the upper Mississippi Valley and Great Lakes region, and south to Costa Rica in 
Central America and southwestern Cuba, including the Isle of Pines. Although they 
are primarily inhabitants of freshwater streams and lakes, several species enter brackish 
and marine waters. L. spatula is a common inhabitant of brackish water in Louisiana, 
and several specimens have been captured from the Gulf of Mexico, on the Gulf side 
of Breton Island and Grand Isle, Louisiana, and at Destin, Florida. The Longnose Gar, 
L. osseus, is somewhat less tolerant of marine waters and is seldom captured outside 
of the brackish estuarine areas. The Spotted Gar, L. oculatus, is a common summer 
inhabitant of the fresher parts of the estuaries in Louisiana. 

One species, L. sinensis (Bleeker, 9: 148, 154), supposedly occurs in China (Wag- 
ner, 68: 738-741). Most standard references fail to mention an extant Asiatic form. 

Geological History. Garfishes are known from the Upper Cretaceous to Lower 
Miocene time; in the Eocene of India and in North America from the Middle Eocene 
to Recent period (Berg, 8: 214, 415; Romer, 62: 548). 

s* 



6 8 Memoir Sears Foundation for Marine Research 

Measurements and Counts. 

Length: measurement from tip of snout to posterior margin of last lateral-line 
scale at base of caudal fin. Unless stated otherwise, Length is used. 

Standard length (SL): nearly equivalent to length. 

Total length (TL): measurement from tip of snout to longest rays of caudal fin. 

Caudal base: defined as a point at end of lateral-line row of scales, located 
approximately at center of caudal fin in a vertical direction; this point is used in: 
Length; Caudal peduncle: length; and Caudal fin: length. 

Caudal peduncle: length — measurement from caudal base to posterior tip. 

Head: mid-dorsal length — measurement from tip of snout to occiput. 

Head: length— distance from tip of snout to posterior margin of bony opercle. 

Head: depth— measurement at occiput. 

Head: width — measurement at occiput. 

Snout: length — measurement from tip of snout to anterior rim of bony orbit. 

Snout: least width— measurement immediately behind perforations usually formed 
by the large anterior teeth of lower jaw; these tooth perforations are immediately 
posterior to the nares and should not be confused. 

Snout: width— measurement at posteriormost teeth. 

Lower jaws: width— measurement at articulation with skull. 

Lower jaws: least width— measurement immediately posterior to symphysis of 
dentaries. 

Postorbital distance to: bony margin — measurement from posterior bony rim 
of orbit to posterior margin of bony opercle. 

Postorbital distance to: fleshy margin^distance from posterior rim of orbit to 
posterior margin of fleshy opercular valve. 

Postsnout: distance — measurement from anterior rim of orbit to free margin of 
bony opercle. 

Frontal bones: least width — measurement at central constriction. 

Scales: lateral line — includes the small scales at base of caudal fin, whether pores 
are discernible or not, so long as they are within the lateral-line scale row. 

Scales: predorsal— includes all median dorsal scales anterior to the paired struc- 
tures (fulcra) on the anterior margin and at the origin of dorsal fin. 

Scales: transverse rows — includes all those in an anteriorly slanted diagonal row 
from the single median anal plate to the mid-dorsal scale, including the anal plate and 
mid-dorsal scale. 

Gill rakers: total— number on left outer arch. 



Family LEPISOSTEIDAE 

Garfishes 
Characters. Those of the Order. 

Genera and Subgenera. All modern Garfishes fall within a single genus, Lepisosteus. 
Rafinesque {55: 69, 71, 76) placed the Gar or Gar-like fishes in three genera: SarMrus, 



Fishes of the IVestern North Atlantic 69 

Lepisosteus, and Litho!epis, the first being based on a young Individual of L. osseus and 
the last on inaccurate information communicated personally by J. J. Audubon. The 
sketch and description (long dorsal and anal, bilobed tail) of the fish in Rafinesque's 
field notes and the description in "Ichthyologia Ohiensis" clearly indicate a fictitious 
fish that should not be recognized with a valid name. Rafinesque subdivided the genus 
Lepisosteus into two subgenera, Cylindrosteus and Atractosteus. 

Fowler recognized two genera, Lepisosteus and Cylindrosteus (22 : 604, 605). Holly 
divided the genus Lepisosteus into three subgenera: Litholepis, Lepisosteus, and Cylindros- 
teus (65: 49). More recently, Moore {6y: 54) and Eddy (ig: 39-41) have recognized 
a single genus, Lepisosteus. When Fowler split the Gars into two genera on the basis of 
presence or absence of enlarged teeth on the palatine {22 : 604), he did not find a second 
row of enlarged teeth on the palatine surface in the smallest specimens of L. osseus \ 
however, small examples of the same form in the Tulane University collection reveal 
an inner row of enlarged teeth on the upper jaw. 

Two subgenera, Lepisosteus and Atractosteus, are recognized here. The nomen- 
clature of fossil Gar material has been confused because of chaotic conditions that 
have existed for many years with regard to the recent forms. Perhaps additional sub- 
genera will have to be recognized for some of the fossil forms. 



Genus Lepisosteus Lacepede 1803 

Lepisosteus Lacepede, Hist. Nat. Poiss., 5, 1803: 331; t)-pe species, Lepisosteus gavialis Lacepede 1803 equals 
Esox osseus Linnaeus. 

Generic Synonyms: 

Esox Linnaeus (in part), Syst. Nat., loth ed., 1758: 313; type species, Esox osseus Linnaeus 1758. 

Litholepis Rafinesque, Amer. Mon. Mag., J, 181 8: 447; type species, Litholepis adamantinus Rafinesque 1818; 

based on a fictitious description personally communicated by J. J. Audubon. 
Sarchirus Rafinesque, J. Acad. nat. Sci. Philad., lip.'), 1818:418; t\'pe species, Sarchirus vittatus Rafinesque 1 8 1 8 ; 

young of Lepisosteus osseus. 
Cylindrosteus Rafinesque, Ichthyol. Ohiensis, 1820: 72; t}'pe species, Lepisosteus platostomus Rafinesque 1820. 
Atractosteus Rafinesque, Ichthyol. Ohiensis, 1820: 75; t\-pe species, Lepisosteus ferox Rafinesque 1820. 
Lepidosteus Koenig, Icon. Foss., 1825: t. 12, emended spelling for Lepisosteus Lacepede 1803; Lepidosteus 

Agassiz, Rech. Poiss. Foss., 2 (2), i843:"'j, emended spelling for Lepisosteus Lacepede. 

Doubtful Synonym: 

Psallisostomus Walbaum, P. Artedi Genera Pise, Emend. Ichthyol., 1792, 581; Psallisostomus Fowler (after 

Walbaum), Fish. New Jersey in Rep. N. J. Mus., Pt. 2, 1905: 8g; Suppl. Acct. Fish. N. J., Pt. 3, 1906: 

263, 387, pi. 83 (after Agassiz, Rech. Poiss. Foss., 1843). 

Characters. Teeth numerous, of various sizes on both jaws and on most bones 
lining roof of oral and pharyngeal cavities; one row of enlarged teeth with radially 
grooved base on lower jaws, and two rows on upper jaws, at least during younger 
stages; outer margin of both upper and lower jaws fortified with a closely set row of 
needle-shaped teeth; a pair of large teeth at anterior end of lower jaws fitting into 
depressions in upper jaws and, in many individuals, perforating anteriormost pre- 



7 o Memoir Sears Foundation for Marine Research 

orbital (maxillary auctorum). A peculiar wedge-shaped Blind pocket in skin of gular 
region. Tongue bifurcate-spatulate. Three Modified scales bordering anus; median 
scale, anterior to anus, a large shield-shaped plate; lateral plates elongate and slightly 
curved. Other characters those of the Order. 

Species. Currently seven species are recognized in North American waters, but 
detailed studies that are in progress may show that two of these forms are only 
subspecifically distinct. There is some indication of hybridization, but verification of 
this must await further study. L. oculatus, osseus, platyrhincus, and spatula are treated 
in the following account. 



Key to Species of the Western North Atlantic 

I a. Total gill raker count on left outside arch, 59-81 (70 specimens); palatines of 

adult with a series of enlarged teeth. Subgenus Atraaosteus Rafinesque 1820. 

2a. Lateral-line scales, 58-62 (25 specimens); diagonal rows between insertion 

of pelvic and origin of dorsal fin, 34—38; predorsal scales, 48—54. 

3a. Total gill rakers on left outside arch, 59-66 (20 specimens); females 

maturing at a larger size, probably not less than 600 mm; anterior body 

scales of 400-mm specimens finely serrated on posterior border. 

spatula Lacepede 1803, p. 83. 
3b. Total gill rakers, 67-81 (5 specimens); females maturing at a smaller 
size, 470 mm; anterior body scales of 400-mm specimens strongly ser- 
rated on posterior border, tristoechus (Bloch and Schneider) 1801. 

Fresh waters of southwestern Cuba and Isle of Pines; not 

yet recorded for salt water but likely to be found there. 

2b. Lateral-line scales, 51-56 (60 specimens); diagonal rows between insertion 

of pelvic and origin of dorsal fin, 28-32; predorsal scales, 43-48 (usually 

44—47). tropicus (Gill) 1863. 

Rio Usumacinta of Guatemala and 

Mexico and tributaries of Lake 

Nicaragua and Rio San Juan in 

Costa Rica on the Atlantic slope, and 

from Pacific drainage of Chiapas, 

Mexico (Miller, 48: 230, 231).* 

Atractosteus bocourti Dumeril 1870; 

mouth of (Rio) Nagualate, I4°N. 

lb. Total gill rakers, 14-33 (h8 specimens); palatines of adult without enlarged 

teeth. Subgenus Lepisosteus Lacepede 1803. 

4a. Snout's least width 13-25.5 times in its length (specimens more than 50mm 

long). osseus (Linnaeus) 1758, p. 75. 

4b. Snout's least width 4.5-1 1 times in its length (specimens more than 50mm 

long). 

8. There are no records from salt water on the Atlantic side but L. tropicus is common in the tidal zones of the 
Pacific drainages. 



Fishes of t lie IV e stern North Atlantic 7 1 

5a. Lateral-line scales, 59-65, usually 60-63 (47 specimens); predorsal scales 
50-55, usually 52 or 53; anterior part of body and head without dark 
spots. platostomus Rafinesque 1820. 

Larger rivers of the Mississippi River 
drainage. 
5b. Lateral-line scales, 53-59, usually 54-58 (249 specimens); predorsal 
scales 45—54, usually 47—50, rarely more than 51 ; anterior part of body 
and head with dark spots and blotches. 

6 a. Adults with bony plates on ventral surface of isthmus, under gill 
membrane (177 specimens); snout and lower jaw longer and nar- 
rower, oculatus Winchell 1864, below. 
6 b. Adults without bony plates on ventral surface of isthmus (76 speci- 
mens); snout and lower jaw shorter and broader. 

platyrhincus DeKay 1842, p. 81. 



Lepisosteus oculatus Winchell 1864 

Spotted Gar 

Figures 12, 14 

Study Material.^ A total of 177 specimens, 32-819 mm TL, including the type 
of L. oculatus Winchell 1864 (dry mount with fins broken; length 700 mm, estimated 
caudal length 119 mm, total length 819 mm, UMMZ 55062), from Michigan, In- 
diana, and Ohio south to the Gulf of Mexico, and from western Florida to central 
Texas along the Gulf coast, in TU, UMMZ, ANSP, USNM, AMNH, and UT 
collections. 

Distinctive Characters. L. oculatus is readily distinguished from L. osseus and L. 
spatula by the profusion of dark spots on the body, head, and fins, and from L. platy- 
rhincus by the plates on the ventral surface of the isthmus. The Spotted Gar has a head 
of medium width, approximately intermediate between the narrow head of L. osseus 
and the very broad head of L. spatula. 

Description. Proportional dimensions in per cent of length, based on 32 specimens: 
from Lake Pontchartrain near New Orleans, Louisiana (5), Sabine R., Texas (3), 
Mermentau R., Louisiana (4), Biloxi R., Mississippi (i). Pearl R., Louisiana and 
Mississippi (19); specimens 180— 512 mm length. 

Body, depth 9.2-12.8; width 8.6-12.6. length 27.1-36.2; depth 6.6-8.3; 

Caudal peduncle: length 10.8-13.4; least width 7.6—10.0. 

depth 5.0—7.3. Snout: length 15.9—23.3; least width 

Head: mid-dorsal length 24.5-33.9; 2.3-3.2 ; width 3.8-5.3. 

9. All of the types listed under Study Material were examined by me. 



72 



Memoir Sears Foundatio?t for Marine Research 





Fishes of the Western North Atlantic 



73 



Orbit: length 2.3—3.3. 

I titer orbital: least bony distance 4.8-5.9. 

Postorbital distance to: bony margin 7.8- 

9.8; fleshy margin 8.9—1 i.o. 
Frontal bones: least width 2.0-3.1. 
Lower jaws: least width 1.7-3.0. 
Mandible: length 13.6-20.8; posterior 

width 4.8-6.9. 
Dorsal fin: depressed length i 3.7-1 7.6. 
Anal fin: depressed length 12.0— 18.7. 
Caudal fin: length 13.5-24.3. 
Pectoral fin: length 8.2-12.8. 



Pelvic fin: length 10. 2-1 5.0. 

Distance from tip of snout to: dorsal origin 
83.2—87.6; pelvic insertion 53.4— 
61.2. 

Distance from pelvic insertion to: anal or- 
igin 23.8—29.9; pectoral insertion 
23.5-29.3; lateral line 4.6-6.8. 

Distance from dorsal origin to: occiput 
52.4—60.2; caudal base 14.0— 17.2; 
lateral line 5.2—7.6. 

Distance from anal origin to : caudal base 
16.8-20.6. 



Proportional dimensions in per cent of head length for 31 specimens, 180— 700 mm 
length, with head lengths 65.1 -196.0 mm. 

Snout: length 57.0—65.0; least width Postorbital distance to: bony mzrg'm 26.4.- 

6.5-10.6. 32.2. 

Orbit: length 8.2-1 i.o. Lower jaws: least width 5.0-10.8. 

Interorbital: least bony distance 15.4- Mandible: length 50.0-57.5. 

21.5. 

Proportional dimensions in per cent of snout length (si) and mandible length (ml). 

Snout: least width 5.1-9.9 times in si. Lower jaw: least width 4.8-1 1.6 times 

in ml. 

Statistics of meristic characters: first number — number of specimens; second and 
third numbers— extremes in range of variation; fourth number— mean figure. 



Scales: transverse rows between pelvic 
and dorsal origins 172, 27-32, 
30.3; transverse rows 170, 18—24, 
20.6; predorsal 173, 45-54, 48.2; 
lateral line 177, 53-59, 56.1. 

Gill rakers: total 32, 15-24, 19.3. 



Fin rays: dorsal 23^ 6-9, 7.4; anal 23-> 
7-9, 7.8; caudal 32, 12 and 13, 
12.5; pectoral on left 32, 9-13, 
10.6, and on right 32, 9-13, 10.7; 
pelvic on left 33, all with 6 rays, 
and on right 33, all with 6 rays 
except I with 5. 



Sexual Dimorphism. It is apparent in this species (Hubbs and Lagler, JJ: 76, 77) 
and occurs in at least one proportion. The female proportionally has a longer snout 
than the male, but this difference is not easily demonstrated because the snout-length/ 
head-length ratio changes with size of individual. 

Color. General coloration darker than in L. osseus; intervening spaces between 
numerous dark spots also dark in some specimens, making such individuals appear 



74 Me^noir Sears Foundation for Marine Research 

black; contrast between dark spots and background coloration usually greater in spec- 
imens living in clear water. Seemingly the intensity of pigmentation and the area of 
body pigmented are somewhat correlated with the color of the water, i.e. how darkly 
stained; streams and bayous that course through pinelands and cypress swamps have 
light brown to nearly black water and their fish inhabitants tend to be darkened also, 
even to the extent of a darkened ventral surface. 

The young are brightly colored as in L. osseus (p. 78), but a few differences in 
color and pattern are apparent. Median dorsal stripe very broad and dark brown in 
contrast to narrow reddish or cinnamon stripe in L. osseus. Dark lateral band nearly 
straight on its upper margin, and narrow reddish-brown stripe above usually separated 
from lateral band. Ventral surface usually a chocolate color as in L. osseus. Mid-dorsal 
stripe and lateral band each breaking up into a single row of spots. Mid-dorsal spots 
developing first and forming in a posterior-to-anterior direction. These spots begin 
to form when the fish is between 100 and 150 mm length. The lateral spots appear 
in the same sequence but do not start to form until the individuals have reached an 
approximate length of 170 mm. 

Size. The 26 oculatus collected from the brackish waters of Lake Pontchartrain 
from July 1953 to February 1955 were 297-690 mm TL. On March 25, 1959, 
four oculatus were seined in one haul from a borrow pit in the Bonne Carre Spillway, 
near Norco, Louisiana. The total length for the largest individual, a female, was 
757 mm; for the three males, 522, 524, and 575 mm. This was probably a spawning 
group, but no spawning activity was observed prior to capture. The female was greatly 
distended at the time of capture because of the enlarged ovaries, but no ova were present 
in the oviducts. 

Development. The scales develop in a posterior-to-anterior manner as described in 
Development (p. (^G)., but the Spotted Gar develops scales and other structures at a small 
size. Specimens 100 mm long have a fully developed lateral line as well as lateral body 
scales forward to the region of the pelvic fins at the same time as the pectoral fin rays 
are just starting to develop. Specimens of 1 30 mm have scales on the body except for the 
anterior belly and breast, and the pectoral fins may have half to all of their rays formed. 
The body scalation is complete in specimens 140— 150 mm. Plates begin to form on the 
ventral surface of the isthmus in individuals of about 200 mm. The enlarged teeth in 
the row on the palatines remain as prominent structures until about 200 mm, and some 
retain them until they reach 300 mm; however, most individuals have only relatively 
small teeth at that size. The caudal appendage may be atrophied in individuals as small 
as 160 mm but may not be completely atrophied until a length of 225 mm. 

Habits. The feeding habits of this species are similar to those described for L. 
osseus (p. 78). In the Lake Pontchartrain area, the diet of the Spotted Gar is composed 
of fishes and Crustacea, the crustacean most often eaten being the blue crab, Callinectes 
sapidus. 

Relation to Man. The Spotted Gar is usually considered an obnoxious fish. 

Range. Lepisosteus oculatus ranges from the Great Lakes south in the Mississippi 



Fishes of the IVestern North Atlantic 75 

Valley to the Gulf of Mexico, and from western Florida to central Texas along the 
Gulf coast. It does not Invade brackish or marine waters as much as either the Longnose 
or Alligator Gars. In the Lake Pontchartrain area, L. oculatus is found most often 
along the marshy shore, and seldom very far out in the lake. 

Specimens In the Tulane University collection were obtained from the tidewater 
section of Deer River on the west side of Mobile Bay, Alabama, and from Choctawhat- 
chee River in Florida. Bailey, et al. reported the Spotted Gar for the tidewater sec- 
tion of the Escambia River, Florida (5: 1 17). Gunter reported the capture of two from 
Copano Bay, Texas (27: 23), and Reid gave records of it for East Bay, Texas {60: 
[1955]: 431). All records cited above indicate that, of the four Gar considered, this spe- 
cies and the Florida Gar {L. pJatyrhincus) are the least tolerant of saline conditions. 

Synonyms, with References to Occurrence in Brackish or Salt Water: 

Lepisosteus oculatus Winchell, Proc. Acad. nat. Sci. Philad., l6, 1864: 183 (Huron R., Michigan); Gunter, 

Publ. Inst. mar. Sci. Texas, I (i), 1945 : 23 (Copano Bay, Texas); Bailey, Winn, and Smith, Proc. Acad. 

nat. Sci. Philad., I06, 1954: 117 (in tidewater, Florida); Reid, Texas Sci., 7 (4), 1955: 431 (East Bay, 

Texas). 
Cy/inJrosteus proJuctus Cope, Proc. Acad. nat. Sci. Philad., ly, 1865: 86 (San Antonio, Texas). 
Cylindrosteus agassizii Dumeril, Hist. Nat. Poiss., 2, 1870: 347-348, 351 (St. Louis, Missouri). 
CylinJrosteus bartonii Dumeril, Hist. Nat. Poiss., 2, 1870: 347-348, 356 (New Orleans, Louisiana). 



Lepisosteus osseus (Linnaeus) 1758 
Longnose Gar 
Figures 13, 15 

Study Material. Many specimens. Including 42, 139-779 mm TL, used for the 
Description\ from tributaries of the Gulf of Mexico from Florida to Mexico; from the 
Mississippi River drainage, Louisiana to Indiana; In USNM, CNHM, UMMZ, 
UI, UL, UT, and TU collections; also, the head only of the type of Lepidosteus lepto- 
rhynchus GIrard 1858, USNM 1002. 

Distinctive Characters. Lepisosteus osseus Is distinguishable from L. spatula, L. 
oculatus, and L. platyrhincus at a glance by its long narrow snout. 

Description. Proportional dimensions in per cent of length, based on 42 speci- 
mens, 1 14-695 mm length, as listed under Study Material. Although the variability of 
most dimensions for a limited size range is not very great, the following figures never- 
theless include allometric variation. 

Body: depth 7.0-10; width 5.9-9.5. Snout: length 21.4-28.9; least width 
Caudal peduncle: length 9.4-13.0; least 1.1-2.1; width 2.6-3.7; '^^^7 "^i"" 

depth 3.7—5.1. row. 

Head: mid-dorsal length 29.4-38.8; Orbit: length 2.2-3.7. 

length 31.8— 41.2; depth 5.4—7.2; Interorbital: least bony distance 3.8 — 

width 5.3-7.6. 5.2. 



76 



Memoir Sears Foundation for Marine Research 




Fishes of the Western North Atlantic 



77 



Postorbital distance to: bony margin 6.6— 
8.9; fleshy margin 7.5—10.0. 

Frontal bones: least width 1.7—2.5. 

Lower jaws: width 3.6—5.2; least width 
0.7-1.3. 

Mandible: length 20.4-26.4. 

Dorsal fin: depressed length 10. 2-14. 8. 

Anal fin: depressed length 11.1-15.9. 

Caudal fin: length 12.0-22.3. 

Pectoral fin: length 2.6-10.3. 

Pelvic fin: length 8.1 -13.2. 



Distance from tip of snout to: dorsal origin 
84.8—89.3; pelvic insertion 56.1 — 
62.1. 

Distance from pelvic insertion to : anal or- 
igin 20.4—26.1; pectoral insertion 
18.7—24.4; lateral line 3.3—4.7. 

Distance from dorsal origin to: occiput 
48.2—56.9; caudal base 10. 6-15. 2; 
lateral line 4.2—5.4. 

Distance from anal origin to : caudal base 
15.3-19.2. 



Proportional dimensions in per cent of head length for 39 specimens, 114.1 — 
695 mm length, with head lengths 81.8—272 mm, including head of type (see Study 
Material). 



Snout: length 67.4-73.8; least width 

3-I-4-5- 
Orbit: length 6.3-9.1. 
Interorbital: least bony distance 10.6— 

I3-4- 



Postorbital distance to: bony margin 17.7- 

23.6. 
Lower jaws: least width 1.8-3.6. 
Mandible: length 64.3-68.4. 



Proportional dimensions in per cent of snout length. 
Snout: least width 12.9—25.7. Postsnout: distance 2.1—3.0. 

Statistics of meristic characters: first number — number of specimens; second and 
third numbers — extremes in range of variation ; fourth number— mean figure. 



Scales: transverse rows between pelvic 
and dorsal origins :},^, 3i-35> 2'^-S\ 
transverse rows 37, 19-24, 21.2; 
predorsal 38, 47—55, 50.8; lateral 
line 40, 57-63, 60.1. 

Gill rakers: total 38, 14-31, 22.6. 



Fin rays: dorsal 40, 6-9, 7.4; anal 40, 
8—10, 8.8 ; caudal 40, i 1-14, 12.8 ; 
pectoral on left 39, 10-13, ii-i> 
and on right 39, 10—13, 11 -2; pel- 
vic on left 40 and on right 40, all 
with 6 rays. 



Color. In general, olivaceous brown above and white below, with specimens from 
clear water showing more contrast in coloration; their backs usually more greenish, 
those from turbid waters more brownish. Dark spots on median fins and on body, 
more clearly defined on specimens from clear water; darkened areas on body frequently 
appearing as blackened margins of scales (as if ink were dropped on the body and then 
wiped off, leaving traces in grooves between the scales); old individuals sometimes 
lacking dark spots on body or fins. Catesby has described and figured "acus maxima 
squamosa viridis" from Virginia as having a pink belly and pink fins (jj: 30, pi. 30). 



7 8 Memoi?^ Sears Foundation for Marine Research 

This peculiar coloration may be indicative of a hemorrhagic condition that is easily 
produced by exposure and rough handling before preservation. 

Young individuals, colorful with various shades of brown to black, and white to 
cream. Broad dusky or dark brown band on the side of body, extending from snout 
through eye to base of caudal fin; upper margin of band scalloped, the elevated portions 
fusing into an interrupted stripe of reddish brown or cinnamon. A narrow mid-dorsal 
stripe of similar color. Ventral surface of small young chocolate colored on midbelly 
and bordered above on either side by a milky to cream stripe, the dark coloration dis- 
appearing in larger young and leaving only a narrow ventrolateral stripe. All fins more 
or less spotted or blotched with dark brown, the dark color of fins usually in the form 
of bars on pelvic, dorsal, and anal. 

Size. The maximum size of osseus is about 1,500 mmTL. Abbott gave a length 
of five feet for a specimen speared in Crossweelssen Creek, New Jersey (jJ: 270). The 
28 specimens collected from July 1953 to February 1955 from Lake Pontchartrain 
were 541-1,180 mm TL. 

Spawning and Development. The early life history of L. osseus is quite well known 
(A. Agassiz, l [1878]; Mark, ^5). The more extensive studies were made by Wilder 
{jl\ 155-163; 'J2'. I -1 2, 192, 195, 10 figs.), Balfour (6: chap. 6), Balfour and Par- 
ker (7), and Eycleshymer {21). Agassiz has described the spawning activities and early 
development of the Longnose Gar of Black Lake, New York [i [1879]: 65—75, 
5 pis.), and many illustrations represent the developmental stages of the fins and ad- 
hesive organs. The adhesive organs on the ventral surface at the tip of the snout on 
newly hatched fry are used for attachment to objects above the bottom silt. Although 
larvae are relatively inactive until the yolk supply is absorbed, they are capable of very 
rapid movements when disturbed. At three weeks of age the yolk has been used, and 
the adhesive organs are reduced to a small swelling. 

The following data give an approximation of the size of individuals at the time 
of atrophy of the fleshy caudal appendage projecting from the upturned end of the 
vertebral column. A caudal appendage has been observed at the following lengths: 191, 
219, 233, 245, 260, and 271 mm.; but none was seen on specimens of these lengths: 
266, 268, 271, 274, 284, and 286 mm. Wilder's observation of a 300-mm specimen 
with the caudal appendage nearly atrophied is in line with measurements given above. 
Our largest specimen (271 mm) is actually 323 mmTL when the caudal fin length 
is added. Apparently no one has yet determined the maximum age attained by Gar- 
fish, but numerous specimens have been kept in aquariums for a number of years. 
Mark, who kept L. osseus for nearly 3^/2 years in aquariums at Cambridge, obtained 
the material from Black Lake, New York, when it was in early stages of segmen- 
tation {45: 5). 

Food and Feeding. The Longnose Gar usually waits for its prey to get within "easy 
reach" and then, with a sudden thrust, grasps the victim (Abbott, 31: 269, 270). 
The prey is often held crosswise in the mouth for several minutes before it is swal- 
lowed. In the brackish waters of Lake Pontchartrain, this Gar feeds on various fishes 



Fishes of the Western North Atlantic 79 

and Crustacea; largescale menhaden {Brevoortia patronus), bay anchovy {Anchoa mit- 
chilli), sea catfish (Galeichthys felis), and the blue crab (Callinectes sapidus) were found 
in the stomachs by Dr. Rezneat M. Darnell. 

Relation to Man. In most areas the Longnose Gar is considered not only an obnoxi- 
ous predatory fish but a destructive one as well, gill and trawl nets often being damaged 
by it in the brackish waters of Louisiana, and probably elsewhere. In some parts of the 
country it is used as food, although most fishermen throw it out on the banks, or 
mutilate it before release back into the water. Smith {64: 59) reported (after Earl, 18: 
485) that in 1880 this was one of the principal foodfishes in the New Bern market. 
The roe, being poisonous, is not used in caviar production as is that of the sturgeons 
and the paddlefishes. 

Range. It frequents waters from Quebec to Florida along the Atlantic except for 
the eastern part of the New England states, and in the west it ranges from the Great 
Lakes region south to northern Mexico. 

The adult is frequently found in the brackish waters of Louisiana; Smith has 
stated that the Longnose Gar sometimes enters salt water and is not rare in Albemarle 
Sound {64: 59). In winter it frequents the deeper waters of Lake Pontchartrain, where, 
during a biological survey, several specimens were collected with trawl nets from dredge 
holes 30 feet in depth. Uhler and Lugger reported it as common in brackish water of 
the Potomac and Patapsco rivers (66; 154), but Hildebrand and Schroeder recorded 
it as not common in Chesapeake Bay (52: 77). Joseph and Yerger reported three 
specimens, 1,000— 1,070 mm TL, for Alligator Harbor, Florida (jp: 120). 

The type of Lepidosteus crassus Cope 1865 was collected from brackish water at 
Bombay Hook, near the mouth of the Delaware River. 

Synonyms, with References to Occurrence in Salt Water: 

Esox osseus (in part) Linnaeus, Syst. Nat., loth ed., 1758: 313; after Artedi, based on "Acus maxima squa- 
mosa viridis," Catesby, 1738: 30, pi. 30 (Virginia). 

Esox virdis Gmelin in Linnaeus, Syst. Nat., 13th ed., I (3), 1789: 1389. 

Lepisosteus gavialis Lacepede, Hist. Nat. Poiss., 5, 1803: 333 (lakes, rivers of both Indies). 

Sarchirus vittatus Rafinesque, J. Acad. nat. Sci. Philad., I (2), 1818: 419, pi. 17, fig. 2 (based on juv.). 

Lepisosteus oxyurus Rafinesque, Ichthyol. Ohiensis, 1820: 73 (Ohio R.). 

Lepisosteus longirostris Rafinesque, Ichthyol. Ohiensis, 1820: 74 (Muskingum R., Ohio, descr. based on head). 

Lepisosteus huronensis Richardson, Fauna Boreal. Amer., J, 1836: 237 (Penitanguishene, Lake Huron). 

Lepidosteus rostratus Cuvier in Richardson, Fauna Boreal. Amer., J, 1836: 238 (Penitanguishene, Lake Huron). 

Lepidosteus semiradiatus Agassiz, Rech. Poiss. Foss., 2, 1836: 2, pi. 2. 

Lepidosteus gracilis Agassiz, Rech. Poiss. Foss., 2, 1836: 3. 

Lepidosteus bison DeKay in Zool. N.Y., 1842: 271, pi. 43, fig. 139 (Lake Erie, Buffalo, New York). 

Lepisosteus lineatus Thompson, Hist. Vermont, 1842: 145 (ill., Winooski R., Burlington, Vermont; based on 
juv.). 

Macrognathus loricatus Gronow in Gray, Cat. Fish. Coll. and Descr. by L. T. Gronow, in Brit. Mus., 1854: 
148 (after Linnaeus). 

Lepidosteus leptorhynchus Girard in Pacif. R. R. Surv., Fish., 20(4), 1858: 351 (Devil R., Texas; head only, 
USNM 1002). 

Lepidosteus otarius Cope, Proc. Acad. nat. Sci. Philad., IJ, 1865: 86 (Platte R. near Fort Riley). 

Lepidosteus crassus Cope, Proc. Acad. nat. Sci. Philad., IJ, 1865 : 86 (type from brackish water at Bombay Hook, 
near mouth of Delaware R.). 



8o Memoir Sears Foundation for Marine Research 




Qi 



■v'h'v 



=^ H 



"g 




-4 ,: 



i" 



Fishes of the Western North Atlantic 8 1 

LepiJosleus treculi Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 327 (Mississippi R.). 

Lepidosteus milberti Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 328 (New York). 

Lepidosteus harlani Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 329, Atlas, pi. 21, fig. i, la, i b (Wabash R.). 

Lepidosteus smithi Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 330 (upper Mississippi R.). 

Lepidosteus ayresii Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 331 (Wabash R.). 

Lepidosteus copei Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 332 (n. North America). 

Lepidosteus lesueurii Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 335 (Wabash R.). 

Lepidosteus elisabeth Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 336 (no local.). 

Lepidosteus lamarii Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 337 (n. North America). 

Lepidosteus clintonii Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 338 (no local.). 

Lepidosteus troostii Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 339 (U.S.A.). 

Lepodosteus piquotianus Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 340 (Lake Erie). 

Lepidosteus horatii Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 341 (n. North America). 

Lepidosteus thompsoni Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 342 (upper Mississippi R.). 

Lepidosteus louisianensis Dumeril, Hist. Nat. Poiss., 2, 1870: 323-324, 343, Atlas, pi. 22, fig. 3 (New Orleans, 
Louisiana). 

Lepidosteus osseus Giinther, Cat. Fish. Brit. Mus., 8, 1870: 330. 

Lepidosteus osseus Jordan and Evermann, Bull. U.S. nat. Mus., ./7(i), 1896: 109; Uhler and Lugger, Rep. 
Comm. Fish. Md., 1876: 154 (brackish water, Chesapeake Bay region); Smith, N.C. geol. econ. Surv., 
2, 1907: 59 (brackish and salt water, N. Carolina); Hildebrand and Schroeder, Bull. U.S. Bur. Fisher., 
^J, 1928: 77 (Chesapeake Bay); Joseph and Yerger, Pap. oceanogr. Inst. Fla. St. Univ. Stud., 22, 1956: 
120 (brackish or salt water, Florida). 

Doubtful Synonyms: 

Lepisosteus stenorhynchus Rafinesque, .A.mer. Mon. Mag., j, 18 18: 447; no description (Ohio, Wabash, Green 

rivers). 
Sarchirus argenteus Rafinesque, Ichthyol. Ohiensis, 1820: 86 (Licking R., Slate Creek; brief descr. based on 

communication from Mr. Owings). 



Lepisosteus platyrhincus DeKay 1842 

Florida Gar, Florida Spotted Gar 

Figure 17 

Study Material. A total of 76 specimens, 187-516 mm TL; and type of Cylindro- 
steus megalops Fowler 1910, 369.4 mm length, 419.0 mm TL, ANSP 25371 (In alco- 
hol); from peninsular Florida, In TU, ANSP, and AMNH collections. 

Distinctive Characters. Like L. oculatus, L. platyrhincus differs from L. osseus and 
L. spatula by the presence of numerous dark spots on the anterior part of Its body 
and head, but It Is distinguished from L. oculatus primarily by its lack of plates on the 
ventral surface of the Isthmus, by its wider snout, and by the wider lower jaws. 

Description. Proportional dimensions in per cent of length, based on 30 speci- 
mens, 157.0— 416.0 mm length; from peninsular Florida. 

Body: depth 10.5-14.3; width 9.2-12.0. length 27.4-36.2; depth 7.2-8.5; 

Caudal peduncle: length 9.8-13.0; least width 8.1-9.8. 

depth S-l~l-S- Snout: length 15.8-22.5; least width 

Head: mid-dorsal length 24.5-33.1; 2.6-3.4; width 4.1-5.4. 

6 



82 



Memoir Sears Foundation for Marine Research 



Orbit: length 2.6-3.5. 

Interorbital: least bony distance 5.0—5.9. 

Postorbital distance to: bony margin 8.3- 

10. 1 ; fleshy margin 9.5—1 1.5. 
Frontal bones: least width 2.2—2.7. 
Lower jaws: width 5.3-6.8; least width 

2.3-3-5- 
Mandible: length 1 3.7-1 9.6. 
Dorsal fin: depressed length 1 3.9-1 6.4. 
Anal fin: depressed length 14. 3-16. 5. 
Caudal fin: length 1 5.1 -19.3. 
Pectoral fin: length 8.8-1 1.8. 



Pelvic fin: length 11. 4-13. 8. 

Distance from tip of snout to: dorsal 
origin 81.0-88.5; pelvic insertion 
52.6-61.3. 

Distance from pelvic insertion to : anal ori- 
gin 24.9-32.0; pectoral insertion 
23.5-28.1; lateral line 4.5-7.5. 

Distance from dorsal origin to: occiput 
53.8-61.3; caudal base 14.2— 16.3; 
lateral line 5.4—7.5. 

Distance from anal origin to : caudal base 
16. 0-19. 8. 



Proportional dimensions in per cent of head length for 30 specimens, 157—416 mm 
length, with head lengths 61.5-118.3 mm. 



Snout: length 57.3-62.3; least width 

7.5-12.0. 
Orbit: length 9.0-1 1.2. 
Interorbital: least bony distance 15.5- 

21.0. 



Postorbital distance to: bony margin 28.0- 

32.2. 
Lower jaws: least width 6.3—12.6. 
Mandible: length 49.3-54.8. 



Proportional dimensions in per cent of snout length (si) and mandible length (ml). 

Snout: least width 4.8-8.2 in si. Lower jaws: least width 4.0-8.5 in ml. 

Postsnout: distance 1.3— 1.6 in si. 

Statistics of meristic characters: first number — number of specimens; second and 
third numbers^extremes in range of variation ; fourth number — mean figure. 



Scales: transverse rows between pelvic 
and dorsal origins 76, 30—33, 
31.7; transverse rows 76, 21—25, 
22.3; predorsal 76, 47—51, 48.6; 
lateral line 76, 54—59, <,6.^. 

Gill rakers: 30, 19—33, 24-9- 



Fin rays: dorsal 30, 7 or 8, 7.3; anal 
30, 7 or 8, 7.6; caudal 30, 12 or 
13, 12.7; pectoral on left 30, 
9-1 1, 9.9 and on right 30, 9-1 1, 
9.7; pelvic 30, all with 6 rays on 
both sides. 



Sexual Dimorphism. As in the case of L. oculatus, sexual dimorphism is apparent 
in this species also, i. e. females attain a larger size and proportionally have a longer 
snout than males. In length of snout, the females of platyrhincus overlap the males of 
oculatus. 

Color. General coloration on back darker than in L. oculatus \ coloration and pattern 
on ventral surface highly variable, with specimens taken from the same place and at 
the same time exhibiting variation from immaculate to solid black. Many specimens 



Fishes of the Western North Atlantic 83 

with a pattern of two or four or more dark stripes on ventral surface. Two other color 
phases also, a brown and an orange (Phillips, 55: 331). 

Young, similar in coloration to color given for L. oculatus (pp. 73, 74). 

Size. Kilby reported observations on specimens taken off Bayport, Florida, a mile 
from shore in the Gulf of Mexico {^i: 91); and three specimens, 169, 445, and 495 mm, 
were collected from the coastal marsh near Bayport. Hammett and Hammett reported 
a maximum of 1,330 mm TL in a sample of 225 specimens (jo: 197). 

Development. In this species the development is probably similar to that in L. ocu- 
latus (p. 74). 

Habits. The diet of this species in brackish and marine waters is unknown. 

Relation to Man. Kilby reported that this species is a nuisance to fishermen 
{41: 191). 

Range. L.platyrhincus ranges from the southern tip of peninsular Florida north- 
ward into the lowlands of Georgia. The paucity of records for salt water seems to in- 
dicate little tolerance of this species for saline conditions, but there may be a lack of 
collections from the areas of its occurrence. 

Synonyms, with Reference to Occurrence in Salt Water: 

Lepisosteus platyr/iinats DeKay, Zool. N.Y., I, 1842: 273, p. 43, fig. 137 (Florida); Kilby, Tulane Stud. ZooL, 

2(8), 1955: 151 (salt water, off Florida). 
Cylindrosteus castelnaudi Dumeril, Hist. Nat. Poiss., 2, 1870: 34.7-348, 355, Atlas, pi. 21, figs. 2, 2 a, 2 b 

(Lake Lafayette, Florida). 
Cylindrosteus megalops¥ow\cx, Proc. Acad. nat. Sci. Philad., 62, 1910: 609, pi. 38, figs. 15, 16 (Bayport, Florida). 



Lepisosteus spatula Lacepede 1803 
Alligator Gar 
Figures 16, 18 

Study Material. A total of 29 specimens: 28, 187—2,159 mm TL^" (dry mounts 
or preserved in alcohol), from western Florida along the Gulf coast to Mexico; in- 
cluding the type of Lepidosteus berlandieri Girard 1858, from Matamoros, Tamaulipas, 
Mexico, USNM 1003; i, 49.5 mm length, 57.3 mm TL, from Hildebrandt Bayou 
near Port Acres, Jefferson County, Texas, TU 22288; all in TU, UMMZ, USNM, 
ANSP, AMNH, CNHM, and UT collections. 

Distinctive Characters. Lepisosteus spatula is separable from L. osseus and L. ocu- 
latus by its large size and broad, short snout. The young of spatula are distinguish- 
able from the young of oculatus, osseus, platyrhincus, and platostomus by the light 
dorsal stripe. 

Description. Proportional dimensions in per cent of length, based on 25 speci- 
mens, 156.5— 1,760.0 mm length; also the 49.5-mm specimen [with measurements 
enclosed in brackets]. 

10. The largest specimen is now at Tulane University (TU Osteol. Coll. 360). 



84 



Memoir Sears Foundation for Marine Research 



Body, depth 9.8-14.2 [12.5]; width 

9.1-14.0 [9.0]. 
Caudal peduncle: length 11. 7-14. 2 

[10.9]; least depth 5.8-7.5 [6.2]. 
Head: mid-dorsal length 21.5—32.4 

[31.5]; length 28.4-36.5 [35-7]; 

depth 8.2-10.6 [9.4]; width 9.4- 

12.9 [9.4]. 
Snout: length 15.7-2 1.4 [19.5]; least 

width 3.9-5.1 [5.0]; width 5.3-6.7 

[6.8]. 
Orbit: length 2.5-4.1 [5.6]. 
Interorbital: least bony distance 7.2—9.0 

[7.2]. 
Postorbital distance to: bony margin 9.7- 

ii.i [10.5]; fleshy margin 10.5- 

12.8 [11. i]. 
Frontal bones: least width 3.7—4.4 [4.8]. 
Lower jaws: width 6.1-9.8 [7.8]; least 

width 4.5-5.5 [4.6]. 
Mandible: length 1 3.7-1 8.5 [15.9]. 



Dorsal fin: depressed length 13.4— 18.4 

[15.1]. 
Anal fin: depressed length 1 3.0-1 7.9 

[H.5]- 
Caudal fin: length 13. 6-19.0 [16.7]; 

filament [19.3]. 

Pectoral fin: length 9.1 -13.3 [5.0]. 

Pelvic fin: length ii.i— 15.0 [8.0]. 

Distance from tip of snout to : dorsal ori- 
gin 82.8-86.0 [82.0]; pelvic inser- 
tion 53.5-59-4 [56-7]- 

Distance from pelvic insertion to : anal ori- 
gin 23.2-30.7 [25.2]; pectoral 
insertion 21.0-32.9 [21.4]; lateral 
line 5.3-8.2. 

Distance from dorsal origin to: occiput 
51.9-59.8 [50.9]; caudal base 
14. 5-17. 5 [16.9]; lateral line 5.8- 

7-5- 
Distance from anal origin to : caudal base 
17.0-20.8 [18. i]. 



Proportional dimensions in per cent of head length for 27 specimens, 156.5- 
1,760.0 mm [49.5], including 2,159-mm specimen, TU Osteol. Coll. 360 (see ftn. 10), 
with head lengths 54.6-562.0 mm [17.7]- 



Snout: length 54.1-59.7 [54.7]; least 

width 13. 0-15. 3 [14. i]. 
Orbit: length 5.4-12. i [15.8]. 
Interorbital: bony distance 23.0-30.5 

[20.3]. 



Postorbital distance to: bony margin 28.6- 

36.5 [29.3]. 
Lower jaws: least width 13.5— 19.2 

[12.9]. 
Mandible: length 46.7—52.4 [44.6]. 



Proportional dimensions in per cent of snout length (si) and mandible length (ml). 



Snout: least width 3.7-4.5 [3.9] times 

in si. 
Postsnout: distance 1.2— 1.5 [1.2] in si. 



Lower jaws: least width 2.7—3.8 [3.4] 
times in ml. 



Statistics of meristic characters: first number — number of specimens; second and 
third numbers — extremes in range of variation; fourth number — mean figure. 



Scales: transverse rows between pelvic 
and dorsal origins 24, 34—38, 35.3; 
transverse rows 27, 23—32, 26.3; 



pectoral 27, 49-54, 51.1; lateral 
line 27, 58-62, 60.3. 
Gill rakers: total 20, 59—66, 62.7. 



Fishes of the Western North Atlantic 8 5 

Fin rays: dorsal 22, 7—10, 7.7; anal 22, and 22, 8 — 16, 12.6 on right; pel- 

7—10, 7.8; caudal 22, 12-14, 12.9; vie 22, all with 6 rays except i 

pectoral 22, 11-15, 12.7 on left with 5 on left. 

Color. In general, dark olivaceous brown above and white to yellowish beneath, 
but some nearly black dorsally; this coloration not unusual for specimens kept in aqua- 
riums. Two small specimens (156.5 and 170.5 mm length) dark brown except for light 
flesh-color on belly, breast, and throat regions. Skin between mandibles finely speckled 
with brown; numerous dark spots on sides — only below lateral line anteriorly, but 
above, below, and along lateral line on peduncle posteriorly. Rays of all fins dark brown, 
the dorsal, anal, and caudal having noticeably darker spots. Individuals of about 500 mm 
have few dark spots on sides, and then usually on peduncle; although only a few dark 
spots are present, there are numerous places where the groove between the scales is 
markedly darker than the surface of the scale; a few dark spots on dorsal, anal, and 
caudal fins; some large individuals from Lake Pontchartrain devoid of spots on fins as 
well as on body. 

Color of 49.5-mm specimen. Light median dorsal stripe extending from tip of 
snout to origin of dorsal, and from posterior insertion of dorsal to upper base of caudal 
fin; this light stripe bordered on either side by a broad dark brown area extending about 
a third of the way down the sides and reaching to dorsal rim of orbit. Dark lateral band 
from anterior tip of mandible to anterior rim of orbit, thence posteriorly from posterior 
rim of orbit to base of caudal. Dark lateral band extending posteriorly along mandible 
from its anterior tip to anterior rim of orbit, and from posterior orbit to base of caudal; 
this dark band on body composed of close-set mottlings, forming irregular borders 
dorsally and ventrally. Venter light except for dark strip along inside of rami of lower 
jaws, thus leaving a light midventral band. All fins with dark brown blotches. 

Size. The smallest Alligator Gar seen is the 49.5-mm specimen described above. 
The largest one on record, a female taken from Belle Island Lake, Vermilion Parish, 
Louisiana, was 9 feet 8.5 inches long and weighed 302 pounds; statistics on other large 
individuals are given by Gudger (26: 1 18-120), Weed {6g: 5, 6), and Hussakof (J4'- 
2). During a recent biological survey of Lake Pontchartrain and adjoining brackish 
waters, 21 specimens, 410—1,472 mmTL," were collected. Additional specimens were 
taken from the Gulf of Mexico at Grand Isle, Jefferson Parish, Louisiana. 

Spawning and Development. Practically nothing is known of the spawning and 
development of the Alligator Gar, but according to the unpublished writings of George 
Powers Dunbar (Wortman, y^: 385), it spawns during December and January. The 
egg is enveloped in gelatinous material, and the egg strands are draped on snags 
and vegetation; by the end of August the young fish has reached 14 inches in length. 
These notes by Dunbar need verification. [Recent observations prove the foregoing to 
be incorrect. The Alligator Gar spawns in April, May, and June in the Louisiana 

II. The remains of several taken from the Gulf side of Breton Island as well as the head of a specimen {ca. 6 ft.) taken 
from the Gulf of Mexico at Destin, Florida, are in the Tulane University osteology collection. 



86 Memoir Sea?^s Foundation for Marine Research 

area. The description of the eggs probably was of some amphibian, as suggested by 
Dr. Edward C. Raney (personal communication)]. 

The 49.5-mm specimen (see Description and Color above) has neither scales nor 
lateral line developed, the pelvics and pectorals are relatively much shorter than in 
older fish, and in the head region, the lower jaws, mandible, and interorbital dimen- 
sion are relatively shorter while the orbit is much larger, as is common in the young 
of many species. In coloration it is distinguished by the light median stripe from the 
snout to the upper caudal except for its interruption by the dorsal. 

Specimens 156.5 and 170.5 mm length have the caudal appendage but lack scales 
in the mid-dorsal region and on the midventral surface except for a partially developed 
anal plate. The rays of the pectoral as well as those of the other fins are fully developed. 

Food and Feeding. L. spatula is credited with eating large numbers of game fishes 
in fresh water (Gudger, 26: 120), but little has been published about its diet in brackish 
and salt water. Raney reported that it ate ducks and water turkey (Anhinga anhingd) 
in Cuartez, Resaca, Texas {^y : 50), and Gunter found mullet (Mugil) in the gut tract 
of 12 out of 24 specimens (27: 24). No other food items were mentioned by these 
authors. Those examined during the Lake Pontchartrain studies contained striped 
mullet {Mugil cephalus) and blue crabs {Callinectes sapidus). 

Jordan (55: 313), Weed {6g: 9), and Gunter {2y. 24) have given evidence that 
the Alligator Gar is a scavenger. 

Relation to Man. It is sold at present for human consumption in the French Market 
in New Orleans, and its scales were used by the Indians (Gowanloch, 24.: 389-392) 
and by the late Mr. Percy Viosca in his jewelry industry. Many unauthenticated ac- 
counts of Gar attacks on humans have appeared in both popular and semipopular liter- 
ature, but swimmers probably need have very little fear of them {24: 389-392; 26: 
120, 121). With the recent development of underwater spear-fishing, all of the Gars 
are desirable targets. Gar fishing rodeos are common annual events in Louisiana. 

Range. The Alligator Gar is a frequent invader of brackish and marine waters, and of 
the four species it is the most tolerant to higher salinities. It occurs in fresh water in the 
Mississippi River and lower parts of its major tributaries from the Ohio and Missouri 
rivers southward to the Gulf of Mexico, and in brackish and salt waters along the Gulf 
coast from Choctawhatchee Bay, Florida, to northern Mexico (26: 118 — 120; 6g: 5, 
6; J^: 2). Gunter recorded Alligator Gars from Copano Bay, Aransas Bay, and from 
a Gulf beach, Texas {2']: 24), and Reid reported the capture of one from East Bay, 
Texas {60 [1956]: 302). Bailey, et al. gave catch records for the tidewater section of 
Escambia River, Florida (5: 1 17), and specimens frequently on display at the Gulfarium 
at Ft. Walton, Florida, are captured in Choctawhatchee Bay. 

Synonyms, with References to Occurrence in Salt Water: 

Lepisosteus spatula Lacepede, Hist. Nat. Poiss., 5, 1803: 333 (no local.); Bailey, Winn, and Smith, Proc. Acad, 
nat. Sci. Philad., 106, 1954: 1 17 (tidewater, Florida); Reid, Texas J. Sci., 1956: 302 (salt water, Texas). 
Lepisosteus ferox Rafinesque, Ichthyol. Ohiensis, 1820: 73 (Ohio River). 

Lepidosteus berlamlieri Girard in Pacif. R. R. Surv., Fish., 10 (4), 1858: 353 (Tamaulipas, Mexico). 
Atractosteus lucius Dumeril, Hist. Nat. Poiss., 2, 1870: 360, 364 (Tampico, Mexico). 



TEXT AND FOOTNOTE REFERENCES 



1. Agassiz (A.), Proc. Amer. Acad. Arts Sci., 

13, \%1%M4, 1879. 

2. Agassiz (L.), Poiss. Foss., 1833-1843. 

J. Agassiz (L.), Amer. J. Sci., (2)26, 1853. 

4. Agassiz (L.), Amer. J. Sci., (2) 77, 1854. 

5. Jdailey, Winn, and Smitii, Proc. Acad. nat. 

Sci. Philad., 106, 1954. 

6. Balfour, Ganoidei (London), 1881. 

7. Balfour and Parker, Philos. Trans., ly^, 1883. 

8. Berg, Class. Fish., Trav. Inst. zool. Acad. Sci. 

URSS., 5 (2), 1940; also reprinting by 
J.W.Edwards, Ann Arbor, Mich., 1947 
(Russ. and Eng.). 

9. Blacker, Ned. Tijdschr. Dierk. Amst., 4, 1873. 

10. Bohlk, et al., Handb. Vergl. Anat. Wirbelt., 

4, 1936. 

11. Bonham, Trans, .^mer. Fish. Soc. (1940), 

1941. 

12. Bridge, Cambr. Nat. Hist, i, 1904. 

2J. Cjatesby, Nat. Hist. Carolina, 1754. 

14. CoUinge, Quart. J. micr. Sci., N.S. 144 (36), 

1894. 
75. CoUinge, Proc. Bgham. nat. Hist. Soc, 8, 1893. 
16. Cope, Trans. ."Imer. philos. Soc, 14, 1871. 

77. DuMERiL, Hist. Nat. Poiss., 2, 1870. 

18. Earl in Goode, et al.. Fish. Fisher. Industr. 

U.S., 2, 1887. 
jp. Eddy, How to Know Freshw. Fish. (Iowa), 

1957- 



20. Emelianov, Zool. Jb., (Anat.) 60, 1935. 

21. Eycleshymer, Publ. Univ. Chicago, (Biol.) JO, 

1903. 

22. PowLER, Proc ."^cad. nat. Sci. Philad., 62, 

1910. 

2J. Cjoodrich, Proc. zool. Soc Lond., 1907. 

24. Gowanloch, Bull. La. Cons. Dep., 2J, 1933. 

2J. Grasse, et al.. Trait. Zool., JJ, 1958. 

26. Gudger, Nat. Hist., N.Y., 4g, 1942. 

27. Gunter, Publ. Inst. mar. Sci. Texas, j(i), 

1945. 

28. Gunther, Cat. Fish. Brit. Mus., 8, 1870. 

29. JTlAMMARBERc, ."^cta ZooL, l8, 1 93 7. 

JO. Hammett and Hammett, Growth, J (2), 1939. 
J7. Hardwicke, Sci. Gossip, 1873. 
J2. Hildebrand and Schrocder, Bull. U.S. Bur. 
Fish., 42, 1928. 
Hubbs and Lagler, Invest. Ind. Lakes. . . 

Streams, 1943. 
Hussakof, Copeia, 1914. 



33 



34- 

35- 
36. 
37- 

38. 
39- 



Jordan, Guide to Study Fish., 1905. 
Jordan, Stanf. Univ. Publ. Biol., J (2), 1923. 
Jordan and Evermann, Bull. U.S. nat. Mus., 

4y{l), 1896. 
Jordan, Evermann, and Clark, Rep. U.S. 

Comm. Fish. (1928), 2, 1930. 
Joseph and Yerger, Pap. oceanogr. Inst. Fla. 

St. Univ. Stud., 22, 1956. 



40. IS-ERR, Proc. zool. Soc. Lond., 1952. 

41. Kilby, Tulane Stud. Zool., 2 (8), 1955. 



87 



Memoir Sears Foundation for Marine Research 



42. J_/ANDACRE and Conger, J. comp. Neurol., 2J, 

I9I3- 
4J. Lankester, Treat. Zool., 9, 1909. 
44. Deleted. 



45. JVIark, Bull. Mus. comp. Zool. Harv., ig, 

1890. 

46. Matsubara, Fish. Morph. Hierarch., Pt. i, 

1955 (Jap-)- 
4y. Mayhew, J. Morph., 38, 1924. 
48. Miller, Copeia, j, 1954. 
4g. Muller, Abh. Akad. Wiss. Berl., Phys.-Math. 

Kl., 1844. 



50. IN icKERsoN, Bull. Mus. comp. Zool. Harv., 24, 

1893. 
5-r. Norman, Synop. Orders, Families, Genera of 

Recent Fish, (mimeo), 1947. 



52. X^FEIFFER, J. Morph., 54, 1933. 

5J. Phillips, Copeia, 1958. 

54. Potter, J. exp. Zool., 4g (i), 1927. 

^^. jKafinesque, Ichthyol. Ohiensis, 1820. 

56. Rafinesque, J. Acad. nat. Sci. Philad., I (2), 



Raney, Copeia, 1942. 

Regan, Proc. zool. Soc. Lond., 1923. 

Regan, Encyc. Britt., ed. 4, 9, 1937. 

Reid, Texas. J. Sci., 7 (4), 1955; 5 (3), 1956. 

Reissner, Arch. Anat. Phys., 1859. 

Romer, Vert. Paleont., ed. 2, 1945. 



63. bcupiN, Arch. Naturgesch., 62 (i), 1896. 

64. Smith, N. C. geol. econ. Surv., 2, 1907. 

65. Das Tierreich (Berlin and Leipzig), 1936. 



66. Uhler and Lugger, Rep. Comm. Fish. Md., 
1876. 



6y. Vertebrates of U.S., 1957. 



Wagner, Zool. Anz., J9, 1912. 
Weed, Leafl. Field. Mus. nat. Hist., 5, 1923. 
Wiedersheim, Zool. Jb., Suppl., 7, 1904. 
Wilder, Araer. Ass. Adv. Sci., 24 B, 1876. 
Wilder, Pop. Sci. Mon., 11, 1877. 
Williamson, Philos. Trans., i^g (2), 1849. 
Williamson, Philos. Trans., 141(2), 1851. 
Wortman, Amer. Nat., 16, 1882. 
Wright, J. Anat., London, ig, 1885. 



Order Isospondyli' 

COMPOSITE AUTHORSHIP 



Characters and Keys to Suborders 
and Families 

HENRY B. BIGELOW 

Museum of Comparative Zoology 
Harvard University 



Acknowledgments. Special thanks are due the following persons for their invaluable 
help and cooperation in preparing the text and keys of this section : William W. 
Anderson^ fames E. Bohlke, Daniel M. Cohen, f. R. Dymond, Robert H. Gibbs, 
fr., W . A. Gosline, Marion Gray, N. B. Marshall, Giles W. Mead, fames E. 
Morrow, fr., George S. Myers, Albert E. Parr, and Robert R. Rofen. 



Characters of Living Isospondyli. Both premaxillary and maxillary bones are present 
in most, but in a few these bones are greatly reduced or lacking (among Argentinoidea, 
Part 4). In most the maxillaries form a part of the border of the mouth (see also 
pp. II, 12, 17, 92, 95). The premaxillaries are only slightly protractile, if at all so, 
except in the Phractolaemidae,^ some Alepocephalidae, and some Stomiatoidea (see 
Bertin and Arambourg, j^: 2247), where they are protractile. The bone in the 
ethmoid position above the vomer is unpaired in most but paired in a few (pp. 95, 96). 

There are four gill clefts, and in a few there is an open slit between the mandible 
and the hyoid arch (Malacosteidae). 

1. With accounts of the included Suborders, Families, Genera, and Species by various authors as listed in the Table 
of Contents. In this series of volumes, the Iniomi, treated as a Suborder of the Isospondyli by some authors, is 
given the rank of Order and is dealt with in Memoir I, Part 5. 

2. A family of the Chanoidea, or so-called milkfishes, of the tropical Pacific. 

89 



go Me^noir Sears Foundation for Marine Research 

Typically the fins have no supporting spines,^ most of the rays being branched; 
however, in a few the rays are unbranched at the origin. Typically the pectorals* are 
posterior to the gill openings, but in the genus Asquamiceps (Alepocephalidae) they are 
well inside the gill cover; in the more primitive genera their line of insertion is low 
down on the side near the ventral profile, but in some of the alepocephalids and argen- 
tinoids they are inserted higher up on the sides; the pectoral bases are far below the 
upper end of the gill openings except in some of the bathylagids, where the upper end 
of the gill opening is greatly restricted. The pelvic fins," if present, are posterior to the 
pectorals, their positions ranging from abdominal to thoracic. In most there is only one 
dorsal fin with jointed bony or membranous rays (lacking in a few); in some there is 
a fleshy "adipose" dorsal fin with unjointed horny rays between the rayed dorsal and 
caudal fins; also, in a few (some Astronesthidae) there is a second adipose fin close in 
front of the anus. The rayed dorsal, or the adipose dorsal if present, is separated from 
the caudal by a definite gap, and in most the anal fin is similarly separated from the 
caudal.* The caudal fin is about the same in width both above and below the projected 
longitudinal axis of the trunk,' and the rear boundary of its fleshy base is symmetrical 
relative to the axis of the trunk; in most, the rear outline of the caudal is more or less 
deeply forked, or at least concave, but in some it is more nearly truncate transversely, 
and in a few it is either rounded or has two or three of the middle rays extending almost 
equally beyond the others (genus Pantodon). 

The radialia of the pectoral fins are articulated basally with the pectoral girdle. 
In the great majority the pectoral girdle has a mesocoracoid element firmly attached 
above to the auditory (otic) region of the skull (see also pp. 2, 6). In the pelvic fins 
the basal radialia are reduced to a few small nodules, and the pelvics are not attached 
to the pectoral girdle. The number of skeletal supports equals the number of rays in 
the dorsal and anal fins. 

The head, though naked in the great majority, is clothed with smooth scales among 
the Esocoidea and in the genus Lepogenys (Alepocephalidae) and with prickly scales 
in Gonorhynchus \ in no living representative of the Order* is the head armored with bony 
plates. And in no living family are the margins of the caudal fin armed with more than 
one enlarged "fulcral" scale, such as is characteristic of the caudal of the sturgeons" 

3. Among the Idiacanthidae the base of each dorsal and anal ray bears on its anterior side a pair of short sharp 
spurs that project through the skin (for details, see Beebe, j: 152, 213). Among the Sternoptychidae, the rayed dorsal 
is preceded by the projecting spine-like tips of approximately 1-7 of the neural processes of an equal number of 
vertebrae (for an early account, see Cuvier and Valenciennes, 75: 395, 402, 419). 

4. The stomiatoid genera Idiacantkus, Pkotostomias, and Tactostoma (described by Bolin, 6: 39), are said to have no 
pectorals; likewise some specimens of Eustomias and Pkotonectes. 

5. Among many families, such as the salmonids, specimens occasionally lack one or both pelvics, but this is an indi- 
vidual anomaly (see Myers, 29: 600-601; 2S: 41). 

6. Not in the freshwater Notopteridae of Africa and the East Indies, or in the genus Coilia (Engraulidae) of the 
tropical warm-temperate Indo-Pacific. 

7. Most of the fin rays, however, are "attached to the lower spines of the hinder vertebrae, which are greatly 
enlarged, and at the same time inclined backwards so as to be more or less parallel to the axis of the body" 
(Norman, 31: 61). 

S. In the fossil Leptolepidae the dermal bones of the skull were covered with an enamel-like substance. 

9. These fulcral scales were "present, but usually small" among the fossil Pholidophoridae (Boulenger, 7: 545)- 



Fishes of the Western North Atlantic 9 1 

and gars (Fig. 3). The body scales, lacking in only a few,'" are thin and overlap each 
other in most instances, but in some they are thick and bony; the exposed surface 
is smooth or only finely striate except in a few where it is either granular and sculptured 
with a network of low ridges" or has fine'- or coarse'^ prickles; in most the exposed 
edges are even, or nearly so, but in a few'* they are pectinate; the exposed portion of 
the scale is not enamel-like in any living family.'^ 

A lateral line is either present or absent, as are luminescent organs. 

The swim bladder, if present, is dorsad in position'" and is either closed or con- 
nected to the anterior part of the alimentary tract by an open duct; it is not con- 
nected to the inner ear by a chain of bonelets (Weberian ossicles). The oviducal 
tracts along which the ova pass to the exterior consist either of closed tubes (oviducts 
complete) or of membranous channels that are open above but are enclosed by lon- 
gitudinal folds of the peritoneum (oviducts incomplete).^' The inner wall of the in- 
testine in a few has a series of circular ridge-like thickenings which suggest the spiral 
valve of elasmobranchs, chimaeroids, and polypteroids; however, they are probably not 
homologous (Cohen, 10: 96). 

The last few vertebrae are turned upward in a few (Salmonidae). If present, the 
lateral processes (parapophyses) that bear the pleural ribs are simply set in pits in 
the vertebral centra in some but are fused with the centra in others. 

Nomenclature. The names Isospondyli'^ and Malacopterygii" have been used inter- 
changeably as ordinal names by various authors for nearly a century. More recently 
the name Clupeiformes, proposed for this Order by Berg (^: 216, 417), has been ac- 
cepted by Bertin and Arambourg, but in this series of publications the name Isospondyli 
is preferred, following Cope (jj: 454, 455). 

10. Among the Alepocephalidae, Stomiatoidea, and Galaxiidae. 

ir. Among freshwater osteoglossids of South America, Malaya, and Australia. 

12. In at least two species of Argentina (see Smitt, ^o\ fig. 229). 

13. In the genus Gonorhynchus of the warm-temperate Indian Ocean and western Pacific. For an excellent illustra- 
tion, see Cuvier and Valenciennes (jj: pi. 568). 

14. In the genus Bre-voortia, p. 342. 

15. The exposed surface of the scales was covered with an enamel-like substance (ganoine) in the fossil family Lep- 
tolepidae, which was abundantly represented from the Upper Triassic to the Middle Cretaceous, and in the 
Pholidophoridae of the Jurassic. 

16. The ventral "sole" of Opisihoproctus, interpreted by Trewavas as a swim bladder (5J: 610), appears to be part 
of a light organ (Bertelsen, 5: 862). 

17. The original idea (Rathke, J7) "that the salmonoids have no oviducts and that the ova are deposited free in 
the abdominal cavity has been handed down to the present day in all literature pertaining to the subject" 
(Kendall, 24: 190). Similarly, according to Cuvier, the ovaries in Argentina discharge the eggs into the abdom- 
inal cavity "comme dans les autres Salmonoides" [14: 410). Trewavas could trace no oviduct in a specimen of 
Opisthoproctus cleared in glycerine and caustic potash before dissection (5J: 610, 611), nor could Beebe do so 
in a mature female of Dolichopteryx [2: 78). It has been reported similarly that the ova are discharged into the body 
cavity in the Galaxiidae, the Haplochitontidae, the Hyodontidae, and the Notopteridae. However, Kendall 
showed from his own dissections that the actual situation, in both the Salmonidae and the Osmeridae, is as 
summarized in the text above, and that the ova cannot be extruded if they be displaced in the abdominal cavity. 
Probably this is equally true of the other bony fishes that have been reported as lacking oviducts. 

18. Proposed in 1871 by Cope (zz: 454-455). 

19. First proposed in 1738 by Artedi (/: i), it was given its first post-Linnaean definition in 1893 by Gill (77: 130- 
131), as was Cuvier 's "Malacopterygiens abdominaux" {12: 159), which also included the Ostariophysi. Mala- 
copterygii was adopted by Boulenger in 1904 (7: 543). 



9 2 Memoir Sears Foundation for Marine Research 

The Order Isospondyli, as defined under Characters of Living Isospondyli, cor- 
responds to: the Order Isospondyli of Cope;^'' the Suborder Isospondyli in part of 
Woodward (54: xxxvi; 55: xvii); the Suborder Malacopterygii plus the Suborder 
Haplomi in part of Boulenger (7); the Suborders Clupeiformes and Esociformes of 
Goodrich (25: 386, 397); the Orders Isospondyli (excluding the Mormyriformes) and 
Haplomi of Regan (j5: 77-78); the Orders Clupeiformes plus Galaxiiformes of Berg 
{4: 216,254,256,417,436,437); the Order Isospondyli of Romer (45: 581,584) 
plus the Suborder Esocoidea, which he referred to the Order Mesichthyes of Hay 
{20: 397); and the Order Clupeiformes of Bertin and Arambourg (18: 210). 

The members of the Order Isospondyli include some of the most generalized of 
living bony fishes, and the fossil record for some of them reaches as far back in geo- 
logic time as the Lower Cretaceous (p. 15). The Order is accepted here as a matter 
of convenience, since the interrelationships of its included units still remain obscure. 

Suborders. The heterogeneous assemblage that is grouped together in Part 5 as the 
Order Iniomi has been treated as a Suborder of the Isospondyli by some authors 
(Schultz and Stern, 47: 233; Bertin and Arambourg, 18: 2269) but as a separate Order 
by Regan {38: 77-78; ^j: 314), Jordan {22: 153), Norman (jo: 317), Berg {4: 242, 
256, 429, 437 — as Scopeliformes), and Marshall (26: 305-336). In accord with pres- 
ent-day tendencies based on evolutionary grounds, discussed below, the Iniomi are 
here classed as a separate Order. 

In 1929 Regan {41 : 313, 314) set the Iniomi apart from the Isospondyli as a 
separate Order on the following grounds: 

I a. Maxillaries (typically) forming a part of upper border of mouth; pectoral girdle 
with mesocoracoid element in the great majority. Isospondyli. 

I b. Maxillaries not forming a part of upper border of mouth; pectoral girdle without 
mesocoracoid element. Iniomi. 

Unfortunately, the separation between Iniomi and Isospondyli is not as clear-cut 
as the foregoing implies. Thus the mesocoracoid is lacking: in some of the Argenti- 
noidea (5J: 609, 612); in the Salangidae (Salmonoidea) ; in the Retropinnatidae, the 
systematic relationships of which remain uncertain; in the Esocoidea (Part 4), the 
Bathylaconoidea (55: 52), and the Aplochitonidae {40: 290), which have sometimes 
been grouped with the Salmonoidea but which seem to deserve the rank of a separate 
Suborder. Furthermore, the Iniomi share the withdrawal of the maxillary bone from 
the upper border of the mouth with some fishes that are isospondylous in other respects: 
i. e. the genus Albula (Albulidae), the Pterothrissidae (here rated as a separate family), 
Nematalosa (44: 465, fig. 127, as "C/iae!oessus"), and the genus Chanos. 

Such members of the Order Isospondyli as are known to occur in the western 
North Atlantic are distributed in Parts 3 and 4 among the Suborders Elopoidea, Clu- 

20. The systematic position of Cope's {ii: 454) Scyphophori (freshwater families Mormyridae — distinguished espe- 
cially by the enormous cerebellum — and Gymnarchidae) remains controversial. 



Fishes of the Western North Atlantic 93 

peoidea, Stomiatoidea (including its subdivisions Gymnophotodermi, Heterophoto- 
dermi, and Lepidophotodermi, named by Parr [jj: 15—17; 24'- 136]), Salmonoidea, 
Argentinoidea,-! Esocoidea, and Bathylaconoidea. 

These Suborders correspond to: the Suborders Clupeoidei, Salmonoidei, Stomia- 
toidei, Opisthoproctoidei, and Esocoidei of Berg (^); the Suborders Clupeoidei, Sal- 
moniformes, Opisthoproctoidea, Stomiatoidea, Gymnophotodermoidea, and Esoci- 
formes of Schultz and Stern (^7); the Suborders Clupeoidea, Stomiatoidea, and Salmo- 
noides, plus the Order Haplomi, of Norman (jj); the Suborders Elopoidei, Clupeoi- 
dei, Salmonoidei, Opisthoproctoidei, Alepocephaloidei, Stomiatoidei, Bathylaconoidei, 
and Esocoidei of Bertin and Arambourg {18: 221 1); and the Suborders Elopina, 
Albulina, Esocina, Clupeina, Chanina, Gonorhynchina, Salmonina, Opisthoproctina, 
and Stomiatina of Matsubara (27: 178-227). 

One major area of uncertainty as to the relationships among the groups of iso- 
spondylous fishes centers around the position of the Elopidae, Albulidae, and Ptero- 
thrissidae. Indeed, within recent years it has been questioned whether Elops is even of 
the same lineage as other living teleosts. Thus Saint Seine has derived it from the 
Halecostomi {46: 297), a group known otherwise from fossils only {18: 2195— 2201), 
and Nybelin has emphasized the holostean nature of certain characteristics of Elops,-''' 
embodying a similar point of view. 

Both the Elopidae and the Albulidae, to be sure, have characteristics that are con- 
sidered archaic by both palaeontologists and neozoologists alike; among these are: the 
retention of the gular plate and of the numerous branchiostegal rays in Elops and Tar- 
pon\ the arrangement of the mucous canals on the top of the snout in Elops (j2: 454, 
fig. i); the presence of close-set teeth on the parasphenoid bone in the roof of the 
mouth (Elopidae, Albulidae, and Pterothrissidae); the presence of roofed post-temporal 
fossae; and the persistence in Albula, Pterothrissus, and Megalops (but not in Elops) of 
two rows of valves in the arterial cone of the heart.^^ 

On the other hand, the elopids and the albulids differ widely and conspicuously 
from living members of Lepisosteus and Amia in the narrow rod-like nature of their bran- 
chiostegal rays and in the nature of the skeleton of their caudal fin ; this latter difference 
is a conspicuous one, externally, for while the rear boundary of the fleshy base of the 
caudal is symmetrical relative to the main longitudinal axis of the trunk in the elopids 
and albulids, it slopes upward-rearward in Lepisosteus (Fig. 3, p. 16) and Amia. Further 
features worth mention are: (i) the invariable presence in the elopoids of a median 
supraoccipital bone, which, to a greater or lesser degree, is involved in the roof-complex 
of the skull; (2) the absence of coronoid teeth in the mandible; (3) a one-to-one rela- 
tionship between the principal caudal rays and hypurals of Amia and Lepisosteus com- 
pared with the considerable consolidation of the hypural elements in the elopids; (4) 
the absence in elopids of the rigid attachment of the premaxillaries to the skull, which 

21. The reason for using this name rather than Opisthoproctoidea is given in Suborder Argentinoidea, Part 4. 

22. "Les Elopidae m'apparaissent etre des Holosteens aussi bien <\\xAmia ou Lepidosteus" (j2: 458). 

23. See Senior for account of the cone for Tarpon {48: 146-151), for Elops {48: 150), and for a summary {4g: 83-84). 



94 Memoir Sears Foundation for Marine Research 

characterizes the living holosteans; and (5) a well-developed articulation with the 
autopalatines in elopids. Indeed, in the development of the whole upper jaw mechanism, 
the elopoids are considerably in advance of the holosteans. 

However, though we accept the traditional view that the elopids and albulids 
(with Pterothrissus) do belong among the Isospondyli, there is much uncertainty as to 
their disposition within that Order. For example, Jordan and Evermann (23: 407), 
followed by Norman (jJ: 63), rated them as families within the Suborder Clupeoidea; 
Berg (4: 221, 222; 419,420) defined them as superfamilies (Elopoidae and Albuloi- 
dae), each to include two families; Fowler {16: 151) and Poll (j6: 8) rated them as 
families of Isospondyli (Clupeiformes of Poll) without the intervention of Suborders. 
However, as early as 1893, Gill {ly : 127—128) made them the basis of a separate Sub- 
order, Elopoidea, in which he was followed by Bertin and Arambourg {18: 221 1), 
who stressed in particular the primitive nature of their caudal fin skeleton. 2* Jordan 
(22: 117— 118), proceeding a step further, proposed the Suborder Elopoidea for the 
elopids and the Suborder Albuloidei for the albulids as contrasted with the clupeoids 
and with other Suborders of Isospondyli; Matsubara^^ has also accepted this arrange- 
ment. 

Exclusion of the elopids and albulids from the clupeoids clearly seems demanded, 
for among living fishes they appear to stand much lower on the evolutionary tree than 
any other teleosts that are at all well known. But it still remains an open question whether 
the features in which the elopids and albulids agree, balanced against those in which 
they differ, are better represented by placing them in two separate Suborders or by 
uniting them in one, as is done here. 

The isospondylids as here defined include the ladyfishes or bigeye herrings, the 
tarpons, the bonefishes, the true herrings, the anchovies, and the salmons, all of which 
have been familiar for generations to seaside dwellers in general in one part of the world 
or another. They also include such of the pikes as enter brackish or salt water, and 
an assemblage of oceanic fishes, many of which have luminescent organs and are 
bizarre in appearance; the latter are seldom seen by ordinary seafarers, commercial 
fishermen, or anglers, and little is known about their mode of life. 

Key to Suborders. Although the general characteristics are now tolerably well 
known for the elopoid, herring-like, salmon-like, and pike-like fishes, we still have so 
much to learn about the internal anatomy of the less familiar groups of Isospondyli 
that their natural affinities remain obscure in many respects. Areas of continuing un- 
certainty are, for example: the relative weights to be accorded one character or 
another from the evolutionary standpoint; the breadth of coverage to be allotted to 
the herring-like fishes and the salmon-like fishes; and the disposition to be made of 
various outlying families such as the Bathylaconidae (Part 4), which do not readily fit 
in any of the major subdivisions that are recognized generally. Furthermore, Suborders 

24. For accounts of the caudal fin skeletons of E lops, Albula, and Megalops, see Regan {jg: 355, 356, fig. i); see Hollister 
{21: 260-276) for Elops and Albula. 

25. Elopina and Albulina, Suborders of Order Clupeida (equivalent to Isospondyli) (57: 180-1S2). 



Fishes of the Western North Atlantic 9 5 

that are easily separable typically may intergrade with one another marginally. For 
instance, the presence of parietal bones among the Clupeoidea and not among the 
Stomiatoidea has been invoked as an alternative between these two Suborders, but 
this is not strictly so, since some stomiatoids do have parietals (see Regan and Tre- 
wavas, 42'- ^gs. 6, 9, 10; Giinther and Deckert, ig: 244). Neither is the arrangement 
of the photophores on the sides in definite longitudinal rows reliably diagnostic for the 
Stomiatoidea, since they are not so in one species of Nconesthes, while in one other 
species the linear arrangement is to be recognized only with difficulty, ^^ as is true of 
some species of Astronesthes as well.-' And Mrs. Marion Grey contributes the infor- 
mation that some species of Cyclothone, Gonostoma, and Malacosteus have no photo- 
phores at all on the sides of the body. 

Under these circumstances the construction of a satisfactory key to the Suborders 
of Isospondyli that would withstand the test of time is not to be hoped for at present. 



Tentative Key to Suborders of Isospondyli of the Western 
North Atlantic 

I a. Parasphenoid bone on roof of mouth with low close-set teeth; larval stage ribbon- 
shaped (leptocephalus-like). Elopoidea, Part 3, p. 107. 
lb. Parasphenoid bone without teeth; larval stage in most not ribbon-shaped.^' 

2 a. Premaxillary bones minute; maxillaries forming almost entire length of upper 
jaw; branchiostegal rays plate-like, the edge of the uppermost attached to 
lower margin of subopercular bone and thus taking part in formation of gill 
cover. Bathylaconoidea, Part 4. 

2b. Premaxillary bones well developed; branchiostegal rays not plate-like; the 
edge of upper branchiostegal ray not joining subopercular and interopercular 
bones and thus not taking part in formation of gill cover. 
3a. Luminescent organs (photophores) absent in most; never present on head, 
but present on eye-tube in a few (Dolichopteryx of Argentinoidea); if 
present on sides of body they are in nonlinear arrangement except along 
base of anal fin {Binghamichthys of Alepocephalidae) ; some with a tubular 
papilla on each shoulder that connects with a voluminous sac beneath 
the skin (Searsiidae). 
4a. No adipose fin between rayed dorsal fin and caudal. 

5 a. Snout noticeably flattened dorsoventrally (depressed), in duck- 
billed form; teeth along rear part of lower jaw large, formidable, 
conspicuous; premaxillary bones far apart in front; mesethmoid 
(proethmoid) bone paired. Esocoidea, Part 4. 

26. Information contributed by Robert H. Gibbs. 

27. According to Zugmayer {56: 4), the photophores are distinguishable only with difficulty in Astronesthes niger (as 
A. myriaster), but Regan and Trewavas {42: 20, fig. 12) picture them clearly for the type specimen of that species. 

28. Certain of the Stomiatoidea have ribbon-shaped larvae; Part 4. 



g6 Memoir Sears Foundation for Marine Research 

5b. Snout not flattened dorsoventrally, its dorsal contour convex; 

teeth along rear part of lower jaw small and inconspicuous ; pre- 

maxillary bones close together in front; mesethmoid (proeth- 

moid) bone not paired. Clupeoidea, Part 3, p, 148. 

4 b. Adipose fin between rayed dorsal fin and caudal present in most, but 

lacking in a few (see 6 b). 

6 a. Swim bladder connected to oesophagus throughout life by 
an open pneumatic duct (physostomic) ; premaxillary bones 
well developed; jaws with well developed teeth except in the 
Coregonidae, which are either toothless or have minute teeth; 
eyes of ordinary type; oviducts incomplete (p. 455), so far as 
is known. Salmonoidea, Part 3, p. 455. 

6 b. Swim bladder not connected to oesophagus during late 
stages of growth (physoclystic);^* premaxillary bones small or 
lacking; jaws toothless; eyes tubular in some but not in others; 
oviducts incomplete in some,^" perhaps complete in others. 

Argentinoidea,^^ Part 4. 

3 b. Luminescent organs (photophores) present on head, on body, on tail 

sectors, or on all three; their arrangement in longitudinal rows on sides 

extending anterior to anus as well as along base of anal fin ; no tubular 

papilla on shoulder. Stomiatoidea (including Suborders Gymno- 

photodermi and Lepidophotodermi^^), Part 4. 

Families. The Isospondyli plus the Iniomi include about 55 families on a world- 
wide basis, about 45 of which are known to occur in the western North Atlantic. At 
first glance, this may seem an appalling number among which to choose when one has 
to run down a given specimen to its proper family, the more so because there is no 
one conspicuous character or combination of characters visible from the outside on 
which anyone but a professional taxonomist can rely to tell him whether his fish be- 
longs among the Isospondyli or among the Iniomi. The task, however, is not as dif- 
ficult as one might expect, for nearly all of the western Atlantic families (or subdivisions 
of families) of the two Orders combined are separable, one from another, by features 
that are visible from the outside, that are susceptible of rather precise definition, and 
that are mutually exclusive, or nearly so. 

However, in numerous instances a family, because of the diverse characters of 
its genera and species, has been split into two or more parts. Hence the same family 

29. Cuvier and Valenciennes seem to have been the first to record this fact {ov Argentina (14: 411). Trewavas' interpre- 
tation of the ventral "sole" of Opisthoproctus as a swim bladder is now known to have been erroneous (5J: 610). 

30. This is the case for Argentina, Opisthoproctus, and Dolichopteryx (Part 4). For Macropinna, however, Chapman 
reported that a "fine duct" extends from the posterior end of each ovary, but he could not determine whether these 
ducts open into the urinary bladder or into the rectum (5: 293). 

31. For a detailed comparison of the Argentinoidea and Salmonoidea, see Argentinoidea, Part 4. 

32. Proposed by Parr, 33: 15-17. 



Fishes of the Western North Atlantic 97 

may be referred to in difterent parts of the same key or in one or more of the other 
keys. For example, such of the Bathypteroidae as have no adipose dorsal fin are 
found in Key A while those with an adipose fin are found in Key B. 



General Key to Families of Isospondyli, Iniomi, and 
Giganturoidei of the Western North Atlantic; Adolescents and Adults^* 

I a. Eyes (if any) of the usual sort, i. e. not at tips of slender stalks or with pupil above 
center, and not capable of being directed upward. 

2 a. No luminescent organs (photophores) either on sides of body, or on head. 

3 a. No adipose fin between rayed dorsal fin and caudal. Key A, below. 

3 b. Adipose fin present between rayed dorsal and caudal. Key B, p. 99. 

2b. Luminescent organs (photophores) present on head, or on body, or on both; 

or with eyes on top of head. Key C, p. loi. 

I b. Eyes at tips of slender stalks, or with pupil above the center and capable of being 

directed upward. Key D, p. 103. 

Key a. No dorsal adipose fin; no photophores on sides of head or body. 

I a. Rays in posterior half of dorsal and in rearmost part of anal much longer than 

those in anterior part of these fins. Macristiidae.^^ 

I b. Not more than one dorsal ray (if any) and no anal ray much longer than the others. 

2 a. Bony (gular) plate present in chin region between branches of lower jaw. 

Elopidae, Part 3, p. 1 1 1. 
2 b. No bony (gular) plate in chin region. 

3 a. Upper jaw reaching rearward far beyond eye. 
4 a. Snout overhanging mouth noticeably. 

Engraulidae (western Atlantic saltwater species). Part 3, p. 152. 
4 b. Snout not overhanging mouth. 

5a. Pectoral fin longer than head; either uppermost pectoral rays 
or outermost pelvic rays much longer than the others; lower lobe of 
caudal much longer than upper in some. 

Bathypteroidae in part. Part 5.^^ 

5b. Pectoral fin shorter than head; none of pectoral or pelvic rays 

prolonged; lower lobe of caudal little, if any, longer than upper. 

6 a. A fleshy ridge present along back anterior to dorsal fin. 

Alepocephalidae in part 
[Anomalopterui)^ Part 3, p. 250.^* 
6 b. No fleshy ridge on back anterior to dorsal fin. 

33. For larval stages of a few, see Key D, i b. In this key no attempt is made to present the author's views as to phylo- 
genetic relationship. 

34. Occur in the vicinity of the Azores, hence may be taken sometime in the western Atlantic. 

35. See Key B, 4a for those with adipose fin. 36. See also Key A, iia; Key C, 15a. 



9 8 Memoir Sears Foundatmt for Marine Research 

7 a. Eyes minute, about same diameter as nostril. 

Ipnopidae in part (^Bathymkrops, 
Bathytyphlops), Part 5.^' 
7 b. Eyes well developed, much larger than nostril. 

Bathysauridae in part. Part 5.^' 
3 b. Upper jaw reaching rearward little (if any) past eye, and ending con- 
siderably short of rear edge of eye in most. 

8a. Snout overhanging mouth; parasphenoid bone rearward along mid- 
roof of mouth thickly set with low teeth. 

9 a. Base of dorsal occupying only about 24—2 5 "/o of distance between 
gill openings and origin of caudal. Albulidae, Part 3, p. 132. 
9b. Base of dorsal occupying about 8o''/o or more of distance be- 
tween gill openings and origin of caudal. Pterothrissidae.^' 
8b. Snout not overhanging mouth; parasphenoid bone without teeth. 
10 a. Rear end of base of dorsal fin separated from upper origin of 
caudal by a distance much shorter than head. 
I I a. Upper jaw not reaching rearward beyond front of eye; 
teeth inconspicuous. 

Alepocephalidae in part. Part 3, p. 250.*" 
I 1 b. Upper jaw reaching rearward about as far as rear edge of 
eye; teeth along rear half of lower jaw large and formi- 
dable. Esocidae, Part 4. 
lob. Rear end of base of dorsal fin separated from upper origin of 
caudal by a distance at least as long as head, and longer in most. 
1 2 a. Midline of abdomen in front of anal fin armed with a 
double series of stiff, pointed scales (scutes). 
Clupeidae in part (Clupeinae, pp. 274—411; Pristigaste- 
rinae, pp. 411— 438; Chirocentrinae, pp. 438— 442; Do- 
rosomatinae, pp. 443—451), Part 3, Key pp. 259— 262.*^ 
12b. Midline of abdomen in front of anal fin rounded; not 
armed with a double series of scutes. 
13 a. Origin of dorsal fin considerably anterior to mid- 
length of trunk; pelvlcs posterior to dorsal; max- 
imum depth of trunk equal to about 30 "/o of dis- 
tance from gill opening to base of caudal ; head about 
25''/o as long as distance from snout to origin of 
caudal. Clupeidae in part (Dussumierinae, 
pp. 262-274), Part 3, Key pp. 259-262.*^ 

37. See also Key C, la. 38. For those with adipose fin, see Key B, 13b. 

39. Sometimes included among the Albulidae. It has not yet been reported for the western side of the Atlantic, 
but it may be expected there because it is plentiful at moderate depths along the coast of tropical West Africa. 
For details and description, see Poll (j6: 16-25, 28, fig. 9). 

40. See also Key A, 6 a; Key €,153. 41. See also Key A, 13a. 42. See also Key A, 12 a. 



Fishes of the Western North Atlantic C)() 

1 3 b. Origin of dorsal fin considerably posterior to mid- 
length of trunk; pelvics below or anterior to dorsal; 
maximum depth of trunk only about i5''/o as great 
as distance from gill opening to base of caudal; 
head between i'j-2o''/o as long as distance from 
snout to base of caudal. 

Argentinidae in part (Microstomatinae), Part 4.*^ 



Key B. a dorsal adipose fin present close in front of caudal; no photophores 

ON SIDES OF head OR BODY. 

I a. Well-developed rayed dorsal fin and adipose dorsal fin present. 

2 a. A narrow, posteriorly-pointing fleshy lobe close above base of pelvic fin on 
each side. 

3a. Teeth in jaws well developed, easily felt; 19 or more scales in a transverse 

series from origin of dorsal to lateral line. Salmonidae, Part 3, p. 457. 

3b. Teeth in jaws minute, if present; not more than 13 scales in a transverse 

series from origin of dorsal to lateral line. Coregonidae, Part 3, p. 547. 

2 b. No fleshy lobe on side above base of pelvic fin. 

4 a. Some of upper pectoral rays longer than others and separate from lower 
part of fin, thus forming a separate division. 

Bathypteroidae in part. Part 5.^* 
4 b. Pectorals not divided into an upper and lower division. 

5 a. Dorsal high, sail-like, its base extending from close behind gill 

openings nearly to adipose fin. Alepisauridae, Part 5. 

5 b. Dorsal not high and sail-like, its base occupying not more than 

about half the distance between level of gill openings and adipose fin. 

6a. Upper jaw not reaching rearward as far as front of eye. 

7a. Mouth at least twice as long as eye; eye not more than 
about i7''/o of hi (less than this in most); anal base at least 
1.5 times longer than base of dorsal; teeth well developed, 
long and fang-like in some. Paralepididae, Part 5. 

7b. Mouth not longer than eye; eye 27— 40''/o of hi; anal base 
not appreciably longer than dorsal base; teeth minute, if 
detectable at all. 

8a. Lateral-line scales not extending out onto caudal fin; 

adipose dorsal anterior to rear end of base of anal fin. 

9a. Gill membranes separate; branchiostegal rays 5 

or 6. Argentinidae in part (Argentininae), 

Part 4.« 

43. See also Key B, 8 b and 9a; Key D, 4b. 44. For those without dorsal adipose fin, see Key A, 5a. 

45. See also Key D, 4 b. 

7* 



lOO Memoir Sears Foundation for Marine Research 

9b. Gill membranes broadly united; branchiostegal 
rays 2. Bathylagidae, Part 4.** 

8b. Lateral-line scales extending out onto caudal fin; adi- 
pose dorsal fin over or posterior to rear end of base of 
anal fin. Argentinidae in part (Nansenia, 

Microstomatinae in part), Part 4.*' 
6 b. Rear end of upper jaw reaching beyond front of eye. 
10 a. Some of jaw teeth large, fang-like, conspicuous. 

I I a. Eye about 33°/o or more of hi; origin of anal pos- 

terior to origin of dorsal by a distance less than 
5o''/o of hi. Omosudidae, Part 5. 

I I b. Eye only about i5''/o of hi; origin of anal posterior 

to origin of dorsal by a distance nearly equal to hi. 
Evermannellidae in part, Part 5.''* 
lob. None of jaw teeth much enlarged or fang-like. 

12 a. Upper jaw extending rearward past eye for a dis- 
tance at least twice the length of eye.*^ 
13 a. Rear end of base of dorsal anterior to origin of 
anal by a distance at least 50 "/o of hi. 

Synodontidae, Part 5. 

1 3 b. Rear end of base of dorsal only slightly ante- 

rior to origin of anal. 

Bathysauridae in part. Part 5.'" 
12 b. Upper jaw extending rearward past eye for a dis- 
tance no longer than length of eye, and relatively 
shorter than this in most. 

14 a. Dorsal profile of head deeply concave close in 

front of eye; insertion of pelvics anterior to 

origin of rayed dorsal by a distance nearly or 

quite as long as eye. Scopelosauridae, Part 5. 

14 b. Dorsal profile of head close in front of eyes only 

weakly concave at most; insertion of pelvics 

under or posterior to origin of rayed dorsal. 

15 a. Upper jaw extending rearward beyond 

eye for a distance about as long as eye; 

dorsal profile of head weakly concave; 

body deepest about at gill opening. 

Neoscopelidae in part 
{Scopelengyi)^ Part 5.** 

46. See also Key A, 13 b; Key B, Sbj Key D, 4 b. 47. See also Key D, i b. 

48. For species having eyes with pupil above center and capable of being directed upward, see Key D, 5 b. 

49. In small specimens of some species of Trachinocephalus the upper jaw does not extend rearward as far as this. 

50. For those without adipose fin, see Key A, 7 b. 51. For those with photophores, see Key C, 14 b. 



Fishes of the IV es tern North Atlantic loi 

15 b. Upper jaw not extending rearward ap- 
preciably beyond eye; dorsal profile of 
head straight to weakly convex; body 
deepest about at dorsal fin. 
1 6 a. Origin of rayed dorsal nearer to 
gill opening than to adipose dorsal 
by a distance of slightly less to 
considerably more than head 
length. 

17a. Pupil round; entire posterior 
edge of opercular flap formed 
by subopercular bone. 

Aulopidae, Part 5. 
17b. Pupil keyhole-shaped; pos- 
terior edge of opercular flap 
formed by opercle above, 
and by subopercle below. 
Chlorophthalmidae, Part 5. 
1 6 b. Origin of rayed dorsal almost or 
quite as near to adipose dorsal as 
to gill opening, and much nearer 
to adipose dorsal in some. 

Osmeridae, Part 3, p. 553. 
lb. No rayed dorsal fin; only an adipose dorsal fin. Anotopteridae, Part 5. 



Key C. Photophores present on head, or on body, or on both; or with eyes 

on top of head. 
I a. Eyes without lenses, covered by frontal bones, and consisting of large and flat 
plates that occupy more than half of dorsal surface of head. 

Ipnopidae in part, Part 5.^'' 
I b. Eyes with lenses, not covered by the frontal bones and, if dorsally directed, oc- 
cupying less than half of dorsal surface of head. 

2 a. One or more barbels on chin, usually long and fleshy but sometimes very 
small. 

3 a. Dorsal fin mostly or wholly anterior to anal, its origin near midlength 
of trunk. 

4a. Dorsal base extending behind pelvics, its origin behind or only 
slightly before pelvic base; first dorsal ray not prolonged. 

Astronesthidae, Part 4. 

52. For those without these organs, see Key A, 7 a. 



I02 Memoir Sears Foundation for Marine Research 

4b. Dorsal base completely in front of pelvic bases; first dorsal ray much 
prolonged. Chauliodontidae, specimens up 

to about 50 mm long,^' Part 4. 
3 b. Dorsal fin opposite anal, its origin far posterior to midlength of 
trunk. 

5 a. Sides of body with scales, or marked by hexagonal pigment patterns 
resembling scales. Stomiatidae, Part 4. 

5 b. Sides of body scaleless, either unmarked or with parallel vertical 
lines, never with hexagonal areas. 

6 a. Dorsal fin with more than 50 rays and beginning well before 
middle of body; base of each dorsal and anal ray flanked on 
either side by a sharp bony spur that penetrates the skin; length 
of trunk posterior to head more than 20 times greater than great- 
est depth of body. Idiacanthidae in part (adults), Part 4.^* 
6 b. Dorsal fin with 30 rays or less, its origin well behind middle 
of body; no bony spurs flanking bases of dorsal and anal rays; 
length of trunk posterior to head not more than 1 5 times greater 
than greatest depth of body. 

7a. Mouth with a membranous floor; isthmus attached to man- 
dibular symphysis; dorsal corner of gill opening behind 
posterior margin of eye. Melanostomiatidae, Part 4. 

7b. Mouth without a floor; isthmus attached to mandibular 
symphysis by a muscular cord; dorsal corner of gill opening 
above or in advance of posterior margin of eye. 

Malacosteidae in part. Part 4.*^ 
2 b. Chin without fleshy barbel. 

8 a. Each shoulder with a tubular papilla open at the tip and connected 
to a voluminous sac beneath skin. Searsiidae, Part 3, p. 254. 

8 b. Shoulders without tubular papilla or sac beneath skin. 

9 a. Dorsal fin preceded by the protruding, spine-like tips of from one 
to several of the neural processes of a like number of vertebrae. 

Sternoptychidae, Part 4. 
9 b. Dorsal fin not preceded by the protruding spine-like tips of the 
neural processes of any vertebrae. 

10 a. Branchiostegal rays plate-like, conspicuously exposed, the up- 
permost connected to subopercular and interopercular bones. 

Bathylaconidae, Part 4. 
lob. Branchiostegal rays not plate-like, the uppermost not connected 
to subopercular and interopercular bones. 
1 1 a. Dorsal adipose fin present. 

53. For larger specimens, see Key C, 12 a. 54. See also Key D, i b. 

55. See also Key C, 15 b. 



Fishes of the Western North. Atlantic 103 

12 a. Teeth enormous, conspicuous even when mouth 
is closed; first dorsal ray greatly prolonged. 

Chauliodontidae in part, Part 4.*' 
:2b. Teeth small to moderate in size; first dorsal ray 
not greatly prolonged. 

13 a. Upper edge of mouth formed by both max- 
illaries and premaxillaries; upper jaw teeth 
extending outside lower jaw when mouth is 
closed; photophores generally present on 
branchiostegal membrane. 

Gonostomatidae in part. Part 4.^' 
13 b. Upper edge of mouth formed by premaxil- 
laries only; upper jaw teeth contained within 
mouth; no photophores on branchiostegal 
membranes. 

14 a. Distance from pelvics to anal origin not 
more than twice the length of eye. 

Myctophidae in part, Part 5.*' 
14 b. Distance from pelvics to anal origin at 
least 3 times longer than eye. 

Neoscopelidae in part. Part 5.^' 
1 1 b. No dorsal adipose fin. 

15a. Body with at least 20 longitudinal rows of scales; 
mouth not widely distensible; gape not extending 
past eye; teeth small. 

Alepocephalidae in part, Part 3, p. 250.'" 
15b. Body without scales; mouth enormously disten- 
sible; gape extending beyond eye by a distance 
3 or more times the length of eye; teeth large and 
conspicuous. Malacosteidae in part, Part 4.^^ 



Key D. Eyes either tubular (teleoscopic), or with pupil above ccnter and ca- 
pable OF being directed upward, or with eye situated at the tip of a 

SLENDER stalk. 

I a. Eyes not at tips of slender stalks. 

2a. Lower lobe of caudal much longer than upper lobe; gill openings minute; 
base of pectorals wholly above gill openings. Giganturidae, Part 4. 

56. See also Kev C, 4 b. 

57. Exceptions: in Ickthyococcus the upper jaw is formed bv the maxillaries alone, and in Triplophos it is ahnost entirely 
bordered by the premaxillaries. 

58. See also Key D, i b. 59. See also Key B, 15 a. 
60. See also Key A, 6a; Key A, 11 a. 61. Sec also Key C, 7 b. 



1 04 Memoir Sears Foundation for Marine Research 

2b. Lower lobe of caudal about as long as upper lobe; gill openings broad; base 
of pectorals not wholly above gill openings. 

3a. Upper jaw falling far short of eye; teeth minute, if detectable at all. 
4 a. Either an adipose fin between rayed dorsal and caudal, or a dome- 
shaped luminescent organ on eye tube. Opisthoproctidae, Part 4. 
4 b. No adipose fin, and no luminescent organ on eye tube. 

Argentinidae in part {Xenophthalmichthys, 
Microstomatinae in part). Part 4.^^ 
3b. Upper jaw reaching rearward beyond front of eye; teeth fang-like, con- 
spicuous. 

5a. Body strongly compressed sidewise, covered with scales; pectorals 
with 19—21 rays, their line of insertion oblique, wholly above artic- 
ulation of lower jaw with skull (quadrate bone). 

Scopelarchidae, Part 5. 
5b. Body not much compressed, without scales; pectorals with only 12 
rays, their line of insertion nearly horizontal, close to ventral outline 
of body, entirely below articulation of lower jaw. 

Evermannellidae in part. Part 5.'^ 
I b. Eyes at tips of slender stalks. 

Stalk-eyed larval stages of Bathylagidae in part (Part 4);"'' of Idiacanthidae 
in part (Part 4);"* of Myctophidae in part (Part 5);** and probably others 
of unknown parentage. 

62. See also Key A, 13b; Key B, 8b; and Key B, 9a. 63. See also Key B, iib. 

64. See also Key B, 9 b. 65. See also Key C, 6a. 

66. See also Key C, 14a. 



TEXT AND FOOTNOTE REFERENCES 



1. Artedi, Ichthyol., J, 1738. 

2. IJEEBE, Zoologica, N.Y., 16, 1933. 

J. Beebe, Zoologica, N.Y., 76(4), 1934. 

4. Berg, Class. Fish., Trav. Inst. zool. Acad. Sci. 

URSS, 5 (2), 1940; also reprinting by J. W. 

Edwards, Ann Arbor, Mich., 1947 (Russ. 

and Eng.). 
J. Bertelsen, Nature, Lend., 181, 1958. 

6. Bolin, Copeia, 1939. 

7. Boulenger, Cambr. Nat. Hist., 7, 1904. 

8. L-iHAPMAN, Ann. Mag. nat. Hist., 11 (9), 1942. 

9. Cohen, Zoologica, N.Y., 41 (2), 1956. 

10. Cohen, Bull. Fla. St. Mus., j, 1958. 

11. Cope, Trans. Amer. philos. Soc, 14, 1871. 

12. Cuvier, Regne Anitn., 2, 1817. 

JJ. Cuvier and Valenciennes, Hist. Nat. Poiss., ig, 

1846. 
14. Cuvier and Valenciennes, Hist. Nat. Poiss., 21, 

1848. 
Jj. Cuvier and Valenciennes, Hist. Nat. Poiss., 22, 

1849. 

16. TowLER, Bull. Amer. Mus. nat. Hist., yo (i), 
1936. 

ly. Cjill, Mem. nat. Acad. Sci., 6, 1893. 
18. Grasse, et al.. Trait. Zool., JJ, 1958. 
ig. Gunther and Deckert, Zool. Beitr., Berl., N.S. 
-r(3). '955- 

20. IXAY, Bull. U.S. geol. Survey., lyg, 1902. 

21. HoUister, Zoologica, N.Y., 2/(4), 1936. 



22. Jordan, Stanf. Univ. Publ. Biol., j (2), 1923. 
2J. Jordan and Evermann, Bull. U.S. nat. Mus., 
47 {i), 1896. 



24. i\.ENDALL, Bull. U.S. Bur. Fish., 57, 192 1. 

25. 1_/ANKESTER, Treat. Zool., 9, 1909. 



26. iVlARSHALL, 'Discovery' Rep., 2y, 1955- 

27. Matsubara, Fish. Morph. Hierarch., Pt. i, 

1955 (Jap-)- 

28. Myers, Copeia, 1946. 

29. Myers, Proc. Calif. Acad. Sci., (4) JJ, 1944. 

JO. JNoRMAN, Hist. Fish., 193 1; and subsequent 

editions. 
JJ. Norman, Synop. Orders, Families, Genera or 

Recent Fish, (mimeo), 1947. 
32. Nybelin, Ark. Zool., N.S. 10 (4-5), i957- 

JJ. Jtarr, Bull. Bingham oceanogr. Coll., J (3), 

1927. 
34. Parr, Copeia, 1930. 
J5. Parr, Copeia, 1948. 
36. Poll, Res. Sci. Exped...Eaux Coderes Afr. 

Atl. Sud, 4 (2): 1953. 

J7. Jxathke, Neue Schr. Naturf. Ges. Danzig, I 

(3), 1824. 
38. Regan, Ann. Mag. nat. Hist., (8) j, 1909. 
J9. Regan Ann. Mag. nat. Hist., (8)5, 1910. 



105 



io6 Memoir Sears Foundation for Marine Research 



40. Regan, Trans, roy. Soc. Edinb., ^9 (2), 1913. 

41. Regan, Encyc. Britt., 9, 1929, and subsequent 

editions. 

42. Regan and Trewavas, Dana Rep., 5, 1929. 

43. Regan and Trewavas, Dana Rep., 6, 1930. 

44. Ridewood, Proc. zool. Soc. Lond., 2, 1904. 
4^. Romer, Vert. Paleont., ed. 2, 194;. 

46. Oaint Seine, Nouv. Arch. Mus. Hist. nat. 

Lyon, 2, 1949. 

4^. Schultz and Stern, Ways of Fish., Ft. i, 1948. 

48. Senior, Biol. Bull. Woods Hole, 12, 1900. 

4g. Senior, Anat. Rec, I, 1907. 

y). Smitt, Hist. Scand. Fish., 2, 1895. 



57. 1 CHERNAVIN, Nature, Lond., l^j, 1946. 
52. Tchernavin, Feed. Mech. Deep Sea Fish., 

1953- 
5J. Trewavas, Proc. zool. Soc. Lend., 1933. 



$4. Woodward, Cat. Foss. Fish. Brit. Mus., J, 

1895. 
5$. Woodward, Cat. Foss. Fish. Brit. Mus., 4, 

1901. 



^6. i^ucMAVER, Bull. Inst, oceanogr. Monaco, 
25J. 19' 3- 



Suborder Elopoidea 

COMPOSITE AUTHORSHIP 



Characters and Key to Families 



y 



HENRY B. BIGEL0W2 

Museum of Comparative Zoology 
Harvard University 



Characters. Silvery, herring-like isospondylous fishes with the Parasphenoid 
BONE on posterior part of roof of mouth thickly set with low Teeth,^ extending 
nearly to posterior end of basioccipital or beyond, without lateral processes for articula- 
tion with the mesopterygoid. Temporal foramina present. Parapophyses not fused 
with vertebral centra, but simply wedged in pits. Proethmoids not paired (see Eso- 
coidea. Part 4). Pectoral girdle with mesocoracoid element. A Bony (gular) plate in 
chin region between branches of lower jaw in some (Elopidae, including the Megalo- 
pidae of Matsubara) but not in others (Albulidae, Pterothrissidae). Luminescent 
ORGANS absent. Adipose fin lacking between rayed dorsal and caudal. Pelvic fins 
abdominal, standing from about 30% to about 45 "/o of distance rearward from gill slit 
toward lower origin of caudal. Anal fin originating from about 5070 to about 68 "/o of 
distance between pelvics and caudal. Caudal deeply forked. Rearmost dorsal ray 
greatly prolonged in some (Tarpon, Megalops)\ the Rearmost anal ray prolonged as 
well in a few (Dixonina'); none of the other fin rays prolonged. A single large, nar- 
row, fleshy Axillary scale above pectorals and pelvics. Scales thin, cycloid. Eyes 
circular, of the ordinary type. Upper jaw extending rearward far beyond eye in most, 

1. With families Elopidae and Albulidae by Samuel F. Hildebrand, pp. iii and 132, respectively. 

2. The assistance of Dr. W. A. Gosline is gratefully acknowledged. 

3. Ridewood, who described their skulls in detail, has recorded this peculiar dentition (j: 39, for Elops; 44, for Mega- 
lops; 40, for Albula; 53, for Pterothrissus). These observations have been verified for Plerothrissus by Poll {2: 222, 
fig. 7), and for Tarpon by original examination. 

107 



io8 Memoir Sears Foundation for Marine Research 

and tip of Lower jaw projecting as far as tip of upper jaw or slightly beyond it; upper 
jaw in others {Albula, Dixonina, Pterothrissus) falling far short of front of eye, and tip 
of snout overhanging mouth. Maxillaries forming part of margin of upper jaw in 
most but excluded from gape in a few {^Albula^ Pterothrissus). Branch iostegal rays 
slender, rod-like, not forming a part of opercular system. One to three lateral sensory 
Canal-bearing bonelets on either side of snout in front of lacrimals. Some with two 
rows of valves in cone of Heart (yilbula, Pterothrissus, and Megalops, but not E/ops). 

There is a ribbon-like (leptocephalus-like) larval stage. 

Families. Three families are recognized here within the Suborder as defined 
above: Elopidae, Albulidae, and Pterothrissidae, the last of which is extralimital and 
is included only in the Key to Families. 

Fossil History. The Elopidae are known from as far back in geologic time as the 
Lower Cretaceous, the Albulidae (including the Pterothrissidae) from as far back as 
the Upper Cretaceous. 

Key to Families 

I a. A bony (gular) plate in chin region between branches of lower jaw; upper jaw 
extending rearward considerably beyond eye; tip of lower jaw projecting at least 
as far as tip of upper jaw, or a little farther; 23—35 branchiostegal rays. 

Elopidae (including Megalopidae of Matsubara), p. iii. 
I b. No gular plate; upper jaw falling short of front of eye; tip of snout overhanging 
mouth; only about 11— 15 branchiostegals. 

2 a. Base of dorsal occupying only about 2 5''/o of distance between gill opening 
and base of caudal. Albulidae, p. 132. 

2b. Base of dorsal occupying about So"/,, of distance between gill opening and 
base of caudal. Pterothrissidae. 

Plentiful off tropical West Africa 
in depths of 100-250 m and re- 
ported down to 500 m;'' also 
Japan. 



4. According to Cadenat (r: 365) and Poll {2: 23). Not yet reported for the western Atlantic but may be expected 
there. 



TEXT AND FOOTNOTE REFERENCES 



I. Cadenat, Bull. Mus. Hist. nat. Paris, (2) 10, 2. Poll, Res. Sci E.xped.. . . Eaux Cotieres Afr. 

•938- Atl. Sud, 4{2), 1953. 

3. JXiDEWooD, Proc. zool. Soc. Lond., 1904. 



109 



1 1 o Memoir Sears Foundation for Marine Research 



Editorial Comments on the Hildebrand Manuscripts 

The following manuscripts by Dr. Hildebrand had not been worked on by him 
for some years prior to his death in 1949. The Sears Foundation is therefore grateful 
to Dr. George S. Myers for having undertaken the task of bringing some manuscripts 
up-to-date in such ways as he thought necessary, and to the U.S. Fish and Wildlife 
Service for having made this possible through the kind offices of Dr. L. A. Walford. 
Dr. Myer's work, which was both time-consuming and extensive, was confined prin- 
cipally to the insertion of recently described genera and species, the rewriting of certain 
keys, and research on the classification or nomenclature of certain genera (notably 
Pristigaster and Ilishd) which required changes in the names originally used by Dr. 
Hildebrand. Revisions that departed radically from what he believed Dr. Hildebrand 
would have included have been inserted as initialed footnotes. 

Some years ago Dr. John Tee-Van, because of the pressure of administrative duties 
at the New York Zoological Society, was no longer able to continue as Editor-in-Chief; 
Dr. Bigelow was subsequently chosen by the Editorial Board as his successor. Upon 
receipt of the Hildebrand manuscripts from Dr. Myers, Dr. Bigelow made extensive 
revisions and additions in many of the accounts, mainly in the sections dealing with life 
history. Thus Dr. Bigelow added much information to that originally contributed by 
Dr. Hildebrand; his major revisions occur primarily in the sections dealing with the 
Elopidae (p. 1 1 1), Albulidae (p. 132), and Clupeidae (p. 257). The taxonomy remains 
essentially as it was written by Dr. Hildebrand or revised by Dr. Myers, excepting 
minor changes for conformance with general format. The value of Dr. Bigelow's con- 
tributions to the Hildebrand papers, and indeed to Part 3 in many other respects as 
well, can be measured only by the usefulness of this volume to those who will have 
occasion to refer to it. I have made such emendations as seemed desirable for publi- 
cation. 

Explanations of Dr. Hildebrand's procedure of measuring and counting are given 
on pages 154, 258, 343. Dr. Hildebrand's taxonomic accounts are as complete as the 
material and original sources of information permitted. In most instances they are 
based upon his study of extensive material available to him at the U. S. National 
Museum and elsewhere, and perhaps equally important, upon his broad and exten- 
sive knowledge acquired over the years from firsthand observations in both field and 
laboratory. Whether one agrees or disagrees with his treatment of relationships and 
such, there can be no question of the care and meticulousness of his basic work. — 

Y. H. Olsen. 



Family Elopidae' 



SAMUEL F. HILDEBRAND2 



Characters. Body slender, more or less compressed laterally; belly rounded; ordi- 
nary scales covering median line. Eye large, with adipose tissue in large examples. 
Mouth large, terminal, or superior. Maxillaries large, in about three pieces, ex- 
tending far beyond eye in adults. Premaxillaries not protractile. Lower jaw with 
an elongate bony plate between its branches. Teeth all small, often bluntly villiform; 
teeth present on jaws, vomer, palatines, pterygoids, tongue, and basibranchials. Bran- 
chiostegals about 23-35. ^^^^ membranes separate, free from isthmus. Opercular 
BONES with membranous borders. Gill rakers moderately long. Last several segments 
of Spinal column directed upward toward base of upper lobe of caudal. Lateral 
LINE present. Scales large or small, with membranous borders; missing on head; the 
median row in front of dorsal neither enlarged nor modified. Dorsal fin over or 
somewhat behind pelvics. Caudal forked. Pectorals and Pelvics similar, each with 
a rather long axillary scale. 

Remarks. The members of this family are characterized by the presence of the 
bony gular plate between the arms of the lower jaw, the numerous branchiostegals 
(23—35), ^"d *h^ large terminal or superior mouth. 

The genera Tarpon and Elo-ps (pp. 112, 123) sometimes have been assigned to 
two families, Megalopidae and Elopidae. However, the two genera agree in so many 
seemingly basic characters — the gular plate, numerous branchiostegals, the dentition, 
the large maxillaries, and the primitive character of the caudal skeleton — that it seems 
desirable to treat them as one family. 

The members of the Elopidae generally are considered to be among the most 
archaic of existing teleosts. Their antiquity, particularly as indicated by their skulls, 
has been fully discussed (ftn. 3). 

1. Edited and emended by George S. Myers, Henry B. Bigelow, and Yngve H. Olsen. 

2. August ij, 1883-March 16, 1949. 

Ill 



112 Memoir Sears Foundation for Marine Research 

The young pass through a leptocephalus stage, similar to that of the eels. 
Range. The members of this family are widely distributed in the warmer seas. 



Key to Western Atlantic Genera 

I a. Body moderately deep, rather strongly compressed laterally, mouth superior, with 
tip of mandible entering dorsal profile; last ray of dorsal and anal produced, fila- 
mentous; scales large, about 41—48 in a lateral series; anal with about 22—25 rays. 

Tarpon Jordan and Evermann 1896, below. 

lb. Body very slender, not greatly compressed laterally; mouth terminal or nearly 
so; last ray of dorsal and anal not produced into a filament; scales small, 
generally more than 100 in a lateral series in American species; anal with fewer 
than 20 rays. Elops Linnaeus 1766, p. 123. 



Genus Tarpon Jordan and Evermann 1896 
Tarpon 

Tarpon Jordan and Evermann, Bull. U.S. nat. Mus., 47 (j), 1896: 409; type species by original designation, 
Megalops atlanticus Cuvier and Valenciennes, Hist. Nat. Poiss., ig, 1846: 398. 

Generic Synonyms: 

Clupea Bloch, Naturg. ausland. Fische, Pt. 9, 1794: 32; in part, for C. cyprimijes Bloch; Shaw, Genl. Zool., 

5 (i), 1804: 173, for C.gigantea Shaw. 
Megalops Lacepede, Hist. Nat. Poiss., 5, 1803: 289; in part, for M.giganteus Voigt in Cuvier, Das Thierreich 

(German transl.), 2, 1832: 423. 

Characters? Body oblong, rather strongly compressed. Vertebrae about 55. 
PsEUDOBRANCHiAE Undeveloped. Branchiostegals about 23. Mouth quite oblique, 
superior, with prominently projecting mandible. Lateral line decurved, with branched 
tubes. Scales large, not especially primitive, their characters similar to those of Elops 
(cf. Cockerell, ly : 122); scales extending somewhat onto base of anal and caudal but 
not onto dorsal. Dorsal fin high anteriorly, its last ray produced and filamentous. Anal 
similar to dorsal but longer, its last ray produced in large examples, its origin not far 
behind dorsal base. Pelvics inserted well in advance of dorsal fin. Last several Verte- 
brae clearly out of line with main axis, being directed toward the tip of upper lobe of 
caudal. Air bladder large and opening into gullet; within bladder a considerable 
amount of cellular, lung-like tissue; two large and two small strands adhering to bladder 
wall (bladder serving somewhat as an air-breathing organ, much as in gars [pp. 64, 
67]; for description and figures of the modified bladder, see Babcock, i [1936]: 50). 
Alimentary canal much shorter than body; the stomach large, reaching far back 

3. The skull has been fully described and figured by Gregory, who regarded it as a little less primitive than that of 
Elops (jo: 137-142, figs. 31-33). The characters of the tail of Tarpon and related genera have been described by 
Regan {sg: 3J4), and in more detail by Hollister (jp: 264). 



Fishes of the JVestern North Atlantic 



113 



into abdominal cavity as a sort of blind sac, the two orifices rather close together. 
Pyloric caeca numerous, short and branched, held together by connective tissue. 

Remarks. This genus is close to Megalops Lacepede, the oxeye herring represented 
by M. cyprinoides of the Indian Ocean and adjacent Pacific, but it differs from Megalops 
mainly in the more anterior position of the pelvic fins — well in advance of the dorsal 
fin in Tarpon but inserted under the base of the dorsal in Megalops. Thus the dif- 
ference is in the position of the pelvic fins rather than in the dorsal as stated in the 
original description of Tarpon. Hollister, who compared specimens, has stated that 
Tarpon has only 57 vertebrae whereas Megalops has 68, and that Tarpon has 12-15 
dorsal rays whereas Megalops has 19-21 {42: 449). Also, Tarpon grows to a much 
larger size. American ichthyologists generally have accepted Tarpon as a valid genus 
while Old World students have rejected its separation from Megalops. Perhaps in con- 
sideration of the additional differences now known, the separation may seem better 
founded. Compared to Elops, Tarpons body is not as slender, its scales are larger, and 
its dorsal and anal fins have long filaments attached to the last ray of each fin. 

Species and Range. A single species is known, ranging from Cape Cod, Massa- 
chusetts, to southern Brazil. It has been reported also for tropical West Africa. 



Tarpon atlanticus (Cuvier and Valenciennes) 1846 

Tarpon, Tarpum, Sabalo Real, Cuffum, Silverfish, Silverking, Grand Ecaille, 

Jewfish, Palika 

Figure 19 

Study Material. A total of 25 specimens: 14, 78-1,175 mm TL (57-837 mm SL), 
measured in the laboratory, from Boggy and Boca Grande, Florida; Fajardo, Puerto 
Rico; "Cuba"; "West Indies"; and "Amazon River," Brazil; 11 adults, 100- 
1,990 mm TL, from Gatun and Miraflores locks. Canal Zone, examined for food, sex, 
and spawning condition. 

Distinctive Characters. See Remarks^ above. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 14 specimens in the Study Material unless stated otherwise, spec- 
imens 57—837 mm SL. 



Body: depth 23.5-29. 
Head: length 25—31. 
Snout: length 4.5—6.2. 
Eye: diameter 5.3—9.5. 
Maxillary: length 1 5.5-1 8. 
Anal fin: length of base 18-20. 
Pectoral fin: length 19-22.8. 



Scales: 41—48 series. 

Gill rakers : 1 9—2 i + 36—40. 

Fin rays: dorsal 13—15; 

anal 22—25; pectoral 13 or 14. 
Vertebrae: 53-57 (4 specimens). 



114 Me;noir Sears Foundation for Marine Research 

Body with almost vertical sides; the dorsal outline of head nearly straight and 
horizontal, the back somewhat elevated, with ventral outline strongly curved anteriorly; 
depth 3.4—4.3 in SL. Scales large, firm, with crenulate membranous border. Lateral 
LINE complete, decurved anteriorly, the pores branched. Head moderately short and 
deep, its depth at middle of eye not quite twice its width at the same place, 3.2-4.7 
in SL. Snout to rim of upper jaw considerably shorter than eye in small specimens, 
gradually becoming equal to and finally longer than eye during growth, 4.5—5.5 in 
head. Eye 3.3—4.7 in head, much nearer to dorsal than to ventral profile, wholly in 
anterior half of head in adults. Gill rakers rather long and slender, scarcely dentic- 




FicuRE 19. Tarf'on atlanticus, from off New Jersey, USNM 14924; from Goode. Drawn by H.L.Todd. 

ulate. Mouth superior. Mandible projecting far beyond the gape, entering dorsal 
profile in advance of mouth. Maxillary broad, strongly curved, extending under 
posterior part of eye in young (80— 100 mm) but far beyond eye in large examples, 
1.5— 1.9 in head. 

Dorsal fin high anteriorly, its last ray produced in specimens over 100 mm long, 
quite long in large specimens, reaching to end of base of anal fin in some specimens 
seen; its origin about equidistant between base of caudal and anterior margin of eye. 
Caudal deeply forked, the lobes of about equal length, generally somewhat longer 
than head. Anal somewhat elevated anteriorly, its last ray produced in adults, some- 
times reaching the base of lower rays of caudal, its origin at least an eye's diameter 
behind end of dorsal base, its base 4.6—5.6 in SL. Pelvic rather large, inserted more 
than an eye's diameter in advance of origin of dorsal, and somewhat nearer to base of 
pectoral than to origin of anal in large specimens. Pectoral rather long, extending to 
base of pelvic in small specimens, not nearly to pelvic in large ones, 1.2— 1.4 in head. 
Axillary scale of pectoral rather small, only about a third of the fin length, 2.7—3.4 
in head. 

Color. Preserved specimens bluish silvery above, sides anti lower parts bright to 
pale silvery. Pectorals and pelvics pale; other fins all more or less dusky. Longley 



Fishes of the Western North Atlantic 115 

remarked, "The large eye falls below the level of the dorsal green and its iris blends 
with the silvery side. As seen by the diver this great fish appears unsubstantial, a gray 
ghost floating in gray water" {^g: 5). 

Size. The Tarpon reaches a very large size; an individual a little over eight feet 
having an estimated weight of 350 pounds has been reported for Hillsborough 
River Inlet, Florida. A fish up to six feet in length, having a weight of perhaps more 
than 100 pounds, is not very rare. 

Development. The exceedingly large number of eggs produced by a Tarpon indi- 
cates extreme prolificness. For example, a female 80 inches long, weighing 142 pounds 
(caught and reported by Babcock, I [1936]: 41), contained 12,201,984 eggs according 
to a close estimate made by John T. Nichols of the American Museum of Natural His- 
tory. This great number far surpasses the classic example of the prolificness of the 
large codfish often mentioned, which contained 1,839,581 eggs. 

Ovarian eggs, taken from both unripe and spent fish, measured 0.6-0.75 "^'"^ '" 
diameter, a small size for so large a fish. They sank slowly when placed in sea water, 
suggesting that they may be demersal, not buoyant. 

Tarpon eggs that have been spawned naturally and are positively identifiable as 
such have not yet been described, nor have the early stages after hatching. But it 
has been assumed from analogy with its smaller relative {Megalops cyprinoides) of the 
East Indies that the Tarpon, like the ladyfish {Elops saurus, p. 124) and the bonefish 
{Albula vulpes^ p. 134), passes through a ribbon-shaped leptocephalus stage during its 
larval development. This assumption has been corroborated by the capture at Beaufort, 
North Carolina, of a leptocephalus 20 mm long, near the transition stage; this, from 
comparison with the corresponding growth stage in Elops and Albula (jS- 45) ^nd in 
the East Indian Megalops from Java, probably was a Tarpon.'' (One i 8 mm long, re- 
cently pictured and described by Gehringer, certainly was [27: 290]. — H.B.B.) 

The smallest fry definitely reported as Tarpon have ranged between 16 and 19.6 
mm SL (j2: 6, fig. 5, pi. i; see also pi. 2 for fry up to 36-38 mm; for earlier ac- 
counts of 37 and 77 mm fry, see JJ: 229 and 63: 72). Breder's painstaking study of 
extensive series of various sizes from the western coast of Florida has shown that little 
alteration takes place in the body proportions with subsequent growth. 

Rate of Growth. Analyses of the number of checks on the scales of Tarpon from 
the western coast of Florida, where it may be assumed that growth is interrupted during 
the cold months, suggest that in open waters it averages about 3,050—3,060 mm in 
length at one year, about 510-640 mm at two years, and about 1,270—1,525 mm at 
three years. Earlier scale studies (j [1936]: 68) suggest that a Tarpon averages about 
1,376 mm or four feet six inches when it is nine or ten years old, that a fish of 70 pounds 
is about 12 years old, and a fish of 100 pounds is perhaps 13-16 years of age. 

Tarpon kept in the old New York Aquarium grew more slowly than those re- 
ported above, from less than 500 to only 1,220 mm in five years, perhaps as a result of 

4. For an account of the leptocephalus stages of Elops and Albula, see pp. 125 and 136; for the East Indian Megalops, 
see especially Chadabaram and Menon {iS: 756-759) and Hollister (42: 449). 



1 1 6 Memoir Sears Foundation for Marine Research 

confinement (jj). Present indications are that a Tarpon does not ordinarily mature 
sexually until it is about four feet long, i.e. until after the sixth or seventh winter 
{13: 229-247). 

Spawning. Ripe or nearly ripe Tarpon have been captured at Boca Grande, Florida, 
June 5-22 (j [1936]: 41), and several fish containing large roe have been taken from 
the Panama Canal Zone during February, March, and April (jy. 25, 30, 2,S)- These 
observations indicate that at least some of them spawn there during the spring. Captures 
of both green and spent fish have shown that "the time of spawning on the west coast 
of Florida is from May to September" (j [1936]: 43; 13: 226). 

The general belief that spawning takes place along shore, mostly in shoal water, 
is borne out by Babcock's report that ripe females as well as males ejecting milt were 
caught near Boca Grande (j [1936]: 41, 43) and that he received an eyewitness account 
of pairs of Tarpon spawning in Charlotte Harbor, near Fort Myers, Florida. The 
presence of very young fish at points as widely distributed as Texas, Alabama, Florida, 
the Bahamas, Puerto Rico, Haiti, Cuba, and Trinidad seems sufficient evidence that 
the breeding range is extensive — along the shores of both continent and islands from 
the Gulf of Mexico eastward through the West Indies and Caribbean Sea areas. To 
what extent it breeds farther southward is not known. The well-known presence of 
young as well as large Tarpon in Lake Nicaragua also makes it likely that this fish 
spawns there in fresh water. No evidence, however, has been found that the Tarpon 
breeds in the lakes of the Panama Canal (j6: 240). It seems probable, therefore, that 
the ripe fish there migrates to sea to spawn, using the locks of the Canal as a passageway. 

Habits. The Tarpon does not school, at least not habitually like the menhaden, for 
example. When it does congregate in considerable numbers, as at the foot of the spill- 
ways of the dams in the Canal Zone, it does so presumably because food is abundant 
there. In other places it sometimes congregates in pursuit of schools of fish. The fish 
apparently is sensitive to cold water and will move away from it if possible. Even in 
southern Florida it disappears from the inshore waters during cool weather {34: 21 1). 
If it migrates out to sea, as it probably does at times to escape from cold water, it 
seems highly probable that it remains at or near the surface where the water generally 
is warmest and where it can utilize the modified air bladder for breathing. The Tar- 
pon more or less habitually ascends freshwater streams, and in some places, in the 
tropics at least, it occupies freshwater lakes more or less permanently (j6: 240; see 
also Habitat). 

The rolling movements of the Tarpon at the surface of the water are familiar to all 
fishermen. As these movements reveal its presence, anglers watch for this activity. 
Why the fish practices such movements habitually was not understood until it was dis- 
covered that it possesses an air bladder supplied with a large amount of lung tissue 
(j6: 246). After the fish has rolled at the surface and descended, bubbles appear at 
the surface, these presumably being air exhaled below the surface. 

It has been suggested by several writers that the great leaps often executed by 
the Tarpon seem to serve a different purpose from rolling movements: to escape from 



Fishes of the lVester?i Nortfi Atlantic 117 

enemies, to shake off remoras or irritating parasites, or to play. Also, it often executes 
great leaps when hooked, evidently trying to free itself; as most Tarpon fishermen 
know from experience, these efforts are frequently rewarded with success. 

Habitat. The Tarpon, a coastal fish, is seldom if ever seen more than a few 
miles out from the land. To be sure, it has been credited with extensive seasonal off- 
shore migrations, even across the Gulf of Mexico (p. i 18), but not with any definite 
supporting evidence. The Tarpon may be taken in either salt or brackish water, and 
not infrequently it lives in freshwater streams and lakes. It has long been known to 
inhabit Lake Nicaragua, a body of water a hundred feet above sea level, and more 
recently it has been reported as plentiful in the rapid San Juan River, via which the 
Lake discharges into the Gulf of Mexico (j [1936]: 19-20). This fish is exceedingly 
abundant, sometimes at least, in fresh or virtually fresh water at the base of the 
spillway of Gatun Dam and also in Gatun Lake below Madden Dam. It occurs 
regularly also in other parts of Gatun Lake and in the small freshwater lake be- 
tween Pedro Miguel and Miraflores locks (j6: 240). To reach these lakes the 
fish must pass through the locks of the Canal. The smaller fish also abounds in 
fresh and slightly brackish streams, as around southwestern Florida. The young fry 
is often reported as occurring in great numbers in very stagnant brackish pools and 
lagoons that were not connected to the sea when the collections were made {22: 80; 
J: 145; 5: 34; 12: 154). 

Migrations. Knowledge of this aspect of the Tarpon's life is scanty. It is probable 
that the great majority of the larval Tarpon passes through the metamorphosis close 
to the land, and many do so in estuaries, such as the Grande Anasco River, Puerto 
Rico {21: 146), and the Indian River, Florida (j2 : i). However, some drift so far off- 
shore that they almost certainly are lost to the Tarpon population: one, for example, 
was taken about 150 miles to the east of Brunswick, Georgia, well out in the Gulf 
Stream (27: 236). 

The reported scarcity of fish less than three feet or so off the open coast, con- 
trasted with their abundance in estuarine situations, even in drainage ditches, suggests 
that most of them tend at first to work inshore. Indeed, great numbers of small fry 
have repeatedly been encountered in brackish pools in mangrove swamps, in landlocked 
lagoons, and in small pools around Florida, Puerto Rico, Haiti, Jamaica, and on An- 
dros Island in the Bahamas, access to which is possible only during periods of very 
high water. Many of the fish that are trapped in this way are no doubt lost to the 
population when the pools dry up. (For an interesting summary of recorded instances 
of this sort, see Babcock, i [1936]: 24, 32-46.) But many of them probably find their 
way out again. It is only in situations of this sort that the Tarpon fry has ever been 
taken in any great numbers, at least around Florida (information from Dr. J. E. Ran- 
dall). It is to be found in such situations around southwestern and southern Florida 
throughout the year. The postlarval stages have recently been taken in such numbers 
in the salt marshes of the Indian River on the eastern coast of Florida as to show that 
this is an important center of dispersal for them (j2: 9). Little Tarpon i— 1.5 feet 



1 1 



8 Memoir Sears Foundatmi for Marine Research 



long are very plentiful in the narrow headwaters of brackish and freshwater streams, 
and even in drainage ditches. But the growing fish tends to move down to the broader 
waters as it grows. 

There is no apparent reason to credit the older Tarpon with any extensive 
migrations, whether inshore-offshore or along shore, in the southern part of the range, 
for large fish are found there within rivers and along the open coast at all times 
of the year. 

In the northern part of its range, however, the large Tarpon drops out of sight 
with the onset of autumn, not to reappear until the following March or later, according 
to locality. It has often been credited with extensive seasonal migrations, "northward 
along the Florida coast and down the Texan coast, and back across open waters" 
{13: 234). But a more plausible explanation is that the autumnal disappearance of 
the Tarpon from the open coasts of the northern-northeastern periphery of the Gulf 
of Mexico represents nothing more extensive than a temporary withdrawal; it probably 
moves far enough offshore and deep enough to escape winter chilling, such as killed 
many of all sizes along the western coast of Florida in 1885, 1894—95, ^9°5' ^^^ '935 
{74: 123; I [1936]: 21; 66: 640). Appearance of the Tarpon off North Carolina in 
summer and autumn (as happens more or less regularly), and even farther north 
(p. 120), evidently represents so extensive a journey from the nearest wintering ground 
available for a warm-water nonoceanic fish that it seems unlikely that any of those 
ever find the way back to their southern home. — H.B.B. 

Food. The Tarpon, so far as known, is wholly carnivorous. A specimen nearly six 
feet long from Boca Grande, Florida, examined by me, had the remains of six marine 
catfish in its stomach; and a 12-inch specimen from the same place had ingested a 
silverside. The Tarpon has been reported as feeding almost exclusively on cutlassfish, 
e.g. in Calcasieu Pass, Louisiana (7J: 170—172). At times it gathers in great abun- 
dance at the foot of the spillway of Gatun Dam in the Canal Zone and at the spillway 
of Madden Dam in the upper part of Gatun Lake, presumably for the purpose of 
feeding on small fish that are carried over the spillways. Various observers have 
reported that it will follow schools of mullets, anchovies, and other fish, apparently 
feeding on them at will. Young Tarpon, ranging from about two to eight inches in 
length, taken in an isolated lagoon at Port-au-Prince, Haiti, had fed exclusively on 
"aquatic hemiptera which filled the waters of the lagoon" (5: 36). Blue crabs con- 
stitute a favorite bait in Florida. 

Enemies. The chief natural enemy of the large Tarpon is undoubtedly the shark, 
and it has been reported that the porpoise attacks it too (j [1936]: 64); the small 
Tarpon quite certainly falls prey to many predators in the sea. The Tarpon's chief 
protection from enemies is apparently its great swimming speed, whereby it escapes. 

Variation. Whether geographical variations exist can be determined only from 
study of a larger number of specimens, collected from many localities, than have been 
available to me. 

Relation to Man. As a gamefish, the Tarpon scarcely is surpassed. Its reputation 



Fishes of the Western Nort/i Atlantic 119 

is world-wide and mucli has been written about the requirements for capturing and 
subduing this powerful, hard-fighting fish. Its great strength became evident to the 
writer one day when he was helping to land a Tarpon 80 inches long. A man standing 
nearby saw the great fish leap ashore from the seine, and wishing to prevent it from 
jumping back into the water, he threw himself on it, evidently believing that he had 
enough strength and weight to manage the fish. However, with one stroke of the tail the 
fish lifted the man, who weighed at least 150 pounds, and threw him a distance of about 
12—15 '^^^^- It is understandable, then, why considerable strength, skill, patience, and 
determination are required to land a large Tarpon and why Tarpon fishing is exciting. 

About angling for Tarpon, Miss Francesca LaMonte has provided the following 
information (personal communication). The hard-fighting Tarpon, either large or 
small, is a prized catch wherever it occurs in western Atlantic waters, from Nova 
Scotia to Argentina, or in the eastern South Atlantic off Lagos, Nigeria, and at the mouth 
of the Volta River, Ghana. The most famous western North Atlantic grounds for both 
numbers of fish and facilities are: the Florida Keys, centering around Islamorada and 
Marathon ; the western coast of Florida, especially in the passes near Boca Grande and 
Useppa Island; the Rio Panuco, Mexico; and the Rio Encantado, Cuba. 

The number of Tarpon caught is always small in comparison with the number 
present, for the fish puts up a violent and often successful fight. A large one is usually 
fished from outboard motor boats (with the motor cut) by casting, trolling, drifting, 
or still fishing, on i 2-pound line or less, with plugs or with either live or dead bait, and 
with or without a drag on the reel. The first problem is to set the hook in the Tarpon's 
hard mouth, then to prevent the fish from throwing it or breaking the line, both of 
which it will try to do. First it will make a fast, powerful run and a series of high, twisting 
leaps, shaking its head with violence. Failing to throw the hook, it will then try to break 
the line either by charging at the boat if the line is too tight, or, if too loose, taking off 
into the mangroves where the line will eventually break on the roots. The small young 
Tarpon is caught in rivers and canals on fly-fishing or spinning tackle with light mono- 
filament line. Though small, it fights just as hard as the adult and is by no means 
sport for an inexperienced angler. 

There is wide diversity of opinion as to the edible qualities of Tarpon flesh. On 
our own shores "large ones at least are not edible" (Babcock, J [1936]: 63), but in 
Panama it is highly esteemed by the natives and West Indian immigrants (j6: 239); 
there the size of the fish receives no consideration, for an 80-inch Tarpon seemed as 
acceptable as a small one; as stated by me in the paper just cited, a small fishery for 
Tarpon was in operation in Gatun Lake in 1935. ^^ '^^^ been reported that it is highly 
prized as food by the Indians of the Caladonia Bay (Panama) region (jo: 152). In 
the United States the commercial value of the Tarpon is so slight that it has not 
been listed in the statistical reports of the commercial fisheries by the former Bureau 
of Fisheries or by the present U. S. Fish and Wildlife Service. 

Range and Abundance. The regular range of the Tarpon extends from North 
Carolina to central Brazil, and it is probable that the Tarpon frequents the Brazilian 



1 2 o Memoir Sears Foundation for Marine Research 

coast southward to Pernambuco, or even to Bahia,^ though the most southerly 
definite record is for Lake Papary, Rio Grande do Norte, in the general vicinity of 
Natal {6^: 8). Thence northward the range is evidently unbroken around the South 
American coast, where it has been recorded for: Marajo Island at the mouth of the 
Amazon {8\ 298); French Guiana {^y. 164); Trinidad, where it has been reported 
as very plentiful {6(): 26); and Lake Maracaibo, Venezuela, and its tributary Rio Con- 
cha, where several were seen rolling and a school was reported (62: 33). It is equally 
omnipresent in suitable situations all along the Central American littoral, including: the 
Panama Canal to Gatun Lake ; the San Juan River (p. 117); and Lake Nicaragua, 
where it is so plentiful that a recent writer reported seeing "more than a dozen 
rolling at the surface at one time" (52: 184). As every angler knows, the Tarpon 
abounds near river mouths and in so-called "passes" along the coasts of Mexico and 
Texas. Indeed, it is not unusual to find a Tarpon 80 or 100 miles up from the sea in 
some Mexican rivers (jj: 219). In fact, the Bay of Campeche, just west of Yucatan, 
was the site of a more or less regular fishery for it as early as about 1675 (Capt. Wm. 
Dampier, cited from Gowanloch, 29: 145-146). Proverbial centers of abundance are 
the mouth of the Panuco River, near Tampico, Mexico, and Aransas Pass, Texas. 
The Tarpon, as Gowanloch (29: 154) expressed it, is to be found near almost any 
coastal island or bay of Louisiana, Alabama, the western coast of Florida, and the 
eastern coast of Florida, northward in summer about to the vicinity of Daytona. Espe- 
cially renowned fishing grounds are: the Calcasieu River, some 30 miles east of the 
Texas boundary; Grand Isle, Louisiana; Boca Grande and Captiva, western Florida; 
the waterways and rivers among the Thousand Islands, and the interior waterways of 
the Florida Keys (see also p. 119). 

The Tarpon is widespread but seemingly not very abundant except perhaps locally 
throughout the West Indian region in general. More than a century ago it had been 
reported in scientific literature for Guadeloupe, Martinique, Santo Domingo, and 
Jamaica {20'. 399). It has been reported for Haiti (j: 143) and is common around 
the Barbados and Puerto Rico. In some rivers of Cuba it is so abundant that a party 
of anglers has reported hooking 105 large ones in a day in the Rio Encantado (jj: 
205). Among the Bahamas a few are seen, and an occasional one is taken off Bimini 
(5J: 207; also personal communication from Vladimir Walters) ; it is common around 
Andros Island (jj : 66), and is taken or seen occasionally around Bermuda. 

The Tarpon ranges in small numbers northward to North Carolina in some sum- 
mers, and perhaps every summer; but large schools are sometimes seen as far north 
as Cape Lookout. A five-foot fish has been taken off Hatteras, and the North Carolina 
State Museum has one of 1 19 pounds from Beaufort, and another of 176 pounds from 
Wilmington (6^: 115). But it appears only as a straggler and at irregular intervals 
farther north. It has been reported for the lower part of Chesapeake Bay by fisher- 
men (j5: 80); there are odd records of it for New Jersey, near Sandy Hook (2j: 

5. The first published account of the Tarpon, by Marcgrave {50), was probably based on examples from Recife (Per- 
nambuco) or Salvador (Bahia). — G. S. Myers. 



Fishes of the JVestern North Atlantic 1 2 i 

264), for the eastern end of Long Island, New York {^5: 33), and for Rhode Island 
{68: 72). The Tarpon has been reported as being taken in small numbers almost 
every year in the vicinity of Woods Hole, Massachusetts ipj: 74 1). But the Tarpon 
has been reported only once for the Gulf of Maine, when a specimen was taken at 
Provincetown at the tip of Cape Cod on July 25, 1915 (6: 91). Northward from Cape 
Cod there are only five records of it, all for Nova Scotia — one from Isaac Harbor 
(jj: 45), one from Harrigan Cove {jO\ (,"1)^ one of about three feet caught near Ter- 
rence Bay, September 6, 1947, and two taken near Halifax on August 3—4, 1953 
(information from Dr. A. H. Leim). The Tarpon is common on the coast of tropical 
West Africa {2^: 154; 45: 106; i [1951]: 18-19). 

Synonyms and References:" 

Cltipea gigantea Shaw (in part), Genl. Zool., 5 (i), 1 804; 173 (no type or type local, design., diagn., genl. acct., 
habitat).' 

Megalops giganteus Voigt in Cuvier, Das Thierreich (German transl.), 2, 1832: 423 (refs.; cf. Amer. M. giganteus 
and east. M. filamentosus Lacepede); "Disciples' edition" of Cuvier's Regne Anim., I, 1836: 559 
(same as above with somewhat more extended descr.); Eng. ed., 10, 1834: 442 (design. Amer. species 
as M. cyprinoides, with ref. to Bloch's pi. 403 ; pelvics under anterior part of dorsal are as in Megalops 
cyprlnoides Broussenet as now understood, but 15 rays in dorsal suggest Tarpon). 

Megalops atlantkus Cuvier and \'alenciennes. Hist. Nat. Poiss., ig, 1846: 398 (orig. descr., type local. Guade- 
loupe, Santo Domingo, Martinique, Puerto Rico; types in Paris); Poey, Repert. Fisico-Nat. Cuba, 2, 1868 : 
423 (diagn., brackish water of Cuban rivers, attains 5 feet); An. Soc. esp. Hist. Nat., 4, 1875 : 146 (Cuba, 
Jamaica, Puerto Rico, Santo Domingo, Martinique, Trinidad, Mexico, British Guiana); in Gundlach, 
An. Soc. esp. Hist. Nat., JO, 1881; 343 (Puerto Rico, Cuba, Santo Domingo, Martinique, Mexico); 
Henshall, Bull. U.S. Fish Comm. (1889), 9, 1891 : 385 (effect of cold water); Boulenger, Ann. Mag. nat. 
Hist., (6) 20, 1897: 298 (Marajo I., Brazil); Gill, Smithson. misc. Coll., 48, 1905: 35 (descr., habitat, 
game qualities, life hist., common names, etc.); Regan, Ann. Mag. nat. Hist., (8) 5, 1910: 354 (caudal 
skel. cf. M. cyprinoides) \ Fowler, Proc. biol. Soc. Wash., JJ, 1920: 147 (Cape May, Monmouth and 
Ocean counties. New Jersey); Proc. Acad. nat. Sci. Philad., 80, 1929: 609 (Barnegat Inlet, .'Atlantic 
Highlands, and Atlantic City, New Jersey); Copeia, 193 1 : 46 (sport fish, Port Aransas and Point Isabel, 
Texas); Bull. Amer. Mus. nat. Hist., "jo (l), 1936: 154, fig. 61 (African recs.; descr. based on Amer. 
specimens); Puyo, Bull. Soc. Hist. nat. Toulouse, jo, 1936: 65, 163 (diagn., habitat in French Guiana); 
Cadenat, Rev. Trav. Off. Peches marit., 10, Fasc. 4, Liv. 4, Pt. 2, 1937: 441, figs. 5, 6 (refs., remarks); 
Norman and Fraser, Giant Fishes, 1938: 93, fig. 34 (semipopular acct.); Puyo, Faune Emp. Franj., 
12, Poiss. Guijane Fran?., 1949: 151, fig. 75 (descr., Fr. Guiana). 

Megalops elongatiis Girard, Proc. Acad. nat. Sci. Philad. (1858), 1859: 224 (orig. descr., type local. Long Island, 
New York; type not preserved). 

Megalops thrissoides Giinther (not of Bloch and Schneider, which probably is M. cyprinoides). Cat. Fish. Brit. 
Mus., 7, 1868: 472 (diagn., Jamaica, Cuba, Trinidad, Brit. Guiana); Goode, Bull. U.S. nat. Mus., 5, 
1876: 68 (rare in Bermuda); Yarrow, Proc. Acad. nat. Sci. Philad., 1877: 215 (Ft. Macon, North 
Carolina); Steindachner, Denkschr. Akad. Wiss. Wien, jp, 1878: 69 (Magdalena R., Colombia); Jordan 

6. The Tarpon has been discussed in so many popular books and periodicals that it does not seem practical to list all 
references to them. Therefore, only some popular accounts are cited, and onlv those citations in scientific periodicals 
that seem to contain substantial information. African records have been omitted except for references to Fowler 
(25) and Cadenat {14), whose publications contain important references to African literature treating the species. 

7. This name was based by Shaw upon Marcgrave {so)-> who dealt entirely with the Adantic Tarpon, and upon Clupea 
cyprinoides of Bloch, which is a composite of the Tarpon and the Pacific Megalops cyprinoides (Broussonet). Since 
Shaw's account is thus a composite, it devolved upon a subsequent reviser to assign the name gigantea to one of the 
included species. This was done by Cuvier [ig: 324), who definitely assigned the name gigantea to the American 
Adantic species. It would thus appear that the correct name of the Tarpon should be Tarpon giganteus (Shaw), 
but the name atlanticus is so well established that its conservation, under suspension of the International Rules, 
would seem to be worthwhile. — G. S. Mvers. 



12 2 Memoir Sears Fou?zdation for Mari?te Research 

and Gilbert, Bull. U.S. nat. Mus., l6, 1883: 262 (descr., distiib.); Willcox, Bull. U.S. Fish Comm. 
(1886), 6, 1887: 123 (effects of cold); Henshall, Bull. U.S. Fish Comm. (1894), 14, 1895: 211 (abund. 
in Florida; food and game qualities; size; etc.). 

Megalops — Gill and Bransford, Proc. Acad. nat. Sci. Philad., 29, 1877: 187 (Tore Rapids, below Lake 
Nicaragua; eaten by natives). 

Tarpon atlantkus Jordan and Evermann, Bull. U.S. nat. Mus., 47 (r), 1896: 409 (descr., range, synon.); 47 {if), 
1900: fig. 177; Smith, Bull. U.S. Fish Comm. (1897), ij, 1898: 90 (Woods Hole, Massachusetts); 
Evermann and Kendall, Rep. U.S. Comm. Fish. (1899), 1900: 54 (Florida); Linton, Bull. U.S. Fish 
Comm. (1899), ig, 1901 : 437 (parasites); Evermann and Marsh, Bull. U.S. Fish Comm. (1900), 20 (i), 
1902: 80, fig. 10 (descr., small specimens from Puerto Rico); Gregg, Where to Catch Fish. E. Cst. Fla., 
1902: 31, fig. (descr., fishing instruct., nonedible or barely so); Eigenmann, Bull. U.S. Fish Comm. 
(1902), 22, 1903: 22Z (20— 192 mm specimens, no descr., Pinar del Rio, Cuba); Bean, Bull. N.Y. St. 
Mus., 60, Zool. 9, 1903: 177 (synon., descr., New York); Hargraves, Fish. Brit. Guiana, 1904: 4 
(local name CufFum; used but not appreciated as food; fresh and brackish water); Smith, N.C. geol. econ. 
Surv., 2, 1907: 114, fig. 36 (synon., descr., feed., range. North Carolina); Meek, Field Mus. Publ., 
Zool., 7, 1907: 99, in (Lake Nicaragua); Fowler, Proc. Acad. nat. Sci. Philad. (1910), 62, 191 1 : 599 
(New Jersey); Eigenmann, Mem. Carneg. Mus., 5, 1912: 444 (Georgetown, Brit. Guiana); Sumner, 
Osburn, and Cole, Bull. U.S. Bur. Fish. (1911), J-T (2), 1913: 741 (Woods Hole, Massachusetts); 
Cockerell, Bull. U.S. Bur. Fish. (1912), J2, 1913: 122 (scales descr.); Gudger, Proc. biol. Soc. Wash., 
26, 1913: 104 (North Carolina); Halkett, Check List Fish. Canada, 1913: 45 (Isaac's Harbor, Nova 
Scotia); Starks, Stanf. Univ. Publ., Univ. Ser., 1913: 8 (Lake Papary, Rio Grande do Norte, Brazil); 
Fowler, Proc. Acad. nat. Sci. Philad., 6^, 1915: 245 (various Florida locals.; color relative to habitat; 
cleaning of "spawning beds"); also 522 (Trinidad, B.W.L); Coker, Copeia, 1921: 25 (small ones, 
Dauphin L, Alabama); Eigenmann, Mem. Carneg. Mus., 9, 1922: 179 (fresh water, Magdalena R., 
Colombia); Meek and Hildebrand, Field Mus. Publ. Zool., 75(1), 1923: 174 (synon., descr., range, 
Panama); Schroeder, Rep. U.S. Comm. Fish. (1923), Append. 12, 1924: 4 (commerc. val., Key West, 
Florida); Bigelow and Welsh, Bull. U.S. Bur. Fish. (1924), 40 (i), 1925: 91, fig. 38 (descr., range, rare 
in Gulf of Maine); Breder, Zoologica, N.Y., 4 (4), 1925: 140, 152 (numerous at spillway, Gatun Dam, 
C. Z.; enters Caladonia Bay, Panama, in schools; highly prized as food by local Indians); Nichols and 
Breder, Zoologica, N.Y. (1926), 9 (l), 1927: 33, fig. (distr., life hist., size); Beebe, Bull. N.Y. zool. 
Soc, JO (5), 1927: 141 (young adults from brackish lagoon, Haiti); Hildebrand and Schroeder, Bull. 
U.S. Bur. Fish. (1927), 43(1), 1928: 80, fig. 41 (s}-non., descr., range, Chesapeake Bay); Beebe and 
Tee-Van, Zoologica, N.Y., 10 (i), 1928: 33, fig. (descr., Port-au-Prince, Haiti); Whiton and Townsend, 
Bull. N.Y. zool. Soc, jl, 1928: 170 (feed., food in Calcasieu R., Cameron Parish, Louisiana): Breder, 
Field Bk. Mar. Fish. Atl. Cst., 1929: 59, fig. (diagn., distr., habits, food value, size); Jordan, Manual 
Vert. Anim. NE U.S., ed. 13, 1929: 37 (diagn., range, size); Nichols, N.Y. Acad. Sci., JO (2), 1929: 
198, fig. (range, Puerto Rico; diagn., habits); Jordan, Evermann, and Clark, Rep. U.S. Comm. Fish. 
(1928), 2, 1930: 29 (common names, range, synon.); Burr, Bull. Texas Fish Game Comm., No. ;, 
1932: 36 (numbers at Port Aransas and Point Isabel, Texas); Beebe and Tee- Van, Field Bk. Shore Fish. 
Bermuda, 1933: 33, fig. (descr., uncommon in Bermuda); Breder, Bull. N.Y. zool. Soc, 34, 1933". 
65-67 (small ones from pond, Andros I., Bahamas; food); Gregory, Trans. Amer. philos. Soc, 2J 
(2), 1933: 137-142, figs. 31-34 (skull descr., cf E/o/>s and prehist. relatives); Storey and Perry, 
Science, 7J (2022), 1933: 284 (young adults, Sanibel I., Florida, and elsewhere); Gowanloch, 
Bull. La. Conserv. Dep., 21, 1932: 8, 54-65 (genl. acct.); Bull. La. Conserv. Dep., 23, 1933: 144- 
168 (account of 1932 enlarged); Breder, Zoologica, N.Y., 18, 1934: 58, 59 (fresh water, Andros 
I., Bahamas); Hildebrand, Copeia, No. l, 1934: 45 (larval Tarpon descr.); Beebe and Hollister, Zoolo- 
gica, N.Y., Jp (6), 1935: 211 (from Union I., Grenadines, B.W.L); Vladykov and McKenzie, Proc. 
N.S. Inst. Sci., ig (i), 1935: 53, fig. 26 (two Nova Scotian recs.); Babcock, The Tarpon, ed. 4, 
1936: 1-175, illustrated (good genl. acct. and much scientific data on anat., habitats, breed., respir., 
etc.); Hollister, Zoologica, N.Y., 21 (4), 1936: 263 (caudal skel.); and 24 (4), 1939: 449-475 
(caudal skel. and other characters cf. Mega/ops cyprinoides); Miller, Forest and Stream, 41 (i), 1936: 
32-33 (fishery in Rio Frio, tributary of L.Nicaragua); Gudger, Amer. Mus. Novit., 944, 1937: i, 
fig. 2 (albino descr.); Heilner, Salt Water Fishing, 1937: 195-210 (genl. acct.; abund. in Cuba 
and elsewhere); Hildebrand, Sci. Month., 44, 1937: 239-248, 4 figs. (Panama Canal; locally prized as 



Fishes of the Western North Atlantic • 123 

food by natives); Kaplan, Big Game Angler's Paradise, 1937: 89-136 (genl. acct.); Storey, Ecology, 
18, 1937: 12 (always hurt by freezes); Breder, Bull. N.Y. zool. Soc., 41, 1938: 24 (irreg. in New York 
Harbor); 42, 1939: 154-155 (habitat of small ones at Sanibel I., Florida); Butsch, J. Barbados Mus. 
hist. Soc, 7(1), 1939: 18 (common locally); Hcilner, Bk. of Fishes, Nat. Geogr. Soc, 1939: 202, 
327—336 (color plate, genl. acct.); Hildebrand, Zoologica, N.Y., 24 (i), 1939: 19, 25, 30, 35 (in locks 
of Panama Canal; transit through Canal completed); Longley and Hildebrand, Publ. Carneg. Instn. 
Wash., 535, 1941 : 5 (feed.; descr., especially as seen by a diver; Tortugas, Florida); Breder, Zoologica, 
N.Y., 27(1), 1942: I (behavior and respir. in confinement, refs. to papers on same subject); Breder, 
Zoologica, N.Y., 29, 1944: 217-252 (habitat, body proportions, growth, movements, refs.). 



Genus Elops Linnaeus 1766 

Ladyfish 

Elops Linnaeus, Syst. Nat., ed. 12, 1766: 518; type and only included species, Elofs saunis Linnaeus 1766. 

Generic synonyms: 

Argentina, in part, Linnaeus, Syst. Nat., ed. 12, 1766: 519, for A. Carolina Linnaeus. 

Mugilomorus Lacepede, Hist. Nat. Poiss., 5> 1803: 398; type and only included species, Mugilomorus arifia- 

carolina Lacepede equals Elops saurus Linnaeus. 
Trachonotus Rafinesque, Anal. Nat. ..., 181 5: 88; type species Mugilomorus anna-carolina Lacepede; Tracho- 

notus proposed as a substitute for Mugilomorus Lacepede. 

Characters? Body slender. Vertebrae about 75. Pseudobranchiae large. Bran- 
CHiosTEGALs about 34. MouTH nearly horizontal and almost terminal. Lateral line 
straight, with simple tubes. Scales small, not especially primitive (fully described by 
Cockerell, 75: "},% ly. 122), forming a sheath on base of dorsal and anal fins. Dorsal 
rather high anteriorly, the last ray not prolonged. Anal similar to dorsal but smaller, 
placed far behind dorsal. Pelvics inserted near vertical from origin of dorsal. Last several 
segments of spinal column directed upward toward base of upper lobe of caudal fin. 

Air bladder slender, with very transparent walls, occupying full length of ab- 
dominal cavity, adhering to abdominal wall dorsally and to alimentary canal ventrally. 
Alimentary canal, below air bladder, consisting of a straight tube except for a large 
elongated blind sac that projects forward to throat below, and parallel to, the main 
canal; a mass of coeca (firmly bound together) attached to distal end of blind sac 
and forming a close-fitting shield on right side of sac throughout its length; one 
lobe of liver forming a hood over blind end of sac, and another lobe shielding left 
side of sac. 

Remarks. The gular plate distinguishes this genus from Albula. The very slender 
body, the small scales, and the absence of filaments attached to the last ray of the dorsal 
and anal fins distinguish it from Tarpon. 

Range. About seven species have been recognized, but only Elops saurus comes 
within the range of this work. The species are widely distributed in the warmer seas, 
entering estuaries and tidal streams. 

8. The characters of the skull have been discussed by Gregory (jo: 138), who regarded it as rather more primitive 
than that of the tarpon. The ganoid character of the caudal skeleton has been well described and correctly figured 
by Hollister (J9: 260-263). 



124 Memoir Sears Foundation for Marine Research 

Elops snurus Linnaeus 1766 

Ladyfish," Bigeye Herring, Tenpounder, Slcipjack, Macabi, Matejuelo Blanco, 
Matejuelo Real, Long John, Benane 

Figures 20, 21 

Study Material. Many specimens, including an almost complete growth series 
ranging from larvae to adults 550 mm long. 

Distinctive Characters. See Remarks, p. 123. 





^^,,,;.,^— ^>^^' 




Figure 20. Elops saiinis; U. S. National Museum specimen. Drawn by H. L. Todd. 



Description. Proportional dimensions in per cent of standard length, based on 
study specimens. 



Body: depth 14. 5-19. 3. 
Head: length 25—31. 
Snout: length 5.8-7.0. 
Eye: diameter 5.0—6.3. 
Maxillary: length 14. 5-1 7.8. 
Anal fin: length of base 9.8-1 1.8. 
Pectoral fin: length 14-16. 



Scales: 103-120. 

Branchiostegals : 26—32. 

Gill rakers: 6-8+ 10-15 (excluding ru- 
diments). 

Fin rays: dorsal 21—25; '^^'^ I4~i7; 
pectoral 16 or 17. 

Vertebrae: 73-80 (10 specimens). 



Body very slender, only moderately compressed, its depth 5.2—6.9 in SL. Scales 
small, thin, with crenulate membranous border (fully described by Merriman, 5/: 
113). Lateral line nearly straight, extending onto base of caudal. Head low, flat 
above, 3.25-4.0 in SL. Snout rather long, usually a little longer than eye, 4.1-4.75 
in head. Eye near upper profile, wholly in anterior half of head, 4.55-5.2 in head. 
Interorbital flat, 5.7-8.4 in head. Gill rakers about 67% of (^he eye, scarcely 
denticulate, their entire surface rough, generally i or 2 rudiments in large specimens 

9. In scientific literature, as well as in earlier popular books on American game fishes, the name Ladyfish was usually 
applied to Albula 'uulpes (p. 134). But to the angler, who alone takes much interest in either of them, "Ladyfish" 
means Elops while Albula is universally termed "Bonefish." 



Fishes of the Western North Atlantic 125 

and 2 or 3 in small ones, on upper and lower limbs of first arch; the lower counts 
given, for this reason, are invariably based on the smaller specimens. Mouth nearly 
horizontal, almost terminal, the lower jaw slightly included. Maxillary reaching 
somewhere under posterior half of eye in young adults, far beyond eye in large exam- 
ples, 1.7—2.0 in head. 

Dorsal fin rather high anteriorly, its longest rays usually reaching tip of last ray 
if deflexed, its outer margin rather deeply concave, its origin about equidistant between 
base of caudal and anterior margin of eye. Caudal very deeply forked, the upper lobe 
rather longer than the lower one, usually about as long as head. Anal moderately small, 
far behind dorsal, its origin usually about equidistant between base of caudal and base 
of pelvic, its base 2.4—2.9 in head. Pelvic fin not much smaller than pectoral, inserted 
under, or slightly in advance of, origin of dorsal, scarcely nearer to origin of anal 
than to base of pectoral. Pectoral fin rather broad, not falcate, reaching scarcely 
halfway to pelvic in large examples, somewhat more than halfway in small ones, 1.75— 
2.0 in head. Axillary scale of pectoral extending to about the midlength of fin, 
3.0—3.9 in head. 

Color. Preserved specimens uniform bluish gray above, silvery elsewhere; dorsal 
and caudal more or less dusky; other fins pale. 

A large fresh specimen bluish on back, with sides silvery; slightly yellowish below. 
Dorsal and caudal dusky and yellowish; pelvics and pectorals yellowish with dusky 
punctulations (j<§: 78). 

Size. Although a maximum of 900 mm (36 in.) TL has been reported (^5: 410), 
the usual length probably is under 500 mm (20 in.). The largest individual seen by 
me was 730 mm (610 mm SL; 29.2 in.); this large fish occurred among hundreds of 
adults stranded in the Gatun Locks of the Panama Canal in 1935 when the Locks were 
drained. In my file is a record of another large one, 718 mm (28.75 ^"0) caught at 
Punta Gorda, Florida. Although one of the common names is Tenpounder, weight 
records indicate that this fish probably never weighs that much. A specimen 470 mm 
long (18.8 in.) weighed a pound, another of 550 mm (22 in.) weighed 1.6 pounds, 
and still another 625 mm (25 in.) weighed 2.7 pounds; 82 fish, ranging between 400— 
600 mm (16—24 in.), with an average length of about 450 mm (18 in.), weighed 
100 pounds. 

Development. The eggs and the earliest stages of the leptocephalus are unknown. 
Of the many leptocephalus larvae studied, the youngest ones, judging principally by 
the development of the fins, are 34, 35, and 37 mm long. Although the forked 
caudal is well developed, the other fins are virtually undifi:'erentiated ; the alimentary 
canal is very loosely attached to the abdomen, as in larval herrings and anchovies, and 
the vent is less than a head-length from the caudal base. 

Two leptocephali, 34.5 and 36.5 mm long, are slightly more advanced; in these 
a thickening within the finfold where the bases of the dorsal and anal are developing 
has taken place, though rays are not yet definitely formed. The head is strongly depres- 
sed, the mouth is terminal and large, and the snout, viewed ventrally, is rather sharply 



126 



Memoir Sears Foundation for Marine Research 



triangular; these characters at once distinguish the leptocephalus of this species from 
that oi Alhula vulpes. The myomeres are mostly quite distinct, 78 having been counted 
in one of the specimens. A row of faint dark spots, which are much more distinct in 
older leptocephali, is present on each side just above the alimentary canal. 

Three leptocephali, 42, 42, and 44 mm long, are the longest larvae in the collec- 
tions studied and may represent about the maximum length attained in the leptoceph- 
alus stage. Here development has progressed somewhat beyond that of the specimens 
described above, for the rays are evident now in the dorsal and anal fins, though an 





Figure 21. EIops sauriis. above, leptocephalus, 35 mm long; below, young fish, 20 mm TL, recently trans- 
formed. 

accurate enumeration cannot be made. The anal fin is removed from the base of the 
caudal by a distance shorter than the head, and the primitive membrane has remained 
between the two fins. The pectoral fins are represented by tufts of membrane and 
the pelvics merely by thickened places in the abdominal wall where their bases are 
developing. Seventy-seven myomeres were counted in one of the specimens. 

In a considerably older specimen, reduced in length to 27 mm, the body, though 
still quite flat, has become much more substantial than in the younger stages described. 
The dorsal and anal fins are sufficiently developed to permit an accurate enumeration 
of the rays, and the anal fin is now well separated from the caudal. The primitive mem- 
brane has disappeared behind the fin, though a short section of it remains in front. 
The pectorals are rather well developed and show signs of rays. However, the pelvics 
are undifi-'erentiated. 

A virtually young adult only about 1 6 mm long (not in good condition) represents 
the maximum "shrinkage" in length during metamorphosis, among specimens studied. 
Though the body has become rather robust, the head and especially the snout remain 
depressed. The fins, except for the pelvics, are rather well developed. The intestinal 
canal remains loosely attached to the abdomen as in younger fish, and the primitive 
rows of dark dots on each side of the canal are distinct. 

The state of development varies widely among the rather numerous young adults, 
18 mm and upward. For example, a specimen scarcely 20 mm long is fully as well 



Fishes of the JVester7i North Atlantic 127 

developed as others around 30 mm, and another specimen about 30 mm long is just 
as well developed as some between "^fC^ and 40mm. The small and exceptionally well 
developed specimens have advanced especially far in color. A specimen scarcely 20 mm 
long already has some of the silvery color of large adults and all the juvenile markings 
are missing, whereas a 32-mm specimen among the retarded ones is very pale and still 
retains the two rows of dark spots on the abdomen, as in the leptocephali. Even in 
the most retarded specimens, the fins are all well developed at a length of 22 mm, and 
the outline of the gular plate is visible under magnification. Scales begin to appear in 
at least some specimens at a length of 50 mm, and the teeth in the jaws are definitely 
in narrow bands. At a length of 60—65 mm, scalation generally is complete, and the 
fish are very similar to full-grown adults. 

The environment probably aft'ects the development profoundly, for the small, 
exceptionally advanced specimens described in the preceding paragraph were taken 
in brackish pools at Key West, Florida, whereas the notably larger and more retarded 
ones were taken at sea off Ocracoke and Beaufort inlets. North Carolina. 

Spawning. There is relatively little information about the actual time and place of 
spawning. However, in 1939 I reported {jy : 25) that 20 individuals, selected at random 
from among the many left stranded when the Gatun Locks (Panama Canal) were 
drained in February 1935, consisted of seven ripe or nearly ripe males (56.5-62 cm 
TL), and 13 females ((>2-S~l'i ^^^ TL) with large roe, indicating that spawning time 
in that vicinity was at hand. Although leptocephali of Albula have been taken in con- 
siderable seining during the winter and spring months on the Atlantic coast of Panama, 
no young of E. saurus have been reported. On the Pacific coast of Panama, leptocephali 
of the closely related E. affinis were taken during February, August, and "autumn." 
In the collections at hand there are leptocephali from Texas, mostly from the vicinity 
of Corpus Christi, collected in February, March, April, and November. Young adults 
were taken in Aransas Pass, Texas, during June, and at Key West, Florida, in March 
and November. Leptocephali are also at hand from the Florida Keys, taken in No- 
vember, and from Cuba, caught during May. For Beaufort, North Carolina, there is 
a note in my files that a nearly ripe female of 604 mm was taken on October 3, 1912; 
leptocephali were collected there during January, February, March, April, May, Oc- 
tober, November, and December; rather recently-transformed young adults were taken 
in the same vicinity during March, May, June, July, and August. Most of the many 
leptocephali studied were taken with tow nets on the bottom, at sea, in comparatively 
shallow water, though several captures were made in estuaries. The young adults, ex- 
cept for a few lots in the transition stage, were collected with seines in inside waters, 
partly in brackish ponds and pools. 

Habitat. This is an active little fish, often traveling in schools, frequently skipping 
at the surface, and jumping when hooked. Its usual habitat is shallow salt and brackish 
water. The young adult has been taken in brackish lagoons and in ponds and pools, 
perhaps more or less recently and temporarily separated from the sea. In the Gatun 
Locks (see above) it occurred in large numbers and was most numerous in the middle 



12 8 Memoir Sears Foundation for Marine Research 

chamber of the three flights, where the water, at least part of the time, is nearly fresh. 
No evidence of its entrance into the strictly fresh water of Gatun Lake was obtained; 
it seems improbable, therefore, that it uses the locks as a passageway; more probably 
the locks serve merely as a feeding ground. If it migrates to sea to spawn, as might be 
supposed since its young pass through a leptocephalus stage, the location of its 
spawning ground (or grounds) remains unknown. 

Food. Its diet has not been thoroughly studied, but the short alimentary canal 
and the well-armored mouth surely suggest a flesh diet. A dozen or so adults taken 
in one catch at Beaufort, North Carolina, had fed exclusively on shrimp, and other 
examples from the same vicinity had fed on them also {48: 352; 64: 116). It has 
been reported that an individual from Haiti had eaten a shrimp and two small fish of 
its own species (5: 33). A specimen taken in Connecticut had fed on two small fish, 
probably Menidia (5J : 113). According to Kaplan, the Ladyfish feeds on shrimp, 
shiners, and squid {4.6: 91). 

Parasites. The cestode parasite, Rhynchobothrium bulbifer Linton, has been found 
in the viscera of the adult Ladyfish {^48: 352). 

Relationships. This species is closely related to E. affinis Regan of the Pacific coast 
of tropical America, apparently differing only in the number of gill rakers: in 41 spec- 
imens of E. saurus examined, 9 had 6 rakers, 16 had 7, and 16 had 8, exclusive of 
rudiments, on the upper limb of the first arch; and in 88 specimens, 2 had 10 rakers, 
7 had 11,13 had 12, 28 had 13, 35 had 14, and 3 had 15 rakers, exclusive of rudiments, 
on the lower limb. In nine specimens of E. affinis examined, 4 had 10 and 5 had 1 1 
rakers on the upper limb, and in 26 specimens, 2 had 16 rakers, 4 had 17, 6 had 18, 
II had 19, and 3 had 20 rakers, exclusive of rudiments, on the lower limb. 

Elops saurus has been recorded for West Africa by Fowler (25: 155) and by 
others, but Regan {58: 38) has held that African specimens were distinct on the basis 
of fewer vertebrae — 68 or 69 compared to 78 or 79 for American specimens — and 
on the basis of fewer scales — 94-98 for African specimens and 102— 1 18 for American 
specimens. According to counts made by me, these differences seem to be constant. 
Ten American specimens have 73, 74, 75, 75, 75, 77, 78, 79, 80, and 80 vertebrae, 
and nine American leptocephali have 77, 78, 78, 78, 79, 80, 82, 82, and 82 myomeres, 
whereas one adult from Ashantee, Africa, has only 67 vertebrae. Counts of scales in 
22 American specimens resulted as follows: i specimen with 103, 2 with 105, i with 
106, 2 with 107, 2 with 108, I with 109, 2 with no, i with iii, 2 with 112, i with 
113, I with 114, I with 115, 2 with 116, i with 118, and 2 with 120. The three 
specimens from Ashantee, Africa, examined by me have 92, 97, and 100 scales. 
Although the range of counts for American specimens is rather great, it does not 
overlap that given by Regan for African specimens. On the basis of the above data 
given by Regan and me, the African representatives, which Regan named Elops senega- 
lensis, may still be considered distinct, though synonomized with E. saurus by Fowler. 

Variations. No geographical variation in the number of gill rakers has been dis- 
covered, but there appears to be a slight increase in number with age, which seemingly 



Fishes of the Western Nort/i Atlantic 129 

may be attributed to the development of some of the rudiments into rakers as the fish 
grows. Thus, in 29 specimens 150 mm SL and under, 5 had 5, 14 had 6, and 10 had 
7 rakers and 2 or 3 rudiments on the upper limb; among 20 specimens over 150 mm 
SL, 2 had 6, 4 had 7, and 14 had 8 rakers and i or 2 rudiments. Similarly, among 35 
specimens 150 mm SL and under, 5 had 10, 14 had 11, 11 had 12, and 5 had 13 rakers 
and 2 or 3 rudiments on the lower limb; among 32 specimens over 150 mm, 4 had 12, 
7 had 13, 18 had 14, and 3 had 15 rakers and i or 2 rudiments on the lower limb. 
Relation to Man. For its size, the Ladyfish affords good sport to the light-tackle 
angler, but it is virtually of no value as food, its meat being dry and bony. It is seen 
from time to time in the market at Colon, Panama, where it is known as "bonyfish" 
to the West Indian Negro immigrants who speak English; however, the demand for 
it is small. When the Gatun Locks were drained in 1935, hundreds of Ladyfish 
became stranded, and it was estimated that more than 1,000 pounds perished. While 
the employees and the native residents eagerly accepted tarpon, crevalle jack (Caranx 
hippos), and other species, the Ladyfish was entirely neglected and had to be buried 

{37 •■ 25)- 

Range and Abundance. Its range extends from Cape Cod to Brazil. The southern- 
most locality for which it has been recorded seems to be Rio de Janeiro (60: 90; 
26: 132). While E. saurus is not numerous in the West Indies, it is widespread in 
the Caribbean-Gulf of Mexico region, and is common along Atlantic Panama; the 
Study Material includes three specimens from the Gulf of Venezuela. It is common 
around Florida, especially in enclosed waters such as the Indian River and among the 
islands off the southwestern coast. The Ladyfish also occurs around Bermuda, though 
not in any great numbers. To the north, it is present irregularly off Beaufort, North 
Carolina, and it has been reported as common in the Chesapeake Bay region; farther 
to the north its appearances — always in autumn — are sporadic. Considerable numbers 
were taken near Sandy Hook in October 1923; for example, 13 out of one trap on 
the 19th {55: 2,'i)- It has been recorded for the eastern part of Long Island (Orient), 
New York, from October 6 to November 4. Also, it has appeared in considerable 
numbers at various localities in the Woods Hole region in some autumns {6y: 741), 
but there is no dependable record for a Ladyfish north of the elbow of Cape Cod — 
there is no knowing whether one taken at Chatham, Massachusetts, in 1888 (7: 139) 
was from Nantucket Sound or the Gulf of Maine shore of Cape Cod. Nor is there 
any possibility of determining whether an "Albula vulpes," taken at Black's Harbor 
(Charlotte County, New Brunswick) and in poor condition when examined (jj: 45), 
was actually a Bonefish or Ladyfish. 

Synonyms and References: 

Elops saurus Linnaeus, Syst. Nat., ed. 12, 1766: 5 r8 (orig. descr.; type local. "Carolina"; type, a dried skin in 
BMNH; see Goode and Bean, Proc. U.S. nat. Mus. [1885], 8, 1886: 205); Bloch and Schneider, 
Syst. Ichthyol., 1801: 430, pi. 82 (descr., refs.); DeKay, New York Fauna, Fishes (4), 1842: 267, 
pi. 41, fig. 131 (descr.. New York Harbor); Storer, Synop. Fish. N. Amer., 1846: 211 (descr., refs.) ; 
Gronow, Cat. Fish., Gray ed., 1854: 168 (descr., refs., distr.); Poey, Repert. Fisico-Nat. Cuba, 2, 1867: 
423 (diagn., Cuba); Giinther, Cat. Fish. Brit. Mus., 7, 1868: 470 (synon., descr., Cuba, Jamaica, St. 



130 Memoir Sears Foundation for Marine Research 

Croix, "South America"; also several African and oriental locals, that probably are not this species); 
Yarrow, Proc. Acad. nat. Sci. Philad., 1877: 215 (Ft. Macon, North Carolina); Jordan and Gilbert 
(part not of Linnaeus), Bull. U.S. nat. Mus., 16, 1883: 261 (diagn., range); Goode and Bean, Proc. 
U.S. nat. Mus. (1885), 8, 1886: 205 (type, a dried skin cut in two pieces, in BMNH in Lin- 
naean Soc. rooms); Willcox, Bull. U.S. Fish Comm. (1886), 6, 1887: 123 (killed by cold, Florida); 
Henshall, Bull. U.S. Fish Comm. (1889), 9, 1891: 373 (leptocephali, Marco, Florida); B. A. Bean, 
Proc. U.S. nat. Mus., 14, 1891 : 93 (Cape Charles, Virginia); Henshall, Bull. U.S. Fish Comm. (1894), 
lA, 1895: 211 (Key West and Tampa, Florida; not as common as Albula\ no econ. import.); Jor- 
dan and Evermann (part not of Linnaeus), Bull. U.S. nat. Mus., 47(x), 1896: 410 (descr., Amer. 
range includ. Pacif. est., which has distinct species, E. affinis Regan); also 47 (j), 1898: 2806 (confusion 
of species); also 47(^4), 1900: fig. 178; Jordan and Rutter, Proc. Acad. nat. Sci. Philad., 1897: 94 
(Jamaica, "sometimes 20 pounds" evidentally an error); Smith, Bull. U.S. Fish Coram. (1897), ly, 
1898: 90 (common in fall. Woods Hole, Massachusetts); Gunther, Proc. linn. Soc. Lond., 1899: 25 
(t)-pe in Linnaean Soc. coll.); Evermann and Kendall, Rep. U.S. Comm. Fish. (1899), 1900: 54 
(Florida); Evermann and Marsh, Bull. U.S. Fish Comm. (1900), 20 (i), 1902: 81, fig. 1 1 (descr., 
Puerto Rico); Bean, Bull. N.Y. St. Mus., 60, Zool. 9, 1903: 179 (refs., descr.. New York); Schreiner 
and Ribiero, Arch. Mus. nac. Rio de J., 12, 1903 : 90 (Caravelas and Rio de Janeiro, Brazil; Woods Hole, 
Mass.); Hargraves, Fish. Brit. Guiana, 1906: 6 (local name "Long John"); GiD, Smithson. misc. Coll., 
48, 1905: 34 (skull figured, range; Elops cf. Megalops)\ Jordan and Thompson, Bull. U.S. Bur. 
Fish. (1904), 24, 1905: 232 (Tortugas, Florida); Linton, Bull. U.S. Bur. Fish. (1904), 24, 1905: 352 
(food, parasites); Bull. Mus. comp. Zool. Harv., 50, 1906: 156 (Gulf of Mexico at Progreso, Yucatan); 
Smith, N. C. geol. econ. Surv., 2, 1907: 116, fig. 37 (refs., descr., range, North Carolina); Linton, 
Proc. U.S. nat. Mus. (1907), 33, 1908: 86 (Bermuda); Regan, Ann. Mag. nat. Hist., (8) J, 1909: 37 
(distr. limited to Amer. Atlant. est.; Pacif. est. represent. E.afjinis; Afric. represent. E. senegalensis); 
Fowler, Proc. Acad. nat. Sci. Philad., 63, 191 1 : 204 (Nantucket, Mass.; "South Carolina"; West Palm 
Beach, Florida; Santo Domingo; doubtfully from Jamaica, and Rio de J., Brazil); Weymouth, Proc. 
U.S. nat. Mus., 38, 191 1 : 136 (Cameron, Louisiana); Nichols, Bull. Amer. Mus. nat. Hist., 31, 191 2: 
181 (common in markets, Cuba); Sumner, Osburn, and Cole, Bull. U.S. Bur. Fish. (1911), 5^(2), 
1913: 741 (Woods Hole area, Mass.); Starks, Stanf. Univ. Publ., Univ. Ser., 1913: 8 (coast of Brazil; 
Lake Papary, Rio Grande do Norte); Metzelaar, Trop. Atlant. Visschen, 1919: 9 (Curasao, Dutch 
W. L, unimport. as food); Fowler, Proc. biol. Soc. Wash., 33, 1920: 147 (Cape May, New Jersey); 
Breder, Zoologica, N. Y., 2(15), 1922: 336 (Sandy Hook Bay); Meek and Hildebrand, Field Mus. 
Publ., Zool., J5(i), 1923: 175 (synon., descr., range, Panama); Linton, Proc. U.S. nat. Mus., 64, 
1924: 59 (parasites); Bigelow and Welsh, Bull. U.S. Bur. Fish. (1924), 40 (i), 1925: 90, ftn. (questions 
Halkett's [1913: 45] ident. of New Brunswick specimen): Breder, Zoologica, N.Y., 4(4), 1925: 141 
(common at Gatun Dam Spillway, Canal Zone); Nichols and Breder, Zoologica, N.Y. (1926), 9(1), 
1927: 33, fig. (Sandy Hook, New Jersey; Orient and New York, N.Y. ; Woods Hole, Mass. ; a fall visitor) ; 
Hildebrand and Schroeder, Bull. U.S. Bur. Fish. (1927), 43 (i), 1928: 79, fig. 40 (descr., range, Chesa- 
peake Bay); Beebe and Tee- Van, Zoologica, N.Y., -TO (i), 1928: 32, fig. (diagn., range, "apparently 
rather rare" at Port-au-Prince, Haiti); Jordan, Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 
2, 1930: 39 (common names, range, synon.); Breder, Field Bk. Mar. Fish. Atl. Cst., 1929: 60, fig. 
(diagn., range, size); Fowler, Proc. Acad. nat. Sci. Philad., 80, 1929: 609 (Atlantic City, New Jersey); 
Jordan, Manual Vert. Anim. NE U.S., ed. 13, 1929: 36 (diagn., range); Nichols, N.Y. Acad. Sci., 10 
(2), 1929: 119 (probably not uncommon in Puerto Rican waters; diagn.; excellent sport fish with light 
tackle); Fowler, Copeia, No. 2, 1931 : 46 (Corpus Christi, Texas; local name skipjack); also Proc. Acad, 
nat. Sci. Philad., 83: 193 1 : 392, fig- i (Trinidad); Beebe and Tee- Van, Field Bk. Shore Fish. Bermuda, 
1933: 33, fig. (descr., rare in Bermuda); Fowler (part not E. saurus Linn.), Bull. Amer. Mus. nat. Hist., 
70(1), 1936: 155 (synon., includ. Afric. refs. based on E. senegalensis Regan; descr. based in part on 
Afric. specimen); Hollister, Zoologica, N.Y., 2 J (4), 1936: 260-263, figs- (caudal skel.); Hildebrand, 
Sci. Month., 44, 1937: 243 (abund. in Gatun Locks, Panama Canal, when drained); Kaplan, Big 
Game Angler's Paradise, 1937: 91 (size, food, called Ladyfish in Florida); Storey, Ecology, 18, 1937: 
16 (often hurt by freezes); Breder, Bull. N.Y. zool. Soc, 41, 1938: 24 (irregular. New York Harbor); 
Hildebrand, Zoologica, N.Y., 24 {\), 1939: 25 (Gatun Locks, Panama Canal, apparently feed, ground); 
Merriman, Copeia, 1939: 113, figs, a-d (scales); Bigelow and Schroeder, Copeia, 1940: 139 (Chatham, 
Cape Cod, Mass., northernmost definite rec); Longley and Hildebrand, Publ. Carneg. Instn. Wash., 



Fishes of the Western North Atlantic 131 

535, 1941 : 4 (rare at Tortugas, Florida); Fowler, Arqu. Zool. estad. Sao Paulo, J (6), 1941 : 132 (refs.); 

Hildebrand, ]. Wash. Acad. Sci., jj (3), 1943: 90-94, 3 figs, (notes on affinity, anat., develop.). 
Argentina Carolina Linnaeus, Syst. Nat., ed 12, 1766: 519 (orig. descr.; type local. "Carolina"; type lost). 
Mugilomorus anna-caroUna Lacepedc, Hist. Nat. Poiss., 5, 1803 : 398 (orig. descr.; type local. "South Carolina"; 

type in Paris). 
Elops inermis Mitchill, Rep. in part on Fish. New York, 1814: 14 (orig. descr.; type local. New York, specimen 

secured in market; type lost); Mitchill, Trans. Lit. philos. Soc. N.Y., 181 5: 445 (descr.. New York). 
Albula vulpes Gregg (not A. vulpes Linnaeus, but Elops saurus). Where to Catch Fish E. Cst. Fla., 1902: 34 

(descr., instruct, for angling, barely edible). 

Doubtful Reference: 

Eiops saurus Halkett, Check List Fish. Canada, 191 3: 45 (report. New Brunswick; ident. doubtful). 

Negative References :!* 

Elops saurus Gregg, Where to Catch Fish. E. Cst. Fla., 1902: 33, fig. {Albula vulpes, not E. saurus); Bean, 

Bahama Islands: Fishes, 1905: 297 (determined as small Jenkinsia lamprotaenia on re-exam, by present 

writer; not E. saurus; USNM 53076). 



10. To these may be added all mention oi Elops saurus based on specimens not taken on the Atlantic coast of America, 
of which there are many. 

9* 



Family Albulidae" 



SAMUEL F. HILDEBRAND12 



Characters. Body slender, little compressed; belly rounded; ordinary scales on 
median line. Head rather low; flat above. Snout conical, projecting far in advance of 
mandible. Eye of moderate size, with much adipose tissue in adults. Mouth mod- 
erately small, nearly horizontal. Maxillary generally reaching nearly to front of eye, 
or to some point underneath it, but never beyond eye. Premaxillaries not protractile. 
GuLAR plate lacking. Teeth small, present in bands on jaws, vomer, palatines, ptery- 
goids, basibranchials, and sphenoid. Branchiostegals 13 or 14. Gill membranes 
separate, free from isthmus; a membrane across isthmus in front. Opercular bones 
with membranous borders. Pseudobranchiae well developed. Gill rakers very short, 
stout, tubercular in large examples. Last several segments of Spinal column directed 
upward toward base of upper lobe of caudal. Lateral line present, straight. Scales 
of moderate size, more or less quadrate, with membranous border, 3 (rarely 4) 
nearly parallel basal radii, and 4 or 5 coarse basal lobes (fully described by Cockerell, 
75: 3; Jd: 865; ly. 122); scales extending onto the fins; a modified median row of 
enlarged scales in front of dorsal fin; no scales on head. Dorsal fin beginning in ad- 
vance of pelvics. Caudal forked. Anal very small, placed far behind dorsal. Pectorals 
and Pelvics similar, each with an Axillary scale. 

The young pass through a leptocephalus stage like the Elopidae and the eels.^* 

Remarks. The members of this family differ from those of Elopidae principally 

in the absence of a gular plate, and in having a conical snout as well as fewer bran- 

11. Edited and emended by George S.Myers, Henry B. Bigelow, and Yngve H. Olsen. 

12. August 15, 1883-March 16, 1949. 

13. The heart of Albula is reminiscent of the ganoids in having a rudimentary conus arteriosus with two rows of 
valves (9: 548). Several fossil albulids are known, from as far back as the Eocene (5.^: 84). — G.S.M. 

132 



Fishes of the JVestern North Atlantic 133 

chiostegals (13 or 14). From members of the Clupeidae they differ in having a well- 
developed lateral line and undeveloped gill rakers. 

Range. Only two living genera, Albula and Dixonina, are recognized.^* The former 
has a world-wide distribution in warmer seas, and the latter is known from both coasts 
of tropical America. 

Key to Western Atlantic Genera 

I a. Last ray of dorsal and anal not prolonged; snout only moderately conical, pro- 
jecting only about a third of its length beyond mandible; maxillary not reaching 
eye. Albula Scopoli 1777, below. 

lb. Last ray of dorsal and anal much prolonged; snout sharply conical, projecting 
more than a third of its length beyond mandible; maxillary reaching nearly 
to middle of eye. Dixonina Fowler 191 1, p. 143. 



Genus Albula Scopoli 1777 
Bonefishes 

Albula Scopoli, Introd. Hist. Nat., 1777: 450; no species named, but based on Albula (name not binomial; 
followed by long description), in Gronow, Zoophyl., I, 1763: 102; type and only included species 
Albula s. salmo albula Meuschen, in Gronow, J, 1 761, index (not paged), Mediterranean; binomial 
name given to Gronow's (1763) Albula (Whitley, JZ: 303) equals Esox vulpes Linnaeus, Syst. Nat., 
1758: 313. 

Generic Synonyms: 

Esox, in part, Linnaeus, Syst. Nat., I, 1758: 313, for E. vulpes Linnaeus, Bahamas. 

Argentina, in part, Forskal, Descr. Anim ..., 1775: 68, for A. glossodonta Forskal, Arabia. 

Amia, in part, Bloch and Schneider, Syst. IchthyoL, 1 801 : 457, for A. immaculata Bloch and Schneider, South 

America. 
SynoJus, in part, Bloch and Schneider, Syst. IchthyoL, 1801 : 398, for S. argenteus Bloch and Schneider, Tahiti; 

Lacepede, Hist. Nat. Poiss., 5, 1803: 32, pi. 8, fig. 2, for S. vulpes Lacepede. 
Clupea, in part, Bloch and Schneider, Syst. IchthyoL, 1801: 427, for C. brasiliensis Bloch and Schneider, 

Brazil; Lacepede, Hist. Nat. Poiss., 5, 1803: 427, for C.macrocephala LacepMe, Martinique. 
Butyrinus Lacepede, Hist. Nat. Poiss., 5, 1803: 45; type by monotypy, Butyrinus bananus Lacepede equals 

Esox vulpes Linnaeus. 
GlossoJus Cuvier, Mem. Mus. Hist. nat. Paris, I, 181 5: 232; genotype by monoXypy, Argentina glossodonta 

Forskal equals Esox vulpes Linnaeus. 
Engraulis ."^gassiz, L. in Spix and Agassiz, Selecta genera et species piscium. . .Brazil. . ., 1829: pi. 23, fig. 2, 

for E. sericus Agassiz; pi. 24, fig. 2, for E. bahiensis Agassiz, Brazil. 
Conorhynchus Gill, Proc. .^cad. nat. Sci. Philad., SuppL, 1861: 55; genotype by monotypy, Butirinus vulpes 

Storer equals Albula vulpes. 

Characters. Dorsal and Anal fins with the last ray not produced. Maxillary 
falling short of vertical from front border of eye. Snout projecting about a third of its 

14. Dr. Hildebrand evidently excluded from the Albulidae the deep-water genus Pterothrissus, which is known from 
the coasts of Japan and West Africa and which may eventually be found in the western North Atlantic. Pterothrissus 
differs widely from Albula and Dixonina in the very long, many-rayed dorsal fin. — G. S. Myers. 



134 Memoir Sears Foundation for Marine Research 

length beyond tip of lower jaw. Caudal skeleton (jp: 269-275) exhibiting some 
ganoid characters, showing a relationship with that primitive group. Vertebrae: last 
2 or 3 in A. vulpes out of line with central axis, directed somewhat dorsally; massive 
spines (neurals) attached to them and preceding segment, extending into base of upper 
lobe of caudal. Alimentary canal appearing as a straight tube in larvae, becoming 
modified in the adult with development of a very large stomach that occupies fully 
67 7o of the length of abdominal cavity If distended with food; stomach largely a 
blind sac; the cardiac and pyloric orifices rather close together. About 13 Pyloric 
coeca, bound together, forming a sort of sheath over part of stomach. Air bladder 
very long, extending backward far beyond vent, its walls very thin, without lung 
tissue (see tarpon) except for a pair of very small kidney-shaped cellular bodies 
attached ventrally at about midlength. 

Remarks. This genus differs from Dixonina chiefly in that the last ray of both 
dorsal and anal is short and does not become prolonged in adults; see also Key, p. 133. 
From E/ops it is distinguishable by the absence of a gular plate (p. 123). 

Range. Only one modern species, Albula vulpes^ Is generally recognized; this 
occurs in virtually all warm seas. 



Albula vulpes (Linnaeus) 1758 

Bonefish, Grubber, Bananafish 

Figures 22, 23 

Study Material. An almost complete growth series ranging from leptocephall with 
virtually undeveloped dorsal and anal fins to adults 430 mm TL, from both Atlantic 
and Pacific. As the growth series was not complete in the material from the Atlantic, 
some specimens, especially the older stages of leptocephall and first stages of young 
adults, from the Pearl Islands in Panama Bay, were used in describing the young. 

Distinctive Characters. See Remarks., above. 

Description. Proportional dimensions In per cent of standard length, and counts, 
based on Atlantic adults only. 

Body: depth at origin of dorsal 20—23 in Anal fin: length of base 5.1—6.75. 

large specimens, 14—19 in young Pectoral fin: length 15—19. 

adults. 

Caudal peduncle: depth 6.9—8.9. Scales: 65—71. 

Head: length 27.5-32. Gill rakers: 7 or 8+ 9 or 10. 

Snout: length 11. 5-13. 6 in large sped- Fin rays: dorsal 17 or 18, rarely 19; 

mens, 8.5-10 In young adults. anal 8 or 9; pectoral 16 or 17, 

Eye: diameter 5.5—8.0 rarely 15. 

Maxillary: length from tip of snout 9.1— Vertebrae: 72-74 (5 specimens). 

12.7. 



Fishes of the Western North Atlantic 



135 



Body slender, rounder and less compressed in large specimens than in young 
adults, its dorsal profile more convex than ventral profile, its depth at dorsal origin 
4.35-4.9 in SL in large specimens, about 5.25-7.1 in the more slender, recently-meta- 
morphosed young. Scales firm, with crenulate membranous edges. Head low, espe- 
cially in large specimens, flat above, its depth exceeding its width at middle of eye 
by about diameter of pupil, 3.0—3.8 in SL. Snout rather long, conical, especially in 
large specimens, projecting about a third of its length beyond mandible, 2.2-2.5 ■" 
head, about 2.7—3.5 '" young adults. Eye moderately small, 4.4—5.5 in head, its 
center nearer to margin of opercle than to tip of snout. Maxillary not quite reaching to 







Figure 22. Albula vulpes, from off Newport, Rhode Island; from Goode. Drawn by H. L.Todd. 



eye in large specimens, but extending to, or slightly beyond, anterior margin of eye in 
young adults, 2.6—3.15 in head (measured from tip of snout). 

Dorsal fin somewhat elevated anteriorly, its longest rays not reaching to tip of 
posterior ray if deflexed, its origin a little nearer to tip of snout than to base of 
caudal. Caudal deeply forked, the upper lobe somewhat longer than the lower. Anal 
very small, its origin notably nearer to base of caudal than to base of pelvic, its last 
ray, though longer than the preceding one, not especially produced, its base 4.7—5.4 
in head. Pelvic somewhat smaller than pectoral, inserted under or slightly behind 
middle of dorsal base. Pectoral with rounded margin, reaching notably less than half- 
way to pelvic, 1.7—2.0 in head. Axillary scale of pectoral about half as long as 
fin, adherent to body. 

Color. Large specimens bluish above; bright silvery on sides and below; dark 
streaks between rows of scales, at least on dorsal half of side; dorsal and caudal with 
dusky margins, frequently entirely pale in preserved specimens. Very young adults 
(completely metamorphosed and about 30 mm) with a double series of dark spots on 
the back, each just off the median line; these spots soon uniting to form about nine dark 
crossbands on the back, extending down nearly or quite to lateral line; the third band 
crossing back at origin of dorsal; next two situated posteriorly under base of dorsal; 



136 Memoir Sears Foundation for Marine Research 

bands persisting until a length of about 75 mm. The dark longitudinal streaks of the 
adults appearing shortly before crossbands become obscure. 

Sixe. Its maximum length is about 770 mm (31 in.) and its weight about 13 or 14 
pounds. 

'Development}^ The eggs and the earliest stagesof the leptocephali remain unknown. 
The youngest at hand, that is, the least developed larva, is apparently somewhat younger 
than any yet described. It is 57 mm long (51 mm SL), was collected at Tortugas, Flor- 
ida, and is the only one of the many that has the dorsal and anal fins entirely unde- 
veloped. The pectoral fins appear as mere tufts of membrane and the pelvics only as a 
thickening of the body wall where the bases are developing. The caudal fin, however, 
is well developed with segmented rays, is broadly forked, and is about as long as the 
head. This larva is sharply compressed, though not more so than considerably older 
ones; its depth increases gradually backward from the head to about the beginning of 
the distal fourth of the body, where it is contained 11.8 times in SL. Myomeres are 
prominent, except anteriorly and posteriorly, and about 66 may be counted. The head 
is low and moderately broad and is contained 16.4 times in SL. The snout is conical 
and projects slightly beyond the mandible, the maxillary is not definitely formed, the 
slightly oblique gape extends under the anterior part of the eye, and the eye is scarcely 
as long as the snout, being contained in the head 4.1 times. The intestinal tract appears 
to be a straight tube, is rather loosely attached to the body, has a slight groove on each 
side, and extends nearly to the base of the lower lobe of the caudal. This old preserved 
specimen, collected in 191 9, is now entirely without pigment. Several other specimens 
from Tortugas and Cuba are only slightly more developed. 

Although leptocephali 75-87 mm long have been reported {28: 40)," the largest 
one now at hand among several hundred from many localities along the Atlantic and 
Pacific coasts of America is just 70 mm long (57 mm SL). In this specimen, one of a 
large collection from the Pearl Islands, Panama Bay, the dorsal and anal fins are devel- 
oped (showing at least most of the rays), and the fulcra may be fairly accurately enu- 
merated. The dorsal fin is placed over about the beginning of the distal fifth of the body. 
The anal is scarcely separated from the base of the lower caudal lobe. The pelvic fins 
are developed but do not have definite rays and are placed at about midbody length 
without the head, a position retained throughout life. The pectoral fins are fairly long, 
with definite indications of rays. The body is deepest somewhat in advance of the dorsal 
fin, and its depth there is contained 9.25 times in SL; 69 myomeres were counted. 
No prominent changes in the head region are evident. The head is contained 14.4 
times in the SL. The snout and eye are of about equal length, being 3.6 in head. The 
jaws possess minute teeth that project somewhat forward. Two almost continuous dotted 
dark lines bound the intestinal tract from behind the pectoral fins to the vent, being 

15. A detailed account of the development and distribution of young Albula, based on the Dana collections, has 
recently been published by Alexander (Dana Rep., 5J, 1961). 

16. The fact that Albula has a "leptocephalus" larva was first ascertained about 1896 by Prof. C. H. Gilbert of Stan- 
ford University. It seems he never published an account himself, but a synopsis of his findings, with his figures of 
the developmental stages, was made known by Jordan {44) and Gill (2S: 43) — G. S. Myers. 



Fishes of the Western North Atlantic 



137 



interrupted only at the point of insertion of the pelvics; and a vertical series of dark 
dots is present on the caudal base. Living larvae of this size, and considerably older 
ones as well, are as transparent as glass. They appear as shining objects in the net and 
are easily overlooked, only the dark eyes being at all conspicuous. 

Older larvae of 45 mm (39 mm SL), though still sharply compressed, are much 
more substantial than younger ones at a length of 70 mm or so. The dorsal and anal 
fins have moved forward considerably, the anal fin now being separated from the caudal 




Figure 23. Albula vulpes. Pearl Island, Pacific Panama, USNM 128391. a Leptocephalus stage, 68 mm long; 
B older larva in transition stage, 76 mm TL; c older fry 30 mm TL. Drawn by Nancy D. Patton. 



by at least half the length of the head. The greatest body depth, at the origin of the 
dorsal, is contained 7.6 times in SL. The myomeres total 70-72. The head remains 
rather low and broad and is contained 6.0 times in the length; the snout projects beyond 
the mandible more prominently and is now rather longer than the eye, 4.0 times in 
head; and the maxillary is fairly well developed, extending well beyond the anterior 
margin of the eye, 2.4 in head. The color remains as in the younger fish described, 
except that a few dark chromatophores have appeared about the head, and a slightly 
broken dark line has developed on both upper and lower lobes of the caudal. 

Larvae 32 mm long (26 mm SL) are the smallest in the study collections that may 
be classed as leptocephali. The body remains strongly compressed, though of course 
more robust and more substantial than in the 45-mm fish described above, and it has 
become more elongate, the depth at the dorsal origin being contained 8.4 times in SL. 
The dorsal and anal fins have continued to move forward, but they still remain more 
posteriorly placed than in the adult; the origin of the dorsal is somewhat nearer to the 
base of the caudal than to the tip of the snout and is approximately over the pelvic 



138 Memoir Sears Foundation for Marine Research 

fins. The intestinal tract has become more fully invaginated and no longer remains 
merely attached, except posteriorly, as in younger fish; the vent, instead of being sit- 
uated at the origin of the anal, is now well In advance of it, having moved forward 
rather more rapidly than the fin. The color markings remain about as in the younger 
fish described, but additional chromatophores have appeared, principally at the occiput, 
at the base of the dorsal, and on the dorsal. 

The smallest specimen that may be classed as a young adult is 28 mm (2 i mm SL), 
and its length is the least of the many Albula at hand. This fish is quite a little longer 
than the one that diminished from a leptocephalus to a young adult of only 20 mm, 
in an aquarium (jp: 269; 40: 109). The body remains more compressed in this 
and other young adults than In large adults, and it is also more slender, the depth at 
the dorsal origin 6.5 times in SL. The head is contained 3.45 times in SL and there- 
fore does not differ in this proportion from large adults. The eye is rather small, 5.05 
times in head, and is definitely shorter than the snout, which is contained 3.05 times 
in the head and projects rather prominently beyond the mandible; the maxillary ex- 
tends to the anterior margin of the pupil and is contained 2.9 times in head. The dorsal 
and anal fins have continued to move forward, the origin of the dorsal now being equi- 
distant between base of caudal and tip of snout; and the attachment of the pelvic fins is 
now only a little in advance of the middle of the dorsal base. At this stage, and even in 
ones a little older, the two dark lines along the sides of the intestinal tract in the lepto- 
cephali persist, two rows of dark spots on the back (described elsewhere) have appeared, 
there are indications of some elongate dark spots along the side, and some additional 
pigment has appeared on the base of the anal and ventrally on the caudal peduncle. 

Scales first appear in young adults about 35 mm long. At a length of 40— 45 mm, 
scalation generally is complete, except perhaps on the abdomen. At these stages the 
lateral line also is well developed. The dorsal and anal fins are situated approximately 
as in large adults ; that Is, the origin of the dorsal is somewhat nearer to the tip of the 
snout than to the base of caudal, and the anal is separated from the caudal by a distance 
somewhat greater than the length of the snout. The pelvic fins, which, from the time of 
their appearance in the leptocephalus, are inserted a little nearer to the head than to 
the base of caudal, retain that relative position and are now about under the middle 
of the dorsal base. The pectorals are long but not falcate and reach more than halfway 
to the pelvics. The vent has continued to move forward and is now In advance of the 
anal by a distance nearly equal to the length of the snout. 

At 40 mm, pigmentation has become general and the nine crossbands on the 
back, described elsewhere, are prominent. But dark longitudinal stripes do not appear 
definitely until a length of about 75 mm is attained, at which size the crossbands have 
become obscure. 

Spawning. The spawning season and the spawning grounds remain unknown. 
Most of the many larvae and young adults described herein were taken off the coast 
of Panama during February and March, but their occurrence at other seasons is not 
known, as no year-round collecting has been done there. Growth stages ranging from 



Fishes of the JVestern N^orth Atlantic 139 

fairly young leptocephali of 70 mm TL to young adults 52 mm long were exceedingly 
numerous among some of the Pacific islands in Panama Bay in March 1937 (especially 
in the Pearl Islands). Somewhat similar stages, though less numerous, were taken 
in February and March on the Atlantic coast of Panama. Other leptocephali at hand 
were taken as follows: January 23 on the Pacific coast of Colombia, February 13 in 
Puerto Rico, March 13 in the Virgin Islands, April 8 in Cuba, April 23 and 28 in 
Haiti, May 1 9 in Cuba, June 30 at Beaufort, N. C, July 2 i in Bermuda, and August 7 
and November 25 at Tortugas, Florida. Beebe and Tee-Van (5: 37) have reported that 
on 34 evenings of collecting (between February 26 and April 30) with an electric light 
lowered from the gangway of a schooner in Port-au-Prince Bay, Haiti, larvae of Albula 
were absent only two nights and were very abundant 17 nights. With the capture of 
the leptocephali spread over most of the year and at widely spaced localities, there is 
little indication of where and when spawning takes place. 

Although an advanced leptocephalus transformed rapidly in an aquarium under 
observation (59: 269; 4.0: 109), it cannot be concluded that development before 
capture was equally fast, and that the rapid transformation under artificial conditions 
was normal, especially in the light of what is known about the slow development of the 
leptocephali of freshwater eels. At least the possibility exists that the rather advanced 
leptocephali that have been described may have been several months old and were 
caught far from the place where they had hatched. 

Habits. The adult is usually caught in comparatively shallow water, but to what 
extent it enters deep water offshore apparently has not been determined. The larva 
has been taken at night at the surface under an electric light, by which it seems to 
be attracted, and during the day in the shallow water of bays and estuaries with seines, 
at least once in a creek with nearly fresh water (in Panama), and on one occasion in 
the surf on the outer shores of Bogue Banks, North Carolina. During March 1937, 
the larva and young adult were found in great abundance among islands in Panama 
Bay. In one very shallow tide pool with a muddy bottom, off San Jose Island in the 
Pearl Islands group, the water was literally "soupy" with fish and crustaceans. At the 
time there was an exceptionally low tide, and the concentration was so great and the 
water so shallow and warm that many small shrimp and fish perished, but the greater 
part of the aggregation seemed to endure the situation. It was possible to select from 
this pool a growth series ranging from leptocephali 70 mm long, the youngest present, 
to young adults 52 mm long. 

Food. This fish, known as the Grubber in Bermuda (^4.0: 104) grubs up some of 
its food from the bottom with its nose and sometimes turns somersaults in the proc- 
ess. In the West Indies and Florida it may be seen by day (55: 187), along shallow 
sandbanks and among underwater grasses, feeding during incoming tides on worms, 
mollusks, and crabs; but in Bermuda it apparently comes into the shallows at night 
to feed. Bivalves and small squid have been reported as its food (5: 37). In the Pearl 
Islands, young adults were gorged with small shrimp, but a few fish (small atherinids) 
had been eaten also. 



140 Memoir Sears Foundation for Marine Research 

Relationship and Variation. Only one species has been recognized. Insufficient 
material from its wide range has been studied, but the present study of specimens 
from the Atlantic and Pacific coasts of America has shown no constant differences or 
variations of importance. 

Relation to Man. This species is not considered a good foodfish in most of its range 
in America. Schroeder, reporting on the fisheries at Key West, Florida, stated that "it 
is not highly regarded as a foodfish, and its commercial value is negligible" (6j: 4). 
Similar statements are included in publications about the fishes of the West Indies. 
But Hollister, presumably reporting on observations made in Bermuda, stated to the 
contrary that "it surpasses its famous cousin, the Silver King, or tarpon, in possessing 
a combination of qualities, that of gaminess and of being a table delicacy" (40: 104). 
From the writer's own observations in Panama, the Bonefish is considered an inferior 
foodfish, whereas the tarpon is highly regarded as food by the native population. Thus 
tastes seem to differ in different localities. 

However, the Bonefish ranks high with anglers, especially around Florida and 
the Bahamas, for it is so wary when feeding on the flats that careful stalking and -long 
and delicate casting are called for; but once hooked, its runs are longer and swifter 
and its resistance far more vigorous and indefatigable than one might expect from such 
a comparatively small fish. 

Miss Francesca LaMontc of the American Museum of Natural History, long 
acquainted with the sport fisheries, has generously offered the following information: 
This warm-water fish, which provides splendid angling with simple equipment, is 
caught on the flats and can be fished there either from a boat or while wading. The 
boat should be light as well as flat-bottomed and should be provided with an outboard 
motor, oars, and anchors for both bow and stern. The fish is taken by casting or still 
fishing, but the favorite method is casting. Light tackle with plug casting or saltwater 
fly rod provides good sport. As bait, anglers may use conch, crabs or crayfish, feathers, 
lures, or squid. Coming in with the tide, the Bonefish spreads out over the flats and 
among the mangrove stands; when the tide ebbs, it seeks deep holes. On the flats, 
the angler may chum or get his bait into the middle of a "mud" or group of Bonefish 
that is feeding, with tails uppermost while grubbing on the bottom. 

When hooked, the Bonefish makes a very fast and strong run, taking out yards 
of line. In so doing it varies the straight dash by suddenly turning and coming toward 
the boat or by running into the mangroves where the line may be cut on the roots. 
After making several of these runs, the Bonefish usually begins to wear out, begins 
circling the boat, and can finally be picked up in a landing net. 

For many years the angling record stood at 13^/4 pounds for one caught on the 
famous Bonefish grounds off Bimini in the northern Bahama Islands, but this record 
was eventually beaten. The present record is for a fish of 1 8 pounds 2 ounces, caught 
off Kauai, Hawaiian Islands. In the western North Atlantic, the best known Bone- 
fish grounds are the Florida Keys near Miami, the Islamorada-Marathon area, and 
the northern Bahamas. 



Fishes of the Western North Atlantic 141 

Range and Abundance. The Bonefish is known from virtually all warm seas. On 
the Atlantic coast of America it is plentiful around southern Florida, among the 
northern Bahamas, and around Bermuda. But its area of chief abundance evidently 
does not reach much farther to the north, for it is rare even at Beaufort, North 
Carolina. However, odd specimens have been reported for the vicinity of New York 
(2: 183), Newport, Rhode Island {68: 74), the vicinity of Woods Hole, Massachu- 
setts {^j: 741), and perhaps even Charlotte County, New Brunswick, at the mouth 
of the Bay of Fundy (ftn. 15). 

To the south it is present more or less regularly in the West Indies and on 
the Atlantic coast of Panama, though it does not seem to be numerous; it is known 
from as far south as Rio de Janeiro. On the Pacific coast of America it has been re- 
ported as occurring from San Francisco Bay, California, to Talara, Peru. 

Synonyms and References ;!' 

Esox vulpes Linnaeus, Syst. Nat., ed. lo, 1758: 313 (orig. descr.; based on Vulpes bahamensts Catesby, pre- 

Linnaean; type local. Bahamas; no type specimen). 
Albula s. Salmo albula (nee Salmo albula Linnaeus) Meuschen, in Gronow, Zoophyl., J, 1781 : index, unpaged 

(on ^''Albula'' Gronow; type local. Mediterranean; t\pe lost). 
Clupea brasiliensis Bloch and Schneider, Syst. Ichthyol., 1801: 427 (orig. descr. ; t}-pe local. Brazil; based on 

Marcgrave's mss. in Berlin). 
Albula conorhynchus Bloch and Schneider, Syst. Ichthyol., 1801: 432, pi. 86 (orig. descr.; based on Gronow, 

Plumier, and Marcgrave; no special tj'pe local, except Medit. and .'\mer.); Giinther, Cat. Fish. 

Brit. Mus., 7, 1868: 468 (descr., synon., Amer. locals. — West Indies, Jamaica, Belize, Puerto Cabello, 

Bahia, Natal, and Pacif. est. of Centr. Amer.); Yarrow, Proc. Acad. nat. Sci. Philad., 1877: 215 (Ft. 

Macon, North Carolina). 
Albula plumieri Bloch and Schneider, Syst. Ichthyol., 1 801 : pi. 86 (name on plate only; lapsus for conorhynchus). 
Amia immaculata Bloch and Schneider, Syst. Ichthyol., 1801: 451 (orig. descr., after Macabi Parra [1787: 

88, pi. 35, fig., nonbinomial]; tj'pe local. S. Amer.). 
Clupea macrocephala Lacepede, Hist. Nat. Poiss., 5, 1803: 426 (orig. descr.; t}'pe local. Martinique; after 

Plumier fig.). 
Glossodus forskali Agassiz, in Spix and Agassiz, Pise. Brasil, 1829: 49; called EngrauHs serkus on pi. 23 and 

E. bahiensis on pi. 24, fig. 2 (orig. descr.; type local. Bahia, Brazil; t\'pe probably in Munich). 
Albula parrae Cuvier and Valenciennes, Hist. Nat. Poiss., ig, 1846: 339 (orig. descr.; type local. Bahia and Rio 

de Janeiro, Brazil; type in Paris). 
Butirinus vulpes Storer, Synop. Fish. N. Amer., 1846: 212 (diagn., refs.). 
Albula rostrata Gronow, Cat. Fish., Gray ed., 1854: 189 (orig. descr.; type locals. Amer. and Indian oceans, 

Medit. Sea; type probably in London). 
Conorhynchus plumieri Poey, Repert. Fisico-Nat. Cuba, 2, 1867: 423 (descr., Cuba). 
Albula vulpes Jordan and Gilbert, Bull. U.S. nat. Mus., 16, 1883 : 258 (descr., range); Bean, Bull. U.S. Fish 

Comm. (1888), 8, 1890: 206 (not uncommon at Cozumel, Yucatan); Collins and Smith, Bull. U.S. 

Fish Comm. (1891), 11, 1892: 104 (commerc. catch. Gulf coast, Florida, 1890); Eigenmann, Proc. 

U.S. nat. Mus. (1892), J5, 1893: 135 (not abund. at San Diego, California); Henshall, Bull. U.S. Fish 

Comm. (1894), J./, 1895: 211 (common at keys, passes, inlets of s. Florida, good food and game fish); 

Lonnberg, Ofvers. Svensk. Vet. Akad. Forh., 9, 1895: 662 (Cape Haitien, Haiti); Jordan and Ever- 

mann, Bull. U.S. nat. Mus., 47 (-T), 1896: 411; 47 ('^), 1900: fig. 179 (descr., range, synon.); Jordan 

and Rutter, Proc. Acad. nat. Sci. Philad., 1897: 94 (Jamaica); Smith, Bull. U.S. Fish Comm. (1897), IJ, 

1898: 91 (rare at Woods Hole, Massachusetts); Evermann and Kendall, Rep. U.S. Comm. Fish. (1899), 

1900: 55 (Florida); Gilbert, Proc. Wash. Acad. Sci., 2, 1900: 163 (Pernambuco and Maceio, Brazil); 

17. Literature dealing with American specimens only, or in part, is cited. It does not seem necessary to cite Old 
World references here. 



142 Memoir Sears Foundation for Marine Research 

Evermann and Marsh, Bull. U.S. Fish Comm. (1900), 20 (i), 1902: 82 (descr. range, econ. import., 
Puerto Rico); Bean, Bull. N.Y. St. Mus., 60, Zool. 9, 1903: 182 (synon., descr., range. New York and 
elsewhere); Bean, Field Mus. PubL, Zool., 7, 1906; 33 (5 Bermuda locals.); Gilbert and Starks, Mem. 
Cal. Acad. Sci., 4, 1904: 39 (in Panama City market); Jordan and Thompson, Bull. U.S. Bur. Fish. 
(1904), 24, 1905: 232 (Tortugas, Florida; discus.); Gill, Smithson. misc. Coll., 48, 1905: 40 (descr. 
and figs, of skull, develop, of leptocephali, habitat, game qualities; C. H. Gilbert discovered leptocephalus 
stage); Bean, Bahama Is., Fish., 1905: 297 (Spanish Wells, Bahamas); Smith, N.C. geol. econ. Surv., 
2, 1907: 117 (synon., diagn., U.S. coasts, esp. North Carolina); Coles, Bull. Amer. Mus. nat. Hist., 28, 
1910: 344 (Cape Lookout, North Carolina); Cockerell, Smithson. misc. Coll., 56(3), 191 1: 3 (scale 
descr.); Bull. U.S. Bur. Fish. (1912), J2, 1913: 122, fig. 2 (scales not difl^erent from those of Dixoaina; 
similar to Synodus and Esox); Fowler, Proc. Acad. nat. Sci. Philad., 6j, 191 1: 204 (Santo Domingo, 
St. Martin, Jamaica); Nichols, Bull. Amer. Mus. nat. Hist., 31, 191 2: 181 (not uncommon in Havana, 
Cuba, market); Gudger, J. Mitchell Sci. Soc, 28, 191 3: 166 (rare, Beaufort, North Carolina); Starks, 
Stanf. Univ. Publ., Univ. Ser., 1913: 8 (Natal, Brazil); Sumner, Osburn, and Cole, Bull. U.S. Bur. Fish. 
(191 1), 31 {2), 1913= 741 ("re at Woods Hole, Mass.); Halkett, Check List Fish. Canada, 191 3: 45 
(Bay of Fundy, New Brunswick); Weber and de Beaufort, Fish. Indo-Aust. Archipelago, 2, 1913: 7, 
fig. 5 (synon., descr., develop, of leptocephali; figs, of leptocephali labeled "after Gilbert," but without 
citation to Gilbert; figs, of larvae apparently after Gill, 1905); Fowler, Proc. Acad. nat. Sci. Philad., 6y, 
1915: 522 (Trinidad, B. W. L); 68, 1916: 397 (Port Limon, Costa Rica); 6g, 1917: 128 (Col6n, 
Panama); Metzelaar, Trop. Atlant. Vissch., 1919: 8 (St. Martin, and other Windward and Leeward 
islands; not common, not much used as food); Metzelaar, Bijdr. Dierk. K. zool. Genoot. Amst., 1922: 
134 (Curasao); Meek and Hildebrand, Field Mus. PubL, Zool., Jf (i), 1923: 178 (synon., descr., 
range, Panama); Schroeder, Rep. U.S. Comm. Fish. (1923), Append. 12, 1924: 4 (commerc. value, 
Key West, Florida); Mowbray, Book of Fishes, Nat. geogr. Soc, ed 1924: 135-137, color pi. on p. 122; 
ed. 1939: 186-187, color pi. on p. 201 (genl. accts.); Breder, Bull. Bingham oceanogr. Coll., 2 (2), 
1928: 4 (young adult, Sonora, Mexico; adult from Pearl Is., Panama); Nichols and Breder, Zoologica, 
N.Y., 9 (i), 1927: 34 (diagn., distr., life hist., size); Beebe and Tee- Van, Zoologica, N.Y., 10 (i), 1928: 
37 (refs., field characters, size, color, range, uncommon at Port-au-Prince, Haiti); Nichols, N. Y.Acad. 
Sci., 10 (2), 1929: 200 (distr., not uncommon in Puerto Rican waters, diagn., habits); Breder, Field Bk. 
Mar. Fish. Atlant. Cst., 1929: 60, fig. (diagn., abund., range, habits, size); Fowler, Proc. Acad. nat. Sci. 
Philad., 80, 1929: 609 (Barnegat Inlet, Atlantic City, first definite recs. for New Jersey); Jordan, Manual 
Vert. Anim.NE U.S., 1929: 38 (diagn., range); Fowler, Proc. Acad. nat. Sci. Philad., 82, 1930: 269 
(Grenada, B. W. I.); Jordan, Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 2, 1930: 39 (range, 
synon.); Gowanloch, Bull. La. Conserv. Dept., 21, 1932: 44, fig.; also Bull. La. Conserv. Dept., 23, 
1933: 43, fig. (relation., life hist.); Beebe and Tee- Van, Field Bk. Shore Fish. Bermuda, 1933 : 34 (descr., 
distr., Bermuda); Gregory, Trans. Amer. philos. Soc, 23 (2), 1933: 144, fig. 37 (relations, as shown 
by skull); Breder, Zoologica, N. Y., 18, 1934: 59 (Grassy Creek, Andros I., Bahamas); Beebe and 
HoUister, Zoologica, N. Y., ig {(>), 1935: 211 (leptocephali, adults. Union I., Grenadines, B. W. I.); 
Myers, Copeia, 1936: 83 (cf. Dixonina nemoptera); HoUister, Bull. N.Y. zool. Soc, jp, 1936: 104- 
109, figs, (metamorphosis notes); Zoologica, N.Y., 21 {^), 1936: 268-275, ^g^- (caudal skel.); Far- 
rington, Atlant. Game Fishing, 1937: 183—186 (scientific name not given, genl. acct., abund. Bi- 
mini, Bahamas); Fowler, Proc. Acad. nat. Sci. Philad., 8g, 1937: 304 (V'entnor, New Jersey); Heilner, 
Salt Water Fishing, 1937: 174-194 (genl. acct.); Kaplan, Big Game Angler's Paradise, 1937: 
190-195 (genl. acct.); Butsch, J.Barbados Mus. nat. Hist. Soc, 7(1), 1939: 18 (uncommon at Bar- 
bados); Longley and Hildebrand, Publ. Carneg. Instn. Wash., 535, 1941 : 5 (color of leptocephali from 
Tortugas, Florida, described); Alexander, 'Dana' Rep., 53, 1961 (descr. larval develop.; world-wide 
distr. and hydrogr.). 

Atopichthjs esunculus Garman, Mem. Mus. comp. Zool. Harv., 24, 1899: 327, pi. 65, figs. 2, 2a (orig. descr.; 
type local, off Acapulco, Mexico; types in MCZ); Jordan, Evermann, and Clark, Rep. U.S. Comm. 
Fish. (1928), 2, 1930: 54 (ref. to orig. descr.). 

Elops saurus Gregg (not Elops sauriis Linnaeus, but Esox vulpes Linnaeus), Where to Catch Fish. E. Cst. Fla., 
1902: 33, fig. (descr., habitat, feed., instruct, for angling, fairly edible). 

Negative Reference: 

Albula vulpes Gregg, Where to Catch Fish. E. Cst. Fla., 1902: 34 (not Albula vulpes Linnaeus, but Elops 
saurus Linnaeus). 



Fis/ies of the Western North Atlantic 



143 



Genus Dixonina Fowler 1 9 1 i 

Dixonina Fowler, Proc. Acad. nat. Sci. Philad. (igio), 191 1 : 651; type species by original designation, Dixo- 
nina nemoptera Fowler. 

Characters. See species account below. 

Remarks. Dixonina is easily distinguished from Albula by the produced last ray of 
both dorsal and anal fins. Furthermore, Dixonina has a somewhat longer and more 
conical snout and a larger mouth than Albula. 

Range. A single species is known; this has been reported from the West Indies, 
Venezuela, northern Brazil, and from the Pacific coast of Mexico and Central America. 



Dixonina nemoptera Fowler 1 9 1 1 

Shafted Bonefish^* 

Figure 24 

Study Material. One Pacific coast specimen, about 413 mm long, 334 mm SL, 
from Acapulco, Mexico, USNM 75547, recorded by Myers. 





^m!y=^.7 A'i.A..'>^..> u,iv.Vi...'i..toA 



Figure 24. Dixonina nemoptera, 413 mm TL, Acapulco, Pacific coast of Mexico, USNM 75547. From 
drawing by Ann S. Green. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on Pacific specimen 334 mm SL. 



Anal fin: length of base ^.^. 
Pectoral fin: length 15.6. 



Body: depth at origin of dorsal 23. 

Head: length 29.5. 

Snout: length 10.8. 

Eye: diameter 5.4. 

Maxillary: length from tip of snout 

Body slender, the back somewhat elevated, the ventral outline nearly straight; 
depth at origin of dorsal 4.3 in SL. Scales firm, with crenulate membranous edges. 

18. Name suggested by George S. Myers. 



Scales: 76. 

Gill rakers: 5+10. 

Fin rays: dorsal 19; anal 8; pectoral 16. 



144 Memoir Sears Foundation for Marine Research 

Head long and low, flat above, its depth at middle of eye exceeding Its width at the 
same place by diameter of pupil, 3.4 in SL. Snout very long, projecting more than 
a third of its length beyond mandible, 2.7 in head. Eye moderately small, 5.4 in head, 
its center notably nearer to tip of snout than to margin of opercle. Maxillary ex- 
tending nearly to middle of eye, 2.15 in head (measured from tip of snout). 

Dorsal fin moderately high anteriorly, its longest anterior rays about as long as 
snout and eye, its last ray filamentous, almost reaching vertical from origin of anal, its 
origin about equidistant between tip of snout and caudal base. Caudal deeply forked, 
the upper lobe apparently longer than the lower (somewhat damaged). Anal very small, 
its last ray produced, extending nearly to base of caudal fulcra, its origin equidistant 
between base of caudal and base of last dorsal ray, its base '}^.c^ in head. Pelvic not 
much smaller than pectoral, inserted about under middle of dorsal. Pectoral not fal- 
cate, extending halfway to pelvic, 1.9 in head. Axillary scale of pectoral adherent to 
body, not very distinct. 

Color. An old preserved specimen, slightly bluish above, silvery below, narrow 
dark streaks between rows of scales on upper part of side. Live specimens from the Paci- 
fic were described by Beebe (4: 44) as follows: 

Brilliant silver, appearing dark above in certain lights with dark green on the head; this dorsal pig- 
mentation resolves into eight very dark lines along each side of upper half of the body, covering about 
one-fourth of each scale nearest to the adjoining line. Dorsal fin greenish, caudal dusky, anal silvery at base, 
a dusky spot at pectoral base, with yellow spot behind; basal membrane of pectoral apple green, rest of fin 
dusky. These colors fade at death. Iris silvery. 

In the young, at least up to 90 mm in length, two rows of dark spots extend along the side of body, 
the upper close to the mid-back. In a 115 mm fish the spots are less conspicuous and the dorsal dark lines 
begin to be distinct. The spots persist after death. In fuU-grown fish the dark scale lines are fainter than in 
fish of medium size. 

Relationship. The specimen at hand from the Pacific agrees very well with two 
published accounts of Atlantic fish. Some Pacific coast specimens have been described 
as D.pacifica by Beebe {4: 43), who considered them specifically distinct from the 
Atlantic ones. On the basis of available descriptions (not specimens) of Atlantic fish, 
the differences between his Pacific specimens and those from the Atlantic were set forth 
as follows: 

In the former {D.pacifica) there is a greater number of lateral line scales (81 to 84, as compared 
with 76); increased number of vertical lines of scales on dorsal half (9 not 7) and on ventral half (9 not 7); 
decrease in predorsal scales (22 instead of 30); small eye (6.8 to 7, not 4.15); longer pelvic fins (2.2 and 
not 2.7); greater number of giU-rakers (7 + 1 1 instead of 4 + 9). 

Most of the apparent distinctions between the Atlantic and Pacific specimens 
pointed out by Beebe become obscure or are of doubtful value in the light of measure- 
ments and enumerations based on the Acapulco fish. The number of scales in a longi- 
tudinal series seems to differ on the opposite sides of this fish; on the right side I counted 
76 scales with pores (excluding greatly reduced ones on caudal base) and 79 on the left. 



Fishes of the Western North Atlantic 145 

Again, I counted 77 oblique series (running upward and backward) just above the 
lateral line on the right side and 83 on the left. No variation in the number of longi- 
tudinal rows between the lateral line and origin of dorsal, and between the lateral line 
and origin of anal, occurs in the specimen studied, as these are respectively 8 and 6 
full rows on each side. As to the number of scales in advance of the dorsal, the speci- 
men before me has 1 1 modified scales in the median series, and 3 1 oblique series 
of ordinary scales (running upward and backward) on each side of the median row. 

As neither Fowler {24: 652) nor Beebe stated which of the two was counted, I 
made inquiry of Mr. Fowler, who re-examined the type and found 2 i modified scales 
in the median series, and 30—32 ordinary scales In the oblique series. As to the size of 
the eye in proportion to the length of the head, the specimen at hand is intermediate, 
the eye being contained In the head 5.4 times. The pelvic fin also is intermediate. Its 
length 2.35 times In head. I have no doubt that the gill rakers in Dixonina become 
shorter and obscure with age, as they do In Albula, and that In large specimens It Is 
difficult without dissection to see all of the bases, for that Is all that remains of some of 
them. In the specimen from Acapulco, I find 5+10, Including one in the angle with 
the lower limb count, a figure Intermediate between counts given by Beebe and by 
Fowler. 

It may be concluded, on the basis of the evidence presented, that D. pacifica for 
the present at least is a synonym of D. nemoptera. 

Range. This albuHd, as here understood, Is known from the Atlantic from the 
type taken at Santo Domingo and from a specimen collected at Puerto Cabello, Vene- 
zuela." For the Pacific It is known from a specimen taken at Acapulco, Mexico 
(reported by Myers), and from 19 specimens taken off the coast of Costa Rica 
(reported as D. pacifica by Beebe). According to Walford {yi: 1 19), the fish from the 
Gulf of California figured as Albula vulpes by Kumada and Hlyama (47: pi. 5) also 
is this species. 

Synonyms and References: 

Dixonina nemoptera Fowler, Proc. Acad. nat. Sci. Philad. (1910), 191 1: 652, fig. (orig. descr.; type local. 

Santo Domingo, West Indies; 7\NSP 1597); Myers, Copeia, 1936: 83 (Acapulco, Mexico; relation, to 

Albula'); Walford, Copeia, 1939: 119 (specimen from Gulf of California, figured as Albula vulpes in 

^7, is Dixonina). 
Albula nemoptera Metzelaar, Trop. Atlant. Vissch., 1919: 9 (descr.; Puerto Cabello, Venezuela). 
Albula vulpes (not of Linnaeus) Kumada and Hiyama, Mar. Fish. Pacif Cst. Mexico, 1937: 27, pi. 5 in color 

(descr., in part at least, and plate based on Dixonina). 
Dixonina pacifica Beebe, Zoologica, 2^ (8), 1942: 43, figs. 1-5 (orig. descr.; type local. Puerto Culebra, Costa 

Rica; SU 46486; cf. D. nemoptera). 



19. One seen from Brazil. — G. S. Myers. 



TEXT AND FOOTNOTE REFERENCES 



1. IJABCOCK, The Tarpon, 1936; ed. 5, 195 1. 

2. Bean, Bull. N. Y. St. Mus., 60, Zool. 9, 1903. 

3. Beebe, Bull. N. Y. zool. Soc, 30 (5), 1927. 

4. Beebe, Zoologica, N. Y., 2y (8), 1942. 

5. Beebe and Tee-Van, Zoologica, N. Y., 10 (i), 

1928. 

6. Bigelow and Welsh, Bull. U. S. Bur. Fish., 

20 (i), 1925. 

7. Bigelow and Schroeder, Copeia, 1940. 

8. Boulenger, Ann. Mag. nat. Hist., (6) 20, 1897. 

9. Boulenger, Cambr. Nat. Hist., 1904. 

10. Breder, Zoologica, N. Y., 4 (4), 1925. 

11. Breder, Bull. N. Y. zool. Soc, 36, 1933. 

12. Breder, Bull. N. Y. zool. Soc, 4-/, 1939. 

13. Breder, Zoologica, N. Y., 29, 1944. 



14. C_>ADENAT, Rev. Trav. Off. Peches marit., 10, 

Fasc 4, Liv. 4, Pt. 2, 1937. 
75. Cockerell, Smithson. misc. Coll., 56, 19 10. 

16. Cockerell, Science, 191 1. 

17. Cockerell, BuO. U.S. Bur. Fish. (1912), 32, 

1913- 

18. Chadabaram and Menon, Proc zool. Soc 

Lond., Iiy, 1947. 
ig. Cuvier, R&gne Anim., ed. 2, 2, 1829. 
20. Cuvier and Valenciennes, Hist. Nat. Poiss., 

ig, 1846. 



25. Fowler, Bull. Amer. Mus. nat. Hist., 70(1), 

1936. 

26. Fowler, Proc. Acad. nat. Sci. Philad., 9J, 1941. 

27. Cjehringer, J. Fla. Acad. Sci., 22(3), 1958. 

28. Gill, Smithson. misc. Coll., 48 {l), 1905. 

2g. Gowanloch, Bull. La. Conserv. Dep., 23, 

1933- 
30. Gregory, Trans. Amer. philos. Soc, 23 (2), 

1933- 



rUALKETT, Checklist Fish. Canada, 191 3. 
Harrington, Copeia, 1958. 
Heilner, Salt Water Fishing, 1937. 
Henshall, BuU. U. S. Fish Comm., 14, 1895. 
Hildebrand, Copeia, 1934. 
Hildebrand, Sci. Men., N. Y., 44, 1937. 
Hildebrand, Zoologica, N. Y., 24(1), 1939. 
Hildebrand and Schroeder, Bull. U. S. Bur. 

Fish., 43, 1928. 
Hollister, Zoologica, N. Y., 21, 1936. 
Hollister, Bull. N. Y. zool. Soc, 3g (3), 1936. 
De/euJ. 
Hollister, Zoologica, N. Y., 24 (4), 1939. 



43. Irvine, Fish. Gold Coast, 1947. 



21. IjRDMAn, Copeia, i960. 

22. Evermann and Marsh, Bull. U. S. Fish Comm., 

20, 1902. 

23. fowLER, Rep. N.J. Mus. (1906), 1907. 

24. Fowler, Proc. Acad. nat. Sci. Philad., 62, 1910. 



44. Jordan, Guide to Study Fish., 1905. 

45. Jordan and Evermann, Bull. U. S. nat. Mus., 

47 W. 1896. 

46. IVAPLAN, Big Game Angler's Paradise, 1937. 
4y. Kumada and Hiyama, Marine Fish. Pacif. Cst. 

Mexico, 1937. 



146 



Fishes of the Western North Atlantic 



147 



^5. J-/INTON, Bull. U. S. Bur. Fish., 24, 1905. 
^9. Longley and Hildebrand, Publ. Carneg. Instn. 
Wash., 535, 1941. 



50. Marcgrave, Hist.-Nat., Brazil, 1648. 

^I. Merriman, Copeia, 1939. 

52. Morden, Nat. geogr. Mag., 86, 194.4. 

5J. Mowbray, Book of Fishes, 1939. 

5^. Myers, Copeia, 1936. 



61. Schroeder, Rept. U.S. Comm. Fish. (1923), 

1924. 

62. Schultz, Proc. U. S. Nat. Mus., 99, 1949. 
6j. Simpson, Copeia, 1954. 

64. Smith, N. C. geol. econ. Surv., 2, 1907. 
6^. Starks, Stanf. Univ. Publ. Biol., 1913. 
66. Storey and Gudger, Ecology, ij, 1936. 
6^. Sumner, Osburn, and Cole, Bull. U. S. Bur. 
Fish., 31 (z), 191 3. 



SS. Nichols and Breder, Zoologica, N. Y., 9 (i), "55. Tracy, 40th Rep. R. I.Comm.ini. Fish., 191c 



1927. 
56. Deleted. 



^J. XEYo, Bull. Soc. Hist. nat. Toulouse, JO, 1936. 



69. Vincent, Sea Fish. Trinidad, 1910. 
JO. Vladykov and McKenzie, Proc. N. S. Inst. Sci., 
ig, 1935. 



55. IVegan, Ann. Mag. nat. Hist., (8) J, 1909. 

59. Regan, Ann. Mag. nat. Hist., (8)5, 1910. 

60. OcHREiNER and Ribeiro, Arch. Mus. nac. Rio 

de J., 12, 1910. 



JI. Walford, Copeia, 1939. 

J2. Whitley, Rec. Austral. Mus., IJ (6), 1929. 

7J. Whiton and Townsend, Bull. N. Y. zool. Soc, 

31, 1928. 
J4. Wilcox, BuU. U. S. Fish Comm., 6, 1887. 



Suborder Clupeoidea 

COMPOSITE AUTHORSHIPi 



Characters and Key to Families 



HENRY B. BIGELOW 

Museum of Comparative Zoology 
Harvard University 



Characters. Eyes of the usual sort, i.e. not at the tips of short (tubular) cylinders 
or of slender stalks, nor with orbital cavity extended downward. Adipose fin absent. 
Pelvic fins at least no farther forward than midlength of body and farther rearward 
than this in most, but lacking in a few. Fin rays jointed, and most of them branched. 
Scales: the head naked, except in some Alepocephalidae; the trunk scaly in most, 
but naked in some of the Alepocephalidae (p. 250); scales large, thin, their free bony 
edge entire in most, often with crenulated membranous margin, but more or less 
strongly pectinate in a few genera (Brevoonia, p. 342, Fig. 85; and Ethmidium); 
exposed surface smooth or only weakly sculptured; midline of abdomen with scales 
of the ordinary shape in most, but with a double row of larger, sharp-pointed, and 
stiffer scales (scutes) in others (Clupeidae, pp. 257, 259, 260). Barbels absent on chin or 
on throat. Luminescent organs lacking in most, but present in a few (some Alepoce- 
phalidae and all Searsiidae). Lateral line well developed in most families (Engrau- 
lidae, Alepocephalidae, Searsiidae) but not visible in others (Clupeidae) though 
continuing rearward below scales to base of caudal except as it may occasionally 
perforate a few anterior scales (jo: 957). Gill membranes wholly free from isthmus 
in the great majority.^ Bony gular plate absent in chin region between branches of 

1. With the late Samuel F. Hildebrand contributing the family Engraulidae as well as the family Clupeidae, except 
for the genera Harengula by Luis R. Rivas and Dorosoma by Robert R. Miller. 

2. Connected to the isthmus by a thin membrane in the anchovy Cetengraulis (p. 245). 

148 



Fishes of the Western North Atlantic 149 

lower jaw. Gill rakers present in all. Proethmoids not paired. Sides of snout without 
Sensory canal-bearing bonelets. Branchiostegal rays slender, rod-like, not forming 
part of the opercular system. Teeth, if any,' small in most, larger in a few (Chirocen- 
tridae); dental patterns various; teeth never present on the parasphenoid bone in mid- 
line of roof of mouth rearward. Premaxillaries not protractile, except in some Alep- 
ocephalidae and in the Phractolaemidae.* Maxillaries forming part of upper bor- 
der of mouth in the great majority.^ Lower j.aw without predentary or supra-angular 
bones. Temporal foramina present.* Basal radialia of the pectoral fin in one row 
in most, articulating directly with the pectoral girdle; in two rows in a few (among 
the Chirocentridae, ll: 9-11, fig. 23; J: 227, fig. 118, 424). Pectoral girdle with 
a mesocoracoid element. Pelvic girdle not connected to the pectoral girdle. Last 
few Vertebrae not evidently upturned in any; lateral processes bearing pleural ribs 
(parapophyses) simply wedged into pits in vertebral centra, not fused with the lat- 
ter; a lower series of intermuscular bones well developed in many (notably among 
Clupeidae), connected basally to pleural ribs and lying between muscle blocks (myo- 
tomes).' Ovaries with complete Oviducts, so far as known. Swim bladder present in 
most,* dorsal in position, and connected throughout life to the alimentary tract via 
an open pneumatic duct. Stomach with numerous pyloric caeca. Intestine without 
spiral or circular folds in most, but perhaps not in all.' Arterial cone of heart 
never with two rows of valves. Larvae not leptocephalus-like. 

Families. The following families fall within the Suborder as defined above: i 
Clupeidae, including Dussumierinae and Dorosomatinae, p. 257; 11 Alepocephalidae, 
p. 250; III Searsiidae, p. 254; iv Engraulidae, p. 152; v Chirocentridae (regarded by 
Regan \8: 313-314] and Norman [5: 72] as a subfamily of Clupeidae; by Berg 
[j: 227, 422] as a separate Suborder, Chirocentroidei), Indian Ocean, China, and 
Japan; VI Chanidae, tropical Indo-Pacific; vii Phractolaemidae, Niger and Congo rivers; 
VIII Kneridae, rivers of tropical Africa; and ix Cromeriidae, the Nile. 

Clupeidae, Alepocephalidae, Searsiidae, and Engraulidae are represented in the 
North Atlantic. The Dolichopterygidae and the Macristiidae have also been referred 
to the Clupeoidea by one author or another. But Dolichopteryx (Part 4) falls among the 
Argentinoidea, as has long been evident from Parr's excellent illustration of it (6: 37, 
fig. 14). And the affinities of the Macristiidae remain as doubtful today as was the 
case in 191 1, when Regan (7: 204, 205) redescribed the only specimen of the family 
that has yet been seen. 

The Chanidae lack teeth, at least as adults. 

A freshwater family. 

In Nematalosa the mouth is bordered above by the premaxillaries, with the maxillaries covered by the lip, as is 

pictured very clearly by Ridewood (9: 465, fig. 127 as "Chaetoessus"). 

Shown clearly in Ridewood's illustrations; see ftn. 5. 

For a diagram showing the relative positions of the pleural ribs and of the lower intermuscular bones, see Ber- 

tin {4: 704, fig. 462). 

Lacking among the Alepocephalidae; probably among the Searsiidae also. 

A structure suggesting the elasmobranch spiral valve, but probably not homologous, has been reported 

in Chirocentrus (see especially Goodrich, 3: 116, fig. 77) and in Alepocephalus (Cuvier and Valenciennes, 

2: 176). 



150 Memoir Sears Foundation for Marine Research 

Key to Families of Clupeoidea, Known or to be Expected in 
the Western North Atlantic 

I a. Tip of fleshy snout overhanging mouth. Engraulidae, p. 152. 

I b. Tip of fleshy snout not overhanging mouth; at most the upper jaw protruding a 
little beyond lower. 

2 a. Shoulder close behind upper part of gill opening, with a conspicuous pro- 
jecting tubular papilla, open at the tip. Searsiidae, p. 254. 
2 b. Shoulder without projecting tubular papilla. 

3 a. Rear end of base of dorsal fin in advance of origin of caudal by a distance 
as long as head (longer in most); sides silvery; swim bladder well devel- 
oped. Clupeidae, p. 257. 
3 b. Rear end of base of dorsal fin in advance of origin of caudal by a distance 
considerably shorter than head; sides not silvery; no swim bladder. 

Alepocephalidae, p. 250. 



TEXT AND FOOTNOTE REFERENCES 



1. Jderc, Class. Fish., Trav. Inst. zool. Acad. Sci. 

URSS, 5 (2), 1940; also reprinted by J. W. 
Edwards, Ann Arbor, Mich., 1947 (Russ. 
and Eng.). 

2. v^iuviER and Valenciennes, Hist. Nat. Poiss., 

ig, 1846. 

2- VjTooDRicH, Treat. Zool., 1909. 

4. Grasse, et al.. Trait. Zool., JJ (i), 1958. 



5. JNoRMAN, Synop. Orders, Families, Genera of 

Recent Fish, (mimeo.), 1957. 

6. X ARR, Bull. Bingham oceanogr. Coll., J (7), 

1937- 

7. Regan. Ann. Mag. nat. Hist., (8) 7, 191 1. 

8. Regan, Encyc. Brit., 14th ed., 9, 1937. 

9. Ridewood, Proc. zool. Soc. Lond., 2, 1904. 



10. Omitt, Scand. Fish., Pt. 2, 1895. 

11. Starks, Stanf. Univ. Publ. Biol, 6 (2), 1930. 



151 



Family Engraulidae' 



SAMUEL F. HILDEBRAND2 



Characters. Body moderately slender, compressed. Chest and Abdomen in 
American species often carinate, without bony serrae. Mouth horizontal or nearly so. 
Maxillary extending well beyond eye, sometimes nearly or quite to gill opening. 
Snout blunt, generally projecting far beyond tip of mandible. Eye placed well forward, 
in anterior half of head, sometimes with well developed adipose tissue in adults. 
Premaxillaries not protractile, embedded in tissue of snout anteriorly, widely sepa- 
rated medianly. Teeth typically minute (some on jaw enlarged in Lycengraulis)\ present 
on jaws, vomer, palatines, pterygoids, and hyoids. Gill covers generally separate and 
free from isthmus (connected by a thin membrane in Cetengraulis). Pseudobranchiae 
present. Gill rakers generally slender, rarely short, broad, and spiny; increasing in 
number with age in some species (as in Anchovid). Scales thin, cycloid (usually lost 
on preserved specimens), generally if not always forming a sheath on bases of dorsal 
and anal fins; an enlarged scale in axil of pectoral and pelvic fins in American species. 
Lateral line absent. Dorsal fin usually median in position, rarely behind mid- 
length, with about 12-16 rays in American genera. Adipose fin lacking. Caudal 
fin rather deeply forked. Anal with various ray counts, about 15-40 in American 
species. 

Genera. The division of the family into compact and clearly definable genera is 
difficult. Thus there is a divergence of opinion among students concerning the status 
of described genera. Seven genera have been recognized (j), all confined to American 
waters, excepting Engraulis, which, though not represented in the western North 
Atlantic, is very widely distributed in both hemispheres; 75 species have been recog- 

1. Edited, with some revision and expansion, by George S. Myers, Myvanwy M. Dick, Henry B. Bigelow, and 
Yngve H. Olsen. 

2. August 15, 1883-March 16, 1949. 

152 



Fishes of the IVestern North Atlantic 153 

nized, 44 from the Atlantic and 32 from the Pacific, including one species common 
to both coasts of tropical America. For the Atlantic species that come within the scope 
of this work, see the Table of Contents. 

Commercial Importance. Some anchovies, because of their tremendous abundance, 
are of considerable importance. Some are canned whole or are used in the production 
of anchovy paste, and larger species are commonly sold fresh in the market, as in 
South America; they are also used extensively as bait. But their value as a forage fish 
for water birds and creatures that frequent the sea far exceeds their direct value to 
man. Many of the larger commercially important fishes, such as the seatrout, bluefish, 
and mackerels, feed on them extensively. 

Range. Representatives of the family occur in the warmer shore waters through- 
out the world. Some of the species ascend freshwater streams, and probably a few live 
entirely in fresh water. The Engraulidae are most numerous in tropical regions, but 
several species either range into temperate waters or live there permanently. 



Key to Genera of the Western North Atlantic 

I a. Gill covers not connected across isthmus by a membrane. 

2 a. Teeth in jaws small or minute, all of about uniform size. 

3a. Origin of anal fin posterior to origin of dorsal; tip of pelvic fin usually 
below origin of dorsal. 

4a. Maxillary more or less sharply pointed distally, generally reaching 
well beyond joint of mandible, often to margin of opercle, or 
nearly so. 

5 a. Gill rakers close-set, increasing in number with age, 40-50 in 
young, about 130 in adults, on lower limb of first arch; body 
depth usually more than 1^"]^ of length; maximum size about 
250 mm. Anchovia Jordan and Evermann 1896, p. 155. 

5 b. Gill rakers not close-set, only about 15-30 on lower limb, not 
known to increase in number with age; body depth usually 
less than 25^/0 of length; maximum size generally under 
150 mm. Anchoa Jordan and Evermann 1927, p. 162. 

4 b. Maxillary square or rounded distally, rarely reaching joint of man- 
dible. Anchoviella Fowler 191 1, p. 204. 
3b. Origin of anal fin in advance of dorsal origin, rarely under it; tip of 
pelvic fin anterior to a vertical from dorsal origin by a distance equal 
to about 50 "/o of head. 

6 a. Distance of dorsal fin origin from tip of snout much less than 
twice the distance from dorsal origin to caudal base; gill rakers 
short and broad, only about 14 on lower limb of first arch; anal 
with 30-33 rays. Pterengraulis Giinther 1868, p. 228. 



1 54 Memoir Sears Foundation for Marine Research 

6 b. Distance of dorsal origin from tip of snout about twice the distance 

of dorsal origin from caudal base; gill rakers long, slender, about 33 

on lower limb of first arch in young, probably more in adults; anal 

with about 25 rays. Hildebrandichthys Schultz 1949, p. 230. 

2 b. Teeth in jaws, especially the lower ones, notably large, usually unequal in 

size. Lycengraulis Gunther 1868, p. 233. 

I b. Gill covers broadly connected across isthmus by a thin membrane (easily torn). 

Cetengraulis Gunther 1868, p. 245. 



Explanation of Measurements and Counts for Engraulidae. 

Total length : from tip of snout to vertical at rearmost point of longest caudal 
lobe. 

Standard length: from tip of snout to base of caudal fin, i.e. to the last vertebra, 
as nearly as can be determined without dissection. 

Caudal peduncle: least depth unless otherwise noted. 

Length of head: from tip of snout to most distal part of bony opercular margin, 
regardless of whether the straight line between these points is horizontal or oblique. 

Depth of head: vertical distance at joint of mandible. 

PosTORBiTAL LENGTH : from postCHor rim of eye to most distal part of bony oper- 
cular margin. 

Maxillary: from its anterior embedded end (located by touch) to its posterior 
extremity. 

Cheek (smooth bone below and behind eye): distance from rim of eye to its 
distal point; the degree of its posterior angle determined by a protractor. 

Dorsal and Anal fins: the counts include all rays, whether simple and un- 
segmentedj simple and segmented, or branched; the last ray of the anal, even though 
split to the base, was counted as one unless definitely separated at base. 

Length of pectoral: from base of upper ray to tip of longest ray. 

Pectoral axillary scale: distance from base of upper ray of pectoral to tip of 
the modified scale or process. 

Splint: adhering to first ray of pectoral; not considered distinct and not counted. 

Scales or Scale pockets: counted along the side from the upper anterior angle 
of the gill opening to the base of the caudal; as the scales are nearly always lost in 
preserved specimens, they are generally not shown in the illustrations, except for a 
small patch. 

Gill rakers : the counts are given as a formula, e. g. 1 2— 1 6 + 1 5—1 8 (as for Anchoa 
spinifer), meaning a range of 12-16 rakers on the upper limb of the first branchial 
arch and 15—18 on the lower limb. The counts include those rudiments, if present, 
that have a free tip but not those that are mere tubercles. The gill raker at the angle of 
the first arch generally has a plainly visible root, extending either upward or down- 
ward; if the root was directed upward, the raker was included with the upper limb 



Fishes of the Western North Atlantic 155 

count; if directed downward, it was counted with the lower limb count. Rarely, the 
raker at the angle has two roots, one directed upward and one downward; in such 
cases it was included with the lower limb count. 

Vertebrae: total number, including hypural, as determined by dissections. 



Genus Anchovia Jordan and Evermann 1896 

Anchovia Jordan and Evermann, Bull. U. S. nat. Mus., 47(1), 1896: 449; type species by original designa- 
tion, Engraulh macrolepidotus Kner and Steindachner 1864. 

Characters. Body in adults compressed, its depth about 2.6—3.8 in SL. Scales 
firm, rather adherent. Teeth small, not disappearing with age. Gill rakers numerous, 
increasing in number with age, about 40 on lower limb in young, 100 or more in 
large specimens. Anal origin somewhere under anterior 66 "/o of dorsal base. Verte- 
brae 41—43. 

Size. A length of 175 mm and upward is attained by most of the species. 

Species. The genus Anchovia is represented in the Atlantic by A. clupeoides and 
A. nigra, and in the Pacific by three species. 

Range. In the Atlantic, from the West Indies southward to or beyond Pernambuco 
(Recife), Brazil, and in the Pacific from the Gulf of California to Ecuador. 



Key to Species of the Western Atlantic 

I a. Insertion of pelvic fins about equidistant between base of upper ray of pectoral 
fin and origin of anal; snout projecting beyond mandible by about half its length; 
anal fin with 30-35 rays, most frequently with ^'i'l vertebrae 42 or 43. 

clupeoides (Swainson) 1839, below. 
I b. Insertion of pelvic fins nearer to base of upper rays of pectoral fin than to origin 
of anal; snout projecting beyond mandible by about 67"/o of its length; anal fin 
with 28-32 rays, most frequently with 29-31; vertebrae 39-41. 

nigra Schultz 1949, p. 158. 

Anchovia clupeoides (Swainson) 1839 

Sardina Boca Torta, Boc6n, Hachudo 

Figure 25 

Study Material. At least 29 specimens, 75-205 mm TL, from: Puerto Rico; 
Jamaica; Cuba; Trinidad; several places on the Atlantic coast of Panama; the Gulf 
of Venezuela; Laguna de Tacarigua, Venezuela; and Para, Rio Grande do Norte, and 
Pernambuco (Recife), Brazil. Some small specimens, 48 mm TL and upward, are at 



156 Memoir Sears Foundation for Marine Research 

hand, but their proportions are not included because they differ considerably from the 
larger ones, especially in having a more slender body. 

Distinctive Characters. A. clupeoides is closely related to A. rastralis of the Pacific 
coast of Middle America, from which it differs in having a slightly longer anal fin, 
inserted a little farther forward with respect to the dorsal. In specimens of equal size, 
the maxillary is usually a little shorter and tapers more abruptly in clupeoides, and its 
body is somewhat more slender. A. clupeoides apparently grows to a larger size than 
A. rastralis, for none of the latter has been reported to be larger than no mm TL. 
See also A. nigra, p. 161. 




Figure 25. Anchovia clupeoides, 60 mm TL, 44 mm SL, from Gulf of Venezuela, USNM 127552. Drawn 
by Louella E. Cable. 



Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens 75—205 mm TL. 



Scales: ca. 39—43. 

Gill rakers: ca. 35—40 in younger spec- 
imens, 75—1 10 in those of 130 mm 
TL. 

Fin rays: dorsal 13—15; anal 30—35; 
pectoral 13 or 14. 

Vertebrae: 42 or 43 (6 specimens). 



Body: depth 25-31. 
Head: length 27.6-31. 
Snout: length 3.1—4.3. 
Eye: diameter 6.0—8.3. 
Postorbital: distance 1 5.5-1 8.5. 
Maxillary: length 19.5—22. 
Mandible: length 19—20.5. 
Anal fin: length of base 30-33. 
Pectoral fin: length 1 5.4-1 7.3. 

Body strongly compressed, its depth increasing with age, about 3.2—4.0 in SL 
in specimens 75 mm TL and upward {ca. 4.5 at 50 mm). Head 3.2—3.6 in SL, its depth 
about equal to its length without snout in specimens 150—160 mm TL; less in smaller 
ones. Snout short, projecting about half of its length beyond mandible, 6.0—8.0 in 
head. Eye 3.5—4.5 in head. Maxillary abruptly pointed, reaching to or beyond joint 



Fishes of the Western Nortli Atlantic 157 

of mandible, 1.3— 1.5 in head. Mandible 1.4— 1.6 in head. Cheek as long as snout 
and eye in large ones, proportionately shorter in small ones, its posterior angle about 
35°. Gill rakers at angle fully as long as eye in large specimens, shorter in young. 

Dorsal fin originating equidistant between caudal base and anterior half of eye, 
its longest rays often reaching beyond tip of last ray if deflexed. Anal originating about 
under middle of dorsal base, its base 3.0—3.5 in SL. Pelvic small, reaching only a little 
more than halfway to origin of anal, inserted about equidistant between base of pectoral 
and origin of anal. Pectoral fin reaching to or slightly beyond base of pelvic in small 
specimens, often not quite to pelvic in large examples, 1.7— 1.9 in head. Axillary 
scale of pectoral broad, about half as long as fin, 3.0-3.9 in head. 

Color. In alcohol, pale; side of head and lower 75 "/o of body silvery. Young with 
a silvery lateral band, becoming diffuse in specimens about 100 mm long and disap- 
pearing entirely with age. Middle of back with dark streak (missing in some, probably 
due to fading). Fins unmarked. 

Sixe. The largest specimen examined is 1 70 mm TL, but 300 mm has been re- 
ported (7: 413). 

Relation to Man. Its commercial importance is not significant. It occurs oc- 
casionally in the Colon market (Panama) and, according to report, in the Havana 
market (Cuba). 

Range. Its range extends from the West Indies to Panama and at least as far 
southward as Pernambuco (Recife), Brazil, probably to Rio de Janeiro and Sao Paulo. 
All specimens studied were apparently taken in salt or brackish water, with one 
exception; one specimen (CAS 9398) from Lake Rogagua, Bolivia, seems to be this 
species. 

Synonyms and References: 

Engraulis clufeoides Swainson, Nat. Hist. Fish. Amphib. Rept., 2, 1839: 388 (orig. descr.; type local. Pernam- 
buco, Brazil). 

Engraulis froductus Poey, Repert. Fisico-Nat. Cuba, J, 1866: 380 (orig. descr.; type local. Matanzas, Cuba; 
type MCZ 17961); Poey, Repert. Fisico-Nat. Cuba, 2, 1868: 423 (Havana, Cuba); Giinther, Cat. Fish. 
Brit. Mus., 7, 1868: 388 (descr., Cuba and Jamaica); Tortonese, Bull. Mus. Zool. Anat. comp. Torino, 
(3) 47> 1929: 6 (diagn., Puerto Cabello, Venezuela). 

Engraulis surinamensis Steindachner (not of Bleeker), Ichthyol. Beitr., 5, 1879: 55 (descr., Bahia, Brazil). 

Stolephorus clupeoides Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 447 (descr., range, S. surina- 
mensis probably incorrectly put in synon.; see Cetengraulis edentulus, p. 245); Eigenmann, Rep. Princeton 
Exped. Patagonia, 1896-1899, J (4), 1910; 45 (Surinam to Rio Grande do Sul; S. surinamensis Bleeker 
apparently incorrectly put in synon.). 

Stolephorus productus ]oxAir\ and Evermann, Bull. U. S. nat. Mus., 47 (l), 1896: 447 (descr., Cuba and Jamaica); 
Jordan and Rutter, Proc. Acad. nat. Sci. Philad., 1897: 96 (descr., Jamaica); Evermann and Marsh, 
Bull. U.S. Fish Comm., 20 (l), 1900: 90 (descr., Palo Seco, and Ponce, Puerto Rico); Nichols, Bull. 
Amer. Mus. nat. Hist., JJ, 1912: 182 ("Havana market"). 

Anchovia clupeoides Fowler, Proc. Acad. nat. Sci. Philad., 6j, 191 1 : 211 (Santo Domingo and Rio Seco, Puerto 
Rico; E.productus Poey put in synon.); Starks, Stanf. Univ. Publ. Biol., 1913: 9 (notes. Lake Papary, 
Natal, Brazil); Jordan and Scale, Bull. Mus. comp. Zool. Harv., by, 1926: 412 (descr., range, synon., 
partly incorrect); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 50 (range, 
synon., partly incorrect); Fowler, Proc. Acad. nat. Sci. Philad., 5j, 1931 : 393 (diagn., Quaima R., Tri- 
nidad); Howell-Rivero, Bull. Mus. comp. Zool. Harv., 52, 1938: 172 (synon.; the larger of two specimens 



158 Memoir Sears Foundation for Marine Research 

[MCZ 17961, Poey's 36] stated to be holotype); Hildebrand, Bull. Bingham oceanogr. Coll., 5 (2), 

1943: 27, fig. 9 (descr., range, synon.); Schultz, Proc. U. S. nat. Mus., gg, 1949: 39 (synon., locals. 

listed). 
Anchovia froducta Meek and Hildebrand, Field Mus. Publ., Zool., ^5(1), 1923: 210 (descr., Mindi Cut, 

Colon [market], and Porto Bello, Panama); Nichols, Ann. N. Y. Acad. Sci., 10 (2), 1929: 206 (diagn., 

range, "not uncommon about Puerto Rico"). 
Anchovia macrolepidota Fowler (not of Kner and Steindachner), Proc. Acad. nat. Sci. Philad., 1917: 130 (two 

from Colon, Panama, found by present writer to be A. clupeoides). 
Anchoviella clupeoides Fowler, Arqu. Zool. estad. Sao Paulo, J (6), 1941 : 134 (Brazilian rec). 

Doubtful References: 

Stolephorus productus Schreiner and Ribeiro, Arch. Mus. nac, Rio de J., 12, 1903: 93 (Rio de Janeiro, without 

comment). 
Stolephorus clupeoides Eigenmann and Norris, Rev. Mus. pauL, 4, 1904: 360 (one from Sao Paulo, Brazil; 

proportions given indicate ident. probably not correct). 



Anchovia nigra Schultz 1949' 
La anchoa 
Figure 26 

Study Material. Many paratypes, 14—76 mm TL (12— 60 mm SL), and the type, 
125 mm TL (95 mm SL), from Lake Maracaibo and its tributary streams. 

Distinctive Characters. See Relationship and Variation^ p. 161. 

Description. Proportional dimensions in per cent of standard length, not including 
small examples regarded as juveniles because they would distort the picture; dimen- 
sions and counts based on 11 or more specimens from Study Material, 50 mm TL 
and over. 

Body: depth 21-28.9. Pectoral fin: length 16-18. 

Head: length 23-31.3. 

Snout: length 4.15—4.3. Scales: ca. 40—48. 

Eye: diameter 6.1-8.3. ^^^^ rakers: ca. 70+ 140 in adults, fewer 

Postorbital: distance 15—19. in young. 

Mi2«7/<zry: length 18.5-24. Fin rays: dorsal 12-14; ^^^^ 29-31; 

Mandible: length 17—22. pectoral 12 or 13. 

Anal fin: length of base 26—31. Vertebrae: 39—41 (18 specimens). 

Body strongly compressed, rather deep, its depth increasing with age, 3.5-4.75 
in SL in specimens 60-125 "^"^ TL, about ^.c, in specimens around 25 mm. Head 
3.2-4.35, its depth scarcely as great as length of head without snout. Snout short, 
projecting 66»/o of its length beyond mandible, 6.5-7.1 in head. Eye 3.7-4.3. Post- 
orbital length 5.25—6.7. Maxillary abruptly pointed, reaching to or a little beyond 
joint of mandible, i. 25-1. 4 in head. Mandible 1.4-1.5. Cheek moderately long and 

3. Dr. Leonard P. Schultz has ^kindly placed his data on this species in my hands. The proportions and counts used 
in the Description are his and mine. 



Fishes of the Western North Atlantic 



159 



narrow, about equal to length of snout and eye in adults, its posterior angle approxi- 
mately 30°. Gill rakers at angle fully as long as eye in adults, proportionately 
shorter in young. 

Dorsal fin moderately high anteriorly, the longest rays reaching to or beyond tip 
of last ray if deflexed, its origin somewhat variable, generally a little nearer to base of 
caudal than the tip of snout. Anal fin moderately long, its origin somewhat variable, 
sometimes nearly under middle of dorsal base, more frequently anterior to this point, 
its base generally about equal to length of head, 3.2-3.85 in SL. Pelvic fin small, 




Figure 26. Anchovia nigra, type specimen, 125 mm TL, from Maracaibo Basin, Venezuela, USNM 121761. 
Drawn by A. M. Awl. 



reaching a little more than halfway to origin of anal, inserted a little nearer to base of 
pectoral than to origin of anal. Pectoral fin rather large, reaching well beyond base 
of pelvic, 1. 6-1. 9 in head. Axillary scale of pectoral fin broad at base, variable in 
length, usually reaching to middle of fin or beyond, 3.4—4.1 in head. 

Color. In alcohol, back a pale straw color. Sides silvery. Silvery lateral band in 
specimens up to 75 mm TL, the band very narrow in 40-mm specimens, becoming 
proportionately broader and less well defined ventrally, diffused with silvery color on 
lower part of sides in largest example ; dark streak present on middle of back in large 
examples but scarcely evident in small ones in which the dark punctulations that later 
form the band remain separate; scattered punctulations extending down to silvery lateral 
band. Dorsal and caudal fins with dark points; other fins plain translucent; small spec- 
imens, 75 mm TL and under, with dark dots along base of anal fin and on ventral 
surface of caudal peduncle. 

Sixe. The largest specimen examined, the type (USNM 121761), is 125 mm 
(5 in.) TL or 95 mm SL. Since a length of 300 mm (12 in.) has been reported 
for its near relative, clupeoides, it may be assumed that larger examples of nigra will 
be found. 



1 6o Memoir Sears Foundation for Marine Research 

Development. The smallest specimen, only 14 mm, is very slender, its depth 9.15 
and the head 4.0 times in SL. In this specimen the mouth is nearly terminal, the lower 
jaw is very thin, the gape ends under the eye, and the maxillary is imperfectly developed. 
The convoluted intestine, which is usual in larval herring and herring-like fishes, is 
externally visible. All fins, however, are sufficiently developed to have at least rudimen- 
tary rays. The body is pale, no doubt transparent in life, as general pigmentation has 
not taken place. The juvenile markings consist of dark dots, which, exclusive of a few 
spots at the nape, are on the lower part of the body; and a few less definite ones are on 
the side of the head. An elongate median streak extends backward from the isthmus; 
also there are spots along each side of the ventral edge of the abdomen that become 
darker and more concentrated along the base of the anal and are reduced to a single 
row on the midventral line of the caudal peduncle. A row of indefinite dark dots ex- 
tends along the lateral side of the tail from above the anal base to the caudal base ; there 
are a few more dark dots on the caudal fin. 

The body of a fish 20 mm TL is proportionately much deeper than the 14-mm 
example, as its depth is contained c^.c^ times in SL. The proportion of the head does not 
differ greatly from either the small or large ones described, being 3.75 in SL. Much 
progress in the development of the mouth parts has been made, however, for the snout 
is more pointed and it projects prominently beyond the lower jaw. The maxillary is 
definitely visible and extends well beyond the posterior margin of the eye, where it 
ends in a rather blunt point. The intestine is no longer visible externally. Pigmen- 
tation on the lower part of the body remains about the same as in the smaller ex- 
ample, but dorsally, especially on the median line of the back and on the head, the 
dark dots are more numerous. 

The advances in development with growth between 20—30 mm are not great. The 
increase in proportionate depth has continued, with the depth now 5.0 times in SL. 
The snout has become more pointed and projects somewhat more strongly, about as in 
adults. Some of the gill rakers are developed, 10 + 20 having been counted in one spec- 
imen. Pigmentation has advanced somewhat, as dark dots on the back and on the head 
have become larger and more numerous. The dark line behind the isthmus has virtually 
disappeared. The dark spots on the side of the abdomen, along the base of the anal, 
and on the ventral side of the caudal peduncle have become less distinct. A few elongate 
dark markings at the base of the outer rays of both lobes of the caudal, already present 
in the 20-mm specimen, have become more distinct. 

In a fish 40 mm long, the depth is contained 4.5 times in SL, and the head 3.5 
times. The cheek and opercular bones are well outlined. And the maxillary, which 
now reaches the margin of the opercle, is provided with minute teeth along its free 
margin. The gill rakers on the first arch are so numerous and close-set that they are 
difficult to count; about 35-45 seem to be present. Pigmentation continues to advance; 
the dark dots on the back have become more numerous and those on the lower parts, 
exclusive of the ones along the anal base and on the peduncle behind the anal, have 
disappeared. The upper parts of the body are slightly straw-colored and the sides 



Fishes of the Western North Atlantic 1 6 1 

of the head are bright silvery; the lower part of the body shows a somewhat silvery 
sheen, and there is slight evidence of a silvery lateral band on the anterior half of 
the body. 

A specimen 70 mm long is shaped essentially like the type, which is 125 mm TL; 
the body is proportionately about as deep in a 70-mm specimen as in the type ; how- 
ever, the gill rakers on the first arch are fewer, as 44 + 70 were counted, whereas about 
70+ 140 seem to be present in the type. Furthermore, the smaller specimen retains 
the silvery lateral band which the larger specimen has lost; it also retains the juvenile 
dark markings along the base of the anal fin, which are missing in the larger one. 

Relationship and Variation. This species is very close to cltipeoides, from which it 
differs significantly in the number of vertebrae, the range in 1 7 specimens of nigra 
being 39-41 and in 22 specimens of clufeoides 42 or 43. The anal fin may begin 
under the middle of the dorsal base or well in advance of that point in nigra, consequently 
it overlaps with clupeoides, in which this fin begins about under the beginning of the 
second third of the dorsal base. Furthermore, nigra generally has fewer rays in the anal 
fin than clupeoides, 28—32 (av. around 30) in 38 specimens, and 30—35 (av. 23) '" 
58 specimens of clupeoides. The lower average number of rays in the anal of nigra is 
reflected in Its somewhat shorter base, 3.2-3.85 in SL in 38 specimens of nigra, but 
3.0—3.5 in 32 specimens of clupeoides. The pectoral fin of nigra extends well beyond the 
base of the pelvic fin in smaller specimens and only slightly beyond this point in the 
largest example, being variable with age in this respect ; in clupeoides the pectoral gener- 
ally does not extend quite so far back as the pelvic fin; nevertheless, enough overlapping 
occurs so that the character is not diagnostic. However, the place of insertion of the 
pelvic fin generally is diagnostic; in nigra it is nearly always inserted somewhat nearer 
to the base of the upper pectoral rays than to the origin of the anal, whereas in clupeoides 
its insertion generally is almost exactly equidistant between the two points mentioned. 
Finally, the snout projects about 66 "/q of its length beyond the mandible in nigra, and 
only about half of its length beyond the mandible in clupeoides. 

A. nigra is also closely related to A. rastralis of the Pacific coast of tropical America. 
But in nigra the pelvic fins are inserted somewhat nearer to the pectoral base than to 
the anal origin, whereas in rastralis they are inserted notably nearer to the anal origin 
than to the pectoral base. The anal in nigra begins at, or more usually in advance of, 
the vertical from the middle of the dorsal base, whereas in rastralis it generally begins 
at or behind this point; and nigra has 39—41 vertebrae, whereas rastralis (like clupeoides) 
has 42 or 43. 

Habitat and Range. A. nigra is known only from the fresh or slightly brackish 
water of Lake Maracaibo, Venezuela, and from the lower sections of its tributary 
streams. Apparently it does not enter the salt water of the Gulf of Venezuela, where 
clupeoides has been taken. 

Reference : 

Anchovla nigra Schultz, Proc. U. S. nat. Mus., 99, 1949: 39, fig. 4 (orig. descr.; type local. Lake Maracaibo, 
Venezuela; type USNM 121761). 



1 62 Me??ioir Sears Foundation for Marine Research 

Genus Anchoa Jordan and Evermann 1927 

Anchoa Jordan and Evermann, Proc. Calif. Acad. Sci., ^ (16), 1927: 501; type species by original designation, 
Engraulis compressiis Girard 1858. 

Characters. Body usually quite elongate, moderately to strongly compressed. 
Maxillary long, reaching to, or more usually beyond, joint of mandible, frequently 
to or nearly to margin of opercle, always more or less pointed (except in very young). 
Gill rakers not numerous, seldom more than 22 on upper limb or more than 28 on 
lower limb except in cubana among Atlantic species, not increasing in number with 
age and growth. Anal origin under base of dorsal, rarely wholly behind it. Vertebrae 
38-44, rarely 45 (in Atlantic species). 

Remarks. This genus differs from Anchoviella principally in its longer maxillary, 
which is somewhat sharply pointed rather than rounded or squared distally. From 
Anchovia it differs in the fewer gill rakers, the number of which does not increase with 
age. It differs principally from Engraulis (not represented in the western North Atlantic) 
in the more compressed body (also frequently deeper), and usually in the fewer verte- 
brae. 

Species and Range. This genus contains many species that range the Atlantic 
from Massachusetts (rarely Nova Scotia) to Argentina, and the Pacific from California 
to Peru. Some of the species enter brackish or almost fresh water, but none is known 
to run far upstream. 



Table I. Frequency Distribution of Anal Rays in Anchoa. 

^ Number of Rays >^ 

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 

spimfer • • — 2 313 6 2 2 

argenteus — ■ i 

duodecim • i 

trinitatis 2 4 7 5 2 

cayorum 3222212 4 

pectoralis 6 8 2 I 

mitchilli mitchiUi • i 8 22 63 74 51 30 ii i 

miuhilli diaphana — ■ ■ — 2 11 55 83 62 44 14 5 

parva I 8 18 18 3 2 

januaria i 312 4 

choerostoma 5 14 16 9 3 

tricolor 2 93017 6 

cubana 913 7 9 + I 

girtiburgi 3 3 2 I I ■ 

lamprolaenia — 1103143815614 3— 

hepsetus hepsetus 21 28 42 57 20 3 

hepsetus colonensls 3 413 13 3 i 

lyolepis — 2 711 5 i— 

filifera — i 3 3 5 2 



Fishes of the Western North Atlantic 163 

Key to Species of the Western North Atlantic 

I a. Anal fin with 25 rays or more {m. mitchilli rarely with 24); anal base 25-35.60/0 
of SL (j>ectoralis sometimes only 23.4''/o). 

2 a. Origin of dorsal about an eye's diameter nearer to tip of snout than to base 
of caudal; anal with 30 rays or more. 

3a. Origin of anal under or somewhat in advance of middle of dorsal base; 
32 or more anal rays. 

4a. Anal with 35-40 rays; pectoral fin generally reaching far beyond 
base of pelvic fin, 5.0-5.7 in SL; margin of caudal black. 

spinifer (Cuvier and Valenciennes) 1848, p. 167. 
4b. Anal with 32 rays; pectoral fin not reaching far beyond base of pelvic 
fin, 6.5 in SL; margin of caudal not black. 

argenteus Schultz 1949, p. 169. 
3 b. Origin of anal behind dorsal base; 30 anal rays. 

duodecim (Cope) 1869, p. 170. 

2b. Origin of dorsal nearer to base of caudal than to tip of snout; anal usually 

with less than 30 rays (mitchilli and trinitatis sometimes with as many as 30). 

5 a. Origin of dorsal nearer to posterior margin of eye than to base of 

caudal ; longest rays of dorsal reaching to or beyond tip of last ray when 

deflexed; pectoral usually with more than 12 rays (cayorum sometimes 

with only 12). 

6a. Depth of body 22.2-25.50/0 of SL; gill rakers usually 16-19+ 19- 
22 (rarely only 15 on upper limb); anal base 29.4-33.40/0 of SL. 

trinitatis (Fowler) 19 15, p. 171. 

6b. Depth of body 18. 2-22. 20/^ of SL; gill rakers 12-14+ 17-19 

{cayorum rarely with 1 5 on upper limb, pectoralis occasionally with 

20 on lower limb); anal base 23.5-29.40/0 of SL. 

7 a. Maxillary narrowly pointed, extending nearly or quite to margin 

of opercle, 19.2-22.50/0 of SL; origin of anal well in advance 

of middle of dorsal base; pectoral with 12 or 13 rays, generally 

failing to reach base of pelvic fin by about half of diameter of 

eye, 1 3.9-1 6.60/0 of SL. cayorum (Fowler) 1906, p. 173. 

7 b. Maxillary bluntly pointed, extending only slightly beyond 

joint of mandible, 1 7.7-1 9.00/0 of SL; origin of anal generally 

under middle of dorsal base; pectoral with 15-17 rays, reaching 

nearly to base of pelvic fin, 1 7.8-1 8.8 0/0 of SL. 

pectoralis Hildebrand 1943, p. 174. 
5 b. Origin of dorsal nearer to base of caudal than to posterior margin of 
eye, very rarely equidistant; longest rays of dorsal usually failing to 
reach tip of last ray when deflexed ; pectoral with 11 or 12 rays. 

mitchilli (Cuvier and Valenciennes) 1848, p. 176. 



1 64 Memoir Sears Foundation for Marine Research 

8 a. Greatest thickness of body usually exceeding depth of caudal 

peduncle; depth i5.8-25''/o of SL; pectoral often failing to reach 

base of pelvic by diameter of eye, 12-1 8 "/o of SL; vertebrae 39-44, 

most frequently 42. m.mitchiUi Hildebrand 1943, p. 176. 

8 b. Greatest thickness of body often less than depth of caudal peduncle; 

depth I 8. 2-2 6. 5 "/o of SL; pectoral often extending to or nearly to 

base of pelvic, 1 3-2o"/o of SL; vertebrae 38-42, most frequently 40. 

m. diaphana Hildebrand 1943, p. 176. 

lb. Anal fin with less than 25 rays {lamprotaenia occasionally with 25 or 26 rays, 

choerostoma rarely with 25); anal base i7.2-25"/o of SL (Januaria sometimes 26). 

9 a. Pectoral without a long, filamentous ray (rarely with a slight filament in 

lamprotaenia); origin of dorsal fin usually nearer to base of caudal than to 

tip of snout, rarely equidistant. 

10 a. Origin of anal in advance of base of last dorsal ray; depth of head at 
joint of mandible exceeding postorbital length; snout much shorter 
than eye; cheek shorter than snout and eye combined; silvery lateral 
band much narrower than eye. 

1 1 a. Gill rakers on lower limb 2 3-3 3, on upper limb 18-23 (rarely only 1 7). 
12a. Depth of body i8.2-22''/o of SL; pectoral fin 14-18. 5«/o 
of SL; gill rakers rather robust, not closely crowded, not 
difficult to count. 

13a. Origin of anal under or near middle of dorsal base; 
depth of body 20-22 o/^ of SL. 

14a. Maxillary sharply pointed distally, extending 
nearly to margin of opercle, 1.2— 1.35 in head; 
vertebrae 38—40 (rarely 41). 

parva (Meek and Hildebrand) 1923, p. 181. 

14 b. Maxillary bluntly pointed distally, extending 

only to or slightly beyond joint of mandible, 
1. 3-1. 5 in head; vertebrae 41 or 42. 

januaria (Steindachner) 1879, p. 183. 
1 3 b. Origin of anal far behind middle of dorsal base ; depth 
of body 18.2-210/0 of SL. 

15 a. Cheek as long as eye and fully 75 "/o of snout 

length; pectoral fin i 5.6— i8.5''/o of SL; axillary 
scale of pectoral about half of length of fin, 
3.0—4.0 in head; anal with 22—24 rays. 

choerostoma (Goode) 1874, p. 185. 

15b. Cheek scarcely longer than eye; pectoral fin 

14— 14.8 "/o of SL; axillary scale of pectoral fully 

75''/o of length of fin, 1.9—2.75 in head; anal with 

18—22 rays. tricolor (Agassiz) 1829, p. 186. 



Fishes of the Western North Atlantic 165 

12 b. Depth of body i5-i8''/o of SL; pectoral fin 1 2.5-1 3.7 "/o 

of SL; gill rakers very slender, close-set, often difficult to 

count. cubana (Poey) 1868, p. 188. 

lib. Gill rakers on lower limb usually 16-22 (rarely 23 in h. hepsetus), 

on upper limb usually 14—20 (rarely 13 or 21). 

i6a. Depth ofbody 1 7.4-1 8.8 "/o of SL; maxillary bluntly pointed 

distally, its upper margin rounded, extending to joint of 

mandible, 17.2— 18.2 "/o of SL; vertebrae 44 or 45. 

ginsburgi Hildebrand 1943, p. 190. 
i6b. Depth of body i8.5-23*'/o of SL; maxillary usually sharply 
pointed distally, generally extending beyond joint of man- 
dible, sometimes to or nearly to margin of opercle (mod- 
erately short, rather blunt distally, and rounded above in 
lamprotaenia), 1 9-24.5 "/„ of SL; vertebrae 39—43 {h. hep- 
setus sometimes with 44). 

17 a. Longest rays of dorsal fin usually reaching well 
beyond tip of last ray when deflexed; origin of anal 
about under beginning of posterior third of dorsal 
base; axillary scale of pectoral only a little more than 
half of length of fin, 3.0—3.6 in head; cheek fully as 
long as eye and half of snout. 

lamprotaenia Hildebrand 1943, p. 192. 
1 7 b. Longest rays of dorsal fin failing to reach tip of last 
ray when deflexed; origin of anal usually about under 
middle of dorsal base, sometimes posterior to middle ; 
axillary scale of pectoral 2.0—2.5 in head; cheek usually 
not much longer than eye. 

hepsetus (Linnaeus) 1758, p. 194. 
I 8 a. Pectoral fin often failing to reach base of pelvic 
by diameter of eye; silvery lateral band usually 
75''/o of width of eye. 

h. hepsetus Hildebrand 1943, p. 194. 

1 8 b. Pectoral fin frequently extending to or nearly to base 

of pelvic; silvery lateral band often scarcely wider 

than pupil, h. colonensis Hildebrand 1 943, p. 1 94. 

lob. Origin of anal under or behind base of last dorsal ray; depth of head 

at joint of mandible equal to postorbital length; snout scarcely shorter 

than eye, 4.3-5.5 in head; cheek about as long as snout and eye 

combined; silvery lateral band as broad as eye. 

lyolepis (Evermann and Marsh) 1902, p. 200. 

9b. Pectoral with a long filamentous ray; origin of dorsal fin usually nearer to 

tip of snout than to base of caudal. filifera (Fowler) 191 5, p. 202. 



I 66 Memoir Sears Foundation for Marine Research 

Table II. Frequency Distribution of Gill Rakers on Upper Limb of First Arch 

in Anchoa* 
, Number of Gill Rakers 



"> 



12 13 14 15 16 17 18 19 20 21 22 23 

Spin if er i 2 11 14 3 — — — — — — — 

argenteus — — — i — — — — — — — — 



tnmtatis — — — I i 7 9 i — — — — 

cayorum i 12 24 3 — — — — — — — — 

fectoralis i 12 2 — — — — — — — — — 

mitchilli miuhilli — — — 8 38 63 56 5 — — — — 

miuhilH diaphana — — — 9 95 118 44 8 — — — — 

parva — — — — — — 3 25 10 — — — 

januaria — — — — — — — — 2 i 2 2 

choerostoma — — — — — i 9 9 4 — — — 

tricolor — — — — — — i 7 19 8 i — 

cubana — — — — — 2 3 9 12 5 8 5 

ginsburgi — — — — i 2 i — — — — — 

lamprotaenia — 7 ;6 47 13 6 4 — — — — — 

hepsetus hepsetus — — — i 19 52 47 23 9 — — — 

hepsetus colonensis — — — 3 9 24 12 i — — — • — ■ 

lyolepis — — — i 5 8 23 12 6 i — — 

filifera — — — — — 5 4 4 — — — — 

* When the counts exceed the number of specimens listed in the text, the rakers were counted on both ante- 
rior arches. 

Table III. Frequency Distribution of Gill Rakers on Lower Limb of First Arch 

in Anchoa* 
Number of Gill Rakers ^ 



15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 

spinifer 4 11 24 4 

argenteus i 

duodecim • 2 — 

trinitatis i 3 7 8 

cayorum 18 40 22 i 

pectoralis 3 13 4 2 — 

mitchilli mitchilli 421 45 70 54 21 6 

mitchilli diaphana 12 64 125 92 46 10 3 

parva — 7 10 14 13 i — 

januaria — 2 4 4 5 

choerostoma — 10 12 9 6 

tricolor II 20 25 8 2 

cubana — 3 211 6 818 8 7 2 2 2 

ginsburgi 2 3 8 

lamprotaenia — 2 26 94 82 31 8 5 

hepsetus hepsetus 9 27 97 93 47 2 i 

hepsetus colonensis 41525 4 

lyolepis 4 11 14 11 7 5 5 4 

filifera 3 3 12 i 

* When the counts exceed the number of specimens listed in the text, the rakers were counted on both 
anterior arches. 



Fishes of the Western North Atlantic 



167 



Anchoa spinifer (Cuvier and Valenciennes) 1848 

Sardine 

Figure 27 

Study Material. Many specimens, 30— 175 mm TL: from the Atlantic at the 
Gatun Ivocks, Canal Zone; Port-of-Spain, Trinidad; and Bahia, Cachoeira, and Santos, 
Brazil; from the Pacific at Pedro Miguel and Miraflores locks. Canal Zone; Panama 
Bay; and Guayaquil, Ecuador. 




Figure 27. Anchoa spinifer, go mm TL, from Gatun Locks, Canal Zone, USNM 127560. Drawn by Louella 
E. Cable. 

Distinctive Characters. The long anal fin, the dorsal's high rays in the anterior part, 
the dorsal's forward position, the long low head and especially its prolonged postorbital 
portion, the long narrow cheek, and the dark margin of the caudal fin distinguish this 
species from all of its relatives. Large examples generally are proportionally deeper 
than small ones, and generally Atlantic specimens are rather deeper than Pacific ones. 
However, there is so much overlapping that, with the material at hand, specimens from 
the opposite coasts cannot be separated by this character, nor seemingly by any other. 
The pectoral fin also varies greatly in proportional length, but this also seems to be 
an individual variation. Nevertheless, the comparison is not satisfactory because nearly 
all of the Atlantic specimens are old and faded, not in good condition. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 1 1 Atlantic specimens and 15 Pacific specimens, 66—175 "^"^ ^^• 



Body: depth 19.2-25. 
Head: length 25-27.8. 
Snout: length 3.0-4.3. 
Eye: diameter 4.5—6.25. 
Postorbital: distance 14. 9-1 6.6. 
Maxillary: length 20.8-23.2. 
Mandible: length 19.6—20.3. 



Anal fin: length of base 33.1-36.6. 
Pectoral fin: length 17.5-20. 

Scales: ca. 45. 

Gill rakers: 12-16+ 15-18. 

Fin rays: dorsal 14—16; anal 35—40; 

pectoral 12—14. 
Vertebrae: 43-45 (15 specimens). 



1 6 8 Memoir Sears Foundation for Marine Research 

Body quite elongate, rather strongly compressed, its depth 4.0-5.2 in SL, in- 
creasing somewhat with age and growth. Head 3.6-4.0 in SL, its depth scarcely 
exceeding its postorbital length. Snout projecting nearly its full length beyond man- 
dible, 6.6-7.5 i'^ head. Eye small, 4.2-5.0. Postorbital part of head long, 1.5-1.7 
in head. Maxillary long and pointed, reaching to or nearly to margin of opercle, 
1. 1 -1. 25 in head. Mandible i. 22-1. 4. Cheek in large specimens much longer than 
snout and eye, about equal to snout and eye in specimens about 80 mm TL; posterior 
angle sharp, about 30°. 

Dorsal fin high anteriorly, its longest ray reaching far beyond tip of last ray if 
deflexed, its origin generally rather more than an eye's diameter nearer to tip of snout 
than to base of caudal. Anal 2.8-3.0 in SL, its origin generally somewhat in advance 
of middle of dorsal base. Pelvic reaching about 66 "/o of distance to anal, inserted 
about equidistant between origin of anal and base of pectoral. Pectoral large, reaching 
to, and generally beyond, base of pelvic, 1.3-1.5 in head. Axillary scale of pectoral 
broad, reaching somewhat beyond midlength of fin, 2.4—2.9 in head. 

Color. In alcohol, pale above. Lower half of side silvery. Young sometimes with a 
poorly defined silvery lateral band. Back with dusky punctulations. Fresh specimens 
125 mm TL and upward (taken in Pedro Miguel and Miraflores locks. Canal Zone) 
bright orange, this color contrasting conspicuously with black margin of caudal fin. 

Size. The largest specimen seen, about 175 mm TL or 7 inches (caudal fin dam- 
maged) and 138 mm SL, is probably near the maximum size attained. 

Range and Habitat. The range in the Atlantic, so far as known, extends from 
Panama to Santos, Brazil. On the Pacific side it is known from Panama Bay to Guaya- 
quil, Ecuador. The West Indies has been included in the general range (7: 410), but 
I have seen no specimens from that locality and have found no other record. In 
rather extensive collecting on the Atlantic coast of Panama (during the dry seasons), 
only a single small specimen was secured; but on the Pacific side it was exceedingly 
abundant in Pedro Miguel and Miraflores locks when these were drained in 1937; 
however, it was not seen elsewhere during four different seasons of collecting. The 
water ranges from near oceanic salinity in the lower flights of Miraflores Locks (at sea 
level) to fresh in Pedro Miguel Locks. 

Synonyms and References: 

Engraulis spinifer Cuvier and Valenciennes, Hist. Nat. Poiss., 21, 1848: 39 (orig. descr.; type local. Cayenne, 
French Guiana); Gunther, Cat. Fish. Brit. Mus., 7, 1868: 394 (descr., type local.); Steindachner, 
Ichthyol. Beitr., 8, 1879: 58 (descr.; Guiana; Bahia and Cachoeira, Brazil; w. coast of Panama). 

Stolephorus spinifer ]oxd.3.r\ and Everraann, Bull. U. S. nat. Mus., 47 (i), 1896: 448 (descr., range); Eigenmann, 
Mem. Carneg. Mus., 5, 1912: 449 (synon., descr., Georgetown, British Guiana); Puyo, Bull. Soc. Hist, 
nat. Toulouse, "jo, 1936: 65, 163 (diagn., habitat French Guiana); Puyo, Bull. Soc. Hist. nat. Toulouse, 
80, 1945: 105, fig. 3 (descr., French Guiana); Puyo, Faune Emp. Fran^., 12, Poiss. Guyane Franj., 
1949: 156, fig. 79 (descr., French Guiana). 

Anchovia spinifera Gilbert and Starks, Mem. Cal. Acad. Sci., 4, 1904: 46, pi. 8, fig. 15 (descr. based on two 
from Panama Bay); Meek and Hildebrand, Field Mus. Publ., ZooL, J5 (i), 1923: 207 (synon., descr., 
range); von Ihering, Rev. Industr. Anim., Anno i (3), 1930: 233 (ref., diagn.); Hildebrand, Zoologica, 
N. y., 24 (i), 1939: 25, 30, 36 (Panama Canal locks). 



Fishes of the Western North Atlantic 



169 



Anchoviella spinifera Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6^, 1926: 409 (descr., Cachoeira, Bra- 
zil; "Panama" presumably Pacific coast; "West Indies" in range); Jordan, Evermann, and Clark, Rep. 
U. S. Comm. Fish. (1928), 2, 1930: 50 (range); Fowler, Arqu. Zool. estad. Sao Paulo, j, 1941 : 134 
(refs., Brazil). 

Anchoa spinifer Hildebrand, Bull. Bingham oceanogr. Coll., 5 (2), 1943: 38, fig. 10 (synon., descr., range, 
local abund.). 



Anchoa argenteus Schultz 1949 
Figure 28 

Study Material. Known only from the type, ca. 115 mm TL (caudal fin damaged), 
94 mm SL, from Lake Maracaibo, off Pueblo Viejo, Venezuela. 

Distinctive Characters. This species is closely related to A. spinifer., with which it 
agrees in the shape and position of the dorsal fin, the long low head, the long narrow 
cheek, and in many other respects. But it differs significantly from that species (so far 
as can be determined from a single specimen) in having fewer anal rays, in a slightly 
greater number of gill rakers on the lower limb, in the longer mandible, in the shorter 
pectoral fins, and in the absence of a dark margin on the caudal fin. 




Figure 28. Anchoa argenteus, type specimen, 94mm TL, from Maracaibo Basin, Venezuela, USNM 121777. 
Drawn by A. M. Awl. 



Description. Proportional dimensions in per cent of standard length, and counts, 
based on the type specimen, 94 mm SL. 



Body: depth 22.5. 
Head: length 26.5. 
Snout: length 4.25. 
Eye: diameter 5.55. 
Postorbital: distance 18.5. 
Maxillary: length 23.5. 
Mandible: length 20.6. 



Anal fin: length of base 32. 
Pectoral fin: length 15.5. 

Scales: ca. 43. 

Gill rakers : 14+19. 

Fin rays: dorsal 16; anal 32; pectoral 13. 



170 Memoir Sears Foundation for Marine Research 

Body elongate, rather strongly compressed, its depth 4.5 in SL. Head 3.75, its 
depth a little greater than its postorbital length. Snout shorter than eye, projecting 
much more than half of its length beyond tip of mandible, 6.25 in head. Eye small, 4.8. 
Postorbital length 5.4. Maxillary moderately pointed, reaching margin of opercle, 
1. 1 5 in head. Mandible 1.3. Cheek narrow, much longer than snout and eye, its 
posterior angle about 30°. Gill rakers at angle of first arch about 66 "/o of length 
of eye, with strong serrae on inner edge, somewhat expanded at tips. 

Dorsal fin moderately elevated anteriorly, its margin nearly straight, its longest 
rays reaching well beyond the last ray if deflexed, its origin about an eye's diameter 
nearer to tip of snout than to base of caudal. Anal fin fairly long, its origin under middle 
of dorsal base, its base notably longer than head, 3.1 in SL. Pelvic fin reaching a little 
more than halfway to origin of anal, inserted a little nearer to base of pectoral than to 
origin of anal, 3.1 in head. Pectoral fin reaching beyond base of pelvic by about half 
of diameter of eye, 6.5 in SL, 1.7 in head. Axillary scale of pectoral extending to 
beginning of distal third of fin, 2.55 in head. 

Color. In alcohol, somewhat straw-colored on back, changing rather abruptly to 
silvery on the side at level of upper margin of eye; back with dusky punctulations, these 
extending onto basal 66 "/o or so of dorsal fin, and onto outer rays of caudal fin. 

Range. Known only from Lake Maracaibo, off Pueblo Viejo, Venezuela. 

Reference : 

Anchoa argenteus Schultz, Proc. U. S. nat. Mus., gg, 1949: 45, fig. 5 (orig. descr.; type local. Lake Maracaibo, 
off Pueblo Viejo, Venezuela; type USNM 121777). 



Anchoa duodecim (Cope) 1869 
New Jersey Anchovy 

Study Material. Known only from the type, 97 mm long (77 mm SL), taken at 
Beesley's Point, New Jersey, ANSP 1363. 

Distinctive Characters. This species has been considered identical with A. mitchilli. 
However, it differs strikingly from other local forms in its deep body, high dorsal fin, 
the positions of the dorsal and anal fins, and in several other respects. It is of interest 
that the species has never reappeared in collections. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the type specimen, 77 mm SL. 

Body: depth 25. Anal fin: length of base 25. 

Head: length 27. Pectoral fin: length 18.2. 

Snout: length 4.7. 

Eye: diameter 5.8. Scales: partly lost, ca. 38. 

Postorbital: distance 16.3. Gill rakers: 14 or 15 + 20. 

Maxillary: length 22. Fin rays: dorsal 14; anal 30; pectoral 

Mandible: length 21. 12. 



Fishes of the Western North Atlantic 



171 



Body rather short, deep, and well compressed; dorsal profile more strongly convex 
than ventral profile. Head short, deep, its depth equal to its postorbital length and half 
of eye. Snout rather short, projecting a little less than a third of its length beyond 
mandible, 5.8 in head. Eye small, 4.7. Postorbital 1.6 in head. Maxillary fairly 
short, reaching joint of mandible, broad and abruptly pointed, 1.25 in head. Man- 
dible 1.3. Cheek as long as snout and eye, its posterior angle about 35°. Gill rakers 
at angle nearly as long as eye. 

Dorsal fin high anteriorly, its longest rays reaching far beyond tip of last ray if 
deflexed, its origin an eye's diameter nearer to tip of snout than to base of caudal. Anal 
rather long and low, its origin half of an eye's diameter posterior to end of dorsal base, 
and about equidistant between base of caudal and base of pectoral, its base 4.0 in SL. 
Pelvic fin large, shorter than pectoral by diameter of pupil, extending about 66 "/o of 
the distance to origin of anal, inserted a little nearer to joint of mandible than to 
origin of anal. Pectoral reaching base of pelvic, 1.5 in head. Axillary scale of pec- 
toral a little more than half of the fin length, 2.7 in head. 

Color. Old preserved specimen grayish brown above. Lower two-thirds of side 
silvery; no lateral band and no punctulations visible. 

Range. Known only from Beesley's Point, New Jersey. 



405 (orig. descr.; type local. Beesley's 



Synonyms and References: 

Engraulis duodecim Cope, Trans. Amer. philos. Soc, JJ (2), li 

Point, New Jersey; type ANSP 1363). 
Anchovia duodecim Fowler, Rep. N. J. St. Mus. (1905), 1906: 11 1 (descr.); Fowler, Proc. Acad. nat. Sci. 

Philad., 63, 191 1 : 219; Fowler, Proc. biol. Soc. Wash., jj, 1920: 148. 
Anchoa duodecim Hildebrand, BuU. Bingham oceanogr. Coll., 8 {2), 1943: 46 (synon., descr. based on type). 



Anchoa trinitatis (Fowler) 1915 
Figure 29 

Study Material. A total of 19 specimens, including the type, from Trinidad and 
Venezuela, ANSP, UMMZ, USNM. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 10 specimens from Study Material., 55— 1 10 mm TL. 



Body: depth 22-25.6. 
Head: length 24.3—27. 
Snout: length 3.9— 4-35. 
Eye: diameter 6.75—7.6. 
Postorbital: distance 11. 7-1 3.3. 
Maxillary: length 16.6—22. 
Mandible: length 1 6.4-1 9. 



Anal fin: length of base 29.4—33.2. 
Pectoral fin: length 1 6.6-1 9. 

Scales: ca. 38-42. 

Gill rakers: 15-19+ 18-22. 

Fin rays: dorsal 13 or 14; anal 26—30; 

pectoral 13 or 14. 
Vertebrae: 41 (i specimen). 



172 Memoir Sears Foundation for Marine Research 

Body rather strongly compressed, its greatest thickness about equal to depth of 
caudal peduncle, its depth 3.9-4.5 in SL. Head short, 3.7-4.1 in SL, its depth equal 
to its postorbital length and about 66''/(, of diameter of eye. Snout extending about 
66 "/o of its length beyond mandible, 5.7-6.6 in head. Eye 3.4-3.7. Postorbital 
1. 9-2. 1 in head. Maxillary pointed, extending nearly to margin of opercle, 1.2— 1.4 
in head. Mandible i. 4-1. 53. Cheek about as long as eye and half of snout, its 
posterior angle about 40°. 

Dorsal fin moderately high anteriorly, its longest rays reaching beyond tip of 
last ray if deflexed, its origin equidistant between base of caudal and a point at anterior 
half of eye. Anal origin under, or somewhat in advance of, middle of dorsal base, its 
base 3.0-3.4 in SL. Pelvic reaching fully halfway to anal, inserted a little nearer to 
origin of anal than to base of pectoral. Pectoral sometimes not quite reaching to pelvic, 
1.35— 1.52 in head. Axillary scale of pectoral rather broad at base, scarcely extending 
to vertical at midlength of fin, 2.8—3.6 in head. 




-Bp-r^^*^" 







1/ 

Figure 29. Anchoa trinitatis, type specimen, from Port-of-Spain, Trinidad, ANSP 45083, modified after 
Fowler (191 5: 527, fig. 3). Drawn by Alice C. Mullen. 



Color. In alcohol, pale above. Side of head and lower part of body silvery. Silvery 
lateral band faint, narrower than pupil, missing in specimens presumably preserved 
originally in formalin. Back and upper surface of head with dusky punctulations, some- 
times forming two almost continuous lines posterior to dorsal fin. 

Size. The largest individual seen is 1 10 mm TL. 

Range. So far as known, this species is confined to four localities: Port-of-Spain 
and Vessigney, Trinidad; Laguna de Tacarigua (on coast 85 miles east of Caracas), 
and Rio Cano de Sagua, Sinamaica, Venezuela. 

Synonyms and References: 

Anchovia trinitatis Fowler, Proc. Acad. nat. Sci. Philad., 191 5 : 527, fig. 3 (orig. descr.; type local. Port-of-Spain, 
Trinidad; type ANSP 45083). 



Fishes of the Western North Atlantic 



173 



Anchoviella trinitatis Fowler, Proc. Acad. nat. Sci. Philad., 83, 193 1 : 392 (Vessigney, Trinidad). 

Anchoa trinitatis Hildebrand, Bull. Bingham oceanogr. Coll., 5 (2), 1943: 96, fig. 40 (descr. cf. several 

related species, range); Schultz, Proc. U. S. nat. Mus., gg, 1949: 44 (synon., specimens and local, listed; 

counts of rays, gill rakers, vert., and scales). 



Anchoa cayorum (Fowler) 1906 

Key Anchovy, Manjua (Cuba) 

Figure 30 

Study Material. A total of 26 or more specimens, 50-85 mm TL, from Tortugas, 
Florida; Cabanas Bay, Cuba; Cozumel Island, off Yucatan, Mexico; Belize, British 
Honduras; and Hailer's Rock, Florida Keys (type and paratype). 




Figure 30. Anchoa cayorum, 85 mm TL, 68 mm SL, from Cabanas Bay, Cuba, USNM 82341, with pectoral 
fin enlarged. Drawn by Louella E. Cable. 

'Distinctive Characters. Museum specimens generally have been identified with 
A. mitchilli, presumably because of the long anal fin, but cayorum differs in the larger 
size it attains (largest seen 85 mm), in the notably fewer gill rakers, and in the much 
broader and more prominent silvery lateral band. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 50—85 mm TL. 



Body; depth 18-21. 
Head: length 25-28.5. 



Snout: length 4.3-5.0. 
Eye: diameter 5.8—7.4. 



174 Memoir Sears Foundation for Marine Research 

Postorbital: distance 1 3-3-1 5-3- Scales: ca. 39-42. 

Maxillary: length 19.2-22.5. Gill rakers: 13-15+ 15-17 (rarely 18). 

Mandible: length 18.5-20. Fin rays: dorsal 13-15; anal 26-28 
Anal fin: length of base 25-29.3. (rarely 25 or 29); pectoral 12 or 13. 

Pectoral fin: length 1 3.9-1 6.6. Vertebrae: 43 (3 specimens). 

Body slender, moderately compressed, its depth 4.75-5.5 in SL. Head 3.5—4.0 
in SL, its depth equal to its postorbital length and about half of eye. Snout pro- 
jecting about 66 "/o of its length beyond tip of mandible, 5.0-6.0 in head. Eye 3.5- 
4.25. Postorbital 1.65-2.0 in head. Maxillary long and pointed, its upper margin 
straight, reaching nearly or quite to margin of opercle, i.i— 1.35 in head. Mandible 
1. 2-1. 4. Cheek equal to, or a little longer than, eye and 75 "/o of snout, its posterior 
angle rather sharp, about 40°. 

Dorsal fin with the longest rays reaching well beyond tip of last ray if deflexed, 
its origin about equidistant between base of caudal and anterior margin of eye. Anal 
fin with origin under beginning of second fourth of dorsal, its base 3.4—4.0 in SL. 
Pelvic inserted somewhat nearer to base of pectoral than to origin of anal. Pectoral 
falcate, sometimes failing to reach base of pelvic by diameter of pupil, occasionally 
reaching nearly to base of pelvic, the first (uppermost) ray generally as long as the 
second and about twice the last ray, 1.5-1.9 in head. Axillary scale of pectoral 
scarcely reaching beyond midlength of pectoral, 3.0-3.5 in head. 

Color. In alcohol, pale. Side of head silvery. Silvery lateral band bright, about as 
wide as eye. 

Range and Habitat. The range extends from the Florida Keys, where it is appar- 
ently rather rare, to Cuba, Yucatan, and British Honduras. The specimens examined 
were apparently taken in shallow water, near the shore. 

Synonyms and References: 

StolepAorus mitchilU Bean, Bull. U. S. Fish Comm. (1888), 8, 1890: 206 (Cozumel I., off Yucatan, Mexico; 

specimens re-exam, and found to be A. cayorum). 
Anchovia choerostoma cayorum Fowler, Proc. Acad. nat. Sci. Philad., ^8, 1906: 85, fig. 4 (orig. descr.; type local. 

Mailer's Rock, Florida Keys; type ANSP 30613). 
Anchovia cayorum Fowler, Proc. Acad. nat. Sci. Philad., 6j, 191 1 : 219 (types relisted). 
Anchoviella cayorum Jordan, Evermann and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 49 (synon.). 
Anchoa cayorum Hildebrand, Bull. Bingham oceanogr. CoU., 8 (2), 1943: 50, fig. 17 (synon., descr., relation., 

range, type material re-exam.). 



Anchoa pectoralis Hildebrand 1 943 

Study Material. Only the type material, 64 specimens, 43-65 mm TL, from 
Vigia, Brazil, MCZ 18004, listed for "Brazil, Louis Agassiz." 

Distinctive Characters. This species is closely related to A. mitchilli diaphana, 
A.parva, and A. januaria^ from all of which it differs in the more numerous pectoral 
rays. Other differences are evident in the descriptions and tables. 



Fishes of the JVestern North Atlantic 



175 



Description. Proportional dimensions in per cent of standard length based on 
7 specimens, the counts based on many more. As the state of preservation is not very 
good, some of the proportions may not be entirely true. 



Body, depth 19.5-23.0. 
Head: length 23.8-26.2. 
Snout: length 3.35-4.16. 
Eye: diameter 6.7-8.0. 
Postorbital: distance 12.9— 13. 
Maxillary: length 1 7.7-1 9.0. 
Mandible: length 1 5.8-1 7.6. 



Anal fin: length of base 23.4—25.6. 
Pectoral fin: length 1 7.8-1 8.8. 

Scales: ca. 35—40. 
Gill rakers : 13 or 1 4 + 1 7— 1 9. 
Fin rays: dorsal 14—16; anal 25—27; pec- 
toral 15—17 (most frequently 16). 
Vertebrae: 42 (2 specimens). 



Body moderately slender, strongly compressed, its depth 4.35—5.1 in SL. Head 
3.8—4.2 in SL, its depth nearly equal to its length without snout. Snout short, pro- 
jecting about half of its length beyond mandible, 6.0—7.0 in head. Eye 3.2—3.7. Post- 
orbital 1.75—2.0 in head. Maxillary bluntly pointed, its upper margin rounded, 
reaching somewhat beyond joint of mandible, 1.2— 1.4 in head. Mandible 1.4— 1.6. 
Cheek about as long as eye, its posterior angle broad, about 50°. 

Dorsal fin with longest rays reaching to, and occasionally beyond, tip of last ray if 
deflexed, its origin generally a little nearer to base of caudal than to middle of eye. Anal 
with origin generally under middle of dorsal, its base 3.9—4.25 in SL. Pelvic reaching 
only about halfway to anal, inserted equidistant between origin of anal and base of 
pectoral, or slightly nearer the former. Pectoral falcate, reaching nearly to base of 
pelvic, 1.3— 1.4 in head. Axillary scale of pectoral pointed, about 66''/o of length 
of fin, 2.5 in head. 

Color. Old specimens in alcohol, brownish silvery. A prominent silvery lateral 
band present, about as wide as eye. Back with dark dots, in more or less distinct 
longitudinal rows. Base of anal with black dots ; caudal with dark dots and with a dark 
margin. 

Range. Known only from the type material from Vigia, Brazil, collected in 1865 
or 1866 by the Thayer Expedition of Harvard University. [Stieler's Atlas lists the 
name Vigia for two places in Brazil, one at the mouth of Rio Para (Tocantins), the other 
well inland on Rio Jequitinhonha in the Province of Minas Gerais.-Y.H.o.]. 

Synonyms and References: 

Anchoviella mitchilli Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6j, 1926: 405 (descr., distr., MCZ 

18004). 
Anchoa pecloralis Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943: 52, fig. 18 (orig. descr.; type local. 

Vigia, Brazil; type MCZ 35276). 



176 Memoir Sears Foundation for Marine Research 

Anchoa mitchilli (Cuvier and Valenciennes) 1848 

Bay Anchovy 

Figures 31, 32 

Study Material. A total of 100 specimens, 52-102 mm TL; for characters con- 
sidered important, such as body depth, pectoral length, number of vertebrae, gill 
rakers, and anal rays, many more were used; taken at many places from Woods Hole, 
Massachusetts, to Yucatdn, Mexico. 

Distinctive Characters. The nearest relative is A.parva, a southern species that 




Figure 31. Anchoa mitchilli mitchilli, 85 mm TL, from Woods Hole, Massachusetts, USNM 125582. Drawn 
by Louella E. Cable. 



occurs in the West Indies and on the mainland coast from Panama to Venezuela. This 
relationship is discussed in the account dealing vn'Csx parva (p. 181). 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on Study Material. 



Body: depth 16-27. 
Head: length 22-26.5. 
Snout: length 3.7-7.25. 
Eye: diameter 5.8—8.2. 
Postorbital: distance 11— 13.5. 
Maxillary: length 1 7.5-21. 
Mandible: length 16. 5-1 8.5. 
Anal fin: length of base 25-30. 

Body rather slender, moderately compressed, its greatest thickness usually ex- 
ceeding the depth of caudal peduncle, its depth 3.65-6.3 in SL. Head 3.75-4-5) its 
depth equal to its length without snout and fourth of eye. Snout short, projecting not 
more than a fourth of its length beyond tip of mandible, 5.0-7.0 in head. Eye 2.9-3.9. 
Postorbital short, 1.8-2. i in head. Maxillary pointed, extending nearly to margin of 



Pectoral fin: length 12—20. 

Scales : (generally lost in preserved speci- 
mens) ca. 38—44. 

Gill rakers : 15-19 + 20-26. 

Fin rays: dorsal 14—16; anal 24—30; 
pectoral 11 or 12. 

Vertebrae: 38-44 (1,233 specimens). 



Fishes of the Western North Atlantic 



177 



opercle, 1.1-1.3 in head. Cheek short and broad, about as long as eye, its posterior 
angle approximately 60°. Gill rakers somewhat shorter than eye. 

Dorsal fin rather low, with nearly straight margin, its last ray scarcely longer 
than the one immediately before it, its longest rays failing to reach its tip if depressed, 
the origin varying between a little nearer to caudal base than to upper anterior angle 
of gill opening and equidistant between caudal base and posterior margin of eye. Anal 
usually originating somewhat posterior to dorsal origin, its base 3.3—4.0 in SL. Pelvic 
fin very small, not quite reaching halfway to origin of anal, inserted nearer to anal 




Figure 32. Anchoa mitchilH diafhana, 72 mm TL, from Grand Isle, Louisiana, USNM 119790. Drawn by 
Louella E. Cable. 



origin than to pectoral base by diameter of pupil. Pectoral variable, failing to reach 
base of pelvic by a distance equal to, or even greater than, diameter of eye in northern 
specimens, sometimes reaching pelvic base in southern fish, 1.4— 1.85 in head. Axil- 
lary scale of pectoral rather narrow, 2.3—3.0 in head. 

Color. Preserved specimens, straw-colored above. Lower parts pale silvery. Sides of 
head bright silvery. Middle of side with a silvery band, often nearly as wide as eye in 
northern specimens, generally narrower in southern ones. Back with dusky punctula- 
tions, arranged in two more or less definite rows posterior to dorsal fin ; base of anal 
with dark spots continued as a single unbroken median ventral line on caudal peduncle; 
caudal with many dark points. Color in life, greenish with bluish reflections above; 
lower parts pale; abdominal walls translucent. 

Size. A total length of 100 mm (4 in.) is seldom exceeded, the usual length being 
around 75 mm (3 in.). The largest examples measured were taken in New York, where 
this species evidently grows larger than in the southern part of its range, for the largest 
specimen among thousands from the Gulf of Mexico was only 75 mm long. 

Development and Growth. The egg generally is slightly oblong, but occasionally 
round; the major axis is about 0.75 mm and the minor one about 0.62 mm. It is 
transparent and has no oil globule. The yolk consists of separate masses that appear 



178 Memoir Sears Foundation for Marine Research 

as large cells under the microscope, a character that seems to be common to all species 
of Engraulldae whose eggs have been studied. The egg floats at or near the surface 
and hatches in about 24 hours at room temperature, at least at Beaufort, North Carolina, 
during early summer.* 

The newly hatched fish, 1.8-2.0 mm long, is rather slender, perfectly trans- 
parent, and has no pigment spots. The yolksac is absorbed within about two days, and 
the large mouth, which is terminal at this stage, then seems to be functional. The pro- 
jecting snout is not developed until the young fish reaches a length of about 20-25 "^"^• 
But the fins are sufficiently developed at a length of about 1 5 mm to permit a fairly 
accurate enumeration of the rays. The number of anal rays and the relative positions 
of the dorsal and anal fins are very useful in identifying the postlarvae with the adults. 

The early young of the season seem to become sexually mature during their first 
summer, for specimens 45—60 mm long that remained quite transparent, taken late 
in July and during the first half of August at Beaufort, N. C, contained well devel- 
oped roe. 

Spawning. In the vicinity of Beaufort, N. C, spawning takes place at least from 
early spring to midsummer. The eggs were taken in tows there as early as April 2 1 (in 
1930) and as late as July 15 (in 1929), while gravid fish were seen as late as August 12 
(from my unpublished notes). 

Food. The food apparently consists mostly of Mysis and copepods, the latter being 
the sole food of the young. Other items taken are small fish, gastropods, and isopods 
(5:110). 

Enemies. This fish, which is exceedingly numerous in a large part of its range, is 
preyed upon extensively by predatory fish and water birds. 

Variation. The variation within this species is so great that I recognized a 
northern subspecies, A. mitchilli mitchilli, and a southern one, A. mitchilli diaphana 
(5: 87-94). The specimens from the southern part of the range have rather consistently 
deeper and more strongly compressed bodies, fewer gill rakers, fewer vertebrae, a 
longer pectoral fin, and a narrower silvery lateral band. These difi-'erences are shown by 
comparing the characters of a small lot of specimens from Woods Hole, Massachusetts, 
with another lot from Grand Isle, Louisiana, which show the widest divergence 
(Table iv). Complete intergradation of the two subspecies occurs in the midsection of 
their range, as in North and South Carolina. 

Commercial Importance. This Anchovy is used as bait and to a limited extent in 
the preparation of anchovy paste. However, it is important chiefly as forage for many 
important foodfishes. As this fish usually occurs in schools and is a rather sluggish 
swimmer, it is an easy prey of many predators. 

Range and Habitat. Cape Cod, Massachusetts (rarely northward into the Gulf of 
Maine), and southward to Yucatan, Mexico. Brazilian records probably are referable 
either to A.parva or A.januaria. While A. mitchilli occurs along the outer and exposed 

4. The information as to the eggs and young fish is in part from Kuntz (9: 13-19) and in part from my 
field notes. 



Fishes of the Western North Atlantic 179 

Table IV. Comparison of Divergent Specimens of A. mitchiUi. 

Woods Hole, Massachusetts Grand Isle, Louisiana 

Eight specimens, 62-85 ^""1 TL Sixteen specimens, 48-76 mm TL. 

Depth, 19-21 «/o of SL. Depth, 23-260/0 of SL. 

Greatest thickness of body, usually exceeding depth of Greatest thickness of body, less than depth of caudal 

caudal peduncle. peduncle. 

Pectoral fin, 13-13. 5 "/o of ^L. Pectoral fin, 16. 5-17. 5 "/o of SL. 

Gill rakers, 24-28 on lower limb. Gill rakers, 21-23 on lower limb. 

Vertebrae, 43 or 44 in four specimens. Vertebrae, 39—41 in nine specimens. 

Silvery lateral band about as broad as eye. Silvery lateral band scarcely broader than pupil. 

beaches at Beaufort, N. C, at least to a depth of 1 5-20 fms., it is found more often 
in inside waters, especially in areas with muddy bottoms and brackish water. The dis- 
tribution is so general that the fish may be expected also in "grassy" areas and along 
sandy beaches. 

Migrations. It was formerly thought that this species, as well as several other 
anchovies, migrated northward and southward with the seasons. This theory was based 
on the fact that in the northern part of the range, as at Woods Hole, Massachusetts, 
the fish is present only during the summer. However, it has been shown (j: 87—94) 
that virtually every section of the coast within the range of mitchilli has a distinctive 
population, which suggests that whatever migration takes place is an inshore and 
offshore movement. 

Details of Occurrence. The most northern records are for Casco Bay, Maine, and 
for Provincetown at the tip of Cape Cod, Massachusetts; these have been based on 
stray specimens only. But it is described as abundant in the Woods Hole region and 
in Rhode Island waters (both adults and larvae), common in New York waters, exceed- 
ingly abundant (often in large schools) at many New Jersey localities, and second in 
abundance (if not the most abundant) in Chesapeake Bay. It has been recorded for 
North Carolina, South Carolina,^ and Georgia, and it is at least tolerably plentiful in 
the Indian River on the eastern coast of Florida. No certain evidence has been found of 
its presence among the Florida Keys, but it has long been known to be plentiful in the 
inlets and passes all along the western coast of Florida. It has been recorded for so many 
localities in Mississippi, Louisiana (Cameron, Grand Isle, and Lake Ponchartrain), 
Texas (Galveston, Corpus Christi, Dickinson Bayou), and Mexico (Vera Cruz, Rio 
Panuco, Tampico) as to prove it practically universal in suitable situations all around 
the coastline of the Gulf of Mexico to Yucatan. But present indications are that Yucatan 
marks the southern boundary of its range, there being no reliable record of its presence 
anywhere to the southward around the Central or South American shores of the Carib- 
bean. 

Reports of it for Cuba {20: 57; ly. 421; 7: 405) on the strength of two 
specimens from Poey in the Museum of Comparative Zoology seem to have been 
based on a misinterpretation of what Poey wrote about this Anchov)'. Actually the 

5. The Museum of Comparative Zoology has a specimen from Charleston, collected by Louis Agassiz. 



I 8o Memoir Sears Foundation Jor Marine Research 

locality given in the Museum catalogue for these specimens is "United States," to 
which a label in the bottle in my handwriting adds "Sent to Poey by Gill for com- 
parison." 

Synonyms and References: 

EngrauHs mitchilli Cuvier and Valenciennes, Hist. Nat. Poiss., 21, 1848: 50 (orig. descr.; type local. New 
York; mentions EngrauHs louisiana^); Giinther, Cat. Fish. Brit. Mus., 7, 1868: 391 (synon., descr., 
range). 

StoUphorus mitchilli Jordan and Gilbert, Proc. U. S. nat. Mus. (1882), 5, 1883: 248 (descr., synon., includ. 
EngrauHs duodecim Cope, a valid species; Galveston, Texas; Pensacola, Florida; Woods Hole, Massa- 
chusetts); Swain, Bull. U. S. Fish Comm. (1882), 2, 1883: 57 (range; Princetown and Woods Hole, 
Massachusetts; Pensacola, Florida; Lake Ponchartrain, Louisiana; Galveston, Texas; and Cuba, which 
is probably A.farva\ includ. EngrauHs duodecim Cope); Bean, Bull. U. S. Fish Comm. (1887), 7, 
1888: 149 (Ocean City, Beesley's Pt., and Longport, New Jersey; local abund.); Henshall, Bull. U. S. 
Fish Comm. (1889), 9, i8gi : 373 (very abund. in most inlets and passes of Florida west coast); Smith, 
Bull. U.S. Fish Comm. (1892), 12, 1894: 361 (abund. in New Jersey; relation, to commerc. fishes); 
Evermann and Kendall, Bull. U. S. Fish Comm. (1892), 12, 1894: 105 (Galveston, Corpus Christi, 
Dickinson Bayou, Texas); Moore, Bull. U. S. Fish Comm. (1892), 12, 1894: 359 (near Sea Isle City, 
New Jersey); Lonnberg, OefVers. Svensk.Vet. Akad. Forh.,57, 1894: 1 15 (St. Petersburg, Florida); Jordan 
and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896-. 446 (descr., range, synon., includ. EngrauHs duodecim 
Cope); Evermann and Bean, Rep. U. S. Comm. Fish. (i8g6), 1897: 241 (many at Cocoa and St. Lucia 
on Indian R., Florida); Smith, Bull. U. S. Fish Comm. (1897), 77, 1898: 92 (Woods Hole, Massa- 
chusetts); Smith and Bean, Bull. U. S. Fish Comm. (1898), 18, 1899: 184 (Potomac R. at Bryants 
Pt., Maryland); Evermann, Rep. U. S. Comm. Fish. (1898), 1899: 309 (Baldwin Lodge, Mississippi; 
Lake Lapourde, Louisiana); and (1899), 1900: 56 (Florida); Bean, Rep. Forest Comm. N. Y. (1901), 
1902: 311 (refs., range, New York, econ. import.); Bean, Bull. N. Y. St. Mus., 60, ZooL 9, 1903: 
218 (descr., range, a forage fish; New York); Latham, Copeia, 191 7 (Orient, Long Island, New York); 
Breder, Zoologica, N. Y., 2, 1922: 338 (Sandy Hook Bay, New Jersey). 

Anchovia mitchilli Fowler, Rep. N. J. St. Mus. (1905), 1906: 1 10, fig. (descr., Delaware R. off Ft. Delaware, 
Delaware; Beesley's Pt. and Ocean City, New Jersey); and (1906), 1907 : 267 (Sea Isle City, New Jersey; 
taken for bait); Smith, N. C. geol. econ. Surv., 2, 1907: 134 (synon., descr., range, Morehead City, 
North Carolina); Jordan and Dickerson, Proc. U. S. nat. Mus., j^, igo8: 12 (diagn., mouth of Rio 
Panuco, Tampico, Mexico); Kendall, Occ. Pap. Boston Soc. nat. Hist., 7, 1908: 42 (Casco Bay, Maine; 
Provincetown and Woods Hole, Massachusetts; Rhode Island); Weymouth, Proc. U. S. nat. Mus., ^8, 
1910: 136 (Cameron, Louisiana); Fowler, Proc. Acad. nat. Sci. Philad., 65, 191 1: 219 (Woods Hole, 
Massachusetts; Seaside Park, Beesley's Pt., Carson's Inlet, New Jersey; Ft. Delaware, Delaware; Tol- 
chester Beach, Maryland); Sumner, Osburn, and Cole, BuU. U. S. Bur. Fish. (191 1), JJ (2), 1913: 743 
(Woods Hole, Massachusetts); Kuntz, Bull. U. S. Bur. Fish. (1913), JJ, 1914: 13-19, figs. 25-46 
(spawn., Beaufort, North Carolina; develop, of eggs and larvae) ; Bigelow and Welsh, Bull. U.S. Bur. Fish. 
(1924), .^0(1), 1925: 124 (descr., size, range. Gulf of Maine); Nichols and Breder, Zoologica, N. Y., 
9, 1927: 44 (diagn., range. Orient, New York; life hist, notes). 

Anchoviella mitchilli Jordan and Seale (in part; not of Cuvier and Valenciennes), BuU. Mus. comp. Zool. Harv., 
6"] (11), 1926: 405 (synon., includ. EngrauHs duodecim Cope; descr., range; Brazilian and West Indian 
specimens not this species); Hildebrand and Schroeder, BuU. U. S. Bur. Fish. (1927), ./J (i), 1928: 109 
(synon., descr., food, spawn., eggs descr., size, abund., habitat, Chesapeake Bay); Breder, Field Bk. 
Mar. Fish. Atl. Cst., 1929: 71 (genl.); Jordan, Manual Vert. Anim. NE U.S., ed. 13, 1929: 43 
(descr., range) ; Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 48 (synon., 
includ. EngrauHs duodecim Cope; range); Hildebrand and Cable, BuU. U. S. Bur. Fish., 46, 1930: 388 
(econ. import., spawn., Beaufort, North Carolina; develop, of eggs and young cf. A. hepsetus; distinguish, 
characters this species and A. hepsetus; growth rate); Perlmutter, 28th Rep. N. Y. Conserv. Dept. 

6. Said to have been described and figured by LeSueur, but since no reference to E. louisiana LeSueur is given and 
since none can now be found, the name presumably was taken from a manuscript, and Hildebrand, in his review 
of the American EngrauUdae regarded it as having no standing in modern nomenclature [3). 



Fishes of the Western North Atlantic i 8 1 

(1938), Suppl., 15 (2), 1939: 18 (around Long Island, New York; spawn.); Greeley, 28th Rep. 

N. Y. Conserv. Dept. (1938), Suppl., 15 (2), 1939: 83 (Long Island, New York; a forage fish). 
Anchoa tnitchilli mitchilH Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 19+3: 87, fig. 37 (synon., descr., 

cf. A. mitchiUi diafhana, range). 
Anchoa mitchiUi diaphana Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943: 91, fig. 38 (orig. descr. of 

subspecies A. m. diaphana; relation., range); Gunter, Publ. Inst. mar. Sci. Texas, I (l), 1945: 33 

(depth, temp., salinity, season, spawn., growth, Texas coast). 

Negative References: 

Stolephorus mitchiUi Bean, Bull. U. S. Fish Comni. (1888), 12, 1890: 206 (many at Cozumel Island, Yucatan; 

USNM 37053, 37105 are h. cayorum); Schreiner and Miranda-Ribeiro, Arch. Mus. nac, Rio de J., 

12, 1903: 93 (Rio de Janeiro, where this species does not occur). 
Anchoviella mitchiUi Fowler, Arqu. Zool. estad. Sao Paulo, 3 (6), 1941 : 135 (refs. Brazilian recs., where this 

species does not occur). 
Anchovia mitchiUi von Ihering, Rev. Industr. .'\nim., Anno i (3), 1930: 232 (Brazil, where the species does 

not occur). 

Anchoa parva (Meek and Hildebrand) 1923 

Manjua (Cuba) 

Figure 33 

Study Material. At least 34 specimens, 35-60 mm TL, and many more used for 
some measurements and counts, from the following localities: many from Jamaica; 
two from Cuba; many from Colon and Porto Bello, Panama, and from Gatun Locks, 
Canal Zone; one from Laguna de Tacarigua, Venezuela; and one from Trinidad; in 
USNM, MCZ, UMMZ, BOC, Museo Poey collections. 

'Distinctive Characters. A. parva is rather close to both subspecies of mitchiUi^ 
differing principally in having fewer anal rays and more numerous gill rakers, though 
there is slight overlapping in each instance (see Tables i, 11, and iii), in the smaller 
anal fin (see Description)., in the origin of the anal, which is somewhat farther back 
under the base of the dorsal, and in the more strongly projecting snout (Figs. 31-33). 
For the relationship oi parva to januaria, see p. 183. Its closest relative is curta of 
the Pacific coast of Central and South America, from which it is scarcely more than 
subspecifically distinct (j: 84). 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 34 specimens in Study Material, 35-60 mm TL. 

Body: depth 20-22. Pectoral fin: length 14. 3-1 7.4. 
Head: length 23.3-27.8. 

Snout: length 4.0-5.5. Scales: ca. 38-42. 

£j5: diameter 6.9-9.5. Gill rakers: 18-20 + 23-27. 

Poj/orto^/: distance 1 1. 1 -1 4.3. Fi" rays: dorsal 13-16; anal 21-23 

Maxillary: length 17.3-2 1.7. (rarely as many as 25); pectoral 12 

Mandible: length 15-19. °r n- 

y/««/>: length of base 21-25. Vertebrae: 38-40, rarely 41 (61 spec- 
imens). 



I 82 Memoir Sears Foundation for Marine Research 

Body moderately deep, strongly compressed, its depth 4.5-5.0 in SL. Head 
3.6-4.3, its depth usually equal to its length without snout. Snout 4.5-6.0 in head, 
extending about half of its length beyond mandible. Eye 3.0-3.5. Postorbital 1.8-2.3 
in head. Maxillary pointed, reaching to margin of opercle, 1.2-1.6 in head. Man- 
dible 1. 4-1. 75. Cheek scarcely longer than eye, its posterior angle about 60°. 

Dorsal fin with the longest rays failing to reach tip of last ray if deflexed, its origin 
generally equidistant between base of caudal and about middle of eye. Anal with origin 
under, or more frequently slightly anterior to, middle of dorsal base, its base 4.0-4.8 
in SL. Pelvic reaching scarcely halfway to anal, inserted about equidistant between 
origin of anal and base of pectoral. Pectoral rather variable in length, usually failing 




Figure 33. Anchoa farva, paratype, 59 mm TL, from Fox Bay, Colon, Panama, USNM 79554- Drawn by 
Louella E. Cable. 

to reach pelvic by diameter of pupil, 1.5-1.7 in head. Axillary scale of pectoral 
generally reaching to beginning of distal third of fin, 2.3-3.0 in head. 

Color. In alcohol, very pale (presumably preserved originally in formalin). Silvery 
lateral band scarcely as wide as pupil (indistinct or missing in some specimens). Back 
with dusky punctulations, at least posterior to dorsal fin and extending onto base of upper 
rays of caudal; larger dark dots usually present at base of anal and continued as a single 
row onto median line of caudal peduncle. 

Size. It is apparently small, as the largest specimen seen is only 60 mm (2.4 in.)TL. 

Range and Habitat. Known from the West Indies and the Atlantic coast of Panama 
and Venezuela. Records of A. mitchilli for the West Indies, in part at least, are refer- 
able to this species. This fish inhabits salt and brackish water, and fresh water at least 
to a limited extent. It was very numerous in all three flights of Gatun Locks, Canal 
Zone, in 1935. The water in the lowermost flight was quite salty, that in the middle one 
brackish, and that in the uppermost flight, fresh. It has been reported also as occur- 
ring in fresh water in Venezuela. 

Synonyms and References: 

Anchovia farva Meek and Hildebrand, Field Mus. Publ., ZooL, 25(1), 1923 : 202, pi. 10, fig. 2 (orig. descr.; 
type local. Porto Bello, Panama; type USNM 81767; also recorded from Col6n); Hildebrand, Zoologica, 



Fishes of the IVestern North Atlantic 



1H3 



N. Y., 24. (l), 1939: 25, 35 (numerous in Gatun Locks, Canal Zone; erroneously reported from Mira- 

flores Locks, Canal Zone, which is A. curtd). 
Anchoviella niitchiUi Jordan and Scale (in part; not of Cuvier and Valenciennes), Bull. Mus. comp. Zool. Harv., 

6^, 1926: 405 (part of Cuban specimens, A.farvd). 
Anchoviella farva Jordan, Evermann, and Clark, Rep. U. S. Coram. Fish. (1928), 2, 1930: 49 (ref to orig. 

descr.). 
Anchoa farva Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943 : 83, fig. 35 (refs., descr., relation., range); 

Schultz, Proc. U. S. nat. Mus., gg, 1949: 43 (synon.; Venezuela; notes, counts). 

Doubtful Reference: 

Anchoviella parva Fowler, Proc. Acad. nat. Sci. Philad., 83, 1931 : 392 (98-mm specimen from Trinidad much 
larger than any seen by me; also anal rays more numerous). 




Figure 34. Anchoa januaria, 75 ram TL, from Pernambuco, Brazil, MCZ 18012. Drawn by Alice C. Mullen. 



Anchoa januaria (Steindachner) 1879 
Figure 34 

Study Material. A total of 28 specimens, 30-75 mm TL, from Amuay' and 
Salinas bays, Gulf of Venezuela; Pernambuco (Recife), Natal, and Rio de Janeiro, 
Brazil; in USNM and MCZ collections. 

Distinctive Characters. From both subspecies of mitchilli, januaria differs in a some- 
what greater number of gill rakers, fewer anal rays, the more posterior position of the 
anal fin, and the more strongly projecting snout. Compared to both mitchilU and farva., 
it has a shorter and blunter maxillary; and from parva it differs further in having more 
numerous gill rakers on the upper limb and in the slightly more posterior position of 
the anal fin. It differs from pectoralis in having fewer pectoral and anal rays and in the 
more numerous gill rakers. 

7. Schultz {ig: 44) has claimed that the somewhat mutilated specimens from Amuay Bay, Gulf of Venezuela, identified 
by me as this species, are A. blackburni. 



I 84 Memoir Sears Foundation for Marine Research 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 15 or more specimens from 33 mm SL upward. 



Body: depth 18-20. 
Head: length 22-26.2. 
Snout: length 4.15—5.7. 
Eye: diameter 6.9—8.0. 
Postorbital: distance 10. 5-14. 2. 
Maxillary: length 1 6.4-1 9.3. 
Mandible: length 1 5.3-1 7. 



Anal fin: length of base 22-26. 
Pectoral fin: length 1 5.6-1 7.5. 

Scales: ca. 36—40. 

Gill rakers: 20-23 + 23-26. 

Fin rays: dorsal 14 or 15; anal 21-24; 

pectoral 12 or 13. 
Vertebrae: 41 or 42 (3 specimens). 



Body strongly compressed, its depth 4.3-5.0 in SL. Head short, 3.8-4.5 in SL, 
its depth equal to its length without snout. Snout 5.0-6.0 in head, projecting fully 
half of its length beyond mandible. Eye 2.9-3.6 in head. Postorbital 2.0-2.3 •" 
head. Maxillary rather bluntly pointed, its upper margin somewhat rounded, reaching 
to, or slightly beyond, joint of mandible, 1.3-1.5 in head. Mandible i. 4-1. 55. Cheek 
about as long as eye, its posterior angle about 50°. 

Dorsal fin with its longest rays failing to reach tip of last ray If deflexed, its origin 
equidistant between base of caudal and about middle of eye. Anal low, its origin 
under or slightly posterior to middle of dorsal base. Its base scarcely shorter than head, 
3.8-4.5 in SL. Pelvic very small, reaching much less than halfway to anal, inserted 
about equidistant between origin of anal and base of pectoral. Pectoral small, failing 
to reach ventral by more than half of diameter of eye, 1.4-2.0 In head. Axillary 
SCALE of pectoral about 75% of length of fin, 2.4 in head. 

Color. In alcohol, pale silvery. Silvery lateral band distinct, about 0.5-0.66 of width 
of eye. Back with dark punctulations, arranged in two rows in some specimens, probably 
faded in others. 

Size. So far as known, this Anchovy reaches a length of only about 75 mm 
(3 in.). 

Range. It apparently ranges from the Gulf of Venezuela to Rio de Janeiro, Brazil. 



Synonyms and References: 

Engraulls januarius Steindachner, Ichthyol. Beitr., 8, 1879: 58 (orig. descr.; type local. Rio de Janeiro, Brazil; 

cf. E. mitchilli). 
Anchovia januaria Starks, Stanf. Univ. PubL, Univ. Ser., 1913: 9 (Natal, Brazil). 
Anchoviella januaria Jordan and Seale, Bull. Mus. comp. Zool. Harv., 6^, 1926: 206 (descr., Rio de Janeiro, 

Brazil, and "Brazil"); Fowler, Arqu. Zool. estad. Sao Paulo, J (6), 1941: 134 (refs. to Brazilian 

recs.). 
Anchoviella mitchilli Jordan and Seale, Bull. Mus. comp. Zool. Harv., 6"], 1926: 405 (Pernambuco, Brazil; 

MCZ 1 801 2 are A. januaria). 
Anchoa januaria Hildebrand, Bull. Bingham oceanogr. Coll., 8 {z), 1943: 8i, fig. 34 (synon., descr., range; 

cf. mitchilli, farva, and pectoralis). 



Fishes of the Western North Atlantic 



Anchoa choerostoma (Goode) i 8 74 
Figure 35 

Study Material. Many specimens in nine collections from Bermuda. 
Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 22 specimens, 55-75 mm TL. 



Body: depth 18.2-20.8. 
Head: length 27.8-30.2. 
Snout: length 5.1-6.4. 
Eye: diameter 6.5—8.4. 
Postorbital: distance 1 3.3-1 5-9- 
Maxillary: length 21.2-24.3. 
Mandible: length 19.2-22. 



Anal fin: length of base 20-22.6. 
Pectoral fin: length 1 5.6-1 8.5. 

Scales: ca. 38—40. 
Gill rakers: 1 7-20 + 23-26. 
Fin rays: dorsal 13-15; anal 22-24 
(rarely 21 or 25); pectoral 12-14. 
Vertebrae: 41 or 42 (2 specimens). 




Figure 35. Anchoa choerostoma, 68 mm TL, from Bermuda, USNM 21976. Drawn by Louella E. Cable. 



Body strongly compressed, moderately deep, its depth 4.8-5.8 in SL. Head 
3.3-3.6 in SL, its depth equal to postorbital and about half of eye. Snout short, 
projecting rather more than half of its length beyond mandible, 4.7-5.8 in head. Eye 
3.3-4.2. Postorbital 1.85-2. i in head. Maxillary reaching nearly to gill opening, 
not sharply pointed, its upper margin rounded, 1.2-1.3 in head. Mandible 1.3-1.5 
in head. Cheek length equal to eye and fully 75 "/o of snout, its posterior angle 
about 30°. 

Dorsal fin rather high anteriorly, its longest rays reaching to or beyond tip of 
last ray if deflexed, its origin about equidistant between base of caudal and anterior 
margin of eye. Anal with origin somewhere under posterior half of dorsal, but well in 
advance of the base of last ray, its base 4.4-5.0 in SL. Pelvic failing to reach anal by 
nearly an eye's diameter, inserted a little nearer to origin of anal than to pectoral base. 
Pectoral moderately long, reaching nearly or quite to pelvic, 1.5-1.9 in head. 



I 86 Memoir Sears Foundation for Marine Research 

Axillary scale of pectoral reaching somewhat beyond midlength of fin, 3.0—4.0 
in head. 

Color. In alcohol, generally pale. Side with a silvery band (often missing in speci- 
mens preserved a long time in formalin), about 75 "/o of width of eye. Upper surface 
of head and back with dusky dots, usually arranged in two longitudinal series 
posterior to dorsal fin. 

Size. The largest seen is 75 mm (3 in.) TL, which may be near the maximum 
size attained. 

Reproduction. Specimens taken on June 24, 1927 at Nonsuch, Bermuda contain 
well developed eggs, indicating that at least some reproduction takes place there during 
the early part of summer. 

Range and Habitat. The range seems to be limited to Bermuda, where no other 
species of anchovy has been found. West Indian and Panamanian (Atlantic) records 
are mostly referable to A. lyolepis. The collections studied, according to the information 
furnished, were either seined near shore or taken at night under an electric light. This 
species has been reported as fairly common about Bermuda (numerous in Hamilton 
Harbor). Although it has been reported as not at all common during July, it was abun- 
dant in August when seined for bait in Bailey's Bay and Flatts Inlet. 

Synonyms and References; 

Engraulis choerostoma Goode, Amer. J. Sci., 8, 1874: 125 (orig. descr.; type local. Bermuda; types lost); Ann. 

Mag. nat. Hist., (4) 14, 1874: 380 (as in Goode, 1874). 
Stolephorus choerostoma Jordan and Everraann, BuU. U. S. nat. Mus., 47 (i), 1896: 444 (descr. after Goode; 

ref.); Barbour, Bull. Mus. comp. Zool. Harv., 46, 1905: 113 (abund. in Bermuda; distrib., includ. Puerto 

Rico, where it apparently does not occur; refs.); Bean, T. H., Field Mus. Publ., ZooL, 7, igo6; 34 

(several locals in Bermuda). 
Anchoviella choerostoma Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 48 (ref., 

Hamilton Harbor, Bermuda); Fowler, Proc. biol. Soc. Wash., ^J, 1930: 146 (72 from Bermuda). 
Anchoa choerostoma Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943 : 71, fig. 28 (refs., descr., distrib.). 

Negative References: 

Anchoviella choerostoma Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6y, 1926: 404 (not A. choerostoma 
[Goode]; descr. evidently based on A. lyolepis [Evermann and Marsh]; those from Puerto Rico and e. 
coast of Panama are lyolepis); Beebe and Tee- Van, Field Bk. Shore Fish. Bermuda, 1933: 38, fig. (not 
A. choerostoma [Goode]; fig. and descr. at least in part from Beebe and Tee- Van [1928: 47]; distrib. 
includ. West Indies, where this species apparently does not occur); Beebe and Tee- Van, Zoologica, 
N. Y., JJ (7), 1933: 136 {A. choerostoma var. atlantica Borodin [1928: 7] from e. coast of Panama put 
in synon., but A. choerostoma [Goode] apparently does not occur there. Borodin's fish not ident. from 
his descr.). 



Anchoa tricolor (Agassiz) 1829 
Piquitinga 
Figure 36 

Study Material. Many specimens, from: Gulf of Venezuela (i); Natal (8), Rio de 
Janeiro (52), Sao Francisco (5), and Sambara River (20), Brazil; Montevideo, Uruguay 



Fishes of the Western North Atlantic 



187 



(i); and Puerto de (or Rio) Quequen, Mar del Plata, Argentina (2), in USNM, MCZ, 
CNHM, MHNBA, UMMZ, SU collections. 

Distinctive Characters. This species generally has been identified in museum col- 
lections with brownii = hepsetus, which it resembles in size and general appearance, but 
it differs sharply from it in the shorter and differently shaped maxillary and in the more 
numerous gill rakers (Tables 11 and iii). 




Figure 36. Anchoa tricolor, 1 10 mm TL.Jgi^mm SL, from Rio de Janeiro, Brazil, USNM 87721, with pectoral 
fin enlarged. Drawn by Louella E. Cable. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 41 specimens, 63-1 18 mm TL. 



Body: depth 1 8.2-21. 
Head: length 23.8-26.7. 
Snout: length 4.15-5.9. 
Eye: diameter 6.7—8.1. 
Postorbital: distance 1 2.5-1 3.5. 
Maxillary: length 18.2-2 1.5. 
Mandible: length 1 5.6-1 8.3. 



Anal fin: length of base 17.2-20. 
Pectoral fin: length 14- 14. 8. 

Scales: ca. 42—45. 
Gill rakers : 18-22 + 24-28. 
Fin rays: dorsal 14—16; anal i 
pectoral 13—15. 

Vertebrae: 40-42 (5 specimens). 



Body quite slender, moderately compressed, its depth 4.75-5.5 in SL. Head 
3.75-4.2 in SL, its depth equal to its postorbital length and fully half of eye. Snout 
projecting about half of its length beyond mandible, 4.75-5.8 in head. Eye 3.25-3.75. 
Postorbital 1.8 5-2. 15 in head. Maxillary not sharply pointed, its upper margin 
rounded, reaching to, or a little beyond, joint of mandible, 1.3-1.4 in head. Mandible 
1. 3-1. 5. Cheek rather broad, scarcely longer than eye, its posterior angle about ^f. 



I 8 8 Memoir Sears Foundation for Marine Research 

Dorsal fin with longest rays failing to reach tip of last ray if deflexed, its origin 
about equidistant between base of caudal and middle of eye. Anal with origin some- 
where under posterior fourth of dorsal, its base 5.0-5.8 in SL. Pelvic reaching scarcely 
halfway to anal, inserted about equidistant between origin of anal and pectoral base. 
Pectoral scarcely falcate, failing to reach pelvic by diameter of pupil to 75''/o of dia- 
meter of eye, 1.75-2.0 in head. Axillary scale of pectoral often shorter than fin by 
diameter of pupil, 1.9—2.75 in head. 

Color. In alcohol, generally rather pale. Side of head silvery. Lateral band bright 
silvery, about 75% of width of eye. 

Size. The largest measured is 1 1 8 mm (4.7 in.) TL, which may be near the 
maximum length attained. 

Commercial Importance. Most of the specimens examined were obtained in markets, 
indicating that this Anchovy has at least some direct economic value to man. 

Range. The range, so far as known, extends from the Gulf of Venezuela to Mar 
del Plata, Argentina, but it apparently does not include streams. This Anchovy's 
frequent appearance in the Rio de Janeiro market (Brazil) seems to indicate that 
it is common locally. 

Synonyms and References: 

Engraulis tricolor Agassiz, in Spix and Agassiz, Pise. Brasil., 1829: 51, tab. 23, fig. I (orig. descr.; type local. 

Bahia, Brazil; plate labeled Engraulis pijuitinga); Goeldi, Boll. Museu Paraense, 2(3), 1898: 466 

(names; from tj'pe local.). 
Stolephorus brownii Jordan and Evermann, Bull. U. S. nat. Mus., 47(1), 1896: 443 (descr., range, synon., 

includ. E. tricolor Agassiz). 
Jnchvia brownii Starks, Stanf. Univ. PubL, Univ. Ser., 1913: 10 (Natal, Brazil; local abund.; descr. tricolor 

by my exam.). 
Anchoviella salvatoris Fowler and Bean, Proc. U. S. nat. Mus., 6j (19), 1923: 6 (orig. descr.; type local. Rio 

de Janeiro, Brazil; type USNM 83165, in bad condition, apparently tricolor'); Fowler, Arqu. Zool. 

estad. Sao Paulo, J (6), 1941 : 134 (refs.). 
Anchoviella epsetus Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6^, 1926: 396 (Rio de J. and "Sambaia" 

[probably Sambara R.], Brazil; Montevideo, Uruguay; all exam, and arc tricolor); Jordan, Evermann, 

and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 48 (includ. tricolor); Fowler, Arqu. Zool. estad. 

Sao Paulo, 5 (6), 1941: 134 (refs. to Brazilian recs., in part tricolor). 
Anchoviella bonaiensis Marini, Physis, Rev. Soc. Argent. Cienc. nat., II, 1935 : 446 (nomen nudum; type local. 

Mar del Plata, Argentina; type in MHNBA; exam, by me and found to be tricolor). 
Anchoa tricolor Hildebrand, Bull. Bingham oceanogr. Coll., 5(2), 1943: 74, fig. 30 (synon., descr., relation., 

distrib.); Schultz, Proc. U. S. nat. Mus., g<), 1949: 43 (synon.). 



Anchoa cubana (Poey) 1868 

Cuban Anchovy, Bocon, Manjiia (Cuba) 

Figure 37 

Study Material. A total of 39 specimens, 45-70 mm TL, including two from 
Poey's type material, MCZ 17958, and two paratypes of Stolephorus astilbe Jordan 
and Rutter, MCZ 31392; many more for the gill raker counts; from Melbourne Beach 



Fishes of the Western North Atlantic 



189 



and Marco, Florida; Mobile Bay, Alabama; oft" Grand Isle, Louisiana; Progreso, 
Yucatan, Mexico; Puerto Barrios, Guatemala; Cuba; Jamaica; and St. Thomas. 

Distinctive Characters. This species dift'ers from A. hepsetus and from nearly all 
other local forms in a very slender body and more numerous gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 39 specimens, 45—70 mm TL. 




Figure 37. Anchoa cubana, 63 mm TL, 53 mm SL, from St. Thomas, West Indies, USNM 83802. Drawn 
by Louella E. Cable. 



Body, depth 15-18. 
Head: length 22.7-27.8. 
Snout: length 3.85-4.7. 
Eye: diameter 6.1—7.75. 
Postorhital: distance 10.5-12.5. 
Maxillary: length 18.7—22.2. 
Mandible: length 1 5.6-1 8.8. 
Anal fin: length of base 18.5-22.2. 



Pectoral fin: length 1 2.5-1 3.7. 

Scales: ca. 40—43. 

Gill rakers: 17—23 + 23—33, usually 19— 

22 + 25—28 (Tables 11, in). 
Fin rays: dorsal 14—16; anal 20—24; 

pectoral 13 or 14. 
Vertebrae: 42 or 43 (14 specimens). 



Body slender, rather strongly compressed, with a rather sharp edge on chest and 
abdomen, its depth 5.5-6.7 in SL. Head small, 3.6-4.3 in SL, its depth equal to its 
postorbital length and about a third of eye. Snout short, extending about 66 "/o of its 
length beyond tip of mandible, 4.8—6.5 in head. Eye small, 3.5—4.0. Postorbital 
short, 1.9—2.3 in head. Maxillary bluntly pointed, its upper margin nearly straight, 
usually extending nearly to margin of opercle, i.i— 1.25 in head. Mandible 1.4— 1.75. 
Cheek, somewhat longer than eye, its posterior angle rather broad, about 45°. Gill 
RAKERS slender, close-set, often difficult to count (probably accounting for wide 
range given). 

Dorsal fin with the longest rays failing to reach tip of last ray if deflexed, its origin 
about equidistant between base of caudal and some point over posterior half of eye. 
Anal with origin under or slightly in advance of middle of dorsal base, its base 4.5— 
5.4 in SL. Pelvic reaching somewhat less than halfway to origin of anal, inserted a 



I go Memoir Sears Foundation for Marine Research 

little nearer to origin of anal than to pectoral base. Pectoral failing to reach base 
of pelvic by nearly an eye's diameter, i. 75-1. 9 in head. Axillary scale of pectoral 
about 75 "/o of length of fin, 2.4-3.0 in head. 

Color. In alcohol, rather pale. Side of head silvery. Lateral band bright silvery, 
about as wide as pupil (often missing in specimens preserved in formalin). Back usually 
with dusky punctulations not arranged in definite longitudinal series. 

Size. The largest fish seen was only 70 mm (2.8 in.) long, and most of the numer- 
ous specimens examined were under 65 mm TL. 

Range and Habitat. A. cubana ranges from both coasts of Florida, along Alabama 
and Louisiana, and through the West Indies, to Yucatan, Mexico, and Puerto Barrios, 
Guatemala. It is also recorded for Puerto Rico {18: 344) and Grenada (j: 269). 
It seems to be common around Jamaica, probably also around Cuba, but apparently 
it is rather rare on our southern coast. The specimens from the West Indies, in 
part at least, were taken in shallow water, whereas those from the United States were 
taken with trawls hauled at depths as great as 30 fms. 

Synonyms and References: 

Engraulis cubanus Poey, Repert. Fisico-Nat. Cuba, 2, 1868: 420 (orig. descr.; type local. Cuba; MCZ 17958); 

Poey in Gundlach, An. Soc. esp. Hist. Nat., 10 (4), 1881 : 344 (Puerto Rico). 
Stokpkorus cubanus Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 442 (descr. from orig., range); 

Evermann and Marsh, Bull. U. S. Fish Comm. (1900), 20 (i), 1902: 88 (after Jordan and Evermann, 

1896). 
Stokphoru! astilbe Jordan and Rutter, Proc. Acad. nat. Sci. Philad., 1897: 95 (orig. descr.; type local. Kingston, 

Jamaica; type SU 4854; paratypes MCZ 31392); Jordan and Evermann, Bull. U. S. nat. Mus., 47 (3), 

1898: 2815 (descr. from orig., synon.). 
Anchoviella cubana Jordan and Scale, Bull. Mus. comp. Zool. Harv., 67, 1926: 399 (synon., descr., Cuba, 

includ. t>'pe material, MCZ 17958); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 

2, 1930: 48 (synon., range); Fowler, Proc. Acad. nat. Sci. Philad., 82, 1930: 269 (Grenada, B.W.I.); 

Howell-Rivero, Bull. Mus. comp. Zool. Harv., 82, 1938: 172 (4 specimens in type material, the largest 

said to be type, MCZ 17958). 
Anchoviella astilbe Jordan and Seale, Bull. Mus. comp. Zool. Harv., Gj, 1926: 402 (synon., descr., paratypes 

MCZ 31392); Jordan, Evermann, and Clark, Rept. U. S. Comm. Fish. (1928), 2, 1930: 49 ("West 

Indies"); Fowler, Proc. Acad. nat. Sci. Philad., 82, 1930: 269 (Grenada, B.W.I.). 
Anchovia cubana Nichols, N. Y. Acad. Sci., 10 (2), 1929: 204 (synon., diagn., Cuba and Puerto Rico). 
Anchoa cubana Hildebrand, Bull. Bingham oceanogr. Coll., 8 {i), 1943: 76, fig. 31 (synon., descr., extension 

of range; cf. paratj'pes oi cubana and astilbe). 



Anchoa ginsburgi H.ildehra.nd 1943 
Figure 38 

Study Material. Type material, 16 specimens, from juveniles 28 mm long to adults 
or near adults 63 mm TL, from Gulf of Venezuela. 

Distinctive Characters. This species differs from hepsetus chiefly in the shorter and 
blunter maxillary and in the more numerous vertebrae when compared with hepsetus 
from southern localities, especially from Yucatan and Panama (j: ^6, tab. 7). Among 
specimens of equal size, this species seems to be more slender. 



Fis/ies of the Western North Atlantic 



191 



Description. Proportional dimensions in per cent of standard length, and counts, 
based on 13 specimens, 43—63 mm TL. 



Body: depth 1 7.4-1 8.8. 
Head: length 25-28.5. 
Snout: length 4.9—5.9. 
Eye: diameter 6.6—9.1. 
Postorbital: distance 1 4.5-1 5.8. 
Maxillary : length i 7 . 2 - 1 8 . 2 . 
Mandible: length i 5.5-1 7.8. 



Anal fin: length of base 19.2-22. 
Pectoral fin: length 1 3.7-1 5.3. 

Scales: ca. 42—45. 

Gill rakers: 16—18 + 20-22. 

Fin rays: dorsal 14 or 15; anal 18—20; 

pectoral 13-15. 
Vertebrae: 44 or 45 (2 specimens). 




Figure 38. Anchoa ginsburgi, paratype, 60 mm TL, 52 mm SL, from Gulf of Venezuela, USNM 1 19789. 
Drawn by Louella E. Cable. 



Body slender, moderately compressed, its depth 5.3-5.75 in SL. Head 3.5-4.0 
in SL, its depth equal to its postorbital length and about half of eye. Snout short, 
projecting a little more than half of its length beyond mandible, 4.5—5.25 in head. 
Eye 3.25—4.0. Postorbital moderately long, 1.3— 1.9 in head. Maxillary short, its 
upper margin rounded, reaching joint of mandible, 1.3-1.5 in head. Mandible 1.5— 
1.65. Cheek broad, a little longer than eye, its posterior angle about 45°. 

Dorsal fin low, the longest rays failing to reach tip of last ray if deflexed, its origin 
usually somewhat nearer to base of caudal than to anterior margin of eye. Anal short, 
with origin commonly slightly posterior to middle of dorsal, its base 4.5—5.2 in SL. 
Pelvic small, inserted equidistant between origin of anal and base of pectoral, or a little 
nearer to the latter. Pectoral scarcely falcate, failing to reach pelvic by about half of 
diameter of eye, 1.7-2.0 in head. Axillary scale of pectoral moderately pointed, failing 
to reach tip of fin by about 75 "/o of diameter of eye, 2.8-3.1 in head. 

Color. In alcohol pale (originally preserved in formalin). Side of head silvery. 
Lateral band bright silvery, about 75 "/o of width of eye. 

Range. Known only from the holotype and 15 paratypes, all from the Gulf of 
Venezuela, partly from Estauques Bay and partly from Salinas Bay. 



192 Memoir Sears Foundation for Marine Research 

References : 

Anchoa ginsburgi Hildebrand, Bull. Bingham oceanogr. Coll., <S (2), 1943: 55, fig. 20 (orig. descr.; type local. 
Estauques Bay, Gulf of Venezuela; type USNM 1 19788); Schultz, Proc. U. S. nat. Mus., gg, 1949: 
43 (ref., types, locals.). 

Anchoa lamprotaenia Hildebrand 1943 

Longnose Anchovy, Manjiia (Cuba) 

Figure 39 

Study Material. A total of 80 specimens, 46-92 mm TL, collected at Palm Beach, 
Miami, Key West, and Tortugas, Florida; Progreso, Yucatan, Mexico; Cuba; Puerto 
Rico; British Honduras; and the Atlantic coast of Panama. 

Distinctive Characters. This species until recently has been consistently identified 
with A. hepsetus in museum collections. Since these species are not easily separated by 
any one character, a combination of characters is sometimes necessary. Generally 
A. lamprotaenia is distinguished by its longer and more falcate pectoral fin, higher 
dorsal fin, shorter and broader axillary scale of the pectoral, and by the shorter and more 
bluntly pointed maxillary with its upper rounded margin. The average number of anal 
rays is greater and the average number of pectoral rays, gill rakers, and vertebrae is 
fewer (see Description and Tables i— iii). 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 46-92 mm TL. 

Body: depth 19-23. Pectoral fin: length 14. 3-1 8. 
Head: length 25-30. 

Snout: length 8.1-9.5. Scales: ca. 38—41. 

Eye: diameter 7.5—8.3. Gill rakers: 13— 18 + 16— 20 (usually 14 
Postorbital: distance 14. 3-16. or 15+18 or 19). 

Maxillary: length 21.5—24.5. Fin rays: dorsal 13—15; anal 19—26 
Mandible: length 18.5—22. (usually 21—24); pectoral 11-15 

Anal fin: length of base 20.5—23.5. (most frequently 12 or 13). 

Vertebrae: 39—42 (98 specimens). 

Body strongly compressed, its greatest thickness about equal to depth of 
caudal peduncle, its depth 4.3-5.25 in SL. Head 3.3—4.0, its depth about equal to 
its postorbital length and half of eye. Snout pointed, projecting about 66*/o of its 
length beyond tip of mandible, 4.5—5.75 in head. Eye 3.5—4.0 in head. Postorbital 
1.75—2.0 in head. Maxillary not sharply pointed, usually distinctly rounded above, 
reaching to, or more usually somewhat beyond, joint of mandible, 1.2-1.4 in head. 
Mandible 1.35— 1.6 in head. Cheek as long as eye and fully half of snout, its 
posterior angle about 35°. Gill rakers about 75°/o of length of eye. 

Dorsal fin rather high anteriorly, its longest rays usually reaching beyond tip of 
last ray if deflexed, its origin equidistant between some point over snout or anterior 



Fishes of the Western North Atlantic 



193 



half of eye and base of caudal. Anal fin with origin under or near beginning of posterior 
third of dorsal base, its base 4.25-4.9 in SL. Pelvic usually inserted slightly nearer to 
origin of anal than to base of pectoral. Pectoral long, falcate, usually failing to reach 
pelvic by only diameter of pupil, its longest ray in a few specimens with a slight filament, 
the rays decreasing rapidly in length after the third one, the shortest ray scarcely reach- 
ing past midlength of first (uppermost) one, 1.5-1.9 in head. Axillary scale of pec- 
toral broad at base, scarcely reaching beyond midlength of fin, 3.0-3.6 in head. 




Figure 39. Anchoa lamprotaenia, paratype, 6omniTL, 48 mm SL, from Key West, Florida, USNM 35000, 
with pectoral fin enlarged. Drawn by Louella E. Cable. 



Color. Preserved specimens pale. Side of head bright silvery. Silvery lateral band 
prominent, not quite as wide as eye. Punctulations on back not prominent; a few dark 
dots at base of anal; dorsal and caudal with a few dusky points; other fins unmarked. 

Size. A. lamprotaenia apparently attains a smaller size than A. hepsetus, the largest 
seen being only 92 mm (3.7 in.) TL. 

Reproduction. Specimens taken at Key West, Florida, June i i, 1936, and at Bush 
Key (Tortugas), Florida, July 23, 1937, contained well-developed roe, showing that 
at least some spawning takes place in southern Florida during midsummer. 

Range and Habitat. From southern Florida through the West Indies to Panama, 
and probably to northern Brazil as shown by an imperfect specimen. The collections 
studied were apparently taken in shallow water near shore. The species is common in 
the Florida Keys, where A. hepsetus seems to be very rare. 

Synonyms and References: 

StoUphorus brownii Jordan, Proc. U. S. nat. Mus., 7, 1885: 106 (abund. at Key West, Florida; specimens 
USNM 35000 are A. lamprotaenia); Evermann and Marsh, Bull. U. S. Fish Comm. (1900), 20 (i), 

'3 



1 94 Memoir Sears Foundation for Marine Research 

1902: 88 (descr., range, distr. in Puerto Rico; synon.; specimens from Fajardo [USNM 73755], at 
least, are A. lamprotaenia). 

Anchovia brownii Meek and Hildebrand, Field Mus. Publ., Zool., i'5(i), 1923: 204, pi. 14, fig. i (synon., 
descr., range, specimens from Colon [CNHM 8217] and Porto Bello [CNHM 8216], Panama, in part 
are A. lamprotaenia). 

Anchoviella epsetus Jordan and Seale, Bull. Mus. comp. Zool. Harv., 6y, 1926: 396 (synon., descr., range; 
specimens from Tortugas [MCZ 17944] and Key West [MCZ 18035], Florida, are A. lamprotaenia):, 
Pierce, Copeia, 1936: 123 (sample of specimens used in feed, experiments are A. lamprotaenia); Long- 
ley and Hildebrand, Publ. Carneg. Instn. Wash., 535, 1941 : 12 (notes on structure; distr. at Tortugas, 
Florida; specimens preserved are A. lamprotaenia). 

Anchoa lamprotaenia Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943 : 62, fig. 23 (orig. descr.; type local. 
Key West, Florida; type USNM 11 7661; range cf. A. hepsetus). 

Doubtful References: 

Anchovia brownii Jordan and Thompson, Bull. U. S. Bur. Fish. (1904), 24, 1905 : 233 (Garden Key [Tortugas], 
Florida; probably part, if not all, are lamprotaenia); Fowler, Proc. Acad. nat. Sci. Philad., ^8, 1906: 
84; and 6j, 191 1 : 216 (from several Florida Keys; probably part, if not all, are lamprotaenia). 



Anchoa hepsetus (Linnaeus) 1758 

Striped Anchovy (U.S.), Manjua (Cuba) 

Figures 40, 41 

Study Material. At least 83 specimens, 55— 140 mm TL, and many more, taken 
at various places from New Jersey southward to Uruguay. 

Distinctive Characters. A. hepsetus is related closely enough to A. lamprotaenia so 
that they are not always separable by any one character, a combination of characters 
sometimes being necessary. In general, hepsetus has a shorter and less falcate pectoral 
fin, a lower dorsal fin, a longer and narrower axillary scale on the pectoral, and a rather 
longer and more sharply pointed maxillary. Also, its average number of anal rays is 
lower, and its average number of pectoral rays, gill rakers, and vertebrae is higher. 
It is related also to A. ginsburgi., from which it differs especially in the longer and more 
sharply pointed maxillary, and generally in having fewer vertebrae. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 83 specimens, 55—140 mm TL. 

Body: depth 18.5-22. Scales: ca. 37-43. 

Head: length 25-30. Gill rakers: 15-20+ 18-24 (usually 16— 

Snout: length 4.4-6.3. 19+ 19—22). 

Eye: diameter 7.15—8.7. Fin rays: dorsal 13-16; anal 18—23; 

Postorbital: distance 13-5— 15-5. pectoral 13—15 (most frequently 

Maxillary: length 19-24.5. 14 or 15). 

Mandible: \ength. 16.5—20. Vertebrae: 40—44 (399 specimens), de- 

yf«a/^«: length of base 19-22.5. creasing somewhat in number 

Pectoral fin: length 14. 5-1 7.5. southward at least to Panama. 



Fishes of the Western North Atlantic 195 

Body slender, moderately compressed, its greatest thickness generally exceeding 
the depth of caudal peduncle, its depth 4.5—5.4 in SL. Head 3.3-4.0 in SL, its depth 
equal to postorbital and about a fourth of eye. Snout projecting about 66 "/o of its 
length beyond tip of mandible, 4.5—6.0 in head. Eye 3.25-4.0 in head. Postorbital 
1.8— 2.1 in head. Maxillary quite pointed, its upper free margin generally straight, 
reaching nearly to margin of opercle, 1.2-1.3 in head. Mandible 1.4-1.6 in head. 
Cheek a little longer than eye. Its posterior angle rather broad, about 45°. Gill rakers 
at angle of first arch about 75''/o of length of eye. 

Dorsal fin low, its margin somewhat concave, its longest rays failing to reach tip 
of last ray if deflexed, its origin usually about equidistant between base of caudal and 
middle of eye. Anal fin with origin under middle to posterior fourth of dorsal base, its 
base 4.4—5.2 in SL. Pelvic reaching about halfway to origin of anal, inserted nearly 
equidistant between origin of anal and pectoral base. Pectoral not strongly falcate, its 
four uppermost rays usually of about equal length, its first or uppermost one some- 
times a little shorter than the second one, its last or shortest ray usually reaching well 
past midlength of the first one, the fin often failing to reach base of pelvic by diameter 
of eye, though sometimes nearly reaching it, its length 1.7—2.0 in head. Axillary 
SCALE of pectoral slender, usually failing to reach tip of fin by less than diameter of 
pupil, 2.0—2.5 '" head. 

Color. Preserved specimens, pale above. Lower parts pale silvery. Side of head 
silvery. Middle of side with a bright silvery band, variable in width, often about 75''/o 
of width of eye above base of anal, sometimes scarcely wider than pupil. 

Size. The largest specimen seen, from Beaufort, North Carolina, is 153 mm 
(6.1 in.) TL, which probably is near the maximum size attained. A specimen 150 mm 
long, weighing one ounce, has been reported for Chesapeake Bay. The usual size 
ranges between about 100-125 mm TL. 

Development and Growth. The egg is elliptical, with a major axis of about 1.4- 
1.6 mm and a minor one of about 0.7-0.85 mm. It is transparent and has no oil 
globule. The yolk, as in other anchovies studied, is separated into masses having 
the appearance of large cells under magnification. The egg floats at or near the surface 
and hatches within about 48 hours at a water temperature of about 66°-70°F. 

The newly hatched fish, very slender and thread-like, is about 3.6-4.0 mm long. 
It is transparent, has a slight greenish shade on the head, and has no definite chromato- 
phores; the head is slightly concave. The myomeres are very distinct. It is quite 
active. The yolksac generally is absorbed within 24 hours after hatching at 65°-70°F. 
The fish is then about 4.0-4.5 mm long, at which stage the mouth is open and ap- 
parently functional. The fish is still transparent. 

The first chromatophores, situated in the midventral line of the chest, appear 
when the fish is about 5.0 mm long, and at this stage the mouth is large, terminal, 
and oblique. The dorsal and anal fins become more or less definitely differentiated 
at a length of 5.0—6.0 mm, but the rays cannot be counted accurately until the fish 
is about 10 mm, at which length it has become considerably more robust, the 



196 Memoir Sears Foundation for Marine Research 

notochord Is definitely bent upward in tlie base of the caudal fin, and the caudal has 
Indefinite rays. 

At a length of 1 5 mm the caudal fin is well developed and forked, the notochord 
Is no longer visible, and the mouth remains terminal and oblique. Some additional 
chromatophores have appeared along the ventral surface, and dark pigment has ap- 
peared along the upper margin of the eye. 




Figure 40. Anchoa hepsetus hepsetus, 114 mm TL, from Atlantic City, New Jersey, USNM 127525, with 
pectoral fin enlarged. Drawn by Louella E. Cable. 

Little external development takes place while the fish grows from a length of 
15 to 25 mm. However, when the fish reaches 35 mm, the mouth has become almost 
horizontal, as In adults, and the conical snout projects prominently. Pigmentation has 
Increased and the silvery lateral band of the adult, though narrow and indistinct at this 
point, has appeared. Although the body has now become fairly robust, it remains more 
slender than in adults, its depth being contained in the length about 6.0 times. 

When It reaches a length of about 40-45 mm It may be definitely considered 
a young adult. Although It remains somewhat more slender than larger fish. It is fully 
pigmented, has the characteristic silvery lateral band well developed, is fully scaled, 
and is easily identified with the adult (4: 389-393; also unpublished field notes of 
the writer). 

Spawning. In the vicinity of Beaufort, North Carolina, spawning begins early In 
the spring. April 16 is the earliest date on which eggs have been taken. Toward the 
middle of June, eggs diminished in number In the tows, though a few were taken 
throughout July. The young, 22 mm and less, were collected in Beaufort Inlet as early 
as April 11. (4: 388; also unpublished data of the writer). 

Food. When the fish is about 20 mm the food consists largely of copepods. As 



Fishes of the Western North Atlantic 



197 



the fish grows, this diet is supplemented by gastropods, foraminifera, an occasional 
ostracod, and annelid worms. The diet of the adult is essentially the same (Linton, 
11: 440; 12: 353; 4: 395). 

Parasites. The principal parasites are nematodes, cestodes {Scolex polymorphus and 
Rhynchobothrium sp.), and trematodes {Distomum appendiculatum and Distomum sp.) 
{11: 440; 12: T,S2)- 




Figure 41. Anchoa hepsetus colonensis, 70 mm TL, 56 mm SL, from Colon, Panama, USNM 11 7664, with 
pectoral fin enlarged. Drawn by Louella E. Cable. 



Variation. So great is the variation in this species that certain specimens from 
Colon, Panama, and northward to the Gulf of Mexico were recognized (5: 60, fig. 22) 
as a subspecies, A. hepsetus colonensis. The greatest divergence is apparent in speci- 
mens from Colon, Panama, with complete intergradation in specimens on the Gulf 
coast of the United States. As typical hepsetus occurs through the range of h. colonensis, 
the latter is not a geographical variant but one that occurs within the range of the 
typical species. 

The variant, or subspecies colonensis, differs from typical hepsetus in having a some- 
what deeper and more strongly compressed body, its greatest thickness scarcely ex- 
ceeding the depth of the caudal peduncle. Its pectoral fin is longer and more pointed, 
generally extending nearly or quite to the base of the pelvic fin. Its silvery lateral 
band is narrower, often scarcely exceeding the width of the pupil. 

Commercial Importance. Each year large quantities are marketed as food for man. 
In 1939 about 21,000 pounds and in 1940 about 29,000 pounds were handled in the 
New York City markets. It is economically important, however, chiefly as a forage fish. 



1 9 8 Memoir Sears Foundation for Marine Research 

It is preyed upon by several commercially important fishes, especially the seatrouts 
{Cynoscion). Water birds also feed on it extensively. 

Range. Plentiful from Chesapeake Bay to the West Indies; north irregularly to 
southern New England and as a stray to Maine and the outer coast of Nova Scotia; 
south at least as far as Montevideo, Uruguay. It has also been recorded for Cape 
Verde, Africa, but that record is in need of verification. 

In some years it is numerous at Beaufort, North Carolina, appearing in large 
schools, but in other years less abundant. At Beaufort and in Chesapeake Bay it was 
taken chiefly with seines near the shore and rarely with trawls in somewhat deeper 
water; however, in the Gulf of Mexico it was taken in deeper water more commonly. 
Collections made with otter trawls, some of them in water as much as 30—40 fms. 
deep, are at hand from off Corpus Christi, Texas, off Grand Isle, Louisiana, and off 
St. Vincent Island and Apalachicola Bay, Florida. 

Details of Occurrence.^ Stray specimens of this Anchovy were reported for Halifax, 
Nova Scotia, In 1931 {21 : 3), and for the mouth of the Penobscot River, Maine, 
in 1930, when many others were reported as seen {8). The next most northerly 
locality of record is the vicinity of Woods Hole, southern Massachusetts, where 
it has been reported as abundant, at least in some years. But only one specimen 
of this particular Anchovy has been reported for Rhode Island. It is less common 
in New York waters than its relative mitchilU (p. 179), but it has been reported as 
being locally abundant there; this applies equally to New Jersey waters and to Ches- 
apeake Bay as a whole. 

Farther to the south, hepsetus appears in large schools in some years along the 
coast of North Carolina (see above), where mitchiUi has been reported only once (p. 1 79). 
It is known from Georgia, the Florida Keys region, various localities along the western 
coast of Florida from Cape Sable to Pensacola, the coast of Mississippi, Grand Isle 
and Cameron, Louisiana, Galveston and Mustang Island, Texas, on the Mexican 
coast off Tampico, Vera Cruz, and Yucatan, and the Atlantic coast of Panama. It is 
widespread throughout the West Indies also, where it has been recorded for the 
northern coast of Cuba, St. Thomas, Puerto Rico, Haiti, Santo Domingo, Jamaica, 
Cura9ao, St. Kitts, Martinique, and Barbados. On the coast of Brazil it is known at 
Mamanguape (near Pernambuco), Bahia, and Rio de Janeiro. 

Synonyms and References: 

Esox hepsetus Linnaeus, Syst. Nat., ed. 10, 1758: 314 (diagn.; type local. "America"; ref's. to 'Piquitinga' 

Marcgrave, Brazil, and Menidia Browne, Jamaica, non binomials). 
Atherina epsetus Bonnaterre, Tabl. Encyc. Meth. IchthyoL, 1788: 175 (descr., same fin ray counts as Linnaeus; 

ref. to Linnaeus, but apparently not ed. 10). 
Atherina brownii Gmelin, Syst. Nat., j, 1788: 1397 (based on Menidia Browne, pre-Linnaean). 
C/itpea vittata Mitchill, Trans. Lit. philos. Soc. N. Y., j, 1 8 1 5 : 456 (orig. descr.; type local. New York; type 

lost); De Kay, New York Fauna, Ft. 4: Fishes, 1842: 254 (descr.); Storer, Mem. Amer. Acad. Arts Sci., 

1846: 205, in separate (brief descr.. New York). 
Engraulis brownii Cuvier and Valenciennes, Hist. Nat. Poiss., 21, 1848: 41 (descr.. New York, West Indian 

8. Added by H. B. Bigelow. 



Fishes of the Western North Atlantic 199 

locals., Vera Cruz, Rio de Janeiro; also several Indian O. locals, where this species does not occur); 
Giinther, Cat. Fish. Brit. Mus., 7, 1868: 389 (descr., West Indies, Gulf of Mexico, and Bahia; also 
from "Libertad," Pacific Central America, and Ceylon where the species does not occur; synon. includ. 
E. tricolor, which is a valid species); Poey, Repert. Fisico-Nat. Cuba, 2, 1868: 419 (descr., Cuba); Poey, 
Enumerat., Pise. Cubana, 1875: 149 (relations., Cuba, Jamaica, and San Domingo); Yarrow, Proc. 
Acad. nat. Sci. Philad., 1877: 215 (Ft. Macon, North Carolina; "not common"). 

Argentina menidia Gronow, Cat. Fish., 1754: 141 (based on Menidia Browne, pre-Linnaean). 

Engraulis vittattis Jordan and Gilbert, Proc. U. S. nat. Mus., I, 1879: 385 (common, Beaufort Harbor, North 
Carolina). 

Engraulis hiulcus Goode and Bean, Proc. U. S. nat. Mus., 2, 1880: 343 (orig. descr.; type local. Clearwater 
Harbor, Florida; USNM 23632; in bad condition, about 57 mm long, not 47 mm as stated by Goode 
and Bean). 

Stolephorus perthecata Goode and Bean, Proc. U. S. nat. Mus., 5, 1883: 434 (orig. descr.; type local. Pensacola, 
Florida; type USNM 30483); Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 442 (descr.). 

Stolephorus browni Jordan and Gilbert, Bull. U. S. nat. Mus., 16, 1883: 273 (descr., range, synon.); Swain, 
Bull. U. S. Fish Comm. (1882), 2, 1883: 56 (descr., several locals, between New York and Brazil; 
synon., includ. tricolor); Jordan and Swain, Proc. U. S. nat. Mus., 7, 1885: 230 (Cedar Keys, Florida); 
Bean, T. H., Bull. U. S. Fish Comm. (1887), 7, 1888: 149 (common arjund Great Egg Harbor, New 
Jersey; eaten by weakfish); Henshall, Bull. U. S. Fish Comm. (1889), 9, 1891: 373 (Cape Sable Cr., 
Cape Romano, and Big Gasparilla, Florida); Bean, B. A., Proc. U. S. nat. Mus., 14, 1892: 93 (abund., 
Cape Charles City, Virginia); Evermann and Kendall, Bull. U. S. Fish Comm. (1892), 12, 1894: 105 
(Galveston, Texas); Moore, Bull. U.S. Fish Comm. (1892), 12; 1894: 359 (abund., size. Sea Isle 
City, New Jersey); Jordan and Evermann, Bull. U.S. nat. Mus., 47 (i), 1896: 443 (descr., range, 
synon.); Smith, Bull. U. S. Fish Comm. (1897), 77, 1898: 92 (abund., season. Woods Hole, Massa- 
chusetts); Evermann and Kendall, Rep. U. S. Comm. Fish. (1899), 1900: 56 (Florida locals.); Gilbert, 
Proc. Wash. Acad. Sci., 2, 1900: 163 (Mamanguape, Brazil); Linton, Bull. U. S. Fish Comm. (1899), 
ig, 1901: 440 (food, parasites); Bean, T. H., 7th Rep. Forest Comm. N. Y. (1901), 1902: 310 
(synon., forage fish); Evermann and Marsh, Bull. U.S. Fish Comm. (1900), 20, 1902: 88 (descr., 
range, Puerto Rico; synon.); Schreiner and Miranda-Ribeiro, Arch. Mus. nac, Rio de J., 12, 1903: 
93 (Rio de Janeiro); Bean, T. H., Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 214 (descr., synon.. New 
York); Linton, Bull. U. S. Bur. Fish. (1904), 24, 1905: 353 (food, parasites); Barbour and Cole, Bull. 
Mus. comp. Zool. Harv., ^o, 1906: 156 (Progreso, Mexico); Breder, Zoologica, N. Y., 2 (15), 1922: 
338 (Sandy Hook, New Jersey); Metzelaar, Bijdr. Dierk. Amst., 1922: 134 (Curasao). 

Anchovia brozvnii Fowler, Rep. N. J. St. Mus. (1905), 1906: 109 (descr., New Jersey); Fowler, Rep. N.J. 
St. Mus. (1906), 1907: 267, pi. 85 (ill.); Smith, N. C. geol. econ. Surv., 2, 1907: 133 (descr., range, 
abund., forage fish); Jordan and Dickerson, Proc. U. S. nat. Mus., 24^ 1908: 12 (Tampico, Mexico); 
Weymouth, Proc. U. S. nat. Mus., j5, 1910: 136 (Cameron, Louisiana); Fowler, Proc. Acad. nat. Sci. 
Philad., 6j, 191 1: 216 (New Jersey, North Carolina, Florida Keys, San Domingo, Puerto Rico, 
Rio de Janeiro); Sumner, Osburn, and Cole, Bull. U. S. Bur. Fish. (1911), 31, 1913: 742 (habitat, 
spawn, time, food, parasites, Woods Hole, Massachusetts); Fowler, Copeia, 1915: 50 (San Domingo); 
Latham, Copeia, 1919: 55 (Orient, Long Island, New York); Meek and Hildebrand, Field Mus. Publ., 
Zool., r5(i), 1923: 204, pi. 14, fig. I (synon., descr., range, Colon and Porto Bello, Panama); Breder, 
Zoologica, N. Y., .#(4), 1925: 152 (large schools, Caledonia Bay, Panama); Nichols and Breder, Zoo- 
logica, N. Y., 9 (i), 1927: 43 (descr., range, abund. New York to Woods Hole, Massachusetts; size); 
Nichols, N. Y. Acad. Sci., 10 (2), 1929: 205, fig. 38 (diagn., range, abund., Puerto Rico). 

Anchoviella epsetus Jordan and Seale, Bull. Mus. comp. Zool. Harv., Gj (11), 1926: 396 (synon., descr., range; 
Rio de Janeiro, "Sambaia" [probably Sambara], Brazil, and Montevideo, Uruguay, specimens are tricolor; 
specimens from Key West and Tortugas, Florida, are lamprotaenia); Hildebrand and Schroeder, Bull. 
U. S. Bur. Fish. (1927), ^j (i), 1928: no (descr., food, spawn, season, size, range, Chesapeake Bay); 
Beebe and Tee-Van, Zoologica, N. Y., 10 (l), 1928: 46 (descr., range, Port-au-Prince, Haiti); Breder, 
Field Bk. Mar. Fish. Atl. Cst., 1929: 71, text fig. (range, spawn., food, abund.); Jordan, Manual Vert. 
Anim. NE U. S., ed. 13, 1929: 43 (descr., range); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. 
(1928), 2, 1930: 48 (synon., range); Hildebrand and Cable, Bull. U. S. Bur. Fish., 46, 1930: 388 
(econ. import., spawn., develop, of eggs and young, growth rate, food); Kendall, Bull. Boston Soc. 
nat. Hist., 58, 1931 : 1 1 (Portland, Maine); Vladykov, Proc. N. S. Inst. Sci., ig (i), 1935 : 3- 55> ^g- 3' 



2 00 Memoir Sears Foundation for Marine Research 

(Halifax Harbor, Nova Scotia); Hubbs, Publ. Carneg. Instn. Wash., 457, 1936: 175 (discus, ident., 
Rio Champoton and Campeche, Yucatdn); Bigelow and Schroeder, Bull. U. S. Bur. Fish., 48 (20), 1936: 
328 (Portland, Maine; Halifax Harbor, Nova Scotia); Howell-Rivero, Handbk. Jamaica, 1936: i, in 
reprint (Port Antonio, Jamaica); Gunter, Ecol. Monogr., 5, 1938: 337 (spawn., abund., Louisiana); 
Gunter, Amer. Nat., 72, 1938: 78, 79; Gunter, Ecology, 22(2), 1941: 203-208 (killed by freeze, 
Texas); Fowler (in part, not of Linnaeus), Arqu. Zool. estad. Sao Paulo, J (6), 1941 : 134 (Brazilian 
recs., in part tricolor). 

Anchoa hepsetus hepsetus Hildebrand, Bull. Bingham oceanogr. Coll., 8 {l), 1943: 57. fig- 21 (synon., descr., 
range). 

Anchoa hepsetus colonensis Hildebrand, Bull. Bingham oceanogr. Coll., 5(2), 1943: 60, fig. 22 (orig. descr.; 
type local. Col6n, Panama; type USNM 1 17664; cf typical hepsetus; range). 

Anchoa hepsetus Gunter, Publ. Inst. mar. Sci. Texas, J (l), 1945: 32 (shallow water, temp., salinity, length 
frequencies by months, sex. develop, in spring, Texas). 

Doubtful References: 

Anchotiia brownii Jordan and Thompson, Bull. U. S. Bur. Fish. (1904), 24, 1905: 233 (Garden Key, Florida; 
some, if not all, were lamprotaenia); Fowler, Proc. Acad. nat. Sci. Philad., 58, 1906: 84 (Florida Keys; 
in part, if not all, lamprotaenid). 

EngrauUs brozonii Lampe, Fisch. Dtsch. Siidpolar Exped., 1901-1903, III, Die Hochsee und Kiisten Fische, 
J5, Zool. 7, 1914: 216 (Cape Verde, So. Africa; ident. needs verification). 

Engraulis hepsetus Fowler, Bull. Amer. Mus. nat. Hist., 'JO (i), 1936: 183 (descr. based on New Jersey speci- 
mens; rec. for Cape Verde after Lampe). 

Negative References: 

Stolephorus brovinii Jordan, Proc. U. S. nat. Mus., 7, 1885: 106 (abund. Key West, Florida; USNM 35000 
listed; these exam, and found to be lamprotaenia); Puyo, Bull. Soc. Hist. nat. Toulouse, 80, 1945: 
100, fig. I (descr., French Guiana; descr. and ill. appear to be unnamed new species oi AnchovieI/a)\ 
Puyo, Faune Emp. Franc, 12, Poiss. Guiane Frang., 1949: 154 (see Puyo, 1945). 

Anchovia brownii Starks, Stanf Univ. Publ., Univ. ser., 191 3: 10 (Natal, Brazil; specimens are tricolor). 



Anchoa lyolepis (Evermann and Marsh) 1902 

Dusky Anchovy, Manjiia (Cuba) 

Figure 42 

Study Material. At least 22 specimens, 45-70 mm TL; juveniles, including the 
type material, also at hand for comparison; from off St. Vincent Island, Florida; Mobile 
Bay, Alabama; Mississippi Sound, Mississippi; Grand Isle, Louisiana; Galveston and 
Aransas Pass, Texas; several places in Puerto Rico, Cuba, and St. Martin, West Indies. 

Distinctive Characters. This species is quite distinctive. It is probably nearest to 
lamprotaenia, from which it differs rather prominently in the lower head, the longer 
snout, longer and narrower cheek, and in the more posteriorly placed anal fin. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on Study Material, 45-70 mm TL. 

Body: depth 16.6-20. Eye: diameter 5.5-7.0. 

Head: length 27-32.2. Postorbital: distance i 5.4-16.7. 

Snout: length 5.9-6.6. Maxillary: length 21-27. 



Fishes of the JVestern North Atlantic 



20I 



Mandible: length 20-22.2. 

Anal fin: length of base 19-22.5. 

Pectoral fin: length 14— 15.4. 

Scales: ca. 40—44. 



Gill rakers: 15— 21 + 19— 26 (most fre- 
quently 18 or 19 + 20-22). 

Fin rays: dorsal 12—15; ^"^' I9~23; 
pectoral 12—14. 

Vertebrae: 41—43 (9 specimens). 



Body very slender, rather strongly compressed, its depth 5.0-6.0 in SL. Head 
3.1—3.7 in SL, its depth about equal to its postorbital length. Snout extending nearly 
its full length beyond tip of mandible, 4.3—5.5 in head. Eye scarcely longer than 
snout in large specimens, 4.25—5.0 in head. Postorbital part of head slightly longer 




Figure 42. Anchoa /yo/epis, 63 mm TL, from Hucares, Puerto Rico, USNM 125580. Drawn by Louella 
E. Cable. 

than snout and eye, 1.7—2.0 in head. Maxillary rather bluntly pointed, its upper 
margin gently rounded, not quite reaching margin of opercle, 1. 15- 1.3 in head. Man- 
dible 1.25— 1.55. Cheek narrow, about as long as snout and eye, its posterior angle 
acute, about 25°. 

Dorsal fin moderately high anteriorly, its longest rays reaching nearly to tip of 
last ray if deflexed, its origin usually a little nearer to base of caudal than to tip of 
snout. Anal with origin under or slightly behind base of last dorsal ray, its base 4.45— 
5.3 in SL. Pelvic inserted equidistant between origin of anal and pectoral base, some- 
times a little nearer to the latter. Pectoral rather falcate, failing to reach pelvic by 
50-75 "/o of diameter of eye, the last ray only about half as long as the first or upper- 
most ray, 1.75— 2.1 in head. Axillary scale of pectoral rather short and broad, about 
50-750/0 of length of pectoral, 2.9-3.5 '" head. 

Color. In alcohol pale. Side of head silvery. Lateral band bright silvery, fully as 
wide as eye (often dusky in specimens preserved in formalin); back and both dorsal 
and caudal fins with dusky dots, these frequently present at base of anal also. 

Size. The 70-mm (2.8 in.) specimen measured is the largest one seen, and it is 
probably near the maximum size attained. 

Reproduction. Specimens of two different lots taken during February in Puerto 
Rico contain well-developed roe. 



2 02 Memoir Sears Foundation for Marine Research 

Variation. Specimens from the Gulf of Mexico have more numerous gill rakers 
than those from farther south, but they seem to agree in other respects. 

Commercial Importance. So far as known, this Anchovy is not used directly as food 
by man. However, where it is common, as in Puerto Rico, it no doubt furnishes con- 
siderable forage for larger predatory fishes that are used as food by man. 

Range and Habitat. The range extends from the Gulf coast of the United States 
through the West Indies to the Gulf of Venezuela. Although some of the specimens 
examined were taken with trawls hauled at depths down to 30 fms., most of them 
were seined along the shore. 

Synonyms and References: 

Stolephorus lyolepis Everraann and Marsh, Bull. U. S. Fish Comm. (1900), 20 (i), 1902: 89, fig. 13 (orig. 

descr. based on juveniles, 30-37 mm long; type local. Culebra, Puerto Rico; type USNM 49528). 
Stolephorus choerostoma Evermann and Marsh, Bull. U. S. Fish Comm. (1900), 20 (i), 1902: 88 (descr., synon., 

specimens in part re-exam., not E. choerostoma Goode). 
Anchovia platyargyrea Fowler, Proc. Acad. nat. Sci. Philad., 63, 191 1 : 216, fig. 4 (orig. descr.; type local. St. 

Martin, West Indies; type ANSP 1416; cf. E. choerostoma Goode; type material, exam, by me, is 

S. lyolepis E. and M.). 
Anchoviella choerostoma ]oTAa.n and Seale, Bull. Mus. comp. Zool. Harv., 6j, 1926: 404 (specimens from e. 

coast of Panama and Puerto Rico probably this species); Beebe and Tee-Van, Zoologica, N. Y., 10 (i), 

1928: 47, fig. (refs., descr., distr., abund. Port-au-Price, Haiti; descr. and fig. show it to be S. lyolepis 

E. and M., not E. choerostoma Goode). 
Anchoviella lyolepis Beebe and Tee- Van, Zoologica, N. Y., 10 (i), 1928: 47, fig. (refs., descr., Port-au-Prince, 

Haiti; descr. based on juveniles); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 

1930: 49 (synon., Puerto Rico only). 
Engraulis platyargyreus Fowler, Proc. Acad. nat. Sci. Philad., 80, 1928: 468 (St. Lucia, B.W.I.). 
Anchovia choerostoma Nichols, N. Y. Acad. Sci., 10 (2), 1929: 205, fig. 39 (refs., diagn., Puerto Rico; S. lyolepis 

E. and M., not E. choerostoma Goode). 
Anchoviella platyargyrea Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 49 (ref. to 

orig. descr.); Beebe and Hollister, Zoologica, N. Y., Jp (6), 1935: 211 (Union I., Grenadines, West 

Indies). 
Anchoa lyolepis Hildebrand, Bull. Bingham oceanogr. Coll., <S (2), 1943: 65, fig- 25 (synon., descr., relation., 

range); Schultz, Proc. U. S. nat. Mus., ()g, 1949: 43 (synon.). 

Doubtful Reference: 

Engraulis {Stolephorus) argentivittatus Regan, Ann. Mag. nat. Hist., (7) JJ, 1904: 257 (orig. descr.; type local. 
Pas Penas, Jalisco, Mexico; type in BMNH). 



Anchoa fiUfera (Fowler) 191 5 
Figure 43 

Study Material. At least 1 1 specimens, ranging between 55—75 mm TL. In addi- 
tion to the three types from Port-of-Spain, Trinidad, 16 additional specimens from 
Kingston, Jamaica, were examined. 

Distinctive Characters. This species in general resembles A. lyolepis, from which it 
differs: in the produced ray of the pectoral, the fin itself being longer; the more anterior 
position of the dorsal and anal fins; and the slightly fewer vertebrae. It differs from 



Fishes of the Western North Atlantic 



203 



A. howelli Hildebrand (j: 105), known only from Santos and Rio de Janeiro, Brazil, 
in having fewer anal rays, gill rakers, and vertebrae, longer pectoral and pelvic fins, 
and a slightly more posteriorly inserted pelvic fin. For details, compare the descriptions 
of the two species. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, SS~1 S "^"^ ^L. 




Figure 43. Anchoa fiUfera, 70 mm TL, from Kingston, Jamaica, USNM 127611. Drawn by Alice C. Mullen. 



Body: depth 17.8-20. 
Head: length 28-29.3. 
Snout: length 5.17-5.85. 
Eye: diameter 5.9—7.0. 
Postorbital: distance 14.7— 16. 
Maxillary: length 20.5-23. 
Mandible: length 18-21. 
Anal fin: length of base 18.5— 21, 



Pectoral fin: length (without filament) 
15. 5-16. 6. 

Scales: ca. 39 or 40. 
Gill rakers: 17-19 + 20-23. 
Fin rays: dorsal 13—15; anal 1 9—23 ; pec- 
toral 13 or 14. 
Vertebrae: 39 or 40 (3 specimens). 



Body strongly compressed, its depth 5.0—5.6 in SL. Head 3.4-3.6, its depth 
slightly exceeding the postorbital length. Snout long, projecting nearly its full length 
beyond mandible, only a little shorter than eye, 5.0—5.5 in head. Eye small, 4.2—4.8. 
Postorbital part of head 1.6— 1.8 in head. Maxillary not sharply pointed, its upper 
margin rounded, reaching nearly to margin of opercle, 1.25— 1.3 in head. Mandible 
1.4— 1.5. Cheek narrow, equal to eye and snout, its posterior angle sharp, about 30°. 

Dorsal fin rather high anteriorly, its longest rays reaching nearly to tip of last 
ray if deflexed, its origin slightly nearer to tip of snout than to base of caudal. Anal 
with origin nearly under the beginning of the last third of dorsal base, the base 4.8 — 
5.4 in SL. Pelvic reaching much more than halfway to anal, inserted a little nearer 
to pectoral base than to anal origin. Pectoral with upper ray produced, its filament 
(often broken) extending to, and possibly beyond, tip of pelvic in perfect specimens, 
the fin, without filament (second ray), reaching base of pelvic, 1.6— 1.8 in head. 



2 04 Memoir Sears Foundation for Marine Research 

Axillary scale of pectoral about 66"/o of length of fin (without filament), 2.8—3.5 
in head. 

Color. In alcohol pale. Side of head and lower part of side silvery. Silvery lateral 
band as wide as eye (often dusky in specimens preserved in formalin). Upper parts of 
head and back with numerous punctulations not arranged in longitudinal series. 

Size. The longest specimen seen (type) is 75 mm (3 in., 61 mm SL). 

Range. This species, as here understood, is known from Puerto Rico, Jamaica, 
Haiti, and Trinidad. 

Synonyms and References: 

Anchovia filifera Fowler, Proc. Acad. nat. Sci. Philad., 191 5: 524, fig. 2 (orig. descr.; type local. Port-of-Spain, 
Trinidad; type ANSP 45080; cf. A. plalyargyrea = A.lyolepis and A. choerostoma. Gill rakers given as 
25 + 22, total anal rays 25, depth 4.75. My counts and measurements give gill rakers 18 + 22, anal 
rays 23, depth 5.0. Orig. fig. shows mandible projecting too far in advance of vertical from anterior 
margin of eye; anal fin rather too far back, its origin actually well in advance of vertical from base of last 
dorsal ray). 

Anchoviella filifera Fowler, Proc. biol. See. Wash., ^J, 1930: 146 (Jamaica); Jordan, Evermann, and Clark, 
Rep. U. S. Comm. Fish (1928), 2, 1930: 50 (ref.). 

Anchoviella longipinna Beebe and Tee- Van, Zoologica, N. Y., Jo(i), 1928: 48, fig. (orig. descr.; type local. 
Bizoton, Haiti; type NYZS 7460; also Port-au-Prince, Haiti; type material not seen; descr. and fig. 
agree well with A. filifera from Jamaica and Trinidad). 

Anchoa filifera Hildebrand, Bull. Bingham oceanogr. Coll., 5(2), 1943: 105, fig. 45 (synon., descr., range). 



Genus Anchoviella Fowler 1 9 1 i 

Anchoviella Fowler, Proc. Acad. nat. Sci. Philad., 42, 191 1 : 211; genotype by original designation, Engraulis 

perfasciatus Poey 1861. 
Amplova Jordan and Seale, Copeia, 1925: 31; type species Anchovia irevirostris Meek and Hildebrand 1923 

equals Amplova balboae Jordan and Seale 1925. 

Characters. Body usually slender, its depth less than 2 5'/o of SL (except in -pal- 
lida^ 2 6-2 8. 5 "/o). Gill rakers usually fewer than 35 on lower limb (except in pal- 
lida, which may have as many as 45), not known to increase in number with age. 
Anal fin rather short, with fewer than 30 rays, its origin usually under or posterior to 
middle of dorsal base (somewhat in advance of middle of dorsal in blackburni and 
jamesi). Vertebrae 39—45. 

Size. Although a maximum length of 155 mm has been reported for eurystole, 
the largest in the collections studied is only 1 30 mm TL, and the other species are 
smaller. 

Remarks. Anchoviella obviously is closely related to Anchoa, from which it differs 
in the shorter and distally square-to-round maxillary. This structure, variable in length 
in both genera, may reach only a little beyond the vertical from the posterior margin 
of the eye, as in Anchoviella jamesi, or it may reach to the joint of the mandible, as in 
Anchoviella eurystole. Indeed, Jordan and Seale (6: 31) erected a new genus, Amplova, 
on the basis of the very short maxillary in A. balboae, a species common on the Pacific 



Species 



Fishes of the Western North Atlantic 205 

Table V. Frequency Distribution of Anal Rays in Anchoviella* 

. Number of Anal Rays >^ 

15 16 17 r8 19 20 21 22 23 24. 21; 26 27 



Species 



eury stole 6 8 17 7 i — — — — 

estauquae i i 2 — — — — — 

perfasciata 6 9 22 9 2 — — — — — — — — 

brevirostris — 2 i i i — — — — — 

guianensis — — — i i — — — — — — — 

alleni — — — 4 4 — — — ' 

pallida — — — — — — — — — I I I — 

elongata — — — — — — — 4 5 5 — 

lepidentostole — — — — — — — — ' — ' — — 

blackburni — — — — — — — — — — 2 2 3 

* Anal rays given as 9-1 1 tor cayennensis, see p. 212. 

Table VI. Frequency Distribution of Gill Rakers on Upper Limb in Anchoviella* 

Number of Gill Rakers on Upper Limb of First Arch n^ 

^P"'" 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 

eurystole 4 11 12 7 4 

estauquae • — ' — " 

perfasciata 112 20 16 5 3 

brevirostris — 4 2 

guianensis i i 

alleni i 7 2- 

pallida I • 2 I 

elongata 7 ^ 

lepidentostole — 3 i — ' 

blackburni i i i 

jamesi 1 i — 

* When the number of counts exceeds the number of specimens listed in the text, the gill rakers on both an- 
terior arches in at least some of the specimens were counted. 

Table VII. Frequency Distribution of Gill Rakers on Lower Limb in Anchoviella* 
Number of Gill Rakers on Lower Limb of First Arch 



15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 



eurystole — ■ 2101310 3 5 2 

estauquae 2 i i 

perfasciata 6 19 17 9 

brevirostris ....- 2 4 

guianensis i i " 

alleni i 5 3 ' 

pallida • ■ 2 I— I I 

elongata 210 3 ■ 

lepidentostole ... 3 i 

blackburni I 4 2 — 

jamesi i i " 

* When the number of enumerations exceeds the number of specimens Usted in the text, the gill rakers have been counted on both 
anterior arches in at least some of the fish. 



2o6 Memoir Sears Foundation for Marine Research 

coast of Panama. However, there are species quite intermediate between the type species 
of Amplova {halboae) and that of Anchoviella (j>erfasciala). For that reason Amplova was 
reduced to subgeneric rank (j: io8). 

The very short maxillary in some species of Anchoviella furnishes a near transition 
from the Engraulidae to the Clupeidae; but other family characters, such as the pro- 
jecting snout and the anteriorly more-or-less embedded and widely separated premaxil- 
laries, remain as in other Engraulidae. 

Range and Habitat. The genus seems to be restricted to the two coasts of America, 
ranging in the Atlantic from Woods Hole, Massachusetts, at least to southern Brazil, 
and in the Pacific from southern California to northern Peru. Most of the species enter 
fresh water. In fact, several of the South American species from the Atlantic drainage 
are known only from freshwater lakes and streams, a few of them having been taken 
a great distance from the sea. 



Key to Species of the Western Atlantic 

I a. Maxillary usually extending beyond posterior rim of orbit by a distance equal to 
or greater than diameter of pupil. Subgenus Anchoviella. 

2 a. Origin of anal posterior to middle of dorsal base, sometimes entirely behind 
dorsal; not more than 20 anal rays. 

3 a. Origin of anal clearly posterior to dorsal base, generally under or behind 
tips of dorsal rays if deflexed; gill rakers 24—28 + 27—33. 
4a. Longest rays of dorsal failing to reach tip of last ray if deflexed; 
eye 3.75—4.5 in head, 2.1—2.5 in postorbital; maxillary 15.8— i7.8''/o 
of SL; postorbital 1 3.8-1 5.5 "/o of SL. 

eurystole (Swain and Meek) 1884, p. 208. 
4b. Longest rays of dorsal reaching beyond tip of last ray if deflexed; 
eye 4.4—4.8 in head, 2.5—2.75 in postorbital; maxillary 17.5— i8.7''/o 
of SL; postorbital 1 5.5-1 5.8 "/o of SL. 

estauquae Hildebrand 1943, p. 211. 
3b. Origin of anal under or slightly behind base of last 2 or 3 rays of dorsal; 
gill rakers usually 13—23 + 19-28. 
5a. Only 9— II anal rays; depth of body less than i4''/o of SL. 

cayennensis (Puyo) 1945, p. 212. 
5b. 15 anal rays or more; depth of body more than 15% of SL. 
6a. Pectoral with 14-16 rays; gill rakers 19-23+24-28. 

■perfasciata (Poey) i860, p. 213. 

6b. Pectoral with 11 — 13 rays; gill rakers 18 or fewer on upper 

limb, generally fewer than 25 on lower limb. 

7 a. Gill rakers 13 or 14 + 18 or 19; pectoral failing to reach 

base of pelvic by diameter of eye, 1 1.2— 13.8 "/o of SL, 

1.7—2.0 in head. brevirostris (Giinther) 1868, p. 216. 



Fishes of the Western North. Atlantic 207 

7b. Gill rakers 16—18 +22—26; pectoral failing to reach pelvic 
by a distance considerably less than eye, 14. 7-1 6.8 "/q of 
SL, 1.4— 1.6 in head. 

8a. Eye 7.0-7.5 "/o of SL, 3.2-3.5 in head; cheek fully as 
long as eye, with a definite posterior angle of about 
60°; axillary scale of pectoral about 50 "/o of fin, 2.4 
in head; a median dark line posterior to anal fin. 

guianensis (Eigenmann) 1912, p. 217. 
8 b. Eye 5.5— 6.25 °/o of SL, 3.7— 4.1 in head; cheek scarcely 
as long as eye, without a definite posterior angle; axil- 
lary scale of pectoral about 75''/o of length of fin, 2.0— 
2.2 in head; no median dark line posterior to anal fin. 
alleni (Myers) 1940, p. 219. 
2 b. Origin of anal under, or in advance of, middle of dorsal base; 22—29 ^"^^^ rays. 
9a. Anal with 28 or 29 rays; pectoral extending to midlength of pelvic. 

nattereri (Steindachner) 1879, p. 220. 
9b. Anal with fewer than 28 rays; pectoral usually failing to reach base 
of pelvic. 

loa. Gill rakers 28-34 + 36-45; depth about 26-28. 5''/o of SL; pec- 
toral extending slightly beyond base of pelvic, 19— 2o''/o of SL. 

pallida (Starks) 191 3, p. 221. 
lob. Gill rakers 18 or fewer on upper limb, 24 or fewer on lower limb; 
depth 2 5*'/o of SL or less; pectoral failing to reach base of pelvic, 
i6.7''/o of SL. 

1 1 a. Maxillary extending beyond orbit a distance about equal to 
diameter of eye, i5-2o'*/o of SL; mandible 1 5.8-1 8.8; gill 
rakers 17 or 18 + 19—24; origin of anal under or slightly 
behind middle of dorsal base. 

12a. Depth of body i7.5-2o"/o of SL; head 25.5-28''/o; 
postorbital part of head 12-13. 3''/o; mandible 17.8- 
i8.8''/o ; silvery lateral band poorly defined, narrower than 
eye. eAongata (Meek and Hildebrand) 1923, p. 223. 
12b. Depth of body 22-25«/o of SL; head 22.7-23.70/0; 
postorbital i o. 7-1 1 "/o; mandible 1 5.8-1 6.3 "/o ; silvery 
lateral band well defined, broader than eye. 

lepidentostole (Fowler) 191 i, p. 224. 

1 1 b. Maxillary extending beyond orbit a distance equal to only 

half of diameter of eye, 1 2-1 5 "/o of SL; mandible 12. 3-14. 8; 

gill rakers 10—12 + 15—17; origin of anal well in advance of 

middle of dorsal base, blackburni Hildebrand 1943, p. 226. 

I b. Maxillary extending a little less than diameter of pupil beyond orbit. 

Subgenus Amplova. 
jamesi (Jordan and Seale) 1926, p. 227. 



2 o 8 Memoir Sears Foundation for Marine Research 

Anchoviella eurystole (Swain and Meek) 1884 

Silver Anchovy 

Figure 44 

Study Material. At least 29 specimens, 55-130 mm, from: Woods Hole and 
Menemsha Bight, Massachusetts; Noank, Connecticut; Long Island, New York; 
Brigantine Bay, New Jersey; and Beaufort, North Carolina. 

Distinctive Characters. A. estauquae is the nearest relative of this species. The rela- 
tionship is shown in the account of that species (p. 211). 




Figure 44. Anchoviella eurystole, 75 mm TL, from Woods Hole, Massachusetts, USNM 77778. Drawn by 
Louella E. Cable. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on Study Material, specimens 55-130 mm. 

Body: depth 15. 4-19. Pectoral fin: length 12. 5-13. 7. 

Head: length 25-28.5. 

Snout: length 3.85-5.0. Scales: c«. 40-45. 

£7^: diameter 5.7-7.0. Gill rakers: 24-28 + 27-33 (most fre- 

Postorbital: distance 1 3.8-1 5.5. quently 25 or 26 + 28-30). 

Maxillary: length 1 5.8-1 7.8. Fin rays: dorsal 13-16; anal 15-18 

Mandible: length 1 5-1 8.5. (rarely 19); pectoral 15 or 16. 

Anal fin: length of base 12. 5-16. Vertebrae: 43-45 (3 specimens). 

Body slender, not strongly compressed, its depth 5.25-6.5 in SL. Head 3.5-4.0, 
its depth equal to length of postorbital part of head. Snout projecting about half of its 
length beyond mandible, only a little shorter than eye in large examples, 4.75-6.5 
in head. Eye 3.75-4.5 in head, 2.1-2.5 in postorbital length of head. Postorbital 
part of head 1.6-1.8 in head. Maxillary rather narrow, rounded distally, extending 
beyond orbit by a distance equal to about 1.3 diameter of eye, failing to reach joint of 
mandible by about half of diameter of pupil, 1.5-1.7 in head. Mandible i. 45-1. 75. 
Cheek only a little shorter than snout and eye, its posterior angle about 45°. Gill 
rakers about 75''/o of length of eye, not broad or close-set, with moderate serrations 
on inner edge. 

Dorsal fin low anteriorly, the longest rays failing to reach tip of last ray if de- 



Fishes of the Western North Atlantic 209 

flexed, Its origin usually somewhat nearer to tip of snout than to base of caudal. Anal 
small, its origin wholly behind dorsal base, most frequently about under tips of longest 
deflexed rays, its base 6.25-8.0 in SL. Pelvic inserted nearly an eye's diameter in 
advance of dorsal and usually somewhat nearer to pectoral base than to anal origin, 
reaching much less than halfway to anal. Pectoral usually failing to reach pelvic by 
more than diameter of eye, 1.8—2.2 in head. Axillary scale of pectoral long and 
narrow, generally failing to reach tip of fin by about half of diameter of pupil, 2.25— 
3.0 in head. 

Color. In alcohol, some rather pale above, others bluish. Side of head and lower 
part of side on some silvery. Silvery lateral band about as broad as eye in small speci- 
mens, broader and less distinctly outlined ventrally in large ones. 

Size. Although a maximum total length of 155 mm (6.2 in.) has been reported, 
the largest example in the collections examined in this study is only 130 mm TL. 

Development. The information on the development of the eggs and larvae that 
follows is based mostly on the work of Kuntz and Radcliffe {10 : 116), who took the 
eggs in tows during August 1 9 1 5 off Gay Head, Massachusetts. 

The egg is buoyant, highly transparent, and ellipsoidal, having a long axis of 
1. 1 5— 1.25 mm, a short axis of 0.55—0.8 mm. The yolk, as usual among anchovies, has 
a network of lines on its surface, giving the appearance of large cells. In this anchovy, 
as in other species studied, the blastodisc forms at one end of the elongated egg. The 
egg is similar to that of Anchoa hepsetus but seems to be slightly shorter though 
scarcely narrower. The development, so far as known, is identical in the two species. 

The newly hatched fish is about 3.0 mm long, being slightly shorter than the 
newly hatched larva of A. hepsetus, in keeping with the rather smaller egg. The body is 
very slender, and the vent as usual in the larvae of anchovies is situated very far back, in 
this species about at the beginning of the distal fourth of the body. Black chromato- 
phores occur in a series along the intestine posterior to the yolksac and at the base of 
the ventral finfold behind the vent. A day or so after hatching, the yolk is virtually all 
absorbed and the fish has grown to a length of about 4.0 mm. The young fish, when 
5.2 mm long, has developed a large oblique, terminal mouth, the fins are becoming 
differentiated, and the convoluted intestine, incompletely invaginated, is conspicuous. 
The chromatophores have become less conspicuous than in the earlier stages. The 
development in older stages apparently remains unknown. 

Spawning. Although the duration of spawning has not been determined, it is at 
its height during July and August (75: 44). 

Range and Habitat. This anchovy ranges from Woods Hole, Massachusetts, at 
least as far south as Beaufort, North Carolina. The southern limit of its range remains 
undetermined, for no specimens of the genus are at hand or seem to be reported for 
the coast between Beaufort, North Carolina, and Melbourne Beach, Florida; those from 
the last mentioned locality are A. perfasciata. During some summers this anchovy Is 
common at Woods Hole; during other summers it seems to be missing. Off Connecti- 
cut, New York, and New Jersey It appears only occasionally. It is not recorded for 



2 1 o Memoir Sears Foundation for Marine Research 

Chesapeake Bay; and for North Carolina it is known by only one specimen taken 
during the summer of 1 913 in inside waters at Beaufort, where intensive collecting was 
carried on during many years. It has been stated often that the home of the Silver 
Anchovy is in the Gulf Stream. The basis for this statement is unknown and certainly 
stands in need of verification, especially as it has been rather definitely established 
that the specimens known from North Carolina northward represent a species distinct 
from the examples from the West Indies northward to Melbourne Beach, Florida. 
All records of argyrophana based on northern specimens apparently apply to eurystole 
as herein understood. 

Synonyms and References: 

Stolephorus ferfasciatus Swain (not of Poey), Bull. U. S. Fish Comm. (1882), 2, 1883: 55 (descr. of one from 
Woods Hole, Massachusetts; indicated as orig. descr. oi S. eurystole by Swain and Meek); Jordan and 
Gilbert (not of Poey), Bull. U. S. nat. Mus., 16, 1883: 273 (descr. of one from Woods Hole, Mass., 
apparently the same one used by Swain ; West Indies includ. in the range where eurystole does not occur) ; 
Bean, T. H. (not of Poey), Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 217 (descr., Noank, Connecticut, 
and Gravesend Bay, Long Island, New York; in part exam, by me and found to be j1. eurystole'). 

Stolephorus eurystole Swain and Meek, Proc. Acad. nat. Sci. Philad., 1885 : 35 (descr. by Swain and by Jordan 
and Gilbert above referred to as orig.; type local. Woods Hole, Mass.; type destroyed by fire. This species 
was said to differ from Engraulis perfasciatus Poey chiefly in the greater number of anal rays, a broader 
lateral band, and a larger size; this northern species apparently reaches a somewhat larger size, but at most 
there is only a small average difference in the number of anal rays, and the width of the lateral band varies 
with age. For diagn. char., see Key); Bean, T. H., Bull. U. S. Fish Comm. (1887), 7, 1888: 150, 
pi. 3, fig. 19 (Ocean City and Longport, New Jersey). 

Stolephorus argyrophanus Jordan and Evermann (not of Cuvier and Valenciennes), Bull. U. S. nat. Mus., 47 
(i), 1896: 444 (descr. apparently based on type of E. eurystole; range "Gulf Stream, occasionally north- 
ward"); Smith (not of C. and V'.), Bull. U. S. Fish Comm. (1897), ij, 1898: 92 (occur, assoc. with other 
species round Woods Hole, Massachusetts); Bean, T. H. (not of C. and V.), Bull. N. Y. St. Mus., 60, 
Zool. 9, 1903: 216 (descr., Ocean City, New Jersey, and Fire Island, New York). 

Anchovia eurystole Fowler, Rep. N. J. St. Mus. (1905), 2, 1906; 109 (descr., Ocean City and Longport, New 
Jersey; ident. of this species and Engraulis argyrophanus C. and V. questioned); Fowler, Rep. N.J. 
St. Mus. (1906), 1907: 267, fig. (name); Fowler, Proc. Acad. nat. Sci. Philad., 6j, 191 1: 219 
(Corson Inlet, New Jersey); Fowler, Proc. Acad. nat. Sci. Philad., 64, 191 2: 34 (Long Beach, New 
York). 

Anchovia argyrophana Sumner, Osburn, and Cole (not of C. and V'.), Bull. U. S. Bur. Fish (1911), jl {2), 
191 3: 743 (abund., where and how taken. Woods Hole, Massachusetts; parasites); Kuntz and Rad- 
cliffe (not of C. and V.), Bull. U. S. Bur. Fish., J5, 1917: 116 (embryol., larval develop.); Nichols 
and Breder (not of C. and V.), Zoologica, N. Y., 9(1), 1927: 43, fig. (diagn., distr., life hist., size); 
Hildebrand (not of C. and V.), Copeia, 1941: 224 (Beaufort, North Carolina). 

Anchoviella argyrophana Jordan and Seale (not of C. and V.), Bull. Mus. comp. Zool. Harv., 67, 1926: 402 
(descr. of one from Provincetown, Mass.; argyrophana and eurystole regarded ident. by Jordan from exam, 
of type of former in Paris, which had 19 anal rays instead of 17, as stated in orig. descr. But number of 
anal rays does not distinguish this species from perfasciata); Breder (not of C. and V.), Field Bk. Mar. 
Fish. Atl. Cst., 1929: 70, 71 (defined in key; range; "Gulf Stream species"; eggs, size); Jordan (not of 
C. and v.). Manual Vert. .'^nim. NE U.S., ed. 13, 1929: 43 (descr., range); Jordan, Evermann, and 
Clark (not of C. and V.), Rep. U. S. Comm. Fish. (1928), 2, 1930: 48 (range; eurystole in synon.). 

Anchoviella eurystole Hildebrand, Bull. Bingham oceanogr. Coll., 8 {2), 1943: 112, fig. 47 (descr., ci. per- 
fasciata, range). 

Doubtful References: 

Anchovia perfasciata Nichols and Breder (probably not of Poey), Zoologica, N. Y., 9 (i), 1927: 42, fig. (diagn., 
N. Y. City area; distingu. char, of this species and one rec. as argyrophana not stated correctly; fig. 



Fishes of the Western North Atlantic 



211 



looks like perfasciata; as it is the same one used by Nichols [1929: 204], it may have been based on a 
West Indian specimen). 
Anchoviella perfasciata (ireeley, 28th Rep. N. Y. Conserv. Dept., Suppl., 15 (2), 1939: 83 (rare at Long Island; 
prabably A. eurysto/e). 



Anchoviella estauquae Hildebrand 1943 
Figure 45 

Study Material. Holotype and 3 paratypes, about 77-92 mm TL (caudal fins 
damaged), 65—77 mm SL, collected on a sandy beach in Estauques Bay, Gulf of Vene- 
zuela, by the U. S. S. Niagara, on February 20, 1925. 




Figure 45. Anchoviella estauque, type specimen, 82 mm TL, 72 mm SL, from Gulf of Venezuela, USNM 
1 19795. Drawn by Louella E. Cable. 



Distinctive Characters. This species is very close to A. eurystole, from which it is 
difficult to separate it with the meager material at hand. The number of gill rakers is 
within the range of eurystole, but the rakers seem to be a little longer, certainly broader, 
apparently closer together, and the serrations on their inner edges are slightly smaller. 
The eye appears to be a little smaller and the postorbital part of the head is slightly 
longer. The dorsal fin is higher anteriorly and somewhat different in shape; the anal 
fin is placed farther back. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on type material, 6^—']6 mm SL. 



Body: depth 1 5.3-1 9.2. 
Head: length 26-27.7. 
Snout: length 5.0-5.55. 
Eye: diameter 5.7—6.16. 
Postorbital: distance 1 5.5-1 5.8. 
Maxillary: length 1 7.5-1 8.7. 
Mandible: length 1 7.3-1 8.5. 



Anal fin: length of base i3-3— 15-3. 
Pectoral fin: length 1 2.8-14.3. 

Scales: ca. 44—46. 
Gill rakers: 26-28 + 30-32. 
Fin rays : dorsal 13—15; anal 15—17; pec- 
toral 15 or 16. 
Vertebrae: 45 (i specimen). 



212 Memoir Sears Foundation for Marine Research 

Body not strongly compressed, its depth 5.2-6.5 in SL. Head 3.6-3.7, its depth 
scarcely equal to its postorbital length. Snout projecting rather more than half of its 
length beyond mandible, only a little shorter than eye, 5-0-5.3 in head. Eye 4.4-4.8 
in head, 2.5-2.75 in postorbital length of head. Postorbital part of head long, 1.73- 
1.77 in head. Maxillary rather narrow, rounded distally, extending beyond orbit a 
distance equal to about 1.3 diameter of eye, not quite to joint of mandible, 1.5-1.6 in 
head. Mandible i. 48-1. 53. Cheek nearly as long as snout and eye, its posterior angle 
about 40°. Gill rakers scarcely as long as eye, broad and close-set, with minute ser- 
rations on inner edge. 

Dorsal fin with longest rays reaching beyond tip of last ray if deflexed, its origin 
about equidistant between tip of snout and base of caudal. Anal with origin well behind 
tips of longest deflexed rays of dorsal, its base 6.5-7.5 in SL. Pelvic inserted about 
half of diameter of eye in advance of dorsal and nearly equidistant between base of pec- 
toral and anal origin, reaching rather less than halfway to anal. Pectoral usually failing 
to reach pelvic by rather more than diameter of eye, 1.9-2.5 in head. Axillary scale 
of pectoral long, narrow, failing to reach tip of fin by a distance equal to or less than 
diameter of pupil, 2.3—2.7 in head. 

Color. In alcohol, pale (originally in formalin). Sides of head bright silvery. 
Lateral band bright silvery, fully as broad as eye. 

Range. This species is known only from the Gulf of Venezuela. 

References : 

Anchovlella estauquae Hildebrand, Bull. Bingham oceanogr. Coll., 8 {2), 1943: 115. fig- 4^ (orig. descr.; type 

local. Estauques Bay, Gulf of Venezuela; type USNM 119795; cf A.eurystole and J. perfasciata); 

Schultz, Proc. U. S. nat. Mus., gg, 1949 : 47 (ref, type and paratypes listed). 



AnchovieJla cayennensis (Puyo) 1945 

Jamais-goute 

Study Material. None. 

Distinctive Characters. The short maxillary, extending about a pupil's diameter 
posterior to a vertical at end of eye, the elongate body, with the depth more than 7 
times in SL, and the very short anal, with 9-1 1 rays and with its origin below that of 
the last dorsal ray, show that this species is a very distinctive one, if Puyo's descriptions 
are reliable. 

Description. Based on published accounts and figures by Puyo (see References 
below). Body very elongate, lightly compressed anteriorly and progressively more 
rounded posteriorly (in transverse section); depth a little more than 7 times in total 
body length (SL). Head a little more than 5 times in same; a sort of transverse keel 
present on superior surface of head. Snout somewhat rounded. Eye equal to snout 
(pupil shown as abnormally large in most of Puyo's drawings, probably in error). 
Maxillary rounded at the end, extending about a normal pupil's diameter posterior 



Fishes of the Western North Atlantic 213 

to vertical at posterior border of eye (in Puyo's published figure). Lower jaw slightly 
shorter than the upper. 

Dorsal fin origin midway between tip of snout and caudal base. Anal fin origin 
under base of last dorsal ray. Fin rays: dorsal 12, anal 9-1 i. Pectoral fin slightly 
less than twice the length of ventral, extending somewhat more than half of the distance 
from its origin to that of the pelvic. 

Color (from Puyo). On the whole, body clear blackish (apparently in formalin); 
opercular bones iridescent. A black longitudinal band on each side; band towards 
posterior part of body giving off irregular transverse projections from its lower border; 
lateral band merging gradually into an irregular streak on middle caudal rays; dark 
band on body bordered below by a silvery band wider than eye diameter. Dorsal and 
anal pale. Pectoral and pelvic hyaline. Caudal lighter than the body, its border lightly 
tinted with brown. 

Size. Puyo's two type specimens, the only examples so far recorded, were 1 30 and 
135 mm SL. 

Habits. Said to be confined to estuaries and river mouths. 

Range. Known only from the coast of French Guiana. The types were taken in the 
Cayenne River, as far up as Macouria," but fishermen told Puyo they found this species 
also in the Mahury River and at the mouth of the Kourou. 

References : 

Stolephorus cayennensis Puyo, Bull. Soc. Hist. nat. Toulouse, 8o, 1945: loi, fig. i (descr., Cayenne R., French 
Guiana); Puyo, Faune Emp. Fran?., 12, Poiss. Guyane Fran?., 1949: 157. fig- 80 (descr.). 



Anchoviella perfasciata (Poey) i860 

Flat Anchovy, Manjua (Cuba) 

Figure 46 

Study Material. At least 41 specimens, 60-105 mm TL, from Puerto Rico, 
Jamaica, Cuba, and St. Lucia; also some poorly preserved specimens from off Mel- 
bourne Beach, Key West, Tortugas, and Pensacola, Florida, and from Mississippi 
Sound, probably belonging to this species. 

Distinctive Characters. This species differs from A. eurystole principally in having 
a larger eye, a higher dorsal fin, a more anteriorly placed anal fin, a longer pectoral fin, 
and fewer gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 41 specimens, 60-105 mm TL. 

Body: depth 1 5.4-1 8.2. Snout: length 4.0-5.25. 

Head: length 23.3-25.6. Eye: diameter 6.6-7.7. 

9. Stieler's Atlas shows Macouria to be on the coast, as is Kourou. — Y. H. O. 



2 14 Memoir Sears Foundation for Marine Research 



Postorbital: distance 11.7-13.3, 
Maxillary: length 1 5.2-1 7.3. 
Mandible: length 1 5-1 7.2. 
Anal fin: length of base 13.7— 15.3. 
Pectoral fin: length 12.2— 14.2. 



Scales: ca. 40—44. 

Gill rakers: 19-23+24-28 (most fre- 
quently 19-21+26 or 27). 

Fin rays : dorsal 12-15; ^"^' 1 5~ i ^ ; pec- 
toral 14—16. 

Vertebrae: 42-44 (7 specimens). 



Body not strongly compressed, its depth S-S~^-S '" ^^- Head 3.9—4.3 in SL, its 
depth slightly exceeding its postorbital length. Snout much shorter than eye, extending 
about half of its length beyond mandible, 4.2—5.75 in head. Eye 3.25-4.0 in head, 1.6- 




FiGURE 46. Anchoviella perfasciata, 100 mm TL, 83 mm SL, from Cuba, USNM 37472. Drawn by Louella 
E. Cable. 

2.0 in postorbital length of head. Postorbital 1.9— 2.1 in head. Maxillary slightly 
expanded and rounded distally, generally extending about an eye's diameter beyond 
rim of orbit, failing to reach joint of mandible by nearly diameter of pupil, 1.4— 1.55 
in head. Mandible 1.4— 1.66. Cheek slightly longer than eye, its posterior angle about 
60°. Gill rakers shorter than eye, broad, not close-set, with short coarse serrations 
on inner edge. 

Dorsal fin with longest rays reaching to, or more usually beyond, tip of last ray if 
deflexed, its origin variable, most frequently equidistant between caudal base and tip 
of snout. Anal originating under or slightly behind base of last dorsal ray, its base 6.5— 
7.25 in SL. Pelvic inserted nearly an eye's diameter in advance of dorsal and about 
equidistant between base of pectoral and anal origin, reaching much less than halfway 
to anal. Pectoral failing to reach pelvic base by about an eye's diameter, 1.7—2.0 in 
head. Axillary scale of pectoral narrow, failing to reach tip of fin by less than dia- 
meter of pupil, 1.9—2.3 in head. 

Color. In alcohol, pale. Sides of head silvery. Silvery lateral band distinct, generally 
nearly as broad as eye; some specimens with dusky punctulations on back extending 
onto caudal fin. 

Size. A total length of about 105 mm (4.2 in.) Is attained. 

Variation. Certain poorly preserved specimens from Melbourne Beach, Key West, 
Tortugas, and Pensacola, Florida, in the Study Material above, appear to be more or 



Fishes of the Western North Atlantic 215 

less intermediate between this species and eurystole. The gill rakers, which are difficult 
to count because of the rather small size and poor condition of the specimens, seem to 
be rather more numerous than in West Indian specimens. The shape of the dorsal, the 
length of the pectoral, and the size of the eye all agree with specimens from the West 
Indies, but the position of the anal is rather intermediate. 

Range and Habitat. This Anchovy apparently ranges from the West Indies to 
Florida and the Gulf Coast, and probably to North Carolina. It has been reported for 
New York (75 : 42), but none from New York and neighboring states has been avail- 
able for study. This fish seems to be common, though probably not numerous, in the 
West Indies. Northward it is apparently only a stray. Most of the specimens studied 
were taken in shallow water. A few, one from off Melbourne Beach, Florida, and two 
from Mississippi Sound, Miss., were taken in somewhat deeper water with trawls. It 
is not known to enter brackish water. 

Synonyms and References: 

Engraulis ferfasciatus Poey, Memorias, 2, i860: 312 (orig. descr.; type local. Cuba; type MCZ 17955; '^^■ 
several species); Giinther, Cat. Fish. Brit. Mus., 7, 1868: 391 (descr., San Domingo, Cuba; doubtfully 
distinct from E. argyrophanus Cuvier and V alenciennes). 

Stolephorus ferfasciatus Swain and Meek, Proc. Acad. nat. Sci. Philad. (1884), 1885: 34 (descr., Key West, 
Florida, and Havana, Cuba); Henshall, Bull. U. S. Fish Comm. (1889), 9, 1891: 373 (more abund. 
at Marco than elsewhere in w. Florida); Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 
441 (descr., range); Evermann and Marsh, Bull. U. S. Fish Comm. (1900), 20, 1902: 88 (descr., 
range, Aguadilla, Puerto Rico); Metzlaar, Bijdr. Dierk. Amst., 1922: 136 (Curasao, taken with S. 
hrozvnii and Jenkinsia lamprotaenia). 

Anchovia perfasciata lordan and Thompson, Bull. U. S. Bur. Fish. (1904), 24, 1905: 233 (Tortugas, Florida); 
Fowler, Proc. Acad. nat. Sci. Philad., 6j, 191 1 : 211 (Aguadilla, Puerto Rico); Nichols, New York .'\cad. 
Sci., JO (2), 1929: 204, fig. 34 (diagn., distr., Puerto Rico); Hildebrand, Copeia, 1941 : 224 (Beaufort, 
North Carolina). 

Anchoviella perfasciata Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6"], 1926: 394 (descr. of "type" from 
Cuba; range, Florida Keys to Cuba); Jordan, Manual Vert. Anim. NE U. S., ed. 13, 1929: 44 (descr., 
range, New York to Cuba); Breder, Field Bk. Mar. Fish. .Atl. Cst., 1929: 70, fig. (range; strays north- 
ward to N. Y.); Fowler, Proc. Acad. nat. Sci. Philad., 82, 1930: 269 (Grenada, B. W. I.); Jordan, 
Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 47 (range, Florida Keys to Jamaica); 
Howell-Rivero, Bull. Mus. comp. Zool. Harv., 82, 1938: 172 (types MCZ 17955; 9 specimens, 82- 
103 mm, includ. loo-mm specimen as holotype, Poey's No. 422); Longley and Hildebrand, Publ. 
Carneg. Instn. Wash., 535, 1941 : 13 (Tortugas, Florida; taxon. notes); Hildebrand, Bull. Bingham 
oceanogr. CoU., 8 (2), 1943: 1 16, fig. 49 (descr., cf. A. eurystole, range). 

Doubtful References: 

Engraulis argyrophanus Cuvier and Valenciennes, Hist. Nat. Poiss., 1848, 21: 49 (orig. descr.; t}'pe local, 
"equitorial Atlantic"; cf A. hepsetus; specimen said to have been taken in equatorial Atlantic on a voy- 
age from Europe to Batavia. The specimen, therefore, may not have been taken near America, and the 
description is too inadequate for positive identification with any American species. Thus it seems inadvis- 
able to use this name. Recent writers generally have referred specimens possessing a short maxillary, 
taken between New Jersey and Cape Cod, to argyrophanus, a procedure that in the light of present 
knowledge seems to be incorrect). 

Stolephorus miarchus Jordan, Proc. U. S. nat. Mus. (1884), 7, 1885: ro6 (Key West, Florida; indistingu. from 
types of this species; probably young A. perfasciata). 

Anchovia perfasciata Nichols and Breder, Zoologica, N. Y., 9 (i), 1927: 42, fig. (diagn., vie. of New York 
Cit}% range; distingu. characters of this species and of eurystole \argyrophanus'\ are not stated correctly; 



2 1 6 Memoir Sears Foundation for Marine Research 

the figure looks somewhat like perfasciata; it is the same one used b}' Nichols [1929: 204] and may have 
been based on a West Indian specimen). 

Negative References: 

Stolephorus perfasciatus Swain, Bull. U. S. Fish Comm. (1882), 2, 1883: 55 (descr. of specimen from Woods 
Hole, Massachusetts, which was A. eurystole); Jordan and Gilbert, Bull. U. S. nat. Mus., 16, 1883 : 272 
(descr. of specimen from Woods Hole, Mass., which was A. eurysto/e; West Indies includ. in range, 
where eurysto/e does not occur); Bean, T. H., Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 217 (descr. of 
specimens from Noank, Connecticut, and Gravesend Bay, Long Island, N. Y.; specimens, in part at least, 
exam, by me and found to be A. eurysto/e). 



Anchovtella brevirostris (Giinther) 1868 

Study Material. Two small soft specimens, 26 and 24 mm SL, from Lago Max- 
imus, far up the Amazon Basin, Brazil, and six specimens that have been more or 
less dry, 17-22 mm SL, from Teffe, also far up the Amazon Basin, which may be 
of this species, all in MCZ. Another, 40 mm TL, from Lake Rogagua, Bolivia, also 
may be of this species (part of CAS 9398). Because of the poor condition and small 
size of the specimens and the inadequacy of the original description, identification is 
uncertain. 

Description. Proportional dimensions in per cent of standard length, based on 
two of the larger and better specimens, 26 and 24 mm SL, respectively. 



Body: depth 15.3, 16.1. 
Head: length 20.4, 24.4. 
Snout: length 3.8, 4.15. 
Eye: diameter 7.7, 7.3. 
Postorbital: distance 12.7, 12.5. 
Maxillary: length 15.4, 12.5. 
Mandible: length 15.4, 13.7. 



Anal fin: length of base 19.2, 17.5. 
Pectoral fin: length }, 11.2. 

Gill rakers: 13 or 14+19 (4 specimens). 
Fin rays: dorsal 12 or 13; anal 16—19; 
pectoral 11 or 12 (4 specimens). 



Body slender, its depth 6.2, 6.5 in SL. Head 4.1, 4.9. Snout projecting about 
half of its length beyond mandible, 5.5, 6.3 in head. Eye 3.0, 3.2. Postorbital part 
of head 1.75, 1.9 in head. Maxillary nearly straight, rounded distally, extending 
about an eye's diameter beyond posterior rim of orbit but not quite to joint of man- 
dible, 1.5, 1.75 in head. Mandible 1.57, 1.6. 

Dorsal origin about equidistant between base of caudal and tip of snout. Anal 
origin under base of last 2 or 3 rays of dorsal; anal base 5.2, 5.7 in SL. Pectoral 
failing to reach pelvic by an eye's diameter, 2.0 in head. 

Color. Very old specimens, in alcohol, faded, except for the silvery lateral band, 
which is somewhat narrower than the eye. 

Digest of Giinther s Description, type 8y mm TL. Head 4.5, its depth S-32)- Snout 
much shorter than eye, projecting only slightly beyond mandible. Eye t,.^ in head. 



Fishes of the Western North Atlantic 217 

Maxillary dilated and rounded distally, not extending to joint of mandible. Dorsal 
origin equidistant between base of caudal and tip of snout. Anal origin below last 
rays of dorsal. Pectoral ending "a great distance" from the pelvic. Fin rays: dorsal 
15, anal 18. Scales 41. Sides with a well-defined silvery band, this band nowhere 
wider than eye. 

It is most unfortunate that the number of gill rakers was not stated. 

Range and Habitat. If the specimens examined have been correctly identified and 
are the same species, the range extends from "Caxoeira,!" Province of Bahia," Brazil, 
type locality, to Lago Maximus and TefFe, both far up the Amazon; and probably to 
Rio Chapare and Lake Rogagua, Bolivia. This species, then, seems to live in fresh 
water. That its distribution is as wide as indicated remains to be confirmed from the 
study of more and better specimens. 

Synonyms and References: 

E/tgraulis brevirostris Giinther, Cat. Fish. Brit. Mus., 7, 1868: 392 (orig. descr. ; type local. "Caxoeira, Province 

of Bahia," Brazil); Jordan and Seale, Bull. Mus. comp. Zool. Harv., 6^, 1926: 41 1 (refer it to Amflova, 

herein regarded incorrect). 
Amplova brevirostris Myers, Proc. Calif .^cad. Sci., 2J, 1940: 440 (notes). 
Anchoviella brevirostris Hildebrand, Bull. Bingham oceanogr. Coll., 8 (z), 1943 '■ 119 (descr., range). 

Doubtful Reference: 

Anchoviella carrikeri Fowler, Proc. .'\cad. nat. nat. Sci. Philad., g2, 1940: 73, fig. 28 (orig. descr.; type local, 
mouth of Rio Chapare, Cochabamba, Bolivia; type ANSP 68980; placed here with considerable doubt 
because scales [31-33] and gill rakers on the lower limb [14 or 15] seem to be too few, and the pectoral 
fin as shown in figure may be too long). 

Negative References: 

Anchoviella brevirostris Pearson, Indiana Univ. Stud., 64, 1925: 50 (Lake Rogagua, Bolivia; ct. paratj'pe of 

Stolephorus guianensis Eigenmann by Myers and found to be that species; see ref under A.guianensis); 

Fowler, Proc. .A.cad. nat. Sci. Philad., g2, 1940: 94 (listed, after Pearson). 



Anchoviella guianensis (Eigenmann) 191 2 
Figure 47 

Study Material. A paratype, 50 mm TL, from the Demerara River, British 
Guiana, CM 2449; and a rather poorly preserved specimen, 62 mm, from Itacoatiara, 
Amazonas, Brazil, CNHM 15480. 

Distinctive Characters. This species dift'ers from the specimens herein assigned to 
A. brevirostris chiefly in the greater number of gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on Study Material, 50 and 62 mm, respectively. 

10. Dr. G. S. Myers has advised that "Caxoeira" and "Cachoeira" are alternative spellings in Brazilian Portu- 
guese for "rapids," and that both are pronounced alike. Cachoeira is the generally accepted spelling today. 
— Y. H. O. 



2 1 8 Memoir Sears Foundation for Marine Research 



Body: depth 20, 17.4. 

Head: length 23.7, 24.2. 

Snout: length 4.5, 5.0. 

Eye: diameter 7.5, 7.0. 

Postorbital: distance 10.7, 14.2. 

Maxillary: length 12.5, 13.3. Fin rays: dorsal 13, 14; anal 18, 19; 

Mandible: length 14, 13.8. 

Body compressed, its depth 5.0, 5.7 in SL. Head 4.2, 4.1 in SL, its depth about 
equal to its postorbital part and half of eye. Snout projecting nearly half of its length 



Anal fin: length of base 16.8, 15.5. 
Pectoral fin: length 15.2, 16.8. 

Scales: ca. 4.1, 38. 
Gill rakers: 16 + 22, 1 7 + 24. 
Fin rays: dorsal 13, 14; anal 18, 
pectoral 12, 13. 




Figure 47. Anchoviella guianensis, 62 mm TL, from Amazonas, Brazil, CNHM 15480. Drawn by Louella 
E. Cable. 



beyond mandible, 5.3, 4.9 in head. Eye 3.2, '},.c^. Postorbital rather short, 2.2, 1.8 
in head. Maxillary extending about 75 "/o of diameter of eye beyond orbit, but not 
to joint of mandible, 1.9, 1.8 in head. Mandible 1.7, 1.75. Cheek broad, very 
slightly longer than eye, its posterior angle about 60°. 

Dorsal fin with longest rays reaching somewhat beyond tip of last ray if deflexed, 
its origin equidistant between tip of snout and caudal base. Anal originating under, 
or a little in advance of, base of last dorsal ray, its base 5.95, 6.5 in SL. Pelvic inserted 
equidistant between pectoral base and anal origin. Pectoral failing to reach pelvic by 
diameter of pupil, 1.55, 1.45 in head. Axillary scale of pectoral narrow, reaching 
well beyond midlength of pectoral, 2.4, 2.45 in head. 

Color. In alcohol, pale (probably preserved originally in formalin). Side of head 
silvery. Pale streak, probably silvery in life, on side. Back with dark punctulations 
arranged in two more or less definite rows behind dorsal fin; dusky punctulations along 
base of anal, and a median dark stripe behind anal; paratype with dusky punctulations 
present also on dorsal, caudal, anal, and upper rays of pectoral. 

Range and Habitat. Its range extends from Trinidad, B.W.I., and Venezuela to 
northern Brazil and the upper Amazon Basin in Bolivia. In addition to the holotype 
from Bartica Rocks, British Guiana, Eigenmann had paratypes from the Morawhanna 



Fishes of the Western North Atlantic 



2\C) 



and Demerara rivers, British Guiana. It is recorded for La Brea Beach, Trinidad, 
and for Cano de Guanoco, Venezuela, without comment by Fowler (see reference 
below). Fresh water seems to be its chief habitat. 

S}'nonyms and References: 

Stolephorus guianensis Eigenmann, Mem. Carneg. Mus., 5, 1912: 447, pi. 62, fig. 5 (orig. descr.; type local. 
Bartica Rocks, British Guiana; type in CM; paratypes from Morawhanna and Demerara rivers, Brit. 
Guiana). 

Anchoviella guianensis Fowler, Proc. Acad. nat. Sci. Philad., 5j, 193 1 : 392, 406 (La Brea Beach, Trinidad, and 
Cario de Guanoco, Venezuela); Hildebrand, Bull. Bingham oceanogr. Coll., 5(2), 1943: 122, fig. 52 
(synon., descr., relation.); Schultz, Proc. U. S. nat. Mus., 99, 1949: 48 (refs.). 

Anckovia brevirostris Pearson (not of Giinther), Indiana Univ. Stud., 11 (64), 1924: 50 (two immatures, 30 
and 37 mm, lU 17354, now in CAS, rec. from Lake Rogagua, Bolivia; cf. paratype of Stolephorus 
guianensis Eigenmann by Myers and found to be that species). 

Amflova guianensis Myers, Proc. Calif. Acad. Sci., (4) 2J, 1940: 440 (descr., exam, specimens from Mo- 
rawhanna [paratype]; Lagoa Grande, lower Amazon, near Obidas, Brazil; and Lake "Rogoagua," 
Bolivia; those from last mentioned place are the ones ident. as A. brevirostris by Pearson; see above). 



Anchoviella alleni (Myers) 1940 

Study Material. The holotype and 6 paratypes, 80-84 "ir" TL, from Peruvian 
Amazon drainage system; the holotype and 4 paratypes in CAS, 2 paratypes in 
USNM. 

Distinctive Characters. This species is close to A. guianensis, from which it differs in 
having a somewhat smaller eye, a shorter anal base, a shorter and broader cheek, a 
longer axillary scale on the pectoral, and in color. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the type material noted above, 80—84 "^"^ TL. 



Body: depth 18.8-20. 
Head: length 22.7-25. 
Snout: length 3.4—4.2. 
Eye: diameter 5.5—6.25. 
Postorbital: distance 11. 3- 12 
Maxillary : length 1 1 . 8 — 1 3 . 
Mandible: length 1 2.7-1 3.2. 



Anal fin: length of base 13.3— 15. 
Pectoral fin: length 14.7— 16.5. 

Scales: ca. 44—46. 

Gill rakers: 16—18+23—26. 

Fin rays: dorsal 14 or 15; anal 18 or 19; 

pectoral 12 or 13. 
Vertebrae: 42 (i specimen). 



Body slender, moderately robust, its depth 5.0—5.3 in SL. Head 4.0—4.4 in SL, its 
depth only slightly less than its postorbital length and eye. Snout projecting about 
half of its length beyond mandible, 5.5-6.0 in head. Eye 3.7-4.1 in head, 1.9-2. i in 
postorbital part of head. Postorbital 1.9—2.0 in head. Maxillary slightly expanded 
and rounded distally, extending as much as diameter of pupil bevond posterior rim of 
orbit, failing to reach joint of mandible by about half of diameter of pupil, 1.7—2.0 in 



220 Memoir Sears Foundation for Marine Research 

head. Mandible i. 63-1. 77. Cheek very short and broad, slightly rounded poste- 
riorly, scarcely as long as eye, and scarcely forming an angle. 

Dorsal fin with longest rays reaching to or beyond tip of last ray if deflexed, 
its origin generally a little nearer to caudal base than to tip of snout. Anal originating 
immediately behind vertical from base of last ray of dorsal, its base 6.7-7.4 in SL. 
Pelvic reaching a little more than halfway to anal, inserted nearer to base of pectoral 
than to origin of anal by diameter of pupil. Pectoral moderately falcate, failing to 
reach pelvic by diameter of pupil, 1.4-1.6 in head. Axillary scale of pectoral reaching 
opposite distal third of fin, 2.0-2.2 in head. 

Color. In alcohol, pale, probably preserved originally in formalin. Lateral band 
bright silvery, nearly as broad as eye, bounded above by a dark line. Tip of snout and 
back with dusky punctulations, these sometimes arranged in two more or less definite 
rows posterior to dorsal fin. Caudal with dusky markings and sometimes with a broad 
dusky margin. 

Range. Known only from the Peruvian Amazon country, from Lake Cashiboya, 
and Gosulima Cocha, Upper Rio Morona, Peru, 2,000 miles from the sea. 

Synonyms and References: 

Amflova alleni Myers, Proc. Calif. Acad. Sci., (4)25, 1940: 441 (orig. descr.; type local. Lake Cashiboya, 
Peru; type CAS 6421; cf Stolephorus guianensis Eigenraann and Engraulis irevirostris Giinther; para- 
types from Lake Cashiboya, Rio Ucayale; Gosulima Cocha, Upper Rio Morona; and Rio Morona, 
all in Peru); Eigcnmann and Allen, Fishes West. S. Amer., Univ. Ky., 1942: 333 (add. specimens from 
upper Amazon basin). 

Anchoviella alleni Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943: 124, fig. 53 (descr. of type material; 
cf. Stolephorus guianensis Eigenmann). 



Anchoviella nattereri (Steindachner) 1879 

Study Material. None. 

Distinctive Characters. In general, A. nattereri seems to be related to A. blackburni 
and A. lepidentostole, but it differs from each of them in the rather more numerous anal 
rays, the more posterior origin of the anal in relation to the dorsal, and the much longer 
pectoral fin, which overlaps the pelvic. 

Description. Based on original account, translated and condensed from Stein- 
dachner, of a specimen 50 mm long." 

II. A description with figure of an anchovy from Iquitos, Peru (upper Amazon Basin), designated Engraulis iquitensis 
by Shoji Nakashima (/</ : 6), shows some of the characters of A. nattereri. At least the general shape of the body, 
the number of anal rays, and the long pectoral overlapping the base of the pelvic seem to agree. However, Naka- 
shima did not describe the maxillary, and its outline is not clearly shown in the figure; the number of gill rakers 
is not stated; and the description and figure do not agree in several respects. Therefore, it is impossible to place the 
fish in any genus with a degree of certainty, or to determine the validity of the species. As the author designated 
no type and failed to compare his fish with any known species, the name probably is without standing and no 
further reference will be made to it. 



Fishes of the Western North Atlantic 



221 



Body: depth 4.0. 
Head: length 3.33. 



Scales: ca. 40. 

Fin rays: dorsal 12; anal 28 or 29. 



Body with dorsal profile anterior to dorsal fin less convex than ventral profile. 
Snout reaching well beyond mandible, 4.0 in head. Eye 3.6. Maxillary nearly- 
square distally, not quite reaching to joint of mandible. Longest Gill rakers about 
as long as eye. 

Dorsal origin slightly nearer to base of caudal than to tip of snout. Anal origin 
about under middle of dorsal base. Pectoral slightly longer than postorbital part of 
head, reaching middle of pelvic. The silvery lateral band is definitely marked. 

Range. Known only from the type from Belem (Para), Brazil. 

Synonyms and References: 

EngrauUs nattereri Steindachner, Ichthyol. Beitr., 8, 1879: 56 (orig. descr.; type local. Para, Brazil; cf. En- 

graulis surinamensis Bleeker = E. edentulus Cuvier). 
Anchoviella nattereri Hildebrand, Bull. Bingham oceanogr. Coll., 5(2), 1943: 133 (descr. after Steindachner; 

relation.). 

Negative Reference: 

Anchovia nattereri Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6"], 1926: 413 (descr. based on specimens 
from Para, Brazil, MCZ 18044; exam, by me and found to be Anchovia clupeoides, differing from 
Steindachner's descr. of A. nattereri in the longer pointed maxillary, in the apparently shorter anal 
fin and its more posterior origin with respect to dorsal, and in the much shorter pectoral fin, not extending 
beyond base of pelvic). 



Anchoviella -pallida (Starks) 1 9 1 3 
Figure 48 

Study Material. Three specimens, 72, 80, and 102 mm TL, the first two being 
the holotype and paratype of Anchoviella venezuelae Fowler, the third one, the holotype 
of Anchovia pallida Starks, from Cafio de Guanoco, Venezuela, and Para, Brazil, re- 
spectively. The specimens agree so well that they almost certainly are one species. 

Distinctive Characters. In general shape this species resembles Cetengraulis edentu- 
lus; however, it differs from it in lacking a membrane connecting the gill covers across 
the isthmus, and in having a longer and narrower cheek, a higher dorsal, and a longer 
pectoral. From the other members of its genus it is well differentiated by the deep body, 
long head, long pectoral, and numerous gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the study specimens, 72-102 mm TL. 



Body: depth 25.6-28.5. 
Head: length 28.5-29.4. 
Snout: length 4.5-4.8. 
Eye: diameter 6.36-6.9. 
Postorbital: distance 1 5.8-1 7. 



Maxillary: length 19-20. 
Mandible: length 18.8-20. 
Anal fin: length of base 27.7-28.5. 
Pectoral fin: length 18.8—20. 



222 



Memoir Sears Foundation for Marine Research 



Scales: ca. 36—38. 

Gill rakers : 2 8-34 + 36-45. 



Fin rays: dorsal 12-14; anal 24-26; pec- 
toral 12 or 13. 



Body strongly compressed, its depth 3.5-3.9 in SL. Head 3.4-3.5, its depth 
equal to its postorbital length and about half of eye. Snout projecting about 0.66 of 
its length beyond mandible, 6.0-6.4 i" head. Eye 4.2-4.9. Postorbital part of head 
1. 7-1. 8 5 in head. Maxillary somewhat expanded, rounded distally, extending about 
1.5 times diameter of eye beyond posterior rim of orbit, not quite reaching joint of 




Figure 48. Anchoviella pallida, type specimen, 102 mm TL, from Para, Brazil, SU 22216. Drawn by Alice 
C. Mullen. 



mandible, i. 4-1. 55 in head. Mandible i. 43-1. 55. Cheek narrow, equal to length of 
snout and eye in large examples, somewhat shorter in smaller ones, its posterior angle 
about 35°. Gill rakers close-set, numerous, about as long as eye (possibly increasing 
in number with age). 

Dorsal fin with longest rays reaching beyond tip of last ray if deflexed, its origin 
about equidistant between caudal base and tip of snout. Anal originating about under 
middle of dorsal base, 3.5-3.6 in SL. Pelvic reaching about 0.66 of distance to anal, 
inserted a little nearer to origin of anal than to base of pectoral. Pectoral reaching 
beyond base of pelvic by as much as diameter of pupil in largest specimen examined, 
somewhat shorter in smaller examples, i. 4-1. 8 5 in head. Axillary scale of pectoral 
broad at base, tapering abruptly, about a third to half as long as the fin, 3.3-3.6 
in head. 

Color. In alcohol, plain. No silvery lateral band evident. Side of head silvery. 
Back with dusky punctulations. Caudal fin with a dusky margin in the largest specimen 
examined (holotype), but not evident in the others. 

Variation. The number of gill rakers in the three specimens examined shows a 
rather wide variation, not attributable to age and size; the smallest specimen, according 
to my counts, has 34 + 42 rakers (34 + 36 according to Fowler's, original description 



Fishes of the IVestern North Atlantic 223 

oi A. venezuehie), the largest 34145, and the intermediate one 33 + 40 on one side 
and 28 + 36 on the other. 

Range. Known from only three specimens, two from Cano de Guanoco, Venezuela, 
and the type from Para, Brazil. 

Synonyms and References: 

Anchovia pallida Starks, Stanf. Univ. Publ., Univ. Ser., 1913: 9, pi i (orig. descr.; type local. Pard, Brazil; 

type SU 22216; cf. Engraulis endentulus Cuvier, E.vaillanli Steindachner and Cetengraulis juruensis 

Boulenger). 
Anchoviella venezuelae Fowler, Proc. Acad. nat. Sci. Philad., 83, 1931: 406, fig. 6 (orig. descr.; type local. 

Cano de Guanoco, mouth of Rio San Juan, V'enezuela; type ANSP 53322; cf. Anchovia parva Meek 

and Hildebrand). 
Anchoviella pallida Fowler, Proc. Acad. nat. Sci. Philad., 9J, 1941 : 126 (synon., descr., Ceara, Brazil); Fowler, 

Arqu. Zool. estad. Sao Paulo, J (6), 1941 : 134 (ref. Brazilian rec); Hildebrand, Bull. Bingham oceanogr. 

Coll., <S (2), 1943: 134, fig. 59 (descr., A. venez.uelae in synon.); Schultz, Proc. U. S. nat. Mus., 99, 

1949: 48 (synon., refs.). 



Anchoviella elongata (Meek and Hildebrand) 1923 
Figure 49 

Study Material. Specimens 65-100 mm: the type material from Mindi, Canal 
Zone, one specimen from Porto Bello, Panama, and 22 from Puerto Pilon, Panama. 

Distinctive Characters. This species is readily recognized by the slender body, 
the very large eye, and the moderately few anal rays and gill rakers. In the length and 
shape of the maxillary it approaches the genus Anchoa rather closely. 

Description. Proportional dimensions in per cent of standard length, aud counts, 
based on at least 14 specimens in Study Material. 

Body: depth 17.5-20. Anal fin: length of base 22-23.7. 

Head: length 25.5-27. Pectoral fin: length 15. 3-16. 3. 

Snout: length 4.5—5.1. 

Eye: diameter 8.2-9.0. Scales: ca. 36-40. 

Postorbital: distance 12..1-13.3. Gill rakers: 17 or 18 + 22-24. 

Maxillary: length 19-20. Fin rays: dorsal 13 or 14; anal 22-24; 

Mandible: length 1 7.8-1 8.8. pectoral 12 or 13. 

Vertebrae: 39 (2 specimens). 

Body slender, moderately compressed, its depth 5.0-5.7 in SL. Head 3.6-3.9 
in SL, its depth equal to postorbital part of head and half of eye. Snout projecting about 
half of its length beyond mandible, 5.3-5.8 in head. Eye large, 3.0-3.3 in head. Post- 
orbital part of head 2.0-2.25 '" head. Maxillary moderately blunt distally, reaching 
nearly or quite to joint of mandible (approaching the genus Anchoa in these respects), 
extending nearly an eye's diameter beyond posterior rim of orbit, i. 3-1. 45 in head. 
Mandible i. 45-1. 5. Cheek about equal to length of eye, its posterior angle approxi- 
mately 60°. 



2 24 Memoir Sears Foundation for Marine Researc/i 

Dorsal fin moderately high anteriorly, its longest rays reaching beyond tip of 
last ray if deflected, its origin about equidistant between caudal base and anterior 
margin of eye. Anal originating under or slightly behind middle of dorsal base, its base 
4.2-4.5 in SL. Pelvic generally reaching more then halfway to anal, inserted rather 
nearer to pectoral base than to anal origin. Pectoral usually extending scarcely to 
base of pelvic, i. 6-1. 75 in head. Axillary scale of pectoral about half as long as 
fin, 2.8—3.75 ''^ head. 

Color. In alcohol, pale (originally preserved in formalin). Silvery lateral band 
poorly defined, narrower than eye. Upper surface of head and back with dusky punc- 
tulations; dark spots at base of anal, and a median dark line posterior to anal. 

Size. The largest specimen in the collection, about 100 mm (4 in.) TL, is the 
largest individual reported. 




Figure 49. Anchoviella elongata, loi mm TL, from Porto Bello, Panama, USNM 82093. Drawn by Louella 
E. Cable. 

Range and Habitat. Known only from the Atlantic coast of Panama, where it ap- 
parently is not rare. Specimens have been collected from brackish water at Mindi 
(near Gatun), and from Gatun Spillway, Canal Zone; and from Porto Bello and Puerto 
Pilon, Panama. 

Synonyms and References: 

Anchovia elongata Meek and Hildebrand, Field Mus. Publ., Zool., 25 (i), 1923: 198, pi. 12, fig. 2 (orig. 

descr.; type local. Mindi, Canal Zone; type USNM 81768; cf. A. cuband)\ Breder, Zoologica, N. Y., 

.^(4), 1925: 141 (common at Gatun Spillway, Canal Zone). 
Anchoviella elongata Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 49 (ref. to orig. 

descr.); Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943: 126, fig. 54 (descr., generic relations., 

range). 



Anchoviella lepidentostole (Fowler) 1 9 1 1 
Figure 50 

Study Material. The type, about 96 mm (caudal fin damaged), 84 mm SL, and a 
paratype 76 mm long, 63 mm SL, from "Surinam." 



Fishes of the IVestern North Atlantic 



225 



Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 84 and 63 mm SL, respectively. 



Body: depth 25, 22.2. 
Head: length 22.7, 23.8. 
Snout: length 3.2, 3.7. 
Eye: diameter 6.9, 7.1. 
Postorbital: distance 10.7, 11. 
Maxillary: length 16.4, 15. 
Mandible: length 16.4, 15.8. 



Anal fin: length of base 23.8, 23.8. 
Pectoral fin: length 16.3, 15.8. 

Scales: ca. 38. 

Gill rakers: 17 + 19 right side, 18 + 20 

left side; 17 + 19 both sides. 
Fin rays: dorsal 14, 15; anal 25, 23; 

pectoral 12 (.''), 14. 



Body rather robust, its depth 4.0, 4.5 in SL. Head 4.4, 4.2 in SL, its 
depth nearly equal to its length without snout. Snout extending about half of its 



,.^ 




Figure 50. Anchoviella lepidentoUole, type specimen, 96 mm TL, from "Surinam," ANSP 1346, modified 
after Fowler (191 1 : 214, fig. 3). Drawn by Alice C. Mullen. 



length beyond mandible, 7.0, ^.^ in head. Eye 3.3, 3.3. Postorbital part of head 
2.1, 2.15 in head. Maxillary not quite square distally, scarcely reaching joint of 
mandible, 1.4, 1.6 in head. Mandible 1.35, 1.5. Cheek as long as eye, the posterior 
angle about 50°. 

Dorsal fin with longest rays failing to reach tip of last ray if deflexed, its origin 
a little nearer to caudal base than to tip of snout. Anal originating under middle of 
dorsal base, its base 4.2, 4.2 in SL. Pelvic reaching scarcely halfway to anal, inserted 
a little nearer to anal origin than to pectoral base. Pectoral failing to reach pelvic by 
diameter of pupil, 1.45, 1.5 in head. Axillary scale of pectoral failing to reach tip 
of fin by half of diameter of pupil, 2.0, 2.15 in head. 

Color. In alcohol, brown. Lower parts silvery. Silvery lateral band strikingly 
prominent, much wider than eye. 

Range. Known only from "Surinam." 

IS 



2 2 6 Memoir Sears Foundation for Marine Research 

Synonyms and References: 

Anchovia kfidentomle Fowler, Proc. Acad. nat. Sci. Philad., 6j, 191 1: 214, fig- 3 (orig. descr.; type local. 

"Surinam"; type ANSP 1346; cf. Engraulis brevirostris Giinther and E.januarius Steindachner). 
Anchoviella lefidentostole Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 49 (ref. to 

orig. descr.); Hildebrand, Bull. Bingham oceanogr. Coll., 5 (2), 1943: 127, fig. 55 (descr. of type material). 



Anchoviella blackburni Hildebrand 1943 
Figure 51 

Study Material. Type material, 1 1 specimens, about 33-38 mm TL (caudal fins 
damaged), all from the Gulf of Venezuela. The specimens probably are not full- 




FiGURE 51. Anchoviella blackburni, type specimen 35 mm TL, from Gulf of Venezuela, USNM 119793. 
Drawn by Louella E. Cable. 

grown, but the maxillary, which is shorter and blunter in very young than in adults, 
probably has acquired the approximate length and shape of a mature fish. At least it 
is shorter and much more rounded than in specimens of Anchoa parva and A. mitchilli 
of similar size. 

Distinctive Characters. The long anal fin, the small number of gill rakers, the short 
head, and the very short, blunt snout distinguish this species from related forms. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the 1 1 type specimens, about 33-38 mm TL. 



Body: depth 18.2-20.8. 
Head: length 21.7-25. 
Snout: length 3.6-3.8. 
Eye: diameter 5.0-7.5. 
Postorbital: distance 11-14.3. 
Maxillary: length 12-15. 
Mandible: length 1 2.3-14.8. 



Anal fin: length of base 23.5—26.5. 
Pectoral fin: length 1 3.8-1 6.6. 

Scales: ca. 38—41. 

Gill rakers: 10-12 + 15— 17. 

Fin rays: dorsal 13-15; anal 25-27; 

pectoral 13 or 14. 
Vertebrae: 43 (i specimen). 



Fishes of the Western North Atlantic 227 

Body quite strongly compressed, its depth 4.8-5.5 in SL. Head 4.0-4.6 in SL, its 
depth equal to postorbital part of head and about half of eye. Snout very short, 
projecting little beyond mandible, about 5.0-7.0 in head. Eye 3.5-4.5. Postorbital 
part of head 1.6-1.9 in head. Maxillary slightly expanded and rounded distally, 
extending about half of diameter of eye beyond posterior rim of orbit, not quite to joint 
of mandible, 1.6-2.0 in head. Cheek only a little longer than eye, the posterior angle 
about 60°. 

Dorsal fin with the longest rays failing to reach tip of last ray if deflexed, its origin 
about equidistant between caudal base and middle of eye. Anal originating about 
under origin of second third of dorsal base, its base 3.75-4.25 in SL. Pelvic reaching 
more than halfway to anal. Pectoral failing to reach base of ventral by somewhat less 
than diameter ot eye, 1.25— 1.6 in head. 

Color. In alcohol, pale (originally preserved in formalin), with only a suggestion 
of a narrow silvery lateral band. A row of dark spots along base of anal, continued as a 
median dark line posterior to anal ; a similar row along base of dorsal to upper lobe of 
caudal. 

Range. Known only from the Gulf of Venezuela, where it was taken in Estauques 
Bay and Jacque Point. 

References : 

Anchoviella blackburni Hildebrand, Bull. Bingham oceanogr. Coll., 8 (2), 1943: 129, fig. 57 (orig. descr.; 
type local. Jacque Point, Gulf of Venezuela; type USNM 1 19793; cf. Anchoa parva [Meek and Hilde- 
brand] and Anchoviella lepdentostole [Fowler]); Schultz, Proc. U. S. nat. Mus., gg, 1949: 48 (ref., 
counts). 



Anchoviella {Amplovd) jamesi (Jordan and Seale) 1926 

Study Material. Two paratypes, about 47 and 37 mm TL, 'i^d and 32 mm SL, 
from fresh water as listed under Range:, MCZ 17783 and 17784, respectively. 

Distinctive Characters. This species has an extremely small mouth for an anchovy 
and is thus related to A. balboae Jordan and Seale of the Pacific coast of Panama, which 
was made the type of the genus Amplova Jordan and Seale, herein regarded as a sub- 
genus. But it differs from A. balboae in having fewer anal rays, fewer gill rakers, a more 
posteriorly placed dorsal fin, and apparently a more slender body. A. balboae has 
24—27 anal rays and 19—21 +29—35 S^'' rakers; its dorsal origin is about equidistant 
between the base of caudal and middle of eye; and the depth in specimens 75— 100 mm 
TL is 3.5—4.1 in SL. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens 36 and 32 mm SL, respectively. 

Body: depth 22.2, 23.8. Eye: diameter 5.8, 6.6. 

Head: length 22.2, 22.7. Postorbital: distance .'', 12.8. 

Snout: length 3.04, 3.1. Maxillary: length 11.6, 11.9. 



2 2 8 Memoir Sears Foundation for Marine Research 

Anal fin: length of base 2 1, 18.7. Gill rakers: 13+21, 12+20. 

Pectoral fin: length 16.6, 18.7. Fin rays: dorsal 13, I2;anal 21, 19; 

pectoral 11, 11. 
Scales: ca. 40. Vertebrae: 40, ?. 

Body moderately compressed, Its depth 4.5, ^.■^ in SL. Head 4.5, 4.4 in SL, its 
depth equal to postorbital length and about half of eye. Snout very short, projecting 
about a third of its length beyond mandible, 7.2, 7.3 in head. Eye 3.8, 2)-S' Post- 
orbital part of head 1.85 in head. Maxillary distally rounded, scarcely reaching 
beyond posterior rim of orbit, and not to articulation of mandible, 2.0, 1.9 in head. 
Cheek very short and broad, scarcely forming a triangle, much shorter than eye. 

Dorsal fin with the longest rays reaching far beyond tip of last ray if deflexed, 
its origin nearer to base of caudal than to eye. Anal with a prominent lobe anteriorly, 
its origin slightly in advance of middle of dorsal base, its base 4.8, ^.t, in SL. Pelvic 
reaching more than halfway to anal, inserted about equidistant between base of pec- 
toral and origin of anal. Pectoral nearly reaching base of pelvic, 1.2, 1.25 in head. 
Axillary scale of pectoral missing in specimens examined. 

Color. Old specimens, in alcohol, faded, pale. Silvery lateral band rather diffuse, 
narrower than eye. 

Range. Known from the type material from fresh water, the holotype being from 
Jutahy River, the two paratypes from Lago Alexo, Brazil. Two small specimens also 
are reported from the Rio Ucayali, in Peru. 

Synonyms and References: 

Amplovajamesi Jordan and Seale, Bull. Mus. comp. Zool. Harv., bj, 1926: 410 (orig. descr.; type local. Jutahy 

R., Brazil; type MCZ 18014; two paratypes listed from Lago Alexo, Brazil); Myers, Proc. Calif. 

Acad. Sci., (4) 23, 1940: 441 (descr. after Jordan and Seale; Rio Ucayali, Peru). 
Anchoviella jamesi Hildebrand, Bull. Bingham oceanogr. Coll., 5 (2), 1943: 137, fig. 61 (descr., relation, with 

Amplova balboae Jordan and Seale). 



Genus Pterengraulis Giinther 1868 

P terengrauHs Gunther, Cat. Fish. Brit. Mus., 7, 1868: 384, 398; type species by monotypy, Clupea atherinoides 
Linnaeus 1758. 

Characters. Body rather deep, strongly compressed. Gill rakers short and broad, 
especially in large examples, only about 11 on upper limb and about 14 on lower 
limb. Dorsal origin about equidistant between margin of opercle and base of caudal. 
Pelvic inserted about midway between pectoral base and dorsal origin. 

Size. A large size is attained, specimens 250 mm long being included in the 
collections studied. 

Remarks. The anal fin is long, with about 30-33 rays, its origin being under, 
or more usually in advance of, the dorsal origin. The teeth are small and of nearly 



Fishes of the Western North Atlantic 



229 



uniform size. The maxillary extends to, or nearly to, the joint of the mandible and is 
rounded distally. 

Range. Venezuela to northern Brazil. 

Species. Only one, P. atherinoides. 



Pterengraulis atherinoides (Linnaeus) 1766 

Hareng gras Sardinha 

Figure 52 

Study Material. A total of 20 specimens, 66—250 mm TL, from Cano de Guanoco 
and Rio Apure, San Fernando de Apure, Venezuela; and from the "Amazon River," 
Para, Fortaleza, Cameta, and Arary, Brazil. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 66—250 mm TL. 




Figure 52. Pterengraulis atherinoides, 245 mm TL, from "Amazon River," USNM 72551. Drawn by Louella 
E. Cable. 



Body: depth 22-26. 
Head: length 22.6-28.5. 
Snout: length 2.4-3.1. 
Eye: diameter 4.0—5.1. 
Postorbital: distance 14.3 — 16.7. 
Maxillary: length 1 6.5-1 9.2. 
Mandible: length i 5.4-1 8. 



Anal fin: length of base 30-34.5. 
Pectoral fin: length 20.4—23.7. 

Scales: 42—47. 
Gill rakers : 10-12 + 12-15. 
Fin rays: dorsal 12 — 14; anal 29—35 
(usually 31 or 32); pectoral 13-15. 
Vertebrae: 43 and 45 (2 specimens). 



Body strongly compressed, its depth 3.5-4.4 in SL. Head 3.8-4.5 in SL, its 
depth usually equal to its length posterior to middle of eye. Snout blunt, 8.0-10 in 
head. Eye 4.0-6.0. Postorbital part of head moderately long, i. 4-1. 55 in head. 



230 Memoir Sears Foundation for Marine Research 

Maxillary rounded distally, often failing to reach joint of mandible, 1.3-1.5 in head. 
Mandible 1.3— 1.5. Cheek increasing in proportionate length with growth, much 
longer than eye and snout in large examples, its posterior angle about 40°. 

Dorsal fin with the longest rays reaching beyond tip of last ray if deflexed, its 
origin about equidistant between caudal base and margin of opercle. Anal long, its 
origin under or more usually somewhat in advance of dorsal origin, its base 2.9—3.3 
in SL. Pelvic reaching only about a third of distance to anal, usually inserted a little 
nearer to base of pectoral than to anal origin. Pectoral reaching to or beyond middle 
of pelvic, 1. 0—1.3 '" head. Axillary scale of pectoral reaching to or somewhat beyond 
midlength of fin, 1.8 -2.1 in head. 

Color. In alcohol, pale brownish above. Sides of head and lower parts of body 
silvery. Silvery lateral band broad anteriorly, narrower on caudal portion of body, be- 
coming diffused with the silvery shade on lower part of body anteriorly in large speci- 
mens, apparently persisting posteriorly. 

Size. Examples 300 mm TL have been reported. 

Commercial Importance. This fish has been reported as very common in the market 
at Belem (Para), Brazil. 

Range. Its range extends from Venezuela to northern Brazil, possibly as far south 
as Rio de Janeiro. It ascends streams. 

Synonyms and References: 

C/upea atherlnoides Linnaeus, Syst. Nat., ed. 12, j, 1766: 535 (orig. descr.; type local. Surinam; type lost). 

Engraulis atherinoides Cuvier and Valenciennes, Hist. Nat. Poiss., 21, 1 848 : 3 1 (descr., relations., Surinam) ; 

Giinther, Cat. Fish. Brit. Mus., 7, 1868: 398 (descr., Surinam, British Guiana, and Para, Brazil); 

Steindachner, Ichthyol. Beitr., 8, 1879: 59 (descr., Surinam; British Guiana; Para, Cameta, Gurupa, 

Rio de Janeiro, and Capim R., Brazil). 
Pterengraulis atherinoides Jordan and Evermann, Bull. U. S. nat. Mus., 47(1), 1896: 450 (descr., range); 

Schreiner and Miranda-Ribeiro, Arch. Mus. nac, Rio de J., 12, 1903: 94 (range, Guianas to Brazil); 

Starks, Stanf. Univ. Publ., Univ. Ser., 191 3: 10 (common in market at Para, Brazil); Jordan and Scale, 

Bull. Mus. comp. Zool. Harv., 6"], 1926: 386 (descr.. Para, Cametd, Gurupa, and Arary, Brazil); Fowler, 

Proc. Acad. nat. Sci. Philad., 82, 193 1 '■ 407 (Cano de Guanoco, V'enezuela); Puyo, Bull. Soc. Hist. nat. 

Toulouse, JO, 1936: 170, fig. 36 (descr., habitat in French Guiana); Fowler, Proc. Acad. nat. Sci. 

Philad., 9J, 1941 : 126, fig. 3 (Rio Jaguaribe, Ceard, Brazil; refs.); Arqu. Zool. estad. Sao Paulo, J (6), 

1941 : 135 (refs., Brazil); Campos, Arqu. Zool. estad. Sao Paulo, J (7) 1941 : 200 (synon., descr., range); 

Hildebrand, Bull. Bingham oceanogr. Coll., 1^(2), 1943: 139, fig- 63 (descr., synon., range); Puyo, 

Faune Emp. Fran?., 12, Poiss. Guyane Frang., 1949: 158, fig. 81 (descr., habitat, French Guiana); 

Schultz, Proc. U. S. nat. Mus., gg, 1949 ■■ 48 (San Fernando de Apure, Venezuela). 
Clupea filchardus de Magalhaes (not of Linnaeus), Monogr. Brazil, Peixes Fluviaes, 193 1 : 194, fig. 104 (descr., 

local name, length, habitat). 

Doubtful Reference: 

Engraulis atherinoides Knerr, "Novara," Zool., Fische (i), 1865: 332 (descr. insufficient for ident.; Rio de 
Janeiro, Brazil). 

Genus Hildebrandichthys Schultz 1949 

Hildebrandichthys Schultz, Proc. U. S. nat. Mus., gg, 1949: 49; type species by original designation, Hilde- 
brandichthys setiger Schultz. 



Fishes of the Western North Atlantic 231 

Characters. Body slender, rather strongly compressed, its depth about equal to 
postorbital part of head. Head long and low. Eye entirely in anterior half of head. 
Maxillary rounded posteriorly, not quite reaching joint of mandible. Teeth all small. 
Gill rakers slender, those at angle of first arch about as long as eye, about 33 on lower 
limb. Dorsal fin far back, its origin about twice the distance from tip of snout as from 
caudal base. Anal fin beginning well in advance of origin of dorsal, about 25 rays. 
Pelvic inserted about equidistant between base of pectoral and anal origin. 

Remarks. The numerous long gill rakers and the posterior position of the dorsal 
fin are the chief distinguishing characters of this genus. Hildebrandichthys resembles 
Anchovia in the numerous long, slender gill rakers, but it differs in the very rear- 
ward position of the dorsal fin. In the position of the dorsal it tends to agree more 
or less with Pterengraulis., with which it differs in the numerous, long slender gill 
rakers. 

This genus is founded on one species, which in turn is based on only a single 
small (juvenile) specimen. Though it is quite distinctive, it offers some difficulties be- 
cause the adult generic characters apparently are not fully developed. The maxillary, for 
example, is short and rounded, which, so far as known, is true of all young Engraulidae; 
it may or may not become produced and pointed in the adult, as shown in other generic 
descriptions. 

Range. Known only from the Maracaibo Basin, Venezuela. 



Hildebrandichthys setiger Schultz i 949 
Figure c^-^ 

Study Material. A single small specimen, about 37 mm TL (caudal damaged), 
3 1 mm SL, from the mouth of Rio Cano de Sagua, Lake Maracaibo, Venezuela. This 
specimen retains some of the juvenile color markings, and the maxillary may not have 
acquired its adult length and shape. 

Distinctive Characters. The numerous gill rakers, which may increase in number 
with age, in general agree with the rakers in Anchovia clupeoides ; the posterior position 
of the dorsal fin calls to mind Pterengraulis atherinoides. See also Remarks under genus. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on young specimen, 31 mm SL. 

Body: depth 19.2. Anal fin: length of base 22.6. 

Head: length 32. Pectoral fin: length 12. 

Snout: length 5.1. 

Eye: diameter 8.7. Scales: mostly lost, ca. 38 pockets. 

Postorbital: distance 17. Gill rakers: c^. 23 + 33. 

Maxillary: length 17. Fin rays: dorsal I2; anal 25; pectoral 13. 

Mandible: length 16. 



232 Memoir Sears Foundation for Marine Research 

Body slender, rather strongly compressed, its depth 5.2 in SL. Head 3.1, its 
depth about equal to its postorbital length. Snout projecting about 67''/o of its length 
beyond tip of mandible, 6.25 in head. Eye small but longer than snout, 3.7 in head. 
Postorbital length 5.9. Maxillary rounded posteriorly, not quite reaching joint of 
mandible, 1.9 in head. Mandible 2.0. Cheek a little longer than eye, its posterior 
angle about 60°. Gill rakers slender, those at angle of first arch about as long as 
eye (too close-set for accurate counting), about 23 on the upper limb and about 33 
on lower limb. 

Dorsal fin small, damaged, apparently with a convex margin, the anterior rays 
not reaching beyond tip of last ray if deflexed, the fin placed very far back, its 




Figure 53. Hildebrandichthys seliger, type specimen, 37 mm TL, from Maracaibo Basin, Venezuela, USNM 
121779. Drawn by A. M. Awl. 

origin about twice as far from tip of snout as from caudal base. Anal originating about 
an eye's diameter in advance of dorsal origin and about equidistant between joint of 
mandible and caudal base, its base 4.4 in SL. Pelvic fin noticeably more than halfway 
to origin of anal, inserted about equidistant between pectoral base and anal origin, 
3.33 in head. Pectoral fin failing to reach base of pelvic by a little less than diameter 
of eye, 8.4 in SL, 2.6 in head. Axillary scale of pectoral not evident, lost or un- 
developed. 

Color. This specimen retains some of the juvenile color markings. In alcohol, 
rather pale. Side of head and abdomen silvery. A rather faint narrow silvery lateral 
band present. Dusky dots present on upper surface of head and snout, along base of 
dorsal, and mostly in two rows behind the dorsal ; a series of rather large black dots 
along base of anal, continued as a single median row behind anal fin (presumably 
juvenile markings); dusky dots generally distributed over the caudal fin; other fins 
pale or at most with suggestions of dusky markings. 

Range. Known only from the mouth of the Rio Cano de Sagua, Lake Maracaibo, 
Venezuela. 



Reference: 

Hildebrandkhthys setiger Schultz, Proc. U. S. nat. Mus., gg, 1949: 49, fig. 7 (orig. descr.; type local. Carlo 
de Sagua, Sinamaica, Venezuela; type USNM 12 1779). 



Fishes of the Western North Atlantic 233 

Genus Lycengraulis Giinther 1868 

Lycengrau/is Giinther, Cat. Fish. Brit. Mus., 7, 1868: 385, 399; type species Engraulis grossidens Cuvier 1828, 
designated by Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 451. 

Characters. Body slender, compressed, the depth generally less than 27''/o of SL. 
Mouth very large. Maxillary extending far beyond eye, generally to or beyond joint 
of mandible, rounded or pointed distally. Mandible curved upward near tip. Gill 
RAKERS variable in length, about 10-20+12-25 °" ^'^'^^ arch. Anal originating 
definitely posterior to dorsal origin, with about 'l\—l'i) rays. Pelvic inserted about 
midway between base of pectoral and origin of anal. Young with a silvery lateral band 
that disappears with age. 

SiT-e. About 250 mm is the maximum size attained in this genus. 

Remarks. The genus is distinguished by the prominent teeth in the jaws: un- 
equal in size in older specimens but rather even in young; larger in the lower than in 
the upper jaw, some of them often canine-like; variable in number among the species, 
about 16—25 enlarged teeth in lower jaw and 40—85 in upper jaw. 

Species. Eight species have been recognized, one from the Pacific ranging from 
Panama Bay to Peru, and seven from the Atlantic ranging from the Gulf of Venezuela 
to northern Argentina. Five of the Atlantic species come within the scope of the present 
work. 

Key to Species of the Western North Atlantic 

I a. Gill rakers long and slender, those at angle equal to or longer than pupil. 

2a. Gill rakers 13-20+ 17-25 on first arch; cheek scarcely longer than snout and 
eye in adults. 

3a. Depth of body 23— 24.5''/(, of SL; maxillary reaching margin of opercle; 
pelvic fin usually inserted nearer to origin of anal than to pectoral base, 
occasionally equidistant between these; vertebrae usually 43 or 44 (oc- 
casionally 42 or 45). grossidens (Cuvier) 1829, p. 234. 
3 b. Depth of body 2i-23''/o of SL in specimens upward of about 100 mm 
TL; maxillary not reaching margin of opercle; pelvic fin inserted nearer 
to pectoral base than to anal origin ; vertebrae usually 41 (occasionally 42). 

limnichthys Schultz 1949, p. 237. 

2b. Gill rakers 9-13+ 12 — 15 ^^ ^""^^ arch; cheek much longer than snout and 

eye in adults. batesii (Giinther) 1868, p. 240. 

lb. Gill rakers short and broad, the longest scarcely exceeding half of length of pupil. 

4a. Maxillary sharply pointed distally, extending well beyond joint of mandible; 

origin of dorsal about equidistant between caudal base and middle of eye; 

gill rakers 11 or 12 + 16 or 17. abbotti (Fowler) 191 5, p. 242. 

4b. Maxillary rounded distally, scarcely reaching joint of mandible; origin of 

dorsal much nearer to base of caudal than to posterior margin of eye; gill 

rakers 9—1 i + 12 — 14. barbouri Hildebrand 1943, p- 243. 



2 34 Memoir Sears Foundation for Marine Research 

Table VIII. Comparison of L. grossidens with L. olidus. Proportional Dimensions 
in Per Cent of Standard Length. 





L. grossidens 


L. olidus 




L. grossidens 


L. olidus 




40 specimens 


44 specimens 




40 specimens 


44 specimens 




& others (40- 


(97-240 mm 




& others (40- 


(97-240 mm 




220 mm TL) 


TL) from 




220 mm TL) 


TL) from 




from northern 


southern Brazil, 




from northern 


southern Brazil, 




Brazil and 


Paraguay, 




Brazil and 


Paraguay, 




Venezuela 


Uruguay, and 
.'Argentina 




Venezuela 


Uruguay, and 
Argentina 


BoJy: depth 


21-24.3* 


19.2-22.7 


Scales: 


ca. 40-44 


ca. 40-45 


Head: length 


22.6-26.6 


22.2-25 


Gill rakers: 


13-17 + 17-21 


15-19 + 19-23 


Snout: length 


.V3-4-3 


2-7-34 




(most fre- 


(most fre- 


Eye: diameter 


4.8-7.0 


4-3-5-7 




quently 14+19 


quently 16 + 21) 


Postorbital: 








or 20) 




distance 


13-15 


13-15-5 


Fin rays: 






Maxillary: length 


18. 2-21. 5 


16.5-19.5 


dorsal 


14 or 15 


14 or 15 


Mandible: length 


15. 2-18. 2 


15-5-17-5 


anal 


24-28 


26-28 


Anal fin: length 






pectoral 


14 or 15 


14 or 15 


of base 


23.7-26.5 


22-25.5 


Vertebrae : 


43 or 44 


46-48 


Pectoral fin: length 


16.6-20.0 


15.8-17.8 




(4 specimens) 


(3 specimens) 



23-24.3 in specimens 100 mm upward. 



Lycengraulis grossidens (Cuvier) 1829 

Sardinha Prata (Brazil) 

Figure 54 

Study Material. A total of 40 specimens, 40-220 mm TL, and many others as 
well [presumably from Rio de Janeiro, Brazil and northward, as implied by Dr. Hil- 
debrand under Relationship and Variation— y.h.o.\ Also 44 specimens, 97— 240 mm 
TL, from southern Brazil, Paraguay, Uruguay, and Argentina, representing L. olidus 
(Giinther), used as basis for measurements and counts in Tables viii-x. 

Distinctive Characters. See Relationship and Variation. 

Description. Proportional dimensions and counts, based on 40 specimens from 
Rio de Janeiro, Brazil, and northward, 40—220 mm TL, in Table viii. 

Body strongly compressed, increasing in depth with age, its depth 3.85—4.4 in 
SL in specimens 100 mm or more. Head with depth nearly equal to its length with- 
out snout, 4.1—4.7 in SL. Snout projecting somewhat less than half of its length beyond 
mandible, 5.6-6.9 in head. Eye 4.2—4.9. Postorbital part of head 1.6— 1.7 in head. 
Maxillary little expanded, pointed distally, extending beyond joint of mandible, often 
nearly to margin of opercle, 1. 15— 1.25 in head. Mandible i. 33-1. 6. Cheek some- 
what longer than eye and snout in large examples, shorter than snout and eye in 
specimens less than 100 mm, its posterior angle about 35°. Gill rakers slender, those 
at angle fully as long as pupil. 

Dorsal fin with the longest rays usually reaching about to tip of last ray if de- 



Fishes of the Western North Atlantic 235 

Table IX. Frequency Distribution of Gill Rakers in Lycengraulis.* 



, Number of Gill Rakers on Upper Limb of First Arch v^ 

^P^*^'" 9 10 II 12 13 14 15 16 17 18 19 20 

grossidens — — — — 4 '5 ^ 7 ' — — — 

olidus ______ 315 8 4 I 3 

batesii 3 2 4 5 2 — — — — — — — 

abbotti — — I I — — — — — — — — 

barbouri I 3 4 — ' — — — — — — — 

* When the number of counts exceeds the number of specimens reported in the text, the gill rakers were counted 
on both anterior arches of at least some of the fish. 

Table X. Frequency Distribution of Gill Rakers in Lycengraulis. * 

, Number of Gill Rakers on Lower Limb of First Arch ^ 



Species 



13 14 15 16 17 18 19 20 21 22 23 24 25 



grossidens . — — — — — i 9 '3 '5 ^ — — 

olidus ______ 2 5 6 20 7 5 

batesii. ... — 3 7 2 — — — — — — — — 

abbotti ... — — — — I I — — — — — — 

barbouri . . i 6 i — — — — — — — — — 



* When the number of counts exceeds the number of specimens reported in the text, the gill rakers were counted 
on both anterior arches of at least some of the fish. 

flexed, its origin rather variable, most frequently nearer to base of caudal than to poste- 
rior margin of eye. Anal with base 3.75-4.2 in SL, its origin generally under middle 
of dorsal base. Pelvic small, extending much less than halfway to anal, usually in- 
serted nearer to anal origin than to pectoral base. Pectoral often reaching nearly to 
base of pelvic, i.i 5-1. 35 in head. Axillary scale of pectoral about 0.75 of length of 
fin, 1.5— 1.9 in head. 

Co/or. In alcohol, grayish above. Sides of head and lower 67 "/o of body uniform 
silvery in adults. Young under 90 mm long with a distinct silvery lateral band nar- 
rower than eye, much broader than eye and less distinctly outlined ventrally in speci- 
mens 120 mm long, intergrading completely with the silvery color of lower parts of 
body in somewhat larger individuals. Margin of caudal blackish in adults. 

Size. Examples having a maximum total length of 280 mm (i 1.2 in.) have been 
reported, but the largest in the collections studied is only 220 mm TL. 

Relationship and Variation. L. olidus (Giinther) is a very near relative of L. grossi- 
dens, the two probably intergrading in southern Brazil. If specimens from Venezuela or 
northern Brazil are compared with examples from extreme southern Brazil, Uruguay, 
and Argentina, it is at once evident that those from the more southern localities have a 
more slender and less strongly compressed body, a shorter and less pointed maxillary, 
smaller teeth in the upper jaw but still much larger than those of other American genera, 
a greater average number of gill rakers, and more numerous vertebrae. See Table viii 
for comparison of L. grossidens with L. olidus. 

L. poeyi of the Pacific coast of Panama is also a near relative of the Atlantic species. 



236 Memoir Sears Foundation for Marine Research 

but it differs in having: a shorter and blunter snout that projects little beyond the 
mandible; smaller teeth of more uniform size in the lower jaw; a longer pectoral 
that reaches to, or a little beyond, the base of the pelvic and has 16 or 17 rays 
(i or 2 more than in grossidens and olidus)\ and apparently a dark vertebral band, which 
is missing in the L. grossidens and L. olidus examined. In depth of body and in length 
and shape of the maxillary it agrees with L. olidus rather than with L. grossidens. 

Commercial Importance. As some of the specimens in the collections studied were 
purchased in markets, this species must be of some commercial value. 




Figure 54. LycengrauHs grossidens, 214 mm TL, from Porto Inhauna, Brazil, USNM 100835. Drawn by 
Louella E. Cable. 



Range and Habitat. The range of L. grossidens extends from the Gulf of Venezuela 
south about to Rio de Janeiro, Brazil, and of L. olidus from southern Brazil to northern 
Argentina; both species ascend freshwater streams. The rather numerous records of 
h. grossidens from Uruguay and Argentina are apparently referable to L. olidus. 

Synonyms and References: 

Engraulis grossidens Cuvier, in Spix and Agassiz, Pise. Brazil., 1829: 50, pi. 24, fig. i E (orig. descr.; type 
local. Rio de Janeiro, Brazil); Giinther, Cat. Fish. Brit. Mus., 7, 1868: 399 (descr., synon., British 
Guiana). 

Engraulis Janeiro Agassiz, in Spix and Agassiz, Pise. Brazil., 1829: pi. 24, fig. i (type local. Rio de Janeiro, 
Brazil; name and fig. only). 

Engraulis dentex (Zvivitx ^.Xid. Valenciennes, Hist. Nat. Poiss., 27, 1848: 28 (orig. descr.; type local. Rio de Janeiro, 
Brazil); Kner, Reise "Novara," Zool., Fische, I, 1865: 332 (descr., Rio de Janeiro); Steindachner, 
Ichthyol. Notiz., 6, 1867: 35 (descr., Rio de Janeiro and La Plata; the latter undoubtedly referable 
to L. olidus'). 

LycengrauHs grossidens Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 451 (descr., range, synon.); 
Gilbert, Proc. Wash. Acad. Sci., 2, 1900: 163 (descr., Pernambuco, Brazil); Starks, Stanf. Univ. Publ., 
Univ. Ser., 1913: 11 (descr.. Natal, Brazil); Jordan and Gilbert, Bull. Mus. comp. Zool. Harv., 6'j, 
1926: 383 (descr., Brazil and Uruguay; those from south of Rio de Janeiro, Brazil, probably L. olidus; 
those from Rio Poty, Brazil, are L. barbourt)\ von Ihering, Rev. Indust. Anim., Anno i (3), 1930: 233 
(size, diagn.); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 51 (range, synon.); 
Fowler, .'Vrqu. Zool. estad. Sao Paulo, J (6), 1941: 135 (refs., Brazil); Hildebrand, Bull. Bingham oceanogr. 



Fishes of the Western North Atlantic 237 

Coll., 8 {z), 1943: 143, fig- 64 (descr., range, synon.); Schultz, Proc. U. S. nat. Mus., gq, 1949: 
54 (synon., refs.). 

Negative References: 

Lycengraulis grossiJens Berg, An. Mus. nac, Buenos Aires, 4, 1895: 21 (fairly abund. in markets, Buenos Aires 

and Montevideo; L.olidus); Devincenzi, An. Mus. nac, Montevideo, 2(5), 1924: 189 {L. olidus); 

Thompson, Proc. U. S. nat. Mus., ^O, 1916: 405 (numerous from Montevideo, two from Buenos Aires; 

in part L. olidus). 



Lycengraulis limnichthys Schultz 1949 
Figure SS 

Study Material. Type material of many specimens, 24-150 mm TL, 20-1 16 mm 
SL, from Lake Maracaibo, Venezuela, and the lower parts of its tributary streams. 

Distinctive Characters. This species is related to L. grossidens and L. olidus., the 
latter being extrallmital from southern Brazil to Argentina. L. limnichthys differs from 
grossidens in having a more slender and apparently less strongly compressed body, 
specimens of nearly equal size compared. In limnichthys the maxillary is a little shorter, 
scarcely reaching beyond the joint of the mandible and not to the margin of the opercle, 
whereas in adults oi grossidens it is longer, reaching well beyond the joint of the mandible 
and often to the margin of the opercle. In limnichthys the pelvic fins are inserted nearer 
to the base of the upper pectoral ray than to the anal origin, whereas in grossidens they 
are equidistant between these two points, or more usually nearer to the anal origin. And 
limnichthys differs from both grossidens and olidus in having fewer vertebrae. 

Description. Propertional dimensions in per cent of standard length, and counts, 
based on at least 1 1 study specimens. 

Body: depth 21—23, '" specimens 90 mm Pectoral fin: length 16—19. 

TL and upward. 

Head: length 23-25. Scales: ca. 35-45. 

6'«oa/: length 3.0-4.15. Gill rakers: 15—18 + 20—23 (rarely 24 

£71?: diameter 5.4—6.5. or 25). 

Pw/or^/Va/; distance 12. 5-15. Fin rays: dorsal 13—16; anal 23-25 

Maxillary: length 1 7.2-1 9.2. (rarely 26 or 27); pectoral 14 or 15 

Mandible: \&ngx\v 16. 3-18. 5. (rarely 13). 

Anal fin: length of base 21-23.5. Vertebrae: 41 or 42 (11 specimens). 

Body moderately compressed, its depth 4.35—4.75 in SL in specimens 73—1 16 mm 
SL. Head 4.0—4.35, its depth slightly exceeding its postorbital length. Snout blunt, 
extending about a third of its length beyond tip of mandible, 6.1—7.0 in head. Eye 
3.6—4.5. Postorbital length 6.65-8.0 in SL, 1.6— 1.8 in head. Maxillary scarcely 
pointed, extending somewhat beyond joint of mandible in the large examples but not 
to margin of opercle, 1.25— 1.35 in head. Mandible 1.27— 1.4. Cheek moderately short 



238 Memoir Sears Foundation for Marine Research 

and broad, generally about as long as snout and eye, its posterior angle about 30°. 
Gill rakers moderately slender, those at angle of first arch about as long as pupil, 
rather spinulose along inner margin, scarcely expanded at tip. 

Dorsal fin with longest rays not reaching tip of posterior ray if deflexed, its margin 
generally slightly concave, its origin usually about equidistant between posterior margin 
of eye and base of caudal. Anal with a definite lobe anteriorly, its base 4.25-4.75 in 
SL, its origin about under middle of dorsal base and equidistant between pectoral base 
and caudal base, or a little nearer to the former. Pelvic generally extending a little less 




Figure 55. Lycengraulis limnichthys, type specimen, from Rio Agua Caliente, above Maracaibo Basin, Vene- 
zuela, USNM 121751. Drawn by A. M. Awl. 



than halfway to anal, inserted somewhat nearer to pectoral base than to origin of anal, 
2.4—2.8 in head. Pectoral reaching to or a little beyond base of pelvic in small 
specimens, frequently failing to reach pelvic in large ones, 1.25— 1.4 in head. Axil- 
lary SCALE of pectoral variable in length, reaching somewhat beyond midlength of 
pectoral to beginning of its distal fourth, 1.4—2.25 in head. 

Color. Back pale, with a median dark streak, much more distinct in some speci- 
mens than in others. Sides with a broad silvery lateral band in adults, becoming diffused 
with the silvery color on the lower part of side in large examples, quite narrow in young. 
Dorsal and caudal fins with dusky punctulations, the margin of the latter much darker 
in some specimens than in others; other fins plain. Peritoneum silvery. 

Size. The largest example in the collection studied is 150 mm TL and i 16 mm 
SL, which may be near the maximum size attained. 

Development. The smallest, 1 9 mm SL, is slender and has all fins well developed. 
The caudal fin is as deeply forked as that in the adult, and the other fins have ac- 
quired approximately the same shape and relative position as those in adults. The 
snout remains shorter and projects less strongly beyond the mandible than in larger 
specimens; the maxillary, also much shorter and rounder, fails to reach the joint of the 
mandible; and the teeth in the jaws are scarcely evident. 



Fishes of the Western North Atlantic 239 

The juvenile color of this same specimen and of others of about equal size consists 
of a few dark dots over the nostrils, a group of similar dark dots at the occiput, and 
more or less definite rows on the back. Scattered dark points, present at and on the 
base of the caudal, extend prominently on the outer rays of the fin, which already has 
a dusky margin that remains throughout life. A series of prominent black spots along 
each side of the anal base continues in a single series on the peduncle behind the anal; 
a few scattered dark dots are present on the side of the head and on the isthmus. This 
small example, then, is definitely beyond the larval stage and is perhaps a young adult. 

The juvenile is slender, as is usual among clupeid fishes. In three specimens, 19, 
19.5, and 20 mm SL, the depth is respectively 9.0, 9.5, and 9.3 times in SL; in a 
group of six, ranging between 27-33 mm SL, the proportionate depth already has in- 
creased and is 8.5-6.5 in SL; and in an additional group of five, 32-50 mm SL, the 
depth is 6.5-5.45 in SL. Measurements of larger fish indicate that the body tends 
to become proportionately deeper as long as the fish grows. 

The teeth in the jaws are well developed but remain minute in specimens around 
25 mm SL. They do not become large (for an anchovy) and unequal in size (the chief 
distinguishing characters of Lycengraulis) until the fish attains about 65 mm SL. 

In fish 25 mm SL, the snout already projects well beyond the mandible, about 
as in the adult, and the maxillary is now somewhat pointed and reaches to a point 
approximately opposite the joint of the mandible. The axillary scale of the pectoral, 
not detected in the smaller specimens described, is now evident and reaches about to 
the midlength of the pectoral ; this axillary process, though rather variable in length, 
reaches far beyond the midlength of this fin in adults, generally to or beyond the be- 
ginning of its distal third. 

The juvenile color markings are retained about as described for the 1 9-mm fish 
until a standard length of at least 40 mm is attained. Even then the juvenile markings 
remain in large part. However, the sides of the head and the abdomen, in part, have 
become silvery. A very narrow indistinct silvery lateral band has become evident, and 
this increases rapidly in proportionate width, being prominent and as broad as the eye 
in specimens of about 75 mm SL; in examples of about 85 mm SL, the band is much 
broader than the eye, and its lower margin generally has become indefinite. In still 
larger specimens it usually has become diffused with the silvery color on the lower part 
of the side. The dark dots along the base of the anal persist longer than any of the other 
juvenile markings and are visible in half-grown fish; in fact, they are visible in some 
specimens up to 90 mm SL. 

Spawning. Several females with large eggs were taken in Lake Maracaibo on May 
16 (1942), indicating that at least some spawning takes place in the spring of the year. 

Range. Known only from Lake Maracaibo and from the lower sections of tributary 
streams, in fresh to slightly brackish water. 

Reference : 

Lycengraulis limnichthys Schultz, Proc. U. S. nat. Mus., gg, 1949: 51, fig- 8 (orig. descr.; type local. Rio Agua 
Caliente, above Lake Maracaibo, Venezuela; type USNM 121751). 



240 Memoir Sears Foundation for Marine Research 

Lycengrnulis batesii (Gunther) 1868 
Figure 56 

Study Material. A total of 13 specimens, 35-185 mm TL, the larger ones from 
Brazil, the smaller ones from Venezuela; also 4 juveniles, 21-31 mm TL, from 
Venezuela. 

Distinctive Characters. The slender body, the long narrow cheek, and the rather 
few and moderately long gill rakers distinguish this species. The statement in the 
original description, "Gill rakers very short, like tubercles," has been misleading, for 



--"-^ ^ .%^ 




^^^%. • 





^i^^ 



Figure 56. Lycengraulis batesii, i85mmTL, 155 mm SL, from TefFe, Brazil, MCZ 18006. Drawn by 
Alice C. Mullen. 



they are fully half of the length of eye, a point that was verified from type specimens 
by J. R. Norman. This statement became especially bothersome upon the discovery 
of other species that had notably shorter gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the Study Material. 



Anal fin: length of base 25—26.2. 
Pectoral fin: length 1 5.4-1 8.8, 

Scales: ca. 41—45. 

Gill rakers: 9-13 + 12-1 5. 

Fin rays: dorsal 13—15; anal 26—28; 

pectoral 13 or 14. 
Vertebrae: 47 (2 specimens). 

Body slender, compressed, its depth 5.25—6.0 in SL. Head 4.3—4.7 in SL, its 
depth about equal to its postorbital length. Snout projecting much less than half of its 
length beyond mandible, 7.0—8.0 in head. Eye 4.5—5.4. Postorbital part of head 
1.4— 1.6 in head. Maxillary slightly expanded, scarcely pointed, reaching nearly or 
quite to joint of mandible, 1.2— 1.3 in head. Mandible 1.27— 1.4. Cheek in large 



Body: depth 16. 5-19. 
Head: length 21-23. 
Snout: length 2.75-3.3. 
Eye: diameter 4.2—4.6. 
Postorbital: distance 14.3— 15.6. 
Maxillary: length 17.5— 18.2. 
Mandible: length 1 6.1 -18.2. 



Fishes of the Western North Atlantic 241 

examples much longer than snout and eye, proportionately shorter in small specimens, 
its posterior angle about 35°. Gill rakers at angle fully half of length of eye. 

Dorsal fin with the longest rays extending nearly or quite to tip of last ray if 
deflexed, its origin much nearer to base of caudal than to eye. Anal base 3.8—4.0 in 
SL, its origin slightly in advance of middle of dorsal base. Pelvic failing to reach 
halfway to anal in adults, extending farther back in young, inserted nearer to pectoral 
base than to anal origin. Pectoral scarcely reaching pelvic in adults, rather longer 
in young, i. 25-1. 5 in head. Axillary scale of pectoral about 67 "/o of length of fin, 
2.3—2.7 in head. 

Color. Old specimens, in alcohol, grayish above. Side of head and lower 66 "/o of 
body bright silvery. Small specimens with a broad silvery lateral band and a dark verte- 
bral band. The original description mentioned "an oblong blackish patch on lower half of 
the end of the tail," but such a blackish patch is not present on the specimens studied. 

Size. A total length of at least 185 mm (7.4 in.) is attained. 

Development. The juvenile, as in other herring and herring-like fishes, is more 
slender but rounder than the adult. All juveniles show externally the convoluted 
intestine, which apparently has not become fully covered by the abdominal wall. The 
snout is very short and blunt, and in the smallest specimen it scarcely projects beyond 
the tip of the mandible. The blunt maxillary extends little beyond the posterior margin 
of the eye. Both anal fin and vent move forward with age; the anal fin has its origin 
under the posterior rays of the dorsal in a 21 -mm specimen, about under the middle 
of the dorsal base in a 31-mm fish, and in advance of the middle of the dorsal, as in 
the adult, in a 35-mm fish. 

The smallest fish has a dark line on the isthmus and chest, dark dots along the 
base of the anal, and a single continuous dark median ventral line on the caudal peduncle. 
The dark line on the isthmus and chest disappears at a length of about 30 mm, but the 
markings at the base of the anal and on the peduncle remain until a length of at least 
52 mm is attained. The dark vertebral band, described for larger fish, first becomes 
partially evident at a length of about 32 mm. There is no indication of a silvery lateral 
band up to a length of 52 mm. 

Range and Habitat. This species is known from Rio Apure (a tributary of the 
Orinoco) at San Fernando de Apure, Venezuela; in Brazil it has been taken from 
the Para River, from Teffe on the Solimoes River, and from the Jutahy River, the last 
two streams being tributaries of the Amazon. 

Synonyms and References: 

Engraulis batesii Giinther, Cat. Fish. Brit. Mus., 7, 1868: 399 (orig. descr.; type local. Pari R., Brazil; 
types in BMNH). 

Lycengraulis batesii Jordan and Seale, Bull. Mus. comp. Zool. Harv., 6y, 1926: 385 (descr. not based on batesii 
throughout; 2 1 gill rakers on lower limb given whereas batesii has only 12-15; Rio Grande, TefFe, 
Jutahy [River] and Doce R., all in Brazil; Rio Grande specimen is olidus, the Doce R. specimen schroederi); 
von Ihering, Rev. Industr. Anim., Anno i (3), 1930: 233 (locals., diagn.); Hildebrand, Bull. Bingham 
oceanogr. Coll., 5(2), 1943: 148, fig. 67 (descr., range); Schultz, Proc. U. S. nat. Mus., gg, 1949: 51 
(ref.). 

t6 



242 Memoir Sears Foundation for Marine Research 

Negative Reference: 

Lycengraulis bateiii Starks, Stanf. Univ. Publ., Univ. Ser., 1913: 11 (specimen from market at Pari, Brazil, is 
L. barbouri Hildebrand). 



Lycengraulis abbotti (Fowler) 1 9 1 5 
Figure 57 

Study Material. The type and only specimen known, 180 mm long, from Port-of- 
Spain, Trinidad. 




Figure 57. Lycengraulis abbotti, type specimen, iSommTL, from Port-of-Spain, Trinidad, modified after 
Fowler, (1915- 523. fig- 0' ANSP 45079. Drawn by Alice C. Mullen. 

Distinctive Characters. This fish is related to L. barbouri, having similar short spiny 
gill rakers, but it differs in the number of gill rakers, the longer and more pointed 
maxillary, and the more anterior position of the dorsal fin. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the type, 180 mm TL. 



Body: depth 27. 
Head: length 23.5. 
Snout: length 3.37. 
Eye: diameter 4.95. 
Fostorbital: distance 14.8. 
Maxillary: length 21. 
Mandible: length 18.3. 



Anal fin: length of base 25.6. 
Pectoral fin: length 18.2. 

Scales: ca. 37. 

Gill rakers: 12 + 16 and 11 +17, in- 
cluding rudiments. 
Fin rays: dorsal 13; anal 25; pectoral 16. 



Body fairly robust, its depth 3.7 in SL. Head 4.25 in SL, its depth scarcely 
less than its length without snout. Snout projecting less than half of its length beyond 
mandible, 7.0 In head. Eye 4.8. Postorbital part of head 1.6 In head. Maxillary 
pointed, extending nearly to margin of opercle, 1.13 In head. Mandible 1.3. Cheek 
much longer than snout and eye. Its posterior angle about 30°. Gill rakers short, 
broad, very splnose, partly rudimentary, none exceeding half of diameter of pupil. 



Fishes of the Western North Atlantic 243 

Dorsal fin with the longest rays not quite reaching tip of last ray if deflexed, its 
origin equidistant between base of caudal and middle of eye. Anal origin slightly in 
advance of middle of dorsal base, its base 3.9 in SL. Pelvic reaching less than halfway 
to anal, inserted nearer to base of pectoral than to origin of anal. Pectoral not quite 
reaching base of pelvic, 1.3 in head. Axillary scale of pectoral fully 75% of fin, 
1.8 in head. 

Color. Grayish above. Sides silvery. No lateral band. Caudal with a narrow dark 
margin. 

Range. Known only from the type taken at Port-of-Spain, Trinidad. 

Synonyms and References: 

Anchovia abbotti Fowler, Proc. .'\cad. nat. Sci. Philad., 191 5 : 522, fig. i (orig. descr.; type local. Port-of-Spain, 

Trinidad; t)'pe ANSP 45079; cf. Stolephorus surinamensis Bleeker = Engraulis edentulus Cuvier). 
Anchoviella abbotti Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 50 (name, local.). 
Lycengrau/is abbotti Hildebrand, Bull. Bingham oceanogr. Coll., 8 {2), 1943: 150, fig. 68 {ci. L. barbouri 

Hildebrand and L. ichroederi Hildebrand). 



Lycengraulis barbouri Hildebrand 1 943 
Figure 58 

Study Material. Type material, 3 specimens 237, 230, and 205 mm long, from 
the Rio Poty, Brazil, and a fourth specimen, 220 mm long, from the fish market at 
Para; holotype and one paratype, MCZ 35277 and 18017, respectively; second paratype 
USNM 1 18997; and fourth specimen SU 22086. 

Distinctive Characters. This species differs from L. batesii in having a deeper body, 
a somewhat more posteriorly placed anal fin, and much shorter and broader gill rakers; 
however, the rakers are about equal in number in the two species. L. schroederi Hil- 
debrand, known only from Rio Doce in southern Brazil, is another related species; It 
has more numerous but equally short, broad, spiny, partly rudimentary gill rakers 
(17 or 18 + 21 or 22 on the first arch), a larger eye, about 4.9 In head, and a more 
posteriorly placed anal, its origin under the posterior third of dorsal base. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the four study specimens, 205—237 mm. 

Body: depth 21.2—23.2. Anal fin: length of base 24.3—26.2. 

Head: length 22-25.5. Pectoral fin: length 1 7-1 8.2. 

Snout: length 2.75-3.0. 

Eye: diameter 4.25—4.45. Scales: ca. 40-45. 

Postorbital: distance 14. 5- 17.7. Gill rakers: 9-1 1 + 12-14. 

Maxillary: length 18.2— 21. Fin rays: dorsal 14 or 15; anal 26—28; 

Mandible: length 17.5-20.8. pectoral 14 or 15. 

Body rather strongly compressed, its depth 4.3—4.7 in SL. Head 3.9-4.5 In SL, 
its depth about equal to its postorbital length. Snout projecting much less than half 



244 Memoir Sears Foundation for Marine Research 

of its length beyond mandible, 7.8-9.2 in head. Eye 5.0-5.8. Postorbital part of head 
1.37— 1.5 in head. Maxillary rounded distally, scarcely reaching joint of mandible, 
1.2 in head. Mandible 1.2 — 1.27. Cheek narrow, its length nearly twice snout and 
eye, its postorbital angle about 30°. Gill rakers short, flat, the longest not more than 
half of length of pupil. 

Dorsal fin with the longest rays extending to, or a little beyond, tip of last ray if 
deflexed, its origin about equidistant between base of caudal and upper anterior angle 
of gill openings. Anal with a rather prominent lobe anteriorly, its origin under, or some- 




FiGURE 58. Lycengraulis barbouri, type specimen, 290 mm TL, 237 mm SL, from Rio Poty, Brazil, MCZ 
35277. Drawn by Alice C. Mullen. 

what in advance of, middle of dorsal base, its base 3.8—4. i in SL. Pelvic reaching much 
less than halfway to anal, inserted somewhat nearer to base of pectoral than to origin 
of anal. Pectoral reaching to, or slightly beyond, base of pelvic, i .3— i .4 in head. Axil- 
lary scale of pectoral about 75 "/o of length of fin, 1.95— 2.1 in head. 

Color. Old specimens, in alcohol, grayish above. Lower 75 "/o of body and 
head plain silvery. Fins unmarked, except for an indication of a black margin on 
caudal. 

Range and Habitat. Known only from northern Brazil, from the Rio Poty, a trib- 
utary of the Rio Parnahyba, and from the Belem (Para) market. [That this species was 
obtained in the Belem market, far removed from Rio Poty, seems to indicate that the 
species may well be marine. — y. h.o.] 



Synonyms and References: 

Lycengraulis batesii Starks (not of Giinther), Stanf. Univ. Publ., Univ. Ser., 191 3: 11 (specimen from Pard, 

Brazil; found to be L. barbouri). 
Lycengraulis grossidens Jordan and Scale (in part not of Cuvier), Bull. Mus. comp. Zool. Harv., ty, 1926: 383 

("Rio Puty"= Rio Poty, Brazil; specimens are L. barbouri). 
Lytengraulis barbouri Hildebrand, Bull. Bingham oceanogr. Coll., 8 {2), 1943: 151, fig. 69 (orig. descr. ; type 

local. Rio Poty, tributary of Rio Parnahyba, Brazil; type MCZ 35277; cf. L. batesii). 



Fishes of the Western North Atlantic 245 

Genus CetengrauUs Giinther 1868 

Cetengraulis Giinther, Cat. Fish. Brit. Mus., 7, 1868: 383; genotype Engraulis cdentuha Cuvicr, first designated 
by Jordan and Evermann, Bull. U. S.Nat. Mus., 47 (i), i8g6: 450. 

Characters. Body rather deep in adults but much more slender in young, quite 
strongly compressed. Maxillary short, distally rounded or very bluntly pointed, 
usually extending about to articulation of mandible. Teeth in jaws all very small (ap- 
parently not disappearing with age as has been stated). Gill rakers long, close-set, 
numerous, increasing with age (about 25 on lower limb in small specimens to about 60 in 
large ones). Pelvic inserted under, or a little in advance of, dorsal origin. Silvery lateral 
band present only in specimens about 100 mm TL or less. 

Remarks. Gill covers connected across the isthmus by a thin transparent membrane 
(easily torn); narrow in very young. 

Range. The genus is represented on the Atlantic side by two species, ranging 
from the West Indies to southern Brazil, and on the Pacific side by one species, C. mysti- 
cetus, known from the Gulf of California to Peru. 

Key to Atlantic Species 

I a. Depth of body 3.0-3.4 in length; head 4.6-5.3; eye 3.7-4.6 in head; pectoral 
failing to reach base of pelvic. edentulus (Cuvier) 1829, below. 

I b. Depth of body 5.0 in length; head ■^.^\ eye ^.^ in head; pectoral reaching a little 
beyond base of pelvic. juruensis Boulenger 1898, p. 247. 

Cetengraulis edentulus (Cuvier) 1829 

Bocon (Cuba) 

Figure 59 

Study Material. About 70 specimens, mostly small, 35-160 mm TL; from Puerto 
Rico; Jamaica; Cuba; and New Grenada; Colon, Panama; Gulf of Venezuela; Rio de 
Janeiro, "Barro Santas," "Sambaia," and Cachoeira, Brazil. 

Distinctive Characters. See Key to Atlantic Species, above. 

Description. Proportional dimensions in per cent of standard length, based on 
20 specimens, 70-160 mm, and most of the counts on many more. 

Body: depth 25-34.4. Scales: ca. 40-43. 

//e<?^: length 30-34.5. Gill rakers: ca. 30 + 36 in specimens 

6'wo«/: length 3.8-4.75. around 65 mm, about 52 + 55 in 

Eye: diameter 6.5-8.0. specimens about 130 mm TL. 

Postorbital: distance 18. 2-21. 2. Fin rays: dorsal 14-16; anal 23-26; 

Maxillary: length 1 7.2-1 9. pectoral 14 or 15. 

Mandible: length 1 7.3-1 8.5. Vertebrae: 42 (3 specimens). 

Anal fin: length of base 21.5-23. 

Pectoral fin: length 1 3.3-1 6.7. 



246 



Memoir Sears Foundation for Marine Research 



Body strongly compressed, the ventral outline much more strongly convex than 
the dorsal outline, the depth increasing with age, 2.9-3.3 •" SL in specimens 75 mm 
TL and upward. Head 3.0—3.5 in SL. Snout quite pointed, extending nearly its 
full length beyond mandible, 5.5-8.0 in head. Eye 3.7—4.8. Postorbital part of 
head i .55-1 -75 in head. Maxillary bluntly pointed or rounded distally, i. 72-1. 8 
in head. Mandible i. 67-1. 73. Cheek nearly or quite as long as snout and eye, its 
posterior angle narrow, about 30°. Gill rakers long, slender, close-set, increasing 
greatly in number with age. 




Figure 59. Cetengraulis edentulus, 120 mm TL, 100 mm SL, from Havana, Cuba, USNM 35159- Drawn 
by Louella E. Cable. 

Dorsal fin with the longest rays failing to reach tip of last ray if deflexed, its 
origin generally about equidistant between anterior margin of eye and caudal base. 
Anal with origin usually under last fourth of dorsal base, its base 4.3—4.6 in SL. 
Pelvic inserted nearer to anal origin than to pectoral base, generally reaching fully 
halfway to origin of anal. Pectoral usually failing to reach base of pelvic, 1.9—2.3 
in head. Axillary scale of pectoral broad, with a somewhat thickened ventral edge, 
3.25—3.7 in head. 

Color. In alcohol, bluish gray above; lower 75''/o of side silvery. Small specimens 
with a silvery lateral band, becoming wider and less distinct with age, generally dis- 
appearing at a length of about 100 mm. 

Size. Specimens as much as 165 mm {G.d in.) TL have been reported. 

Range and Habitat. Reported for the West Indies and the Atlantic coast of 
Panama to southern Brazil. It ascends freshwater streams. 

Synonyms and References: 

Engraulis edentulus Cuvier, R^gne Anim., ed. 2, 2, 1829: 323 (orig. descr.; type local. Jamaica); Cuvier and 

Valenciennes, Hist. Nat. Poiss., 2r, 1848: 51 (descr., abund., distr.). 
Cetengraulis edentulus Giinther, Cat. Fish. Brit. Mus., 7, 1868: 383 (descr., range) ; Jordan and Evermann, 

Bull. U. S. nat. Mus., 47(1), 1896: 450 (descr., range, synon.); Meek and Hildebrand, Field Mus. 



Fishes of the Western North Atlantic 247 

Publ.,Zool., /5(i), 1923: 214 (descr., range; Stolephorus garmani Evermann and Marsh and S.gilberti 
Evermann and Marsh put in synon.); Jordan and Scale, Bull. Mus. comp. Zool. Harv., 6j, 1926: 414 
(descr., Cuba and Brazil); Beebe and Tee-Van, Zoologica, N. Y., 10 (i), 1928: 49, fig. (descr., Port- 
au-Prince, Haiti); Nichols, N. Y. Acad. Sci., 10 (2), 1929: 206, fig. (synon., diagn., Puerto Rico); Jor- 
dan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 51 (range, synon.); Fowler, Proc. 
Acad. nat. Sci. Philad., 80, 1931: 393 (Trinidad, B.W.I.); Beebe and HoUister, Zoologica, N. Y., jp 
(6), 1935: 211 (Union I., B.W.I.); Fowler, Arqu. Zool. estad. Sao Paulo, j (6), 1941: 135 (Brazilian 
recs.); Hildebrand, Bull. Bingham oceanogr. Coll., 5 (2), 1943: 155, fig. 71 (descr., synon., range); 
Schultz, Proc. U. S. nat. Mus., gg, 1949: 54 (refs.). 

Engraulis brevis Poey, Repert. Fisico-Nat. Cuba, I, 1866: 379 (orig. descr.; type local. Cuba; cf. E.edentulus 
Cuvier); Howell-Rivero, Bull. Mus. comp. Zool. Harv., 82, 1838: 172 (holotype, MCZ 24296, Poey's 
No. 716). 

Stolephorus robertsi Jordan and Rutter, Proc. .'\cad. nat. Sci. Philad., .^9, 1897: 95 (orig. descr.; type local. 
Jamaica; type SU 4853; cf. Stolephorus opercularis Jordan and Gilbert, a Pacific coast Cetengraulis); 
Jordan and Evermann, Bull. U. S. nat. Mus., 47 (3), 1898: 2815 (descr. after Jordan and Rutter). 

Stolephorus gilberti Evermann and Marsh, Rep. U. S. Comm. Fish. (1899), 1900: 352 (orig. descr.; type local. 
Palo Seco, Puerto Rico; type USNM 49359; cf. Engraulis productus Poey = Engraulis clupeoides Swain- 
son and Stolephorus garmani E. and M.= Engraulis edentulus Cuvier); and Bull. U.S. Fish Comm. 
(1900), 20(1), 1902: 90, fig. I 5 (orig. descr.); Jordan and Evermann, Bull. U. S. nat. Mus., 47 (4), 
1900: 3146 (descr. after Evermann and Marsh). 

Stolephorus garmani Evermann and Marsh, Rep. U. S. Comm. Fish. (1899), 1900: 352 (orig. descr.; type local. 
Puerto Real, Puerto Rico; type USNM 49360; cf. Engraulis productus Poey = Engraulis clupeoides 
Swainson, and Stolephorus gilberti E. and M.= Engraulis edentulus Cuvier); and Bull. U. S. Fish Comm. 
(1900), 20(1), 1902: 89, fig. 14 (orig. descr.); Jordan and Evermann, Bull. U. S. nat. Mus., 47(4), 
1900: 3146 (descr. after Evermann and Marsh). 

Anchoviella robertsi Jordan, Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 2, 1930: 49 (range, synon.). 

Cetengraulis gilberti ]oxdL3.n, Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 2, 1930: 51 (range, synon.); 
Howell-Rivero, Proc. Boston Soc. nat. Hist., ^l (4), 1936: 55 (Cuba). 

Cetengraulis garmani Jordan, Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 2, 1930 : 5 1 (range, synon.). 

Doubtful References: 

Stolephorus surinamensis Bleeker, Tijd. Dierk. Amst. (1865) J, 1866: 178 (orig. descr. of 96-mm specimen; 

type local. Surinam; nothing in descr. inconsistent with C. edentulus of similar size); Eigenmann, 

Mem.Carneg. Mus., J, 1912:448 (descr. of 27- and 37-mm specimens from Bartica Rocks, British Guiana); 

Puyo, Bull. Soc. Hist. nat. Toulouse, 80, 1945: 103, fig. 2 (descr., French Guiana); Puyo, Faune Emp. 

Fran?., 12, Poiss. Guyane Frang., 1949: 155, fig. 78 (descr., French Guiana). 
Stolephorus manjuba Miranda-Ribeiro, Kosmos, Rio de J., Feb., 1908, 5 — no pagination (orig. descr. inadequate; 

type local., Rio de Janeiro, Brazil). 
Anchovia gilberti Fowler, Proc. Acad. nat. Sci. Philad., 1916: 401 (Colon, Panama, without comment). 
Anchoviella surinamensis Fowler, Proc. Acad. nat. Sci. Philad., 80, 193 1 : 392 (three specimens, 34-36 mm 

long, from Icacos Beach, Trinidad, B.W.I. ; notes suggest C. edentulus). 



Cetengraulis juruensis Boulenger 1898 

Study Material. None. 

Distinctive Characters. See Key to Atlantic Species, p. 245. 

Description. Based on original description and figure, which is all that is known. 
Body depth 20 "/(, of SL. Head 28.5 in SL. Snout projecting strongly (extending 
nearly its full length beyond mandible, according to a published figure). Eye c,.c^ in 
head. Maxillary extending to joint of mandible. Gill rakers long, finely denticulate, 
about 40 on lower limb. 



248 Memoir Sears Foundation for Marine Research 

Dorsal origin equidistant between tip of snout and caudal base; 13 rays. Anal 
origin under posterior rays of dorsal; rays 23. Pelvic fin inserted somewhat nearer to 
base of pectoral than to origin of anal. Pectoral extending a little beyond base of 
pelvic, its length a little more than half of length of head. Axillary scale of pectoral 
extending somewhat beyond midlength of fin (according to figure). Scales 38. 

Color. Olive above, silvery on sides and below. Snout blackish above. Fins pale 
orange; caudal rays blackish at tips. 

Range and Habitat. Known only from the holotype, a specimen 140 mm TL, from 
fresh water, from the Jurua River, a tributary of the Amazon in Brazil. 

References : 

Cetengraulis juruensis Boulenger, Trans, zool. Soc. London, 14 {10), 1898: 427, pi. 41, fig- 3 (orig. descr.; 
type local. Rio Jurua, Brazil; cf. C edentulus)\ Eigenmann, Rep. Princeton Exped. Patagonia, 1896- 
1899, J (4), 19 10: 452 (refs. to orig. descr., habitat); Campos, Arqu. Zool. estad. Sao Paulo, J (7), 1941 : 
202, fig. 15 (descr., range) ; Hildebrand, Bull. Bingham oceanogr. Coll., 8(2), 1943: 158 (descr. after 
Boulenger). 



TEXT REFERENCES 



1. JTowLER, Proc. Acad. nat. Sci. Philad., 82, 11. LyiNTON, Bull. U. S. Fish Comm. (1899), ig, 

1930. 1901. 

2. Fowler, Proc. Acad. nat. Sci. Philad., 83, 12. Linton, Bull. U. S. Bur. Fish. (1904), 2./, 1905. 

1931- 

jj. JVIyers, Proc. Calif. Acad. Sci., (4) 2J, 1940. 
J. irliLDEBRAND, BuU. Bingham oceanogr. Coll., 

* (2), 1943- XT 

4. Hildebrand and Cable, Bull. U. S. Bur. Fish., 14- I^akashima (Shoji), Bol. Mus. Hist. nat. 

46, 1930. 'Javier Prado,'5(i6), 1941. 

5. Hildebrand and Schroeder, Bull. U. S. Bur. ^5- Nichols and Breder, Zoologica, N. Y., 9 (i), 

Fish. (1927), ^j(i), 1928. 1927- 



6. Jordan and Seale, Copeia, 1925. 

7. Jordan and Seale, Bull. Mus. comp. Zool. Harv., 

6y (11), 1926. 



16. "earson, Indiana Univ. Stud., JJ (64), 1924. 
77. Poey, Synop. Pise. Cubensis, 1868. 
18. Poey in Gundloch, An. Soc. asp. Hist, nat., 
10, 1881. 



8. J^ENDALL, Bull. Boston Soc. nat. Hist., §8, 

1931. 

9. Kuntz, Bull. U. S. Bur. Fish., jj, 1914. 

JO. Kuntz and RadclifFe, BuU. U. S. Bur. Fish. 
(1915-16), J5, 1918. 



ig. bcHULTz, Proc. U. S. nat. Mus., gg, 1949. 

20. Swain, Bull. U. S. Fish Comm., 2, 1883. 

21. Vladvkov, Proc. N. S. Inst. Sci., ig, 1935. 



249 



Interim Account of 
Family Alepocephalidae 



HENRY B. BIGELOW 

Museum of Comparative Zoology 
Harvard University 



Characters. Essentially as in the Clupeidae, except with the Dorsal fin consider- 
ably posterior to the midlength of body, its base partly or entirely over anal fin and 
preceded in a few {Anomalopterus) by a fleshy fold or ridge. Lateral line present in 
most (described as indistinct or lacking in Xenodermichthys, Leptoderma, and Anomal- 
opterichthys). Swim bladder lacking. Color ranging between dark blue, purple, and 
black, below as well as above; transparent but never silvery. Barbels lacking on chin 
and throat. Bony gular plate absent in throat region. External tubular papilla 
absent on shoulders (present in Searsiidae, p. 254). Abdomen rounded, its midline 
without keeled scales (scutes). Snout less than twice the length of mouth in most, 
not tubular, but tubular and more than twice the mouth in a few {Aulostomatophord). 
Mouth ranging from very small, at tip of tubular snout (see above), to a little more 
than half as long as head [Anomalopterus and Bathytroctes'). Pectoral fins very small; 
rudimentary in a few (Mirognathus). Pelvic fins present, at least in most. Teeth small, 
feeble, present on premaxillaries in most (upper jaw toothless in Anomalopterichthys, 
Leptochilichthys, and Asquamiceps), and on maxillaries, palatines, and mandible in some. 
Scales present or absent; if present, thin, rounded. Luminescent organs (photophores) 
lacking in most but present in a few; situated on nodules of raised skin if present. 
Vent posterior to midlength of trunk in most (anterior to midlength of trunk in Lep- 
todermd). Gill membranes free from isthmus and overlapping. Opercular system 
thin but complete. Gill rakers long, numerous. Branchiostegal rays 5 or more. 
Pyloric caeca 2—21 reported (rudimentary in some). 

Genera. Parr, in his recent synopsis of the family, recognized 22 genera, six of 
which were described by him as new (^: i — 14); to these he added two more new genera 
a year later (5: 263, 265). 

250 



Fishes of the IVestern North Atlantic 



251 



Remarks. The Alepocephalidae differ externally from the true herrings (Clupeidae) 
in their very soft texture, in their uniformly dark, dull coloration (below as well as above), 
and in the position far rearward of their dorsal fin. They differ internally from the her- 
rings in lacking a swim bladder. 




Figure 60. a. Alepocefhaltis productus, from Albatross St. 2030, off New Jersey, 39°3o'N, 7i°43'W, in 588 
fms.; after Goode and Bean. b. Alepocephalus bairdii, from Grand Banks off Newfoundland, at 200 fms.; 
after Goode and Bean. c. Alepocephalus agassizii, from off New Jersey; after Goode and Bean. d. Xenodermich- 
thys copei, from off Chesapeake Bay; after Goode and Bean. e. Aulostomatophora fhosphorops, from the Arabian 
Sea; after Goode and Bean, based on Alcock. 



The members of the family as a whole are creatures ot the deep sea, seldom taken 
in depths less than 250 fms. or so; a cursory examination of the literature revealed 
only two records for shallower depths — 200 fms. for the Grand Banks off New- 
foundland for Alepocephalus bairdii (j: 55) and 180 fms. north of Celebes for Bathy- 
troctes macrolepis (j: 225). The alepocephalids can be described as "universal" in all 
oceans, down the lower part of the slopes and on the ocean floors at depths to at 
least 3000 fms. (5490 m) or so, and very likely deeper still. 



252 Memoir Sears Foundation for Marine Research 



Depth 
(fms.) 

50-250 
251-300 
301-350 
351-400 
401-450 
451-500 
501-550 
551-600 
601-650 
651-700 
701-750 



Table I. 



Total 
Hauls 



Hauls from off Southern Nova Scotia to Virginia 
Containing Alepocephalids 



76 
18 
22 
15 



Hauls with 


"/, of Hauls with 


Av. 


no. Specimens 


Jepocephalids 


Alepocephalids 


in Hauls with 
Alepocephalids 













I 


6 




I 


4 


18 




16 


6 


40 




7 


8 


38 




8 


13 


62 




19 


8 


100 




12 


4 


100 




6 


4 


100 




13 


2 


100 




8 


2 


100 




9 



During the experimental cruises that were carried out from the Woods Hole 
Oceanographic Institution by Mr. William C. Schroeder on the dragger Capn. Bill ii 
in 1 952-1 953 along the continental slope of North America between the offings of 
southern Nova Scotia and Virginia, the shallowest haul that brought up an alepoce- 
phalid was at 251—350 fms.' Alepocephalids, however, of one kind or another were 
taken in more than a third of the hauls from deeper than 351—400 fms. and in all 
of the hauls from deeper than 500 fms., as shown in Table i, contributed by Mr. 
Schroeder. 

A detailed account of the family is unavoidably postponed until a subsequent 
volume. The general appearance of typical members is represented in the accompanying 
outline drawings (Fig. 60, 6 1 a). 



I. A depth of "68" fms. given by Goode and Bean for Concara macroptera should read 687 fms. (2: 40); see 
Townsend (6: 403). 



TEXT AND FOOTNOTE REFERENCES 



1. GooDE and Bean, Proc. U. S. nat. Mus., 2, 4. Parr, Amer. Mus. Novit., 1531, 1951- 

iSyg. 5. Parr, Bull. Mus. comp. Zool. Harv., 707(4), 

2. Goode and Bean, Smithson. Contrib. KnowL, 1952. 

30, 1895. 
J. Giinther, Rep. sci. Res. 'Challenger,' Zool., 

22(57), 1887. 6. 1 owNSEND, Rep. U. S. Comm. Fish., 1901. 



253 



Interim Account of 
Family Searsiidae 



HENRY B. BIGELOW 

Museum of Comparative Zoology 
Harvard University 



Characters. Essentially as in the Alepocephalidae, except that there is a volu- 
minous sac extending forward underneath the skin of each shoulder and opening 
rearward through a Tubular papilla so conspicuous that it affords a convenient and 
reliable field mark by which a member of this family can be recognized. This shoulder 
apparatus appears to be a secretory organ, probably producing luminous mucus. 
Pelvic fins are wanting in a few {Platytroctes). 

Remarks. The type genus Searsia was described by Parr in 1937 (1:12), but it 
was not until 1951 (2:2) that he proposed and defined the family Searsiidae to include 
the genera Searsia Parr 1937, Platytroctes Giinther 1878, Platytroctegen Lloyd 1909, 
Holtbyrnia Parr 1937, Barbantus Parr 1951, Normichthys Parr 1951, Pellisolus Parr 
1 95 1, and Persparsia Parr 1956. [As this paper goes to press, Parr's recent extensive 
and complete monograph of the known Searsiidae is at hand (j). To the genera already 
listed above, he has added Mirorictus Parr 1947, Sagamkhthys Parr 1953, Maulisia Parr 
i960, and Mentodus Parr 1951, with Mirorictus in the subfamily Mirorictinae, Platy- 
troctes and Platytroctegen in the subfamily Platytroctlnae, and all the other genera in the 
subfamily Searsinae. — y.h.o.] 

The Searsiidae, like the Alepocephalidae (p. 250), are deep-sea fish, thus far re- 
ported in the western North Atlantic for the offing of Delaware Bay; the Caribbean; 
the offing of Pernambuco, equatorial Brazil; and the vicinity of Bermuda; also for 
the middle and eastern Atlantic off northwestern Africa, Cape Verde, and the Gulf of 
Gascony; the offing of southern Africa in the eastern South Atlantic; the Bay of 
Bengal; and the eastern equatorial Pacific off the Bay of Panama. The shallowest 
haul from which any searsiid has yet been reported was from 500 fms., in the Bay 
of Bengal. 

254 



Fishes of the Western North Atlantic 



255 



The detailed account of the Searsiidae, like that of the Alepocephalidae (p. 250), 
is unavoidably postponed until a subsequent volume. The general appearance of a 
member of the Alepocephalidae (Fig. 6 1 a) and two of the included genera in 
Searsiidae (Fig. 61 b, c) are shown in the accompanying outline drawings. 




Figure 61. a. Leptoderma macrops (Alepocephalidae), from off northwestern Africa; after Goode and Bean, 
based on V'aiUant. b. Searsia koefoedi, from off the Bahamas; after Parr. c. Platytroctegen minis, from the Bay 
of Bengal; after Lloyd. 



TEXT REFERENCES 



I. Xarr, Bull. Bingham oceanogr. Coll., J (7), 1937. 2. Parr, Araer. Mus. Novit., 1531, 1951. 

J. Parr, 'Dana' Rep., 5J, i960. 



256 



Family Clupeidae' 



SAMUEL F. HILDEBRAND2 

excepting 

Genus Harengula by Luis R. Rivas 
Genus Dorosoma by Robert R. Miller 



Characters. Body usually slender in most but short and deep in a few species; 
slightly to excessively compressed in most but almost round in some. Scales covering 
body; cycloid, serrate, or pectinate; head naked. Chest and abdomen compressed into 
a sharp edge, armed with Bony scutes, except in the Dussumierinae (see i a in Key to 
Genera). Lateral line wanting. Mouth moderately large, terminal, superior or infe- 
rior. Teeth in jaws small, or canine-like in Chirocentrlnae (see 1 1 b in Key to Genera), 
sometimes wanting in large adults of Alosa. Gill rakers usually long, slender, nu- 
merous, increasing in number with age and growth in some species, in moderate 
numbers in others. Gill membranes not united, free from the isthmus. Pseudobran- 
chiae present. Dorsal fin, if present, usually about at midlength of the standard 
length, but sometimes posterior (present in American species). Adipose fin absent. 
Caudal fin forked. Anal moderately long to very long. Pelvic fin moderately large, 
small, or absent. Pectoral fin well developed. Vertebrae in species of the western 
North Atlantic about 39-59. 

Remarks. This large family, composed of about 70 genera and perhaps 150 or 
more species, of which 16 genera and 36 species come within the scope of this work, 
is economically one of the most important groups of fishes in the world. Many of the 
species occur in great schools, making it easy for either man or natural predators to 

1. Edited, with some revision and expansion, by George S. Myers, Henry B. Bigelow, Mywanwy M. Dick, and Yngve 
H. Olsen. 

Some of the subfamilies recognized herein, pai'ticularly the Dussumierinae and Dorosomatinae, have been treated 
as separate families in some recent works. Nevertheless their affinities are with the Clupeidae. For practical purposes, 
therefore, the characters of the subfamilies have been sufficiently indicated in the Key to Genera (p. 259) for the 
major subdivisions to be recognized without further treatment elsewhere. 

2. August 15, 1883 - March 16, 1949. 

>7 
257 



258 Memoir Sears Foundation for Marine Research 

catch them in large quantities; and the supply of some, such as those of the genus 
Clupea, seems to be almost inexhaustible. Fresh herring are used extensively as food 
by man in nearly all parts of the world, and in many areas large quantities are eaten 
after they have been salted and smoked or canned. Great quantities, especially of 
Brevoortia, are reduced to oil, fish meal (feed for livestock and poultry), and fertilizer. 
The smaller species and the young of the larger ones serve still another useful purpose 
in providing extensive forage for larger predatory fishes, many of which enter into 
the diet of man. Aquatic mammals, birds, and turtles also feed on them. All this 
food is provided for man and beast at virtually no cost, as the herring feed almost 
entirely on small organisms that are not suitable food for man or for the other preda- 
tors named. 

Although the members of the family Clupeidae have no lateral line, nearly all 
of the species described in this work have numerous mucous pores and canals about 
the head and shoulders that may in part compensate for the absence of a lateral line. 
An air bladder with a very thin wall is present in all the species of the genera examined, 
namely Clupea, Alosa, Pomolobus, Brevoortia, Opisthonema, Harengula, Sardinella, and 
Etrumeus, and it may be assumed that the other members of the family also have an 
air bladder. Its walls are so thin and delicate that dissections must be made with care 
to find it undamaged. 

Range. Herring (Clupeidae) occur in all seas of the world except those of the Arc- 
tic and Antarctic regions. Many species ascend freshwater streams to spawn, and 
others live in fresh water permanently. 

Preparation of Descriptions and Location of Study Material. Nearly all Descriptions 
were prepared directly from specimens, but a few are included for which no speci- 
mens were available. In such instances it is stated in the accounts that they were 
compiled or quoted from publications. The specimens used are in the USNM unless 
otherwise stated. Freshwater species occurring in the rivers draining into the western 
North Atlantic are not numerous and have been included for the sake of complete- 
ness ; it is possible that all of these forms occasionally enter brackish or salt water. 

Measurements and Counts. The proportions used in Description are based on measure- 
ments made with vernier calipers having sharp straight points. Many of the measure- 
ments of small fish (sometimes even of larger ones) and of the small structures such 
as eye, snout, and maxillary were made under magnification. See also Measurements 
and Counts under Brevoortia, p. 343. 

Total length : distance between margin of snout or tip of lower jaw (whichever 
is longer), more or less along the body axis to a vertical at tip of longest lobe of 
caudal fin. 

Standard length: distance between anterior margin of snout and base of caudal. 

Body depth: greatest depth unless otherwise stated; this may occur over base of 
pectoral or as far back as origin of dorsal fin. 

Head: if used without a modifier, the distance between anterior margin of snout 
and most distal point of bony (not membranous) margin of opercle. 



Fishes of the Western North Atlantic 259 

Head depth: distance from a slight crossgroove at the occiput, between the 
naked head and scaly body, to margin of first ventral scute; this line may be vertical 
or may slope somewhat backward from the occipital crossgroove. 

Snout: distance from anterior margin of snout to anterior margin of orbit. 

Eye: horizontal diameter of eyeball, not of orbit. 

Maxillary: distance from anterior rim of snout to maxillary 's posterior margin. 

Mandible: distance from its tip to its joint. 

Interorbital: width of bone over or between middle of eyes. 

Cheek: the bone, usually with a glossy surface in the herrings, lying below and 
usually somewhat behind the eye, its depth measured from lower rim of orbit to its 
lowest or deepest point. 

Predorsal length: distance between anterior margin of snout and origin of 
dorsal. 

Caudal lobes: distance from midbase of this fin to tip of lobe. 

Pelvic and Pectoral fins: distance from base of outer or upper ray to tip of fin. 

Fin ray count: includes all rays, whether rudimentary, simple, or branched; 
the last ray, even though nearly double, counted as one. 

Scales: range in scale numbers, e.g. 56-62, obtained by counting the oblique 
rows (upward and slightly forward) that cross the side between upper anterior angle 
of gill opening and base of caudal; any deviation from that procedure is stated. 

Gill rakers: number present on first gill arch, including rudiments if present; 
all enumerations based on adult fish unless stated otherwise. 

Ventral scutes: three sets of counts are given: the range of the total number 
present, the number in front of the pelvic fins, and the number behind these fins; the 
scute ending nearest to the insertion of the pelvics is included in the number in 
advance of these fins; the rest compose the number behind. 

Vertebrae: total number, including the hypural. 

Key to Western North Atlantic Subfamilies and Genera of Clupeidae^ 

I a. Body little compressed; abdomen rounded, covered with ordinary scales, without 
bony scutes. Subfamily Dussumierinae. 

Atlantic coast of North America; 
eastern and western Pacific; south- 
eastern coasts of Africa and Australia. 
2a. Pelvic fin much smaller than pectoral, inserted behind base of dorsal; dorsal 
fin with about 17-20 rays; chest round. Etrumeus Bleeker 1853, p. 262. 
2b. Pelvic fin nearly as large as pectoral, inserted under base of dorsal; dorsal 
fin with only about 9-14 rays; chest slightly keeled. 

Jenkinsia Jordan and Evermann 1896, p. 267. 

3. This Key is drawn up simply for the ready identification of representatives of the subfamihes and genera that are 
known from the western North Atlantic without reference to those outside of this range. Therefore, some characters 
are used that apply only to the representatives described in this work and for that reason are not of generic value. 

I7» 



2 6o Memoir Sears Foundation for Marine Research 

lb. Body moderately to very strongly compressed; chest and abdomen compressed 
into a sharp keel armed with bony scutes. 

3 a. Median line of back in front of dorsal fin with ordinary or modified scales, 
thus without a prominent median naked strip; stomach not gizzard-like, but 
wall somewhat thickened in some species. 

4a. Pelvic fin well developed, inserted under base of dorsal, with 7—9 rays; 
mouth gently oblique, the lower jaw usually included but occasionally 
projecting; body moderately compressed. Subfamily Clupeinae. 

Virtually aU temperate and tropical 
waters of the world, with some species 
entering frigid zones; inhabiting salt, 
brackish, and fresh water. 
5 a. Vertical edge on rim of shoulder girdle (cleithrum) underneath free 
edge of gill cover (opercle) smooth or even, without dermal lobes. 
6 a. Back in front of dorsal fin with ordinary scales like those on 
rest of body; pelvic with 8 or 9 rays; pectoral fin without a 
free axillary process; scales not overlapping deeply, their ex- 
posed part not much deeper than long, their margin slightly 
irregular, not definitely serrate or pectinate. 
7 a. Vomer with teeth; greatest depth of body about 4.0—5.0 
in SL; ventral scutes weak, numerous, total number about 
39—46. Clupea Linnaeus 1758, p. 274. 

7b. Vomer toothless; depth of body generally less than 4.0 in 
SL; ventral scutes strong, total number about 27—38. 
8 a. Upper margin of lower jaw within mouth rising gently; 
teeth missing in examples 200 mm and upward in 
length; cheek deeper than long (Fig. 69). 

Alosa Link 1790, p. 293. 

8 b. Upper margin of lower jaw within mouth rising 

abruptly; teeth present, at least in lower jaw, at all 

ages; cheek not deeper than long, definitely longer than 

deep in some species (Fig. 76). 

Pomolobus Rafinesque 1820, p. 312. 

6 b. Back in front of dorsal fin with a row of enlarged modified 

scales on each side of median line; pelvic with 7 rays; pectoral 

fin with a free axillary process; scales overlapping deeply, their 

exposed part much deeper than long, their margin strongly 

serrate to pectinate (Fig. 85). Brevoortia Gill 1861, p. 342. 

5 b. Vertical edge on rim of shoulder girdle (cleithrum) underneath free 

edge of gill cover (opercle) with two dermal lobes (Fig. 100). 

9 a. Last ray of dorsal greatly produced, filamentous, often reaching 

nearly or quite to base of caudal (Fig. 94); anal with 20-25 

rays. Opisthonema Gill 1861, p. 380. 



Fishes of the Western North Atlantic 261 

9b. Last ray of dorsal little if any longer than the preceding ones; 
anal with about 15—20 rays. 

10 a. Last two rays of anal not enlarged; gill rakers about 25— 
40 on lower limb; scales about 34—41 oblique series cross- 
ing middle of side; pelvic with 8 rays. 

Harengula Cuvier and Valenciennes 1847, P- 3^6. 

lob. Last two rays of anal enlarged, finlet-like (Fig. 100); gill 

rakers about 70—130 on lower limb; scales about 41—47 

oblique series crossing middle of side; pelvics with 9 rays. 

Sardinella Cuvier and Valenciennes 1847, P- 397- 

4b. Pelvic fin small or wanting, inserted in advance of dorsal fin if present; 

mouth strongly oblique to nearly vertical, the lower jaw projecting; body 

strongly to excessively compressed. 

1 1 a. Teeth in jaws all small, without canines; vomer toothless. 

Subfamily Pristigasterinae. 
12 a. Maxillary with a short sharp spine near anterior end, just 
behind premaxillary (Fig. 103); anal fin small, beginning far 
behind base of dorsal, with only about 15—18 rays. 

Rhtnosardinia Eigenmann 19 12, p. 411. 
12b. Maxillary without a spine; anal fin much longer, beginning 
in front of dorsal fin, under it, or immediately behind it, 
with more than 30 rays. 

13a. Pelvic fin present; dorsal fin moderately large, with 
about 15—20 rays. Ilisha Gray 1846, p. 415. 

13b. Pelvic fin wanting; dorsal fin with about 10—15 rays. 
14 a. Body with chest and abdomen greatly dilated, 
more or less half-moon shaped (Fig. 108), its 
greatest depth about half of its SL; median line of 
back in front of dorsal fin with several short spines. 
Pristigaster Cuvier 18 17, p. 427. 
1 4 b. Body elongate, with chest and abdomen not greatly 
dilated, its greatest depth less than a third of its 
SL; no external spines on median line of back. 
15a. Body excessively compressed; dorsal outline 
more or less concave over the head; a tooth- 
less membranous section between maxillary 
and premaxillary, the three elements form- 
ing a continuous margin; maxillary greatly 
produced in large adults (Fig. 1 1 1); anal fin 
with about 52—80 rays. 

Odontognathus Lacepede 1800, p. 430. 
15 b. Body rather strongly but not excessively 



262 Memoir Sears Foundation for Marine Research 

compressed ; dorsal outline convex over head ; 
maxillary and premaxillary not separated by 
a membranous section ; maxillary overlapping 
distal end of premaxillary, interrupting mar- 
gin of upper jaw (Fig. 113); anal fin with 
only about 39—48 rays. 
Neoopisthopterus Hildebrand 1948, p. 435. 
lib. Teeth in jaws of unequal size, both jaws with strong canines (Fig. 
115); vomer with teeth. Subfamily Chirocentrinae. 

Chirocentrodon Gunther 1868, p. 438. 
3 b. Median line of back in front of dorsal fin naked, not crossed by scales ; stomach 
with very thick walls, like the gizzard of a fowl. Subfamily Dorosomatinae. 

Dorosoma Rafinesque 1820, p. 443. 



Genus Etrumeus Bleeker 1853 
Round Herrings 

Etrumeus Bleeker, Verh. Batav. Genoot., 2^, 1853: 48; genotype by monoytpy, Clupea micropus Temm'mck 
and Schlegel. 

Generic S}Tionym: 

Perkinsia Rosa Eigenmann, Amer. Nat., 25, 1 89 1 : 153; type species by monotypy, P. othonops Rosa Eigenmann 
equals Clupea micropus Temminck and Schlegel. 

Characters. Body elongate, little compressed, especially in Atlantic species; chest 
and abdomen fully rounded. Bony scutes absent. Scales with smooth edges, deciduous, 
covering chest and abdomen. Snout long, tapering. Mouth terminal. Maxillary 
slender, ending about under anterior margin of eye. Teeth minute, present on jaws, 
vomer, palatines, pterygoids, and tongue. Dorsal fin beginning notably nearer to 
margin of snout than to base of caudal, with about 1 7—20 rays. Anal about half as 
far from base of caudal as from origin of dorsal, with about 10—12 rays. Pelvic fins 
inserted well behind base of dorsal, with a shield of enlarged scales covering them in 
part, each fin with 8 rays. Pectorals of moderate size, also shielded in part by enlarged 
scales, each fin with about 14-16 rays. Pelvic and pectoral each with a long Axil- 
lary PROCESS. 

Range. Represented by one species on the Atlantic and Gulf coasts of the United 
States, and by two or more species in the eastern and western Pacific on the coasts of 
southeastern Africa and southeastern Australia. 



Fishes of the Western North Atlantic 



263 



Etrumeus sadina Wl'itch'ill 18 14 

Atlantic Round Herring 

Figures 62, 63 

Siudy Material. Many specimens, postlarvae to adults, 25-180 mm TL, the 
specimens from New Jersey northward being adults caught near or on the shore, 
those from off Cape Lookout, northern Florida, and the Gulf of Mexico, mostly young, 
taken quite a distance offshore in 5-2 1 fms. depth except for a lot of six small ones 
from Beaufort Inlet, North Carolina, taken in water only about 2—3 fms. deep. 




Figure 62. Etrumeus sadina, female, 140 mm TL, 115 mm SL, Woods Hole, Massachusetts, USNM 143574. 
Drawn by Ann S. Green. 

'Distinctive Characters. The very slender and roundish body, the roundish chest 
and abdomen covered with ordinary scales, the absence of bony scutes, the small anal 
fin situated near the base of the caudal, and the rather small pelvic fin inserted entirely 
behind a vertical from the base of the last dorsal ray, readily distinguish this species 
from other American clupeoids. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 22 specimens from Study Material, $§ mm (young adults)-i8o mm TL, 
45—154 mm SL. 



Body: depth 16-18. 5. 
Caudal peduncle: depth 6.7—8.2. 
Head: length 24—29. 
Snout: length 6.0-9.5. 
Eye: diameter 6.5—8.0. 
Interorbital: width 3.0—4.4. 
Maxillary: length 7.0-10.5. 
Anal fin: length of base 5.6-6.9. 
Pelvic fin: length 8.9-1 1.5. 
Pectoral fin: length 14—17. 



Distance from snout to origin of: dorsal 
42—46. 

Scales: about 48-55, usually lost in pre- 
served specimens. 
' Gill rakers: 32—38 on lower limb, ca. 14 
on upper limb. 

Fin rays: dorsal 16-20, usually 18 or 19; 
anal 10—12, most frequently 11; 
pectoral 14—16. 

Vertebrae: 49 (7 specimens). 



264 Memoir Sears Foundation for Marine Research 

Body almost round, its greatest thickness about 80 "/o of its depth, its greatest 
depth 5.4-6.25 in SL; ventral profile a little more strongly convex than dorsal pro- 
file. Caudal peduncle compressed, its depth 3.3-3.9 in head. 

Scales deciduous, usually lost in preserved specimens, very thin, scarcely deeper 
than long, with even edges. 

Head low, rather long, 3.45-4.15 in SL. Snout equal to or more usually longer 
than eye, tapering, 3.0—3.9 in head. Eye with much adipose tissue in adults, 3.25-4.1 
in head. Interorbital bone 6.25—7.55. Maxillary obliquely rounded posteriorly, 
reaching about under anterior margin of eye, 2.6-3.0 in head. Mandible coterminal 
with snout, 1.85-2.0 in head. Gill rakers rather slender, moderately close-set, those 
at angle somewhat exceeding half of length of eye; apparently not increasing in number 
with age. Teeth small, in a single series on mandible; a few minute teeth on pre- 
maxillaries, and a row on the margin of maxillaries; very small granular teeth on 
vomer, palatines, pterygoids, and tongue.* 

Dorsal fin moderately elevated anteriorly, with concave margin, its longest rays 
generally not reaching tip of posterior ray if deflexed, its origin about equidistant be- 
tween margin of snout and vertical from middle of anal base, its distance from snout 
2.15-2.4 in SL. Caudal fin forked, the lobes of about equal length, generally somewhat 
longer than head without snout. Anal small, rather feebly developed, placed far behind 
dorsal fin, a little nearer to base of caudal than to base of pelvic, its base 3.5—4.5 in 
head. Pelvic fin rather short and broad, with nearly straight margin, inserted a little 
behind vertical from base of last ray of dorsal, generally a little nearer to base of pec- 
toral than to base of caudal, 2.3—3.0 in head; a membranous axillary process present, 
about 67 "/o of the length of fin. Pectoral fin moderately developed, reaching notably 
less than halfway to base of pelvic, inserted fully an eye's diameter nearer to tip of 
mandible than to base of pelvic, 1.5— 1.8 in head; a long membranous axillary process 
present about 0.83 of length of fin. 

Alimentary canal about half of SL. Stomach with a long blind sac and nu- 
merous pyloric caeca. Air bladder long and narrow. 

Color. Fresh specimens olive green above, silvery on lower half of side and below. 
In old preserved specimens, back generally brownish and lower parts pale (sometimes 
silvery). 

Size. A length of 200—250 mm (8—10 in.) is said to be attained by this species. 

Development. The postlarvae and young adults are very slender, as is usual in 
young herrings. In postlarvae 25— 28 mm TL (21— 24 mm SL) the body depth is 
contained about 16-24 times in the SL; the jaws are weakly developed; the teeth 
are not evident; the snout is proportionately longer than that in adults, being 1.5— 
2.0 times longer than the eye; the pectoral fins are not much more than tufts of mem- 
brane; and the pelvics are incompletely developed. The larval pigment spots are still 
retained, chiefly in a single slightly broken dark line on the chest, in two diverging 
dark lines on the abdomen, in a row of dark spots along each side of the base of 

4. To see the teeth clearly it may be necessary to dry the head. 



Fishes of the Western North Atlantic 



265 






Figure 63. Etrumeus sadina. a, larva, 27.5 mm TL, near Pensacola, Florida, b, young, 35 mm TL, near 
Pensacola, Florida, c, young, 42.5 mm TL, Beaufort, North Carolina, USNM 143804. All drawn by Ann 
S. Green; a and B based on camera-lucida sketches by W. W. Welsh. 



anal, and in a single dark median line behind the anal fin; also several dark spots are 
present on the base of the caudal. 

At about 33—35 mm TL, the fins are all rather fully developed and the body 
has become proportionately much deeper, but the pigmentation remains virtually the 
same as in smaller examples. Specimens of this size, then, may be classed as young 
adults. General pigmentation is delayed and is scarcely complete at 55 mm TL. The 
fish continue to Increase in relative depth as long as they grow. 

Spawning. The smallest specimens at hand, postlarvae 25—28 mm TL, are so 
young that they probably were caught not far from where they were hatched; they were 
taken at 30°29'N, 8o°37.5'W, off northern Florida. Still smaller larvae, 7.6-21 mm, 
were listed by the late W. W. Welsh in his unpublished notes, accompanied by seven 
camera-lucida sketches. These were taken by the Grampus at the surface in the Gulf 
of Mexico at 28°2i'N, 89°o6'W, 17 miles or so off the Mississippi Delta. Although 
these specimens are apparently no longer extant, the drawings leave no room for doubt 
as to their Identity. Also in this as well as other collections there are young adults from 
northern Florida south to Cape Canaveral and from the Gulf of Mexico off Mobile 
Bay. The smallest young (Including postlarvae) were taken on January 25 (1940), and 



2 66 Memoir Sears Foundation for Marine Research 

other small young adults were taken in March off northern Florida and in the Gulf of 
Mexico. The smallest young were caught in water 19 fms. deep, and other small 
ones were taken at 21, 20, 16, and 5 fms. From these data it may be concluded that 
spawning in the south takes place during the winter, somewhat offshore, and in water 
several fathoms deep. If spawning takes place northward it seems not to have been 
reported. 

Migration and Habitat. In the northern part of its range, as at Woods Hole, Massa- 
chusetts, this Round Herring is taken along the shore only during summer. It has not 
yet been determined whether this fish migrates northward during summer or whether 
it only comes inshore from deep water during warm weather. In southern waters it 
is not known to occur along the shore, for there it has been taken only offshore in 
water a few to several fathoms in depth. Whether it migrates at all in these waters re- 
mains unknown. 

Relationships. E. sadina is the only representative of the genus in the Atlantic, but 
other species occur in the Pacific: off California, the Philippine Islands, Japan, Hawaii, 
and Galapagos Islands. Those from the western Pacific are more compressed and pro- 
portionately deeper than the Atlantic species and are therefore less typically "round 
herrings." 

Abundance. Generally this Round Herring is rare to scarce from New Jersey to 
New England, where it occurs along the shore, but occasionally it is common to 
plentiful. It was very abundant off Woods Hole, Massachusetts, in 1905 and again in 
1908 (123: 741). A number of specimens was also taken at Campobello Island at the 
mouth of the Bay in September 1 937 (82 : ^\g2: 15). Along the coast of eastern Maine, 
in August-September 1953, something like 200,000 pounds were landed {114: 194). 
But apparently it has been taken only sparingly off the southern Atlantic and Gulf 
coast states, though commonly regarded as a warm-water fish. 

Range. It is known from Passamaquoddy Bay at the mouth of the Bay of Fundy 
and from the eastern part of Maine to northern Florida off Cape Canaveral, and from 
the Gulf of Mexico.^ Although it has been reported as not rare southward (72, 
1896: 420), no records have been found of its occurrence from Delaware southward; 
and there is only one record for the Gulf, the latter being based on specimens taken 
from the stomachs of snappers landed at Pensacola, Florida (7^: 143). Recent col- 
lections by the U.S. Fish and Wildlife Service include: examples from Beaufort Inlet 
and off Cape Lookout, North Carolina; four collections of young from off northern 
Florida southward to Cape Canaveral ; and six collections from the Gulf of Mexico, 
five of these from off Mobile Bay and one from south of New Orleans. These rather 
recent collections are all from offshore waters of 2—21 fms. depth. Those from New 
Jersey northward, all adults, were caught in shallow water near or on the shore. 

5. Benin (jj:9) synonymized E. microps and E.jacksoniensis, previously recognized as valid species, and thereupon 
gave the range as the Atlantic coast of the United States, the coast of southeastern Australia, southeastern Japan, 
Hawaii, and the southeastern coast of Africa. However, from comparison of specimens from the United States and 
Japan, it was at once evident that the latter were different in shape, being especially more compressed. A further 
study of specimens from the widely separated regions where the genus occurs would be desirable. 



Fishes of the Western North Atlantic 267 

Apparently this fish does not occur off southern Florida, for it was not reported 
by Longley and Hildebrand from collections made at Tortugas over many years, in- 
cluding dredgings at various depths down to about 400 fms (<S7). This seems to be 
another instance of a discontinuous range of the genus such as is reported for Pomolobus, 
Alosa^ and Brevoortia (p. 343). However, in Etrumeus the inhabitants of the Gulf and 
the Atlantic appear to be identical, whereas in the other genera named, the inhabitants 
of the two areas differ sufficiently to be recognized as distinct species. 

Synonyms and References: 

Clupea sadina Mitchill, Rep. in part on Fishes of New York, 1814: 21 (orig. descr.; type local, presumably 
New York; type lost); Mitchill, Trans. Lit. philos. Soc. N. Y., j, 1815: 457 (descr.. New York*). 

Alosa teres DeKay, N. Y. Fauna, Fishes, Pt. 4, 1842: 262, pi. 40, fig. 128 (orig. descr. ; type local. New York 
Harbor,- type in State Coll.). 

Etrumeus teres Giinther, Cat. Fish. Brit. Mus., 7, 1868: 467 (synon., notes, range); Jordan and Gilbert, Bull. 
U. S. nat. Mus., 16, 1883: 263 (diagn., refs.); Jordan and Gilbert, Proc. U. S. nat. Mus. (1883), 6, 
1884: 143 (from stomachs of snappers, Pensacola, Florida); Bean, T. H., Bull. U. S. Fish Comm. 
(1887), 7, 1888: 148 (Great Egg Harbor Bay, New Jersey); Bean, T. H., Bull. N. Y. St. Mus., 60, 
Zool. 9, 1903: 189 (synon., descr.. New York); Fowler, Rep. N. J. St. Mus. (1905), 2, 1906: 93 
(diagn., synon.); Kendall, Occ. Pap. Boston Soc. nat. Hist., 7(8), 1908: 36 (refs.. New England); 
Sumner, Osburn, and Cole, Bull. U. S. Bur. Fish. (1911), 52(2), 1913: 741 (refs., parasites, Woods 
Hole, Massachusetts); Fowler, Proc. biol. Soc. Wash., JJ, 1920: 147 (New Jersey); Bigelow and Welsh, 
Bull. U. S. Bur. Fish. (1924), 40(1), 1925: 91, fig. 39 (descr., range, Gulf of Maine); Breder, Field 
Bk. Mar. Fish. Atl. Cst., 1929: 63, fig. (diagn., distr.). 

Etrumeus sadina Jordan and Evermann, Bull. U. S. nat. Mus., 47 (l), 1896: 420 (descr., range, synon.); Jordan, 
Evermann, and Clark, Rep. U.S. Comm. Fish. (1928), 2, 1930: 46 (synon., range); Bigelow and Schroe- 
der, BuU. U. S. Bur. Fish., 48(20), 1936: 327 (Yarmouth R. and Casco Bay, Maine); Bertin, Bull. 
Inst, oceanogr. Monaco, 853, 1943: 9, fig. 4 (synon., descr., distr., etc.); Bigelow and Schroeder, Fish. 
BuU. (74) U. S. Fish Wildl. Serv., 53, 1953: 87 (Gulf of Maine). 



Genus Jenkinsia Jordan and Evermann 1896 
Dwarf Herrings 

Jenkinsia ]oTdia.n and Evermann, Bull. U.S. nat. Mus., 47(1), 1896: 418; genotype by monotypy, Dussumieria 
stoUfera Jordan and Gilbert equals Clupea lamprotaenia Gosse. 

Characters.'' A very small species. Body slender, compressed; abdomen rounded; 
chest with suggestion of a keel. Bony scutes absent. Scales deciduous, very thin, 
with smooth edges, covering abdomen. Snout moderately long, tapering. Mouth 
terminal. Maxillary narrowly rounded, terminating about under anterior margin of 
pupil. Teeth in a single series on the margin of maxillary and premaxillary; a short 

6. Because of such inaccurate statements in the original description as "Mouth. . .toothless," "Lower jaw projecting," 
and "A. 15," it was long questioned whether sadina was available for this species. However, the description con- 
tains pertinent remarks such as "Form neat, tapering, and slender," and especially "the beUy is not at all serrate, 
but quite smooth," which can scarcely apply to any other clupeoid. Therefore, sadina certainly seems available. 
Apparendy Alosa teres DeKay, though much better described and rather well figured, must rest in synonymy, 
under the rules of zoological nomenclature. 

7. Bertin (jj: 17) has stated that Jenhinsia is not separable from Spratelloides Bleeker. — G. S. Myers. 



2 68 Memoir Sears Foundation for Marine Research 

toothless membranous section present between these elements ; no teeth on mandible or 
vomer; at most a few granular teeth on palatines, pterygoids, and tongue. Dorsal small, 
beginning rather nearer to margin of snout than to base of caudal, with about 9—14 rays. 
Anal only a little larger than dorsal and placed far behind it, with about 12-16 rays. 
Pelvic fins nearly as long as the pectorals, each with 8 rays; a scale process present 
between fins at base. Pectoral fins rather small, each with about 10-14 rays. Pelvic 
and pectoral both without a free Axillary process. 

Remarks. Jenkinsia resembles Etrumeus in having a rather roundish, elongated body, 
a rounded abdomen, and no bony scutes on the ventral outline of the chest and abdomen. 
However, in Jenkinsia the chest is very slightly keeled. The rather small anal fin, situated 
far behind the dorsal, and the rather large pelvic fin, inserted under the anterior half of 
the base of dorsal, also help to distinguish this genus. The small size of these herrings, 
about 50—70 mm, also separates them from other herrings. 

Range. Known from Bermuda, doubtfully from North Carolina, the Florida Keys, 
the West Indies, and from some small islands in the Caribbean Sea, off Mexico and 
Central America. Only two species are known. 



Key to Species 

I a. Gill rakers not especially close-set, 19-24 (usually 20 or 21) on lower limb; oblique 
series of scales crossing silvery lateral band between margin of opercle and base 
of caudal, 33-37. lamprotaenia (Gosse) 1851, below. 

I b. Gill rakers close-set, 27-30 on lower limb; oblique series of scales crossing silvery 
lateral band between margin of opercle and base of caudal, 39-42. 

viridis (T. H. Bean) 1912, p. 272. 



"Jenkinsia lamprotaenia (Gosse) 1 8 5 1 

Dwarf Herring, Green Fry 

Figure 64 

Study Material. Many specimens, 28-65 mm TL, 23-55 mm ^L, collected at 
Key West and Tortugas, Florida, the Bahama Islands, Cuba, Isle of Pines, Jamaica, 
Haiti, Puerto Rico, St. Eustatius Island, Old Providence Island, and Yucatan Island 
off Cozumel, Mexico. 

Distinctive Characters. See Remarks, and Key to Species, above. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 16 specimens, 40-65 mm TL, 32— 55 mm SL. 

Body: depth 15-20. Snout: length 6.2-8.0. 

Caudal peduncle: depth 6.5-8.5. Eye: diameter 7.6-9.7. 

Head: length 25-29. Interorbital: width 3.8-5.4. 



Fishes of the Western Nortli Atlantic 



269 



Maxillary: length 9.0-12. 
Mandible: length 1 2-14.5. 
Anal fin: length of base 10-14. 
Pelvic fin: length 11 -15. 
Pectoral fin: length 14-17. 
Distance from snout to origin of: dorsal 
44-50. 



Scales: 33—37, generally missing in pre- 
served specimens. 

Gill rakers: 19—24 (usually 20 or 21) on 
lower limb. 

Fin rays: dorsal 9—13; anal 12—15; P^'^" 
toral 10-13. 

Vertebrae: 39 or 40 (6 specimens). 





Figure 64. Jenkinsla lamprotaenia, 65 mm TL, 53 mm SL, Haiti, USNM 89600. Drawn by Ann S. Green. 



Body moderately compressed, its greatest thickness somewhat greater than half of 
its depth, its greatest depth 5.0—6.65 in SL; ventral outline more strongly convex 
than dorsal outline. Caudal peduncle rather strongly compressed, its depth 3.2—4.1 
in head. 

Scales generally lost in preserved specimens, very thin, with smooth edges; those 
from middle of side below dorsal fin with a wavy ridge on anterior half, about 5 wavy 
radii behind it, and generally several shorter ridges at margin of scale. 

Head rather low, 3.4—4.0 In SL. Snout generally shorter than eye, 3.4—4.2 in 
head. Eye 2.8—3.4. Interorbital 5.5—7.0. Maxillary narrowly rounded posteriorly, 
reaching to or more usually a little beyond vertical from anterior margin of pupil, 
2.25—2.75 in head. Mandible generally included, not projecting, 1.9-2.2 In head. 
Gill rakers moderately close-set, those at angle of first arch about 75 "/o of length 
of eye. Teeth missing on mandible and vomer but present In a single series on 
margin of premaxillary and maxillary, the series interrupted by a soft membranous 
section between these elements; palatines, pterygoids, and tongue at most with a few 
granular teeth. 

Dorsal fin moderately elevated, the margin nearly straight, the longest rays reach- 
ing well beyond the tip of last ray if deflexed, the fin beginning notably nearer to margin 
of snout than to base of caudal, the distance from margin of snout 2.0—2.3 in SL. Caudal 
fin moderately forked, the lobes of about equal length, rather longer than head without 
snout; each lobe with a greatly elongated scale at base. Anal fin much lower than dorsal, 
beginning far behind dorsal fin, about equidistant between insertion of pelvic and base 



270 Memoir Sears Foundation for Marine Research 

of caudal, its base 1.9—2.6 in head. Pelvic fins only a little shorter than pectorals, with 
a large scaly process between them, the fins inserted under about 4th— 6th dorsal ray, 
and generally close to an eye's diameter nearer to base of caudal than to tip of mandible ; 
1.9—2.5 in head. Pectoral fins somewhat pointed, inserted about equidistant between 
tip of mandible and base of pelvics, 1.6— 1.9 in head. 

Alimentary canal short. Stomach with a large blind sac. Pyloric coeca large, 
about 7. Peritoneum pale, with many dark dots. 

Color. In life, faintly greenish above, with a distinct silvery lateral band. Preserved 
specimens straw colored, the silvery band persisting. Margin of mandible brown to 
black. Well-preserved specimens with an elongated silvery area on abdomen behind 
pectoral fin, and a similar though smaller area behind pelvic. Scales above lateral 
band with a brownish line on middle of exposed section, paralleling the curve of the 
margin of the scale; dark dots also present, especially in and above upper margin of 
silvery lateral band; two rows of dark dots on back; generally numerous dots on snout, 
including an elongated one in front of lower half of eye ; an irregular dark spot at base 
of each anal ray, continued as a single median ventral line on caudal peduncle. 

Size. The largest example at hand, (>i, mm (2.6 in.) TL, may be near the maximum 
size attained. 

Spawning and 'Development. Nothing is known definitely concerning reproduction 
in this species, but examples collected in 1937 at Kingston, Jamaica, during our winter 
months, contain well-developed eggs, indicating that at least some spawning takes place 
during that time around Jamaica. 

The smallest examples at hand, a little more than an inch (28—30 mm) TL, are 
fully developed young adults. They differ principally from large examples from the 
same general locality (Tortugas, Florida) in being more slender, the depth being about 
6.4 times in SL; in the large adult this proportion varies between about 2.25-5.6. 
However, there are large specimens in the collections studied that are just as slender 
as the young mentioned. For further discussion of the difference in depth, see Varia- 
tions (p. 271). 

Migration and Habitat. It seems improbable that this little fish travels long dis- 
tances. It does occur in large compact schools that probably consist of millions of fish, 
as observed by me at Key West, Florida, and as reported for Tortugas, Florida (.^9: 
156; 21: i). It moves about more or less like the young Atlantic menhaden, Brevoortia 
tyrannus, without pursuing any definite course. 

Food. This species, like many other clupeoids, feeds on plankton, which it screens 
from the water as the school girates or mills around at the surface. 

Enemies. Many large fish feed on this species. The gray snapper was mentioned 
especially by Gudger, who said that the little fish exhibited little fear; it hovered over 
the snapper in dense swarms but left a clear space around it approximately equal to the 
length of the snapper {4^: 156). Whenever the snapper moved, the rather orderly school 
broke up but soon re-formed. Longley said, "Exceedingly common at Tortugas (Fla.), 
particularly along shore, where predacious fishes often drive them too close for safety. 



Fishes of the Western North Atlantic 271 

and large numbers are cast upon the windward beaches to spring about until a wave 
rescues them, or until they perish" ((§7: 12). 

The copepod, Colliparvus parvus Wilson, was found on the gills of specimens 
taken at Nassau, Bahamas (JJ5: 79), and an unidentified trematode was reported as 
parasitic on it at Tortugas, Florida (^o: 372). 

Relationships. Only two species of Jenkinsia are recognized herein. A third nominal 
species, stolifera, long was thought to be distinct, but recent works on taxonomy agree 
that it and lamprotaenia constitute one variable species; since lamprotaenia is the older 
name, stolifera has become a synonym. J. viridis is closely related to lamprotaenia^ as 
shown in the account of viridis (p. 272). 

Variations. This species, as indicated, is very variable. While the depth of body 
in general increases with age, adult specimens of equal size vary greatly in this respect 
among themselves, as is evident from the wide range shown in the Description. And 
this in turn is reflected in the wide range in the depth of the caudal peduncle and in 
the length of the head. A rather wide variation in the number of dorsal, anal, and pectoral 
rays, and in the number of gill rakers, is also shown in the Description. Although some 
of the earlier counts may have erred because fins are often broken in preserved specimens 
(no doubt so in the original description of stolifera)., this error does not enter into those 
given in the foregoing Description, for ample specimens with undamaged fins were 
chosen for the enumerations. 

Commercial Importance. Although this little Herring has been reported among the 
"white bait" in the market at Port-au-Prince, Haiti (JO: 45), no doubt its chief value 
is that of providing food for foodfishes. 

Range. Known from the Florida Keys, the Bahamas, throughout the West Indies, 
and from Swan Island, Glover Reef, and Cozumel Island in the Caribbean Sea; recently 
reported from Venezuela; doubtfully recorded for Beaufort, North Carolina. Appar- 
ently it is chiefly a shore species. 

Synonyms and References: 

Clupea lamprotaenia Gosse, Naturalist's Sojourn in Jamaica, 185 1: 291, pi. I, fig. 2 (orig. descr.; type local. 
Jamaica; types in BMNH; found in schools with an anchovy, Engraulis brownii = Anchoa hepsetus); 
Giinther, Cat. Fish. Brit. Mus., 7, 1868: 465 (descr., Jamaica). 

Dussumieria stolifera Jordan and Gilbert, Proc. U. S. nat. Mus. (1884), 7, 1885: 25 (orig. descr.; type local. 
Key West, Florida; type USNM 34964; local abund.); Bean, T. H., Bull. U. S. Fish. Comm. (1888), 
8, 1890: 206 (Cozumel, Yucatan). 

Jenkinsia lamprotaenia Jordan and Evermann, Bull. U.S. nat. Mus., 47 (i), 1896: 419 (descr., after Giinther); 
Evermann and Marsh, Bull. U. S. Fish Comm. (1900), 20(1), 1902: 84 (descr., refs., Puerto Rico); 
Metzelaar, Rapp. Vissch. Curasao, ed. by J. Boeke, i, W.Indian Fishes, 1919: 9, fig. (refs., descr., 
range; Leeward Is., West Indies); Beebe and Tee- Van, Zoologica, N. Y., 10 (l), 1928: 43, fig. (refs., 
descr., range, J . stolifera synon.; Port-au-Prince, Haiti); Breder, Field Bk. Mar. Fish. ."^tl. Cst., 1929: 
63, fig. (diagn., range); Nichols, N. Y. Acad. Sci., 10 (2), 1929: 201, fig. 32 (refs., distr., diagn., 
Puerto Rico); Parr, BuU. Bingham oceanogr. CoU., J (4), 1930: 3 (refs.; discus, ident. oi lamprotaenia 
and stolifera; Bahamas, Turks Is.); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 
1930: 46 (range, refs.); Longley and Hildebrand, Publ. Carneg. Instn. Wash., 535, 1941: 12 (discus. 
lamprotaenia and stolifera; variability; Tortugas, Florida); Hildebrand, Copeia, 1941: 224 (extension 
of range, Beaufort, North Carolina); Fowler, Monogr. .\cad. nat. Sci. Philad., 6, 1944: 123, fig. 30 



272 Memoir Sears Foundation for Marine Research 

(descr., locals., pp.434, 456); Martin, Mem. Soc. Cient. La Salle, 75, 1955: 184-188 (descr., ills., 
Venezuela). 

Jenkinsia stolifera Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 419 (descr., ref. to orig. descr.); 
Evermann and Kendall, Rep. U. S. Comm. Fish. (1899), 1900: 5; (refs., Florida) ; Evermann and 
Marsh, Bull. U. S. Fish Comm. (1900), 20 (i), 1902: 84 (descr., refs., Puerto Rico); Bean, B. A. in 
Bahama Islands, Fishes, 1905: 297 (rec. from Spanish Wells); Fowler, Proc. Acad. nat. Sci. Philad., 
5c?, 1906: 81, fig. I (descr.; assoc. with anchovies; Florida Keys); Wilson, Proc. U. S. nat. Mus., JJ, 
1917: 79 (host of copepod, Colliparvus parvus Wilson; Nassau, Bahamas); Breder, Bull. Bingham oceanogr. 
Coll., J(i), 1927: 13 (Isle of Pines, Swan I., and Glover Reef off" British Honduras); Breder, Amer. 
Mus. Novit., 382, 1929: 1—5 (school, behav., Tortugas, Florida); Gudger, Publ. Carneg. Instn. Wash., 
391, 1929: 156, pi. I, fig. 6 (local abund., habits, Tortugas, Florida); Jordan, Evermann, and Clark, 
Rep. U. S. Comm. Fish. (1928), 2, 1930: 46 (range, ref); Breder, Publ. Carneg. Instn. Wash., 435, 
1932: 8 (fed on by needlefish, Strongylura; Tortugas, Florida); Gowanloch, Bull. La. Conserv. Dep., 
23. 1933 ■• 214 (in great schools. Gulf of Mexico; but no records for Gulf coast north of extremes 
Florida, therefore, apparently not of Louisiana fauna). 

Spratelloides larnprotaenia Bertin, Bull. Inst, oceanogr. Monaco, 853, 1943: 17 (diagn., revision). 

False References: 

Jenkinsia stolifera Nichols (not of Jordan and Gilbert), Proc. biol. Soc. Wash., JJ, 1920: 63 {Stolephorus 

viridis in synon., Bermuda). 
Jenkinsia larnprotaenia Beebe and Tee- Van (not of Gosse), Field Bk. Shore Fish. Bermuda, 1933: 37, fig. 

(field char., diagn., distr.); Beebe and Tee-Van, Zoologica, N. Y., 13 (j), 1933'. 136 (refs., Stolephorus 

viridis and J. stolifera in synon., Bermuda); HoUister (not of Gosse), Zoologica, N. Y., 21 (4), 1936: 

276, figs. 40—44 (caudal skel.). 



Jenkinsia viridis (T. H. Bean) 1912 

Bermuda Herring, Dwarf Herring, Green Fry 

Figure 65 

Study Material. A total of 19 specimens, 42—58 mm TL, 32—48 mm SL, as fol- 
lows: 4 "types," from Bermuda, presumably from Hamilton, USNM 74084; the holo- 
type, USNM 100546; and 14 paratypes of Jenkinsia bermudana Rivas, Grampus St. 
10178, off Bermuda, 32°2o'N, 64°2i'W. 

Distinctive Characters. See Relationships (p. 273) and Key to Species (p. 268). 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 32—48 mm SL. 



Body: depth 1 5-1 7.5. 
Caudal peduncle: depth 6.7—8.2. 
Head: length 25—27. 
Snout: length 6.7-8.5. 
Eye: diameter 7.0—9.0. 
Interorbital: width 3.6-4.75. 
Maxillary: length 9.4—10.8. 
Mandible: length 11. 6— 13.5. 
Anal fin: length of base 12—14. 
Pelvic fin: length 12—15. 



Pectoral fin: length 14—16. 
Distance from snout to origin of: dorsal 
44.5-50. 

Scales: about 39—42, generally lost in 
preserved specimens. 

Gill rakers: 27—30 on lower limb. 

Fin rays: dorsal 9—14; anal 14—16; pec- 
toral 12—14. 

Vertebrae: 40 (i specimen). 



Fishes of the IVestern North Atlantic 



273 



Body shaped in all respects as in laniprotaenia, its greatest depth 5.7—6.6 in SL. 
Caudal peduncle 3.2—4.0 in head. 

Scales as in lamprotaenia, except that there appear to be a few more in a lateral 
series. 

Head low, with exceedingly numerous mucous canals and pores, its length 3.7- 
4.0 in SL. Snout equal to, or more usually a little shorter than, eye, 3.2-3.8 in head. 
Eye 2.9-3.5. Interorbital 5.5-7.1. Maxillary narrowly rounded posteriorly, gen- 
erally reaching vertical from anterior margin of pupil, 2.45—2.6 in head. Mandible 
slightly included, about coterminal with the snout, 2.0-2.2 in head. Gill rakers very 




Figure 65. Jenkinsia viriJis, 45 mm TL, from Bermuda, USNM 100546. After Rivas. 

close-set, those at angle rather more than half of length of eye. Teeth apparently in 
all respects as in lamprotaenia. 

Dorsal fin elevated anteriorly, the longest rays reaching far beyond tip of last 
ray if deflexed, the fin generally beginning about half of an eye's diameter nearer to 
margin of snout than to base of caudal, the distance from margin of snout 2.0-2.25 in 
SL. Caudal fin as in lamprotaenia. Anal lower than dorsal, its origin about equi- 
distant between insertion of pelvic and base of caudal, its base 1.8—2.2 in head. Pelvic 
fin only a little shorter than pectoral, inserted under the 4th or 5th dorsal ray and 
about equidistant between anterior margin of eye and base of caudal, its length 1.55— 
2.2 in head. Pectoral fin pointed, inserted a little nearer to tip of mandible than to 
base of pelvic fin, 1.55— 1.8 in head. 

Color. In all respects as in lamprotaenia. 

Size. The largest specimens at hand, 55-58 mm TL (2.2-2.3 in.), may be near 
the usual maximum length attained. 

Food. Although no specific information is at hand, there seems to be no reason to 
believe that the food of this species differs from that of lamprotaenia. 

Enemies. No enemies have been listed. However, there can be no doubt that 
predatory fish and water birds feed on this species as they do on lamprotaenia. 

Relationships. As stated in the account of lamprotaenia, the two species of this 



2 74 Memoir Sears Foundation for Marine Research 

genus recognized herein are closely related. But they differ in the number of gill rakers 
on the lower limb of the first arch, viridis having 27—30 while lamprotaenia has only 
19—24 (usually 20 or 21). That there is a difference between the species in the inser- 
tion of the pelvic fin, as stated by Rivas {108: 3), I am unable to substantiate, as it 
seems to be obliterated by the variations among specimens. However, according to four 
viridis and seven lamprotaenia that remain almost fully covered with scales, in the former 
39—41 oblique rows of scales cross the lateral band between the margin of the opercle 
and the base of the caudal, whereas in the latter only 33-37 series cross this band. 
The specimens of viridis at hand have much more prominent mucous canals and pores 
about the head and on the anterior part of the body, but they may not be specifically 
significant. 

It so happens, then, that even though T. H. Bean placed this species in the 
wrong genus and family {Stolephorus, family Engraulidae), his species still remains 
valid (9: 22). 

Commercial Importance. This species is not known to be of direct commercial 
importance. It has been reported as exceedingly common in Bermuda (9: 123; JJ: 37), 
therefore it must be valuable as a forage fish. 

Range. Known only from Bermuda and vicinity. It congregates in schools like 
lamprotaenia but ventures some distance offshore, as shown by the Grampus catch noted 
above. 

Synonyms and References: 

Stolephorus viridis Bean, T. H., Proc. biol. Soc. Wash., 25, 191 2: 122 (orig. descr.; type local. Bermuda; 

types USNM 74084; common). 
Jenkinsia stolifera Nichols (not of Jordan and Gilbert), Proc. biol. Soc. Wash., JJ, 1920: 63 {Stolephorus 

viridis in synon.). 
Jenkinsia lamprotaenia Beebe and Tee- Van (not of Gosse), Field Bk. Shore Fish. Bermuda, 1933: 37, fig. 

(field char., diagn., distr.); Beebe and Tee- Van, Zoologica, N. Y., JJ (7), 1933: 136 (refs., Stolephorus 

viridis and J. stolifera in synon.; Bermuda); HoUister, Zoologica, N. Y., 21 (4), 1936: 276, figs. 40-44 

(caudal skel.). 
Jenkinsia bermudana Rivas, Smithson. misc. Coll., 106(14), 1946: 1-4, i fig., i pi. (orig. descr.; type local. 

Bermuda; type USNM 100546; cf. lamprotaenia; but S. viridis Bean apparently overlooked, of which 

bermudana becomes a synon.). 



Genus Clupea Linnaeus 1758 

Clupea Linnaeus, Syst. Nat., ed. 10, 1758: 317; type species by implication (unquestioned), Clupea harengus 
Linnaeus. 

Generic Synonym: 

Rogenia Cuvier and Valenciennes, Hist. Nat. Poiss., 20, 1847: 340; type species by monotypy, Rogenia alba 
equals Clupea harengus lAnnatMS 1758. 

Characters. Body quite elongate, compressed. Bony scutes on median line of 
chest and abdomen, weak. Scales with crenulate membranous borders. Maxillary 
with a large paddle-shaped supplemental bone. Teeth on vomer permanent, forming 



Fishes of the IVestern North Atlantic 275 

a small patch. Dorsal fin rather small, without a produced ray. Pelvic commonly 
with 9 rays, rarely with only 8." Vertebrae 45-57. 

Remarks. The very slender body and the presence of vomerine teeth distinguish 
this genus from other genera of the western North Atlantic. 

Range. North Atlantic and North Pacific, ranging on the American side of the 
Atlantic from North Carolina to Greenland and in Europe from the Strait of Gibraltar 
to Spitsbergen. In the Pacific, the genus is represented from southern California to the 
Aleutian Islands and across to Siberia and Japan. A single species occurs in the western 
North Atlantic' The species generally occurs in schools, which are usually seen at or 
rather near the shore. 



Clupea harengus Linnaeus 1758 

Atlantic Herring, Sea Herring, Herring 

Figures 66-68 

Study Material. At least 18 specimens, 75-270 mm TL, 63-224 mm SL, and 
smaller specimens, including postlarvae; from Kings Creek (tributary to Tangier 
Sound), Chesapeake Bay, Maryland; Ipswich River (estuary) and Woods Hole, Mas- 
sachusetts; Sheepscot River and Eastport, Maine; St. Lawrence River; Nova Scotia; 
New Brunswick; and Newfoundland. This American material was compared with 
specimens from Norway and Sweden, and with specimens from the Pacific collected In 
California, Oregon, and Alaska. 

Distinctive Characters. The following conspicuous field marks separate the post- 
larval and adult Atlantic Herring from the shads {Alosd) and from representatives of 
Pomolobus (alewlves, hickory shad) : the point of origin of its dorsal fin is about midway 
of its trunk (considerably farther forward in the others) ; its body Is not as deep, a differ- 
ence shown better In the Illustrations than by words; and the sharp midline of Its belly 
Is less strongly serrate. 

[Critical examination Is required to distinguish the early stages of clupeolds of the 
western North Atlantic, one from another. Larvae of other species with which the 
Herring larva might be confused are the launce (Ammodytes), the so-called rock gunnel 
(Pholis), the capelin {Mallotus\ and the smelt {Osmerus). But the position of the anus, 
about 83 "/o of the way back along the body, sets the larval Herring apart from all 
of these (about 75 °/o of the way back In Mallotus and Osmerus., about 66 "/o of the 
way back In Ammodytes and Pholis). — h. b. b.] 

Description. Proportional dimensions In per cent of standard length, and counts, 
from specimens In Study Material, not including small specimens except for men- 
tion of their much more slender body and fewer gill rakers; specimens 63-224 mm SL. 

8. Pelvics with 7 or 8 in South American species. — G. S. Myers. 

9. The genus Clupea in the present sense is also represented in Australian-New Zealand temperate waters and in 
southern South America. Norman has reviewed the South American forms (102: 37). — G. S. Myers. 



276 Memoir Sears Foundation for Marine Research 



Body, depth 20-25.8. 
Caudal ■peduncle: depth 7.2—10. 
Head: length 22.6—26.4. 
Snout: length 5.6-7.1. 
Eye: diameter 5.3— 7- 7- 
Interorbital: width 3.0—4.5. 
Maxillary: length 10.2- 12.3. 
Pelvic fin: length 9.4.— 11. 3. 
Pectoral fin: length 14.2 — 17.7. 



Scales: ca. 56—62. 

Ventral scutes: ca. 39—46. 

Gill rakers: 37-52. 

Fin rays: dorsal 16-20, usually 17-19; 
anal 16-20, usually 17 or 18; pec- 
toral 17-19. 

Vertebrae: SS~S1 ('^ specimens). 




Figure 66. Clupea harengus, 315 mm TL, 12.62 inches, USNM 16667. Drawn by H. L.Todd. 



Body compressed, its greatest thickness somewhat less than half of its depth, its 
depth 3.9-5.0 in SL, much less in very young, 5.2-9.5 in specimens 40-70 mm. 

Scales only moderately adherent, often lost, especially in small fish; scales with 
crenate membranous border; about 16 or 17 longitudinal rows between ventral edge 
at base of pelvic and anterior ray of dorsal. Ventral scutes weak, often difficult to 
count, about 26—33 i" front of pelvic fin and 13-15 behind. 

Head 3.8—4.4. Snout 3.25—4.1 in head. Eye 3.4-4.25. Interorbital 3.0-4.5. 
Maxillary rounded posteriorly, generally not quite reaching middle of eye, 2.0-2.3 •" 
head. Gill rakers at angle nearly as long as eye in adults, proportionately shorter in 
young; increasing in number with age, the principal increase taking place before the 
specimens reach a length of 100 mm; specimens 36— 70 mm long with only about 
25—36 on lower limb. Teeth missing in upper jaw; the margin of maxillary with minute 
serrae; lower jaw anteriorly with about 6 rather prominent teeth on each side; middle 
of tongue with an elongated patch of small teeth ; vomerine patch elongate, the teeth 
rather few and fairly strong. 

Dorsal fin little elevated anteriorly, with nearly straight margin, its origin some- 
what in advance of pelvics, usually nearer to base of caudal than to tip of snout. Caudal 
with lower lobe rather longer than the upper one, about as long as head. Anal very 
low, its origin about equidistant between base of pelvic and base of caudal. Pelvic 
very near ventral edge, inserted equidistant between base of pectoral and last anal 



Fishes of the Western North Atlantic 



277 



ray, 2.2-3.0 in head. Pectoral inserted a little higher on side than pelvic, and below 
distal margin of opercle, 1.3— 1.6 in head. 

Color. According to Bigelow and Welsh (j6: 93): 

Peritoneum quite dusky. Deep steel blue or greenish blue on the back with green reflections; the sides and 
belly silvery; the change from dark to pale sides often marked by a greenish band. The gill covers sometimes 
glisten with a golden or brassy gloss; indeed, fish just out of the water are iridescent all over with different hues 
of blue, green, and violet, beauties that soon fade, however, 
leaving only the dark back and silvery sides. The ventral and 
anal fins are transparent white; the pectorals, however, are dark 
at the base and along the upper edge; the caudal and dorsal 
dark grayish or shading into blue or green. i^Ts"^ 

Size. The maximum size attained by the 
Atlantic Herring is about 450 mm (18 in.), 
though the usual size is only about 300 mm 
(12 in.). 

Scope of Following Account. Much has been 
written about this Herring's spawning habits, food, 
rate of growth, seasonal migrations, and so on, 
by both European and American authors. Sum- 
maries of the literature about this species on the 
American coast, with frequent references to the 
European literature, have been given by Bigelow 
and Welsh {16: 92—105) and by Bigelow and 
Schroeder (75: 88—100), who drew heavily on 
important papers by Moore {g^: 387—442), Lea 
(7^: 75-164), and Huntsman (6j: 1 65-1 71); 

furthermore, they were able to add much to the already known facts from their own 
studies. A review of all the literature would make this account much too long for 
the present work. Therefore, only a brief summary of what is known about its life 
history, drawn in large part from the accounts mentioned above {l^\ 16), is given here. 
Nearly all of the papers cited provide bibliographies from which references to literature 
containing additional information may be obtained. 

Development and Growth. The eggs of the Atlantic Herring, i.o— 1.4 mm in 
diameter (16: 98) or 1.2— 1.5 mm (2g: 362), are heavier than sea water, and adhesive, 
adhering singly or in clumps to objects in the water or to the bottom. If they fall on 
muddy bottom they probably smother. 

The number of eggs produced by a single female is relatively small, and as usual 
it varies according to the size of the fish, large fish having many more eggs than small 
ones. The number produced by one female may vary between 21,000-47,000 {loi: 
283) or between 20,000-40,000 {16: 94). 

The duration of incubation depends upon the temperature. Norman stated that 
eggs kept at 52— 58°F hatched in eight or nine days, whereas others held in water at 




Figure 67. Clupea harengus (Europe) eggs 
attached to seaweed. After Ehrenbaum. 



278 Memoir Sears Foundation for Marine Research 

32°F hatched in 47 days {lOl: 330). Bigelow and Welsh gave 10-15 days as the 
probable average incubation period for the usual temperatures prevailing in the Gulf of 
Maine during the hatching season; these seem to vary between about 46-55° F(j6: 94). 
The newly hatched fish, transparent and very slender, vary considerably in length, 
with a range of about 5.3-7.0 mm {2g'. 362) or about 5.0-6.0 mm {l6\ 97). The small 




Figure 68. Clupea harengus (Europe), a-d, larval stages; a 7 mm; b 10 mm; c 19 mm; d 29 mm; e young 
fry, 41 mm. 



yolksac is absorbed in several days, when the larvae have attained a length of 8 — 10 mm; 
no marked changes take place in the form of the body during this time. The vent is 
situated so far back that the length posterior to it is less than a fifth of the total body 
length, and the depth at this size is contained about 24 times in the total length. Con- 
siderable advancement in the development of the fins was evident in larvae around 
1 9 mm long, taken in the tow at Helgoland ; rays definitely were present in the dorsal 
and caudal, and the caudal had a concave margin. However, the anal and pelvic fins 
did not become differentiated until a length of about 29 mm was attained. In young 
about 4 1 mm long, the proportionate depth of the body has increased greatly, and the 
vent as well as the anal and pelvic fins have all moved forward ; larvae of this size are 
said to be in the transition stage. However, this stage may be attained by the spring fish 
at a length of 31-44 mm, and by the fall fish at 44-60 mm (29: 363-366, figs. a-g). 



Fishes of the Western North Atlantic 279 

Young fish from the Clyde Sea area of England lost their yolksac and began 
developing the dorsal fin at about 10 mm; the end of the notochord turned up at about 
17 mm; pelvic fins appeared at about 22 mm; the caecal part of the stomach began to 
form and the air bladder became prominent at a length of 30—35 mm; and the least 
length at which metamorphosis took place was 42 mm {gi: 248, 252). Marshall, et al. 
also recorded a regular weekly increase of 3 mm prior to metamorphosis, at a length of 
about 35— 40 mm. This rate of growth of the larvae is in fair agreement with that 
reported for elsewhere in Europe. 

Huntsman observed that, on the American coast, as at Grand Manan Island, 
New Brunswick, larvae hatched in late summer and early autumn had attained a length 
of 1 7-20 mm by the end of November or early in December. Fry 50— 60 mm long 
were abundant in St. Andrews Bay in June. Yearlings were 90—125 mm long in the 
fall of the year in the Bay of Fundy. This rate of growth is in agreement with that ob- 
served by European investigators in examples from Norwegian waters and from the 
North Sea. A one-year old Atlantic Herring, then, is about 88—125 mm long. 

These fish do not fall into definite size groups when more than one year old. But 
their subsequent rate of growth may be determined by segregating the fish into year- 
classes through scale reading, for the so-called winter rings are so pronounced that they 
can be relied upon with considerable confidence to show the age of the individual 
specimen.^" Thus, Meek (pj: 76) has indicated the following average lengths for those 
caught on the Dogger Banks in the North Sea: 98 mm at one year, 183 at two, 220 at 
three, 241 at four, 255 at five, 266 at six, 275 at seven, 281 at eight, 287 at nine, 291 
at ten, 293 at eleven, and 295 mm at twelve years of age. The Norwegian fish spawned 
in 1899 grew more slowly at first than the Dogger Bank fish but about equaled the 
latter in size by their sixth year. Fish up to 19 years old have been found in many 
Canadian samples (<5j: 10) and some 20 years of age have been seen in Europe. How- 
ever, the rate of growth after age 1 2 is extremely slow. According to observations by 
Huntsman and by Lea, the rate of growth is about the same for Atlantic Herring both 
here and in Europe, with Bay of Fundy fish growing a little faster than those spawned 
in the Gulf of St. Lawrence, and with the fish spawned along the outer coast of Nova 
Scotia growing a little faster than those produced in the Bay of Fundy. 

On the coast of Maine, some individuals reach sexual maturity at a length of about 
235 mm (9.5 in.), when three or four years old. Usually they do not spawn until they 
are 250—262 mm (10— 10.5 in.) long, at five to six years of age. Most of the individuals 
in the schools of spawning fish are 300— 325 mm (12—13 i'^O l°"g) the majority of 
which is then perhaps 10 years of age or more. It is believed that Atlantic Herring, 
after reaching sexual maturity, spawn every year as long as they live. 

Atlantic Herring may be divided roughly into three growth stages," each with its 

10. For a detailed study of the scale of the Atlantic Herring in relation to growth, see Lea (75:81-93). 

11. Sometimes a greater number of growth stages has been recognized. Thus, Johnson (69:351), following the nomen- 
clature of the fishermen of Passamaquoddy Bay, listed "brit" 8o-ioomm, "snippers" 1 10-130 mm, "oils" 140- 
180 mm, and "mustards" 190-230 mm. In New England at least, the term "sperling" is used for young fish 75- 
125 mm long; they therefore come within the range of the sardine group, as used by Bigelow and Welsh (j6:99). 



2 8 o Memoir Sears Foundation for Marine Research 

own general distribution and seasonal migration. The first stage, the "sardine," consists 
of one- and two-year-old fish 45— 200 mm long; the second is the "fat" stage, con- 
sisting of sexually immature fish about two years old, 190— 200 mm long; and the 
third or "spawn" stage is a mature fish (see also pp. 283, 284). 

Spawning Seasons. Spawning takes place in the spring, summer, or autumn, or 
in both spring and autumn, depending on the locality. Apparently there is no time in 
the year when spawning, off Europe at least, does not take place somewhere. On the 
southern coast of Newfoundland and in the Gulf of St. Lawrence, they are spring 
spawners, commencing in May or even as early as the latter part of April. In the southern 
and inner part of the Gulf as a whole, the season lasts only a few weeks. But on the 
south coast of Newfoundland, and perhaps on the west coast as well, spawning may 
continue until July, and some fish spawn in the Gaspe-Bay of Chaleur region in Sep- 
tember. To the south, on the Atlantic coast of Cape Breton, Nova Scotia, spawning 
occurs chiefly in May or June, but also in August and September; and on the outer 
coast of this province to the south of Halifax, from August through September (^^ : 
405 ; 60 : I o ; (5j : 1 1 , fig. 6). 

The Bay of Fundy once harbored both spring and summer-autumn spawning 
schools, as did the outer coasts of Nova Scotia (by fishermen's reports); but they seem 
never to have been numerous; it is doubtful whether any of the spring spawners still 
exist anywhere south of the Gulf of St. Lawrence. At the mouth of the Bay of Fundy 
and off the easternmost part of the Maine coast, the heaviest spawning takes place in 
July-August and September in some years, with spawning continuing late in the fall; 
but in other years it does not commence until August, to end early in October. Passing 
westward— southward, the spawning run is progressively later and shorter: from mid- 
August until October in the general vicinity of Mt. Desert Island; in October along 
the east coast of Massachusetts (where only a few spawn); and early November near 
Woods Hole, Massachusetts. 

Off northern Europe, where two principal races are recognized, namely coastal 
herring and sea herring, the coastal fish spawn mostly in brackish water and princi- 
pally in the spring months, whereas the sea herring spawn in the open sea over widely 
distributed times of the year, but mostly in the late summer, autumn, and winter. The 
spawning periods of the different stocks are so varied that one may find spawning 
fish somewhere at almost any time of the year (29: 362). The Pacific Herring has 
only one spawning period, which may vary between December and June; the earliest 
spawning takes place in California and the latest in northern Alaska (iii: 278). 

Spawning Grounds and Depths. In northern European waters, as already indicated, 
at least some fish spawn in shallow inshore brackish water whereas others spawn in the 
open sea. But in the western side of the Atlantic they are not known to spawn in water 
less saline than about 31.9 "/oo or more saline than about 33 "/oo. In the Gulf of Maine, 
spawning takes place at temperatures of about 46—55°. For the more northern part 
of the American range, however, precise information in this regard is lacking. 

Off America, Atlantic Herring spawn in greater or fewer numbers throughout 



Fishes of the Western North Atlantic 281 

their geographic range, from north to south. Around Newfoundland they spawn chiefly 
within the bays. However, from the southern side of the Gulf of St. Lawrence south- 
ward as far as southern Maine there is no considerable sector of the coast, except a 
part of the Bay of Fundy (p. 290), that does not see a more or less plentiful spawning 
run. Moore has spoken of the northern coastline of the Gulf of Maine as "a con- 
tinuous spawning ground," succeeded by scattered grounds wherever the bottom is 
suitable, extending as far southward as Block Island off the Rhode Island coast {q^ : 408). 
However, local spawning grounds may be abandoned for a term of years — a common 
occurrence {g^\ 430). 

In the Gulf of St. Lawrence, Canada, they have been described as spawning in 
water so shallow that the eggs are sometimes washed up on the beach by the waves 
{()^: 411), as happens every year in northern European waters {l^\ 99). In general, 
however, our Atlantic Herring spawn at depths of, say, 2-30 fms. Thus spawning takes 
place both along shore and on the various ledges that lie between 5-25 miles off the 
coast, where eggs are often found sticking to the anchor lines of fishing vessels. But the 
nature of the bottom makes it Improbable that any eggs are laid at a depth greater 
than 75 fms. on the American side, though spawning at a depth of 100 fms. has been 
reported for Scandinavian waters {q^: 411). Whether, or how regularly, the fish may 
spawn on the shallower parts of the offshore fishing areas, from Georges to Grand 
banks, remains to be learned. 

Habits?-'^ Herring are primarily wanderers of the open sea. While they also enter 
bays and estauaries freely, they seem never to have been reported reliably from water 
that Is more than slightly brackish; along the American coast a salinity of perhaps 
2.8 "/oo is about the lower limit of tolerance. 

As a rule, all sizes of Atlantic Herring congregate In schools of hundreds or 
thousands of individuals, with a given school usually consisting of Individuals of about 
the same age and size. How long a school may preserve Its identity as such Is not 
known, for while Fridriksson and Aasen have found that specimens tagged and released 
together may be recaptured at widely separated localities (40: 22), this may not apply 
to schools that have assembled under natural conditions. 

When a school is at the surface In the daytime, as often happens In calm 
weather. It is betrayed by a fine rippling of the water; but the fish do not ordinarily 
"fin" or lift their noses above the water, as Is characteristic among menhaden 
(p. 354). At night they are betrayed at the surface by their luminous trails If the 
water Is "firing." 

A school of Atlantic Herring Is not as easily frightened as a school of mackerel 
by an approaching boat. Neither is it usual for them to leap clear of the water, except 
when young fry are pursued from below by larger pelagic fish. 

A school that Is actively feeding is likely to be more or less stationary, drifting as 
a whole with the current. At other times schools are seen traveling as If with purpose- 
ful Intent, all of the members swimming side by side In the same direction. Schools 

12. Condensed from Bigelow and Schroeder {i$\ 89). 



282 Memoir Sears Foundation for Marine Research 

of sardine-sized fry may often be seen streaming past some rocky headland, seemingly 
in an endless procession. What the directive stimulus may be we have yet to learn, for 
as Huntsman has pointed out, "There is no indication that herring swim against 
the current unless the water is somewhat turbulent" {65: 83). Whether they make or 
lose headway over the bottom while doing so depends on their swimming speed relative 
to the strength of the current. 

No information is available as to how rapidly a school may travel at sea. But Frid- 
riksson and Aasen have found that specimens held alive in nets swam constantly at a rate 
of about 0.2-0.25 sea miles per hour (6-8 m/min.) when not disturbed. It Is certain 
that they are capable of long journeys, for Individuals tagged on the east coast of 
Iceland have been recaptured In southern Norway, and vice versa (40: 26, 27). 

How active Atlantic Herring are at any time and place depends primarily on 
the water temperature. In the Bay of Fundy region, for example, they "move very 
sluggishly when the water is coldest. In February and March," I. e. when the upper 20 
fms. or so have cooled to about 32-36°. They become active again when the water has 
warmed to about 40-43°. 

While the vertical range of this species extends down to an extreme depth of 
about 100 fms., the vast majority lives at depths shallower than 40-50 fms., living 
pelagically in the upper water layers for most of the year but close to the bottom during 
the coldest months. 

Relation to Light. That light is an Important factor in controlling their vertical 
distribution, at least in southern New Brunswick, was discovered by Johnson in 1939 
{6g: 349-354)- He found that at night, in the absence of appreciable light, all sizes 
were quite near the surface. During weak light, as at dawn and dusk, all sizes still were 
a foot or so from the surface. However, as the sun rose above the horizon they left 
the surface, the largest fish, 190-230 mm long, descending to a depth greater than 
10 feet and those 140-180 mm long to a depth of 10 feet or more; the smaller ones in 
general were nearer the surface. Also, on cloudy days all sizes were closer to the surface 
than on clear days. 

Migratory Movements. Because of the economic value of this species, knowledge 
of the migrational habits, which would make it possible to predict runs in advance of 
their occurrence, has been and continues to be sought on both continents. In Europe 
at least, these studies have been complicated by the presence of a large but In- 
determinate number of local populations or races (p. 287), each with its own mi- 
gratory pattern, a fact not known until comparatively recently and even now not well 
understood. As Norman (jo J: 263) wrote: 

At some seasons herrings may be found in huge numbers in a given locality (in Europe), at others they 
will disappear almost entirely; in other places they may be caught all the year round, but the numbers caught 
on a given ground may exhibit an immense amount of variation from one season to another. 

To explain these long known and very obvious facts, an older view, widely held 
until about 1825, was that Atlantic Herring as a whole withdraw in winter to Arctic 



Fishes of the Western North Atlantic 283 

waters, to return to more temperate coasts in spring, some of them to the eastern side 
of the North Atlantic, others to the western side (for further details, see Meek, pj: 
67-87). But Perley wrote as early as 1852 that this idea was "supposed to be wholly 
imaginary," it being generally believed then "that the herring fattens in the depths 
of the ocean, and approaches the shore in shoals merely for the purpose of depositing 
its spawn" {104: 206). Successive and long continued studies at many hands have 
since proven that in reality we are not dealing with any widespread mass migrations, 
but with short-range movements (inshore and offshore) of local bodies of fish, each 
with its own area of occurrence and each of which may include subpopulations that 
spawn at different times of year, as Norman has pointed out. It is equally clear that the 
ranges of the races of European fish overlap — highly probable at least that more or 
less interchange is constantly taking place between adjacent races. 

However this may be, the basic migratory pattern, essentially the same for all of 
them, may be conveniently divided into the three successive phases noted previously — 
sardine, fat, and spawn. 

Those of the sardine stage, 45-200 mm long, tend to remain near their birth- 
place. They probably spend the winters on the bottom in a few fathoms, appear in the 
inshore waters of New England, New Brunswick, and Nova Scotia in the spring as 
one- and two-year-old fish, and remain there through the summer and autumn. 
How far they may journey during this period and in what direction depends chiefly 
on the movements of the water. For example, every summer untold millions of sardines 
congregate in the Passamaquoddy region at the entrance to the Bay of Fundy (p. 290), 
and it is now known^' that the responsible factor is the prevailing movement shoreward 
of the subsurface water in which the sardines are swimming; this shoreward flow is 
motivated by the offshore movement of surface flow from tributary streams combined 
with superficial offshore currents set up by the wind. Sardine-sized fish, in other 
words, drift with the current much as do any planktonic animals, such as the euphausiid 
shrimps and copepods on which they feed. In technical language, they are denatant, 
which applies equally wherever young Atlantic Herring are produced. 

The fat stage consists of sexually immature fish about two years old and 1 90— 
200 mm long that have accumulated a large amount of fat around the viscera and 
among the body tissues. During the year's period that precedes their sexual maturity, 
they disperse much more widely than do the sardines. Whether this wide scattering 
is brought about wholly by transport of water movements or whether directive swim- 
ming plays an important role remains a mystery. In either case they are encountered 
anywhere and everywhere throughout the range of this species. In the Gulf of 
Maine, for example, where more attention has been paid to the movements of these 
fish than has been the case anywhere else in the western North Atlantic, the fat 
stage has been encountered indifferently close to shore, as at the mouth of the Bay 
of Fundy in the northeast, and offshore in the Massachusetts Bay region in the 

13. Studies carried out from the Atlantic Biological Station at St. Andrews, chiefly under Dr. A. G. Huntsman's 
leadership (65:95, 96)- 



284 Memoir Sears Foundation for Marine Research 

southwest, in the central parts of the Gulf, and over Georges Bank. Those picked 
up during the warmer months either in mackerel seines or in otter trawls are mostly 
very fat and show no signs of approaching sexual maturity. It seems that these "fat" 
fish, like the sardines, merely sink down close to the bottom for the winter wherever 
they may be. 

About all that is known of the mature spawn stage in the western Atlantic is that 
these fish live mostly some distance offshore near bottom and appear in vast num- 
bers in their spawning areas at spawning time. In the Gulf of Maine, most of them 
seem to drop out of sight after having spawned; the large ones caught there out of 
spawning season fall in the fat category. But there is no reason to suppose that they 
travel far during the interval (between spawning time and winter) when they are 
feeding greedily to recover condition sufficiently for spawning again the next year. In 
fact, it is probable that they, like the other two stages in the Gulf of Maine, merely 
descend into deeper water to winter, as has long been known to be their custom in 
European waters. How deep the great body of them may go is not known, but the 
contour of our continental shelf is such that no fish need swim any great distance to 
reach water deeper than 50—75 fms. anywhere between Cape Cod and the Gulf of 
St. Lawrence. It seems likely also that they spend the winter in deep water in the 
southern part of the Gulf of St. Lawrence, but definite information in this regard 
is lacking. On the southern and western coasts of Newfoundland, however, where 
spawning takes place in the spring, the mature fish "move off into deeper water in 
summer" (J25: 2,^) but return late in the autumn to their natal bays, where they 
have long supported important November— January fisheries. 

Food and Feeding. Atlantic Herring feed on plankton — first on diatoms and other 
unicellular organisms, then on copepods and their eggs following the yolksac stage 
(9J: 252). In summing up her rather extensive investigations at Plymouth, England, 
Lebour {jg: 463) listed the following by stages of growth; before disappearance of 
the yolksac: larval gastropods, green food (probably diatoms and flagellates), larval 
bivalves, nauplii and other young stages of small Crustacea, as well as the eggs; after 
disappearance of the yolksac up to about I2 mm: the same, but with small adult 
copepods added; after 12 mm and probably until metamorphosis: copepods; after 
metamorphosis: copepods, decapod crustaceans, amphipods, and fish. 

Only two kinds of food were found in about 1 500 examples taken off East- 
port, Maine [g^: 401). Moore said that "One of these [foods] consisted of copepods 
('red seed'), which appeared to constitute the sole food of the small herring, the so- 
called brit, and a considerable portion of that of the larger individuals from 5^/2 inches 
upward." But the principal food of the larger fish was euphausiid crustaceans 
{Meganyctiphanes norwegicd), known to local fishermen as "shrimp." These are so Im- 
portant In the diet of this species that it seems likely that the appearances and disap- 
pearances of large fish in the open waters of the Gulf are correlated with the presence 
or absence of euphausiids of one kind or another. Concerning their capture, Moore 
remarked : 



Fishes of the Western North Atlantic 285 

To capture such prey requires some address on the part of the herring; they frequently throw themselves 
almost clear of the surface and their splashings. . . are audible at a considerable distance. When feeding on 
copepods the movements of the herring are less impetuous. They swim open-mouthed, ^^ often with their snouts 
at the surface crossing and recrossing on their tracks and eventually straining out the minute crustaceans by means 
of their branchial sieves. 

The sieves were described by Moore (95: 401) as follows: 

An examination of the mouth cavity of the herring wiU disclose a series of long bristle-like processes, 
the gill rakers, projecting from the anterior face of each gill arch, like the teeth of a comb. When the mouth is 
opened widely the tips of the gill rakers stand apart, but when it is closed or partly closed they become more 
closely approximated and each series is pressed closely against the inner face of the series attached to the arch 
next in front. There is thus formed a beautifully fine and effective sieve, capable of retaining small bodies contained 
in the water taken in at the mouth and discharged through the gill slits. 

The amphipod genus Euthemisto is also an important food for this species off 
Europe ; that this is not so in the Gulf of Maine no doubt reflects the scarcity of this 
pelagic crustacean in our coastal waters. In default of an abundant supply of Crustacea, 
and sometimes even when these are plentiful, the fish feed on whatever other kinds of 
smaller planktonic animals and plants may be available. Small fish of various kinds 
have been found in their stomachs: capelin, especially in Newfoundland waters in 
winter; and grass shrimps (Crago), schizopod shrimps (Mysis), amphipods (Gammarus), 
small squid, annelid worms, and crab larvae at Woods Hole, Massachusetts. But the 
smaller planktonic plants, whether diatom or peridonidian, have never been found in 
specimens longer than 1 5 mm, probably because their gill sieves are not fine enough 
to retain them. 

Enemies. Their chief natural enemies include all kinds of marine predators, such 
as cod, pollock, haddock, silver hake, mackerel, dogfish and other sharks, finback 
whales, and the common squid {16: 104). An attack on a school of young Atlantic 
Herring by squid has been described by Hildebrand and Schroeder {^g : 82) as follows: 

We had occasion to watch the wholesale destruction of 2 to 4 inch herring during June 1925, on the flats 
about Provincetown, Mass. Schools of 10 to perhaps 50 squids circled around a school of herring until they had 
bunched their prey into a compact mass. Individual squids then darted in and seized one, sometimes two, and then 
darted back for more. Along the beach there remained a silvery streak of dead herring. 

A similar firsthand account of silver hake {Merluccius bilinearis) driving a school of large 
Atlantic Herring up on the beach in Massachusetts Bay has been reported {15: 90). 
Parasites. Among the European fish, "cestode larvae and trematodes were fre- 
quently found as parasites in the gut. The latter were common and the incidence ap- 
peared to increase as the herring grew, as many as 29-30 per cent being parasitized 
in the large catches of May 2 1 and 30. Cestode larvae were less numerous, and they 
too were more frequent in the older fish" (Marshall, et al., gi: 253). No fatalities 
were specifically ascribed to these parasites; Sindermann and Rosenfield have found 
that infestation with trematode larvae may be fatal — at least under experimental con- 

14. That the larger herring "swim open-mouthed" when feeding on copepods has been questioned by Batde, et al., 
who are of the opinion that the fish seize the organisms individually (7:411). However, as to the organisms ingested 
there is general agreement. 



2 86 Memoir Sears Foundation for Marine Research 

ditions — "provided they are present in numbers large enough to overwhelm the host" 
{liy. 12). Furthermore, it has been reported that nematodes, cestodes, and trematodes 
occur in the viscera and sporozoa in the muscles of small fish taken at Woods Hole, 
Massachusetts {8^: 473). 

Mass Mortalities and Diseases. The literature that deals with the Atlantic Herring 
contains many accounts of their great destruction on both sides of the North Atlantic. 
For June and July 19 14, Cox reported widespread mortality on both the southern and 
western sides of the Gulf of St. Lawrence {28: 82—85). -^ ^^^ years later, a spectacular 
event occurred in Cohasset Harbor, Massachusetts Bay, on October 5, 1920, when an 
estimated 20,000 barrels of fish 100—125 mm long became stranded on the mud with 
a falling tide {16: 104). On October 10 a second but smaller run and on October 15 
a run as large as the first were stranded similarly. The exact reason for the stranding 
and mass destruction was not determined, but fishermen thought that the fish had 
been driven into the nearly landlocked harbor by silver hake, Merluccius bilinearis. 
Other mass strandings of small individuals have also been reported, for Manchester 
Harbor, Massachusetts {l^\ 327). Pursuit by predatory fishes is the most likely cause 
of simultaneous mass strandings within small areas; for an eyewitness account of one 
such event, see page 285. 

But enough evidence has now accumulated to make it practically certain that 
when mortality is very widespread it has been caused by disease of one kind or 
another. Thus Cox found that fish from the Gulf of St. Lawrence case of 19 14 (above) 
were so evidently in a diseased condition and were so heavily infested with a myxo- 
sporidian protozoan that it seems practically certain that the latter had been responsible 
for the mortality {28: 82—85). It is known now that those along the coasts of New 
England, and presumably in more northern waters, suffer from fungous disease caused 
by Ichthyosporidium, from ulcer disease caused by protozoa, from pigment spot disease 
caused by larval trematode worms, and from tail rot caused by bacteria; it is known 
also that fungous disease in particular, and probably the others as well, sometimes 
reaches epidemic proportions, with mortality so widespread and so severe that the sardine 
fishery is very seriously affected. ^^ 

Relationships. Although the Pacific herring has been recognized as being specifi- 
cally distinct from the North Atlantic Herring and has for many years been designated 
Clupea pallasii Cuvier and Valenciennes, it is difficult in the light of rather recent studies 
to find distinguishing characters. This difficulty was pointed out by Rounsefell, who 
made some studies of the relationship himself and reviewed the findings of several 
European investigators {ill: 243). The distinguishing character given by Regan in 
his revision of the Clupeidae is "Ventral scutes all keeled" in harengus and "Ventral 
scutes in front of pelvic fins not keeled" in pallasii {106: 227). According to the material 
now at hand, the scutes in pallasii are merely less strongly developed. ^^ 

15. For a general survey, with photographs, of diseased fish, and references, see Sindermann and Rosenfield {iiy). 

16. Relationships with the species of temperate southern hemisphere waters have not been adequately worked out. 
Regan, the most recent revisor, had inadequate material (jo6:228). Norman did not attempt to compare the South 



Fishes of the IVestern North Atlantic 287 

P'ariations. It has long been generally appreciated that Atlantic Herring in the east 
do not constitute a homogeneous population, since a number of regional races is in- 
cluded (pp. 282, 283). Thus "it is possible," writes Norman, "to recognize North Sea, 
Baltic, Norwegian, Icelandic herring" (lOi). These races differ from each other in such 
morphologic characters as average proportional dimensions and average numbers of 
fin rays and vertebrae. But we ought to caution the nontechnical reader that these are 
numerical averages for many specimens and are not used for the identification of an 
individual specimen, or of even a few fish; only when the number of specimens examined 
is large are such figures significant. For example, if the average number of vertebrae 
was given as 56.3-56.5 for one locality and as 56.6-56.8 for another, it means simply 
that in the first case there were more fish with 56 vertebrae and less with 57; and in 
the second case, more fish had 57 and less had 56. No fish ever has a fractional num- 
ber of vertebrae." 

Each of these races has its own chief center of abundance, its own pattern of sea- 
sonal migration, and its own spawning season or seasons, for some of the races include 
both early and late spawners. The races differ in the average size to which their mem- 
bers grow. A knowledge of the times and localities where these populations are catch- 
able in greatest amount is of so much commercial importance that these matters are 
under continuous survey by the fisheries services of the countries chiefly concerned. 

In 1914, Hjort found (60: 9-12) in the western Atlantic Herring: that the average 
number of keeled ventral scales is larger (13—14) for the autumn spawners of outer 
Nova Scotia and the Gulf of Maine than for the spring spawners of the Gulf of St. 
Lawrence; that the number of vertebrae is greatest (55-59, av. 56.8) in those from the 
west coast of Newfoundland; that the number of dorsal and anal rays averages higher 
in those caught in the open sea than in those from the more enclosed waters of the 
Gulf of St. Lawrence and Bay of Fundy; and that, among the spring spawning popula- 
tions, the average number of vertebrae, fin rays, and keeled scales is higher in those 
from the west coast of Newfoundland than in those from the southern part of the 
Gulf of St. Lawrence. But the fishery for "fat" and for "spawn" fish plays so small a 
role in the general welfare of Canada and the United States (p. 288) that no additional 
information seems to have been contributed to the racial question on this side of the 
Atlantic. From analyses of extensive data, however. Day, Leim, and Tibbo showed that 
Canadian Atlantic Herring represent six more or less distinct populations, differing 
from each other in rate of growth and in average number of vertebrae : one in the northern 
part of the Gulf of St. Lawrence, a second in the Bay of Chaleur, a third in the southern 
part of the Gulf, a fourth oft" the outer Nova Scotian coast, a fifth along southern New- 
foundland, and a sixth along western Newfoundland (see Leim, 83: 11, figs. 6, 12). 

Commercial Importance. From an economic standpoint, these fish are perhaps the 
most important in the world. In addition to their great value as food for man, many 

American forms with the northern ones (102:37). Regan's Ctupea holodon from New Zealand is now known as 
Clupea antipodum (Hector). — G. S. Myers. 
17. For a general survey of the races of eastern Atlantic Herring, see especially Lubbert and Ehrenbaum (Sp : 14-19); 
for a detailed study of the average number of vertebrae in the different races, see LeGaU (50:167-170). 



2 88 Memoir Sears Foundation for Marine Research 

are used for bait, and they constitute an important item in the diet of other fish that 
are of great economic importance. The catches have fluctuated greatly from year to year, 
but they have not always provided a criterion of the supply available. In America, at 
least, the catches have often been governed by demand and profit. Therefore the fishing 
efforts from year to year have not been uniform. Through 1952, the annual catches 
on the American side have fluctuated between 172,000,000-359,000,000 pounds,'* 
the major part of the catch from Canada and Newfoundland. The catch in Europe in 
1948 was 4,150,000,000 pounds (1,881,713,000 kilo; 24: 6), and in 1956, the latest 
year for which information is conveniently available, it was 5,858,650,000 pounds 
(2,590,715,000 kilo). European waters, in short, have yielded annually something like 
15 times the catch (in pounds) taken in American waters. However, the apparent dis- 
crepancy in relative productivity does not loom so large when we note that the bulk 
of the American catch is concentrated within a latitudinal belt of about 8° 30' at most, 
while the European catch is distributed along something like i9°3o', or a distance 
nearly 2.5 times greater. 

Small fish, 75— 125 mm or so in length, are canned in large quantities as sar- 
dines, not only in Norway but along the easternmost coast of Maine and in the region 
of Passamaquoddy Bay at the mouth of the Bay of Fundy on the New Brunswick side; 
these last account for the concentration of Atlantic Herring landings there. Elsewhere 
the American catch consists chiefly of the "fat" and "spawn" stages. Some of the large 
ones are consumed fresh, others are salted or pickled, some are salted and smoked in 
various ways, and still others are canned. 

The three principal types of smoked Herring are the kipper, bloater, and red her- 
ring. The kipper is split down the back, Immersed In brine for a very short period of 
time, slightly dried, and then smoked for several hours. The bloater is cured without 
splitting, only the belly being cut open for cleaning; It is lightly salted and smoked 
only long enough to dry the flesh. The red herring, like the bloater. Is not split, but It 
is much more heavily salted (buried in salt for at least five days) and then smoked for 
ten days or so. Thus the last is a fairly durable product, which can be shipped to distant 
markets, whereas the kipper and bloater are perishable. The several curing processes 
are practiced much more extensively In Europe than in America and are sometimes 
carried out aboard fishing vessels. Large quantities are also frozen, to be used as bait 
by the hand-and-long-line fishermen: for example, 32,000,000 pounds in Canada (In- 
cluding Newfoundland) in 1954. While corresponding information is not conveniently 
available for the United States, this time-honored method of fishing Is now far out- 
stripped by the otter trawlers. The landings by line fishermen in Maine and Massa- 
chusetts in 1954 were 9,000,000 pounds, the greater part of which was hooked on 
frozen herring. The offal at the canneries and occasionally surplus catches of whole 
fish are reduced to oil, fish meal, and fertilizer. 

Fluctuations in Abundance and Catch ; Year-Classes. It Is notable that a fishery such 
as the one for this species in European waters, exploited for centuries, has apparently 

18. Statistics from publications issued by the governments of the respective countries named. 



Fishes of the Western North Atlantic 289 

shown no positive or permanent decline. Of course it has been claimed from time to 
time that signs of exhaustion were evident, but such claims seem to have been based 
on the scarcity or disappearance of the fish in certain areas, not on the population as a 
whole. There certainly is insufficient information concerning the American stocks to 
furnish a basis for an answer. If a decline has taken place in American waters, it could 
scarcely be ascribed to overfishing, as the supply of fish generally has exceeded the de- 
mand. The fluctuations in the annual catches, then, do not necessarily mean that the 
fish were scarce when a small quantity was caught, or were abundant when a large catch 
was taken. Neither does a small annual catch necessarily indicate that the demand was 
light. It may mean only that the fishermen were operating in an area where the fish 
are usually abundant but were absent there that year. The fish are influenced in their 
migrations by the weather and by the presence of food, which seem to account for their 
abundance in one area one year and elsewhere the next. At present it is impossible 
to predict the time of arrival of schools in any given area. 

From his studies of Atlantic Herring from the English Channel, Ford wrote, 
"If it could be shown that the variation from season to season followed some rule, it 
might yet be possible to discover that rule from the study of past data and hence to predict 
what Is likely to occur in the future" (j6: 369). But all uniformity seems to be lacking. 
"It can be regarded as certain that fishing results are dependent upon the weather, 
but In the present state of meteorological science, we cannot hope to receive much 
advanced information regarding weather conditions during a forthcoming season." 
Nevertheless, he found that daily forecasts had some value, for a gale warning not only 
indicated an interruption of operations, but It also suggested heavy catches when fishing 
was resumed. The prediction of calm weather, on the other hand, suggested lighter 
catches. 

The many age analyses of Atlantic Herring, chiefly of those from European waters, 
have shown that a very large crop of young may survive in some years while very 
few do In other years, especially In more northerly spawning regions. Consequently, 
those spawned In a favorable year may dominate the population for several years to 
come. The classic example is the year-class produced In 1904 that dominated the 
Norwegian catches for the next six years and dominated catches on the west coast of 
Newfoundland until 19 14 and even 191 5. In the Gulf of St. Lawrence, of late years, 
the broods of 1939, i94i> I943) and 1944 were highly productive, whereas those of 
1940, 1942, and 1945 contributed but little to the population (<5j: 107). There is 
also the well-documented case of a particular body of Bay of Fundy fish that did 
not receive any significant recruitment for some 10 years, by which time the few re- 
maining fish seem to have dropped out of the picture due to old age {l$: 93, 99). Con- 
sequently, the successes or failures of reproduction In given years are reflected In the 
catches. Thus Ford (36: 370) was able to predict which year-classes would predominate 
in the catches of certain years, as well as the probable proportion of large fish, for as 
a rule old fish are larger than the younger ones, though the fish In some areas grow 
faster than those In others. Ford concluded his discussion on predictions as follows: 



290 Memoir Sears Foundation for Marine Research 

In practice it is possible to obtain advance information of the relative richness of broods and hence of the 
relative success of future fisheries. But there still remain possible influences which militate against a forecast 
of yield based entirely on this evidence, chief among these being the factors which govern migrations of herring. 

Methods of Fishing. Most of the fish are caught in traps known as wiers, in floating 
traps, stop seines, purse seines, and drift nets; in the American bays and harbors, large 
numbers, especially of the smaller sizes, are taken by "torching," i.e. by netting fish 
attracted by a light. 

Range. Atlantic Herring in American waters range from Greenland to North 
Carolina, but usually not in commercial abundance south of New Jersey; they are most 
abundant north of Cape Cod. In Europe, where certain races enter brackish water, 
they range from Spitsbergen, Novaya Zemlya, the Kara Sea, the White Sea, and 
northern Norway south to the Straits of Gibraltar." 

Occurrence in the Western North Atlantic. Stray specimens have been reported for 
as far south as Cape Hatteras, the lower part of Chesapeake Bay, Maryland, and New 
Jersey {^g: 82; lig: 120; jy. 93); and only irregularly do a few visit the coasts of 
New York and Rhode Island. But they are seasonally abundant and generally distrib- 
uted throughout the Woods Hole region, though not in any great numbers {123: 741). 
Although they appear sporadically in the Massachusetts Bay region, they are plentiful 
enough in some years to support local fisheries. Along the coast of eastern Maine, and 
in the Passamaquoddy region at the mouth of the Bay of Fundyj^" they are not only 
far more plentiful but more regular in their comings and goings. But farther up the 
Bay of Fundy, on the New Brunswick side and at its head, they are far less plentiful, 
probably because of the turbidity of the water there. However, along the Nova Scotian 
side, toward the mouth of the Bay, they are sufficiently abundant so that they yielded 
between five and six million pounds in 1946 (Annapolis and Digby counties); indeed in 
the general vicinity of Yarmouth on the west coast of the province they are so plenti- 
ful that great numbers of giant tuna congregate there summer after summer to harry 
and destroy the schools in the local tide rips.-^ Thence eastward along the outer coast of 
Nova Scotia they are abundant and generally distributed; witness a reported catch be- 
tween the vicinity of Cape Sable and Cape Breton of something more than 4 1 million 
pounds in 1946, and 21 million pounds in 1956, the latest year for which pertinent 
statistics are readily available. — h.b.b. 

There are fewer, it seems, along the Nova Scotian shore line of the Gulf of St. 
Lawrence; landings there in 1946 were 1,330,000. But thence westward these fish 
are numerous again throughout the southern side of the Gulf,^- around Anticosti 
Island (described there as in "bancs enormes"; lid: 285) and along the Gaspe shore 
in the lower part of the St. Lawrence estuary. But they are progressively fewer up the 

19. "With the favourable climatic conditions in Arctic regions, the Herring has spread to Jan Mayen, Spitzbergen, 
Nowaja Zemlya and Kara Sea" (Jensen, 67:64). 

20. For details as to their regional and seasonal occurrence in the Gulf of Maine, see Bigelow and Schroeder (X5:93)- 

21. The annual "Tuna Tournament" is held at Wedgeport, Nova Scoda, not far from Yarmouth. 

22. 1946 catches, 36,103,600 for the New Brunswick shore of the Gulf; 6,080,100 for Prince Edward Island; 15,078,000 
for the Magdalens; 15,525,000 for Bonaventure and Gaspe counties, Quebec. 



Fishes of the Western North Atlantic 291 

St. Lawrence, and we find no report of them beyond Trois Pistoles, Quebec, about 
opposite the Saguenay River. Although the area of abundance does not extend to the 
northern shore of the Gulf, they are widespread there as well as in the Strait of Belle 
Isle; the 1946 catch for this entire coastline was only 923,700 pounds. They have 
been reported also for the banks in the northeastern corner of the Gulf between New- 
foundland and Labrador (125). 

They are so plentiful along the Newfoundland shores of the Gulf that they have 
long supported important fisheries in and off Bay of Islands and at Bonne Bay. Indeed, 
"it seems proven . . . that practically all the Newfoundland bays are frequented by 
spawning herring often in great numbers" (Thompson, 12 5: 36). Thus Fortune and 
Placentia bays on the eastern part of the southern coast have been the sites of produc- 
tive Atlantic Herring fisheries since the days of the early cod fishery; the local catch 
is used chiefly as cod bait. 

The reported commercial catch of Atlantic Herring in 1946 for Newfoundland as 
a whole was only about 6^ "/o of that for the southern part of the Gulf of St. Lawrence 
on the Canadian side^^. This, taken at face value, suggests that the cold waters of eastern 
Newfoundland produce fewer Atlantic Herring than the waters of the south and west 
coasts. However this may be, they are at least moderately plentiful on the east coast 
of Newfoundland right up to the entrance of the Strait of Belle Isle; Jeffers not 
only reported them for Rahleigh but stated that they were numerous enough at Quirpon 
to supply the local inhabitants with winter food for their sledge dogs {66: 206). They 
have also been reported for the northern side of the Strait at Blanc Sablon (121 : 124), 
and repeatedly for the Strait without specified locality. 

But a review (perhaps not sufficiently searching) of the various lists of Labrador 
fishes has not yielded any reference to their presence anywhere to the north of the 
Strait, apart from Stearns' report that they are abundant "north of Blanc Sablon, 
growing more and more so all along the Labrador coast, the farther down which are 
the greatest catches" (121: 124). 

The offshore boundary of their range follows in general the break in slope at the 
outer edge of the American continental shelf. Schools are occasionally seen at the surface 
or are picked up by trawlers here or there on Georges and Browns banks, fronting the 
Gulf of Maine (J5: 95). It seems the situation is similar on the Nova Scotian Banks 
eastward to Banquereau and on the Newfoundland Banks, where they are often taken 
by otter trawlers in quantities {125: 36). All this has long been common knowledge 
among the banks fishermen. During the early years of the Georges Bank cod fishery, 
it was common practice to set herring drift nets for bait (j6: 102). And it may well 
prove that they are much more plentiful on the banks than is now realized, for while 
the numbers reported (75: 95, for Georges Bank) have been insignificant, no 
serious attempt has yet been made to sample the offshore population with drift nets, 
midwater trawls, or bottom trawls of a kind adapted to the capture of this partic- 
ular fish. 

23. Data on Canadian catches are published annually in "Fisheries Statistics of Canada." 

i9« 



292 Memoir Sears Foundation for Marine Research 

It has been known since Fabricius' day (j^) that there were Atlantic Herring in 
western Greenland waters, in small numbers. They are now known to be widespread 
along the southwestern and western coasts, northward at least to 7 2° 30' N, though not in 
any great numbers if judged by the standards of northern Europe or the American 
coast. Their eggs have also been found in the Julianehaab district. The spawning 
extends through August and into September, and the presence of young fry is proof 
of successful local reproduction. In 1932 Hansen reported the presence of schools on 
the eastern coast in the neighborhood of Cape Walloe and Grittenfeldts Island (5J: 42); 
he also reported the capture of a specimen near Angmagssalik. The evidence, in short, 
is strong that this species has expanded its range in Greenland waters during the past 
30 years or so, as have other boreal sub-Arctic fishes (for details as to their occurrence 
in Greenland waters, see Jensen, 6^ : 61—64). 

Synonyms and References i^"* 

Clupea harengus Linnaeus, Syst. Nat., ed. 10, 1758: 317 (diagn.); Cuvier and Valenciennes, 20, 1847: 30, 
pi. 591 (descr., anat., life hist., distr., etc., includ. review of lit.); Giinther, Cat. Fish. Brit. Mus., 7, 
1868: 415 (refs., descr., Europe and N. Amer.); Jordan and Gilbert, Bull. U. S. nat. Mus., 16, 1882 
265 (descr.); Goode, et al.. Fish. Fish. Industr. U. S., i (3), 1884: 549-568, pi. 204 (abund., import, 
distr., migr., reprod., food, capture, uses); Jordan and Evermann, Bull. U. S. nat. Mus., 47(1), 1896 
421; 47(4), 1900: fig. 18 (descr., synon.); Smith, BuU. U. S. Fish Comm. (1897), IJ, 1898: 91 
(spawn., growth rate. Woods Hole, Massachusetts); Linton, Bull. U. S. Fish Comm. (1899), ig 
1901: 437 (food, parasites); Bean, 7th Rep. Forest Comm. N. Y. (1901), 1902: 298, with fig. (econ 
import., spawn., range); Bean, BuU. N. Y. St. Mus., 60, Zool. 9, 1903: 192 (synon., refs., descr. 
import., abund., spawn., distr.); Fowler, Rep. N. J. St. Mus. (1905), 1906: 93, fig. (diagn.. New 
Jersey); Smith, N. C. geol. econ. Surv., 2, 1907: 120 (descr., range); Ehrenbaum, Eier und Larven Fisch., 
Nord. Plankt., 2, 1909: 361 (develop, of eggs and larvae); Sumner, Osburn, and Cole, Bull. U. S. 
Bur. Fish. (191 1), JJ, 1913: 741 (refs., season, abund., spawn., food, parasites, Woods Hole, Massa- 
chusetts); Cox, Contr. Canad. Biol. (1914-16), 1916: 81 (disease. Gulf of St. Lawrence, 1914); Meek, 
Migrations of Fish, 1916: 67-88 (genl. acct.); Regan, Ann. Mag. nat. Hist., (8) ig, 1917: 227 (diagn. 
in key); Lebour, J. Mar. biol. Ass. U. K., N. S. 12 (3), 1921 : 459 (food of young clupeoids); Bigelow 
and Welsh, Bull. U. S. Bur. Fish. (1924), 40, 1925: 92, fig. 40 (descr., range, breed, habits, growth, 
size, season, movements, food, enemies, destruct. by natural causes, annu. fluct.. Gulf of Maine); Jordan, 
Fishes, 1925: 275 (distr., abund.); Hildebrand and Schroeder, Bull. U. S. Bur. Fish. (1927), ./J, 1928: 
81 (refs., descr., spawn., import., rare in Chesapeake Bay); Breder, Field Bk. Mar. Fish. Atl. Cst., 
1929: 64, fig. (range, import., habits, size); Jordan, Manual Vert. Anim. NE U.S., 1929: 39 (diagn., 
range); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 41 (range, synon.); 
Perlmutter, 28th Rep. N. Y. Conserv. Dept., Suppl., 15(2), 1939: 17 (spawn., eggs, young caught. 
Long Island, New York); Greeley, 28th Rep. N. Y. Conserv. Dept., Suppl., i 5 (2), 1939: 82 (Long 
Island, New York); Bigelow and Schroeder, BuU. U. S. Bur. Fish., ./cS, 1940: 327 (range extension; 
catch on Georges Bank; distr. of young); Warfell and Merriman, BuU. Bingham oceanogr. CoU., 9 (2), 
1944: 51 (near New Haven, Connecticut; sizes and spawn.); Anonymous, Fish. Resources U. S., 
Senate Doc. No. 51, 1945: 61-62, with figs, (commerc. import., range, Ufe hist., fluct.); Bigelow and 
Schroeder, Fish. BuU. (74) U.S. Fish Wildl. Serv., 5J, 1953: 88-100 (descr., Ufe hist.. Gulf of Maine). 

24. Numerous accounts pertaining whoUy or in part to this Herring have been written, for it has been a principal 
subject of extensive study for centuries. A complete bibliography would occupy many pages. Therefore, the refer- 
ences given herein are limited to the papers actuaUy quoted (directly or indirectly), to taxonomic works of a general 
nature, or to faunal lists of the American coast. Many of these also contain bibliographies. Furthermore, references 
are given to publications containing descriptions of the species itself and of several "species" now considered synony- 
mous with Clupea harengus. For additional references bearing especiaUy on the biology of this species, see 
Scattergood {114)- 



Fishes of the Western North Atlantic 293 

Clupea elongata LeSueur, J. Acad. nat. Sci. Philad., I (2), 1818: 234 (orig. descr.; type local. Marblehead 
and Sandy Bay, Massachusetts). 

Herring, Moore, Rep. U.S. Comni. Fish. (1896), 22, 1898; 387-442 (life hist., econ. import., Passamaquoddy 
Bay); Huntsman, Canad. Fish. Expcd. (1914-191 5), 1919: 165-171 (growth of young); Lea, Canad. 
Fish. Exped. (1914-1915), 1919: 75-165 (age and growth, Canad. waters); Norman, Hist. Fishes, 
1931 : 42, 65, 80, 91, 135, 183, etc. (genl. acct.); Anonymous, Rep. U. S. Tariff" Comm., (2) 69, 1933: 
78, 106, 117, 143, 149-151, 157, 161, 163-165, 218 (statist, fishery products, Europ. and Amer. 
countries); Ford, J. Mar. biol. Ass. U. K., N. S. ig, 1933-34: 305 (econ. of fishery, Plymouth, Eng- 
land; life hist, data; forecast, catch); Battle, Huntsman, Jeff"ers, Johnson and McNairn, J. biol. Bd. 
Canad., 2(4), 1936: 401 (fatness and food of young, Passamaquoddy Bay); Marshall, NichoUs and 
Orr, J. Mar. biol. Ass. U. K., N. S. 22, 1938: 245 (growth, feed., Clyde); Johnson, J. Fish. Res. Bd. 
Canad., 4, 1939: 349, 392 (effects of light on movements, feed.). Leim, Tibbo, and Day, Bull. Fish. 
Res. Bd. Canad., 3, 1957: i — 184 (investig. e. Canada, growth, abund. populations, year-classes). 

Doubtful References: 

Clupea halec Mitchill, Rep. in part, Fishes of New York, 1814: 20 (orig. descr.; type local. New York; type lost; 

possibly a species of Pomolobui); Mitchill, Trans. Lit. philos. Soc. N. Y., I, 181 5: 451 (descr.). 
Clupea coerulea Mitchill, Trans. Lit. philos. Soc. N. Y., I, 181 5: 457 (orig. descr.; type local. New York; 

type lost; possibly a species of Pomolobus). 

Genus Alosa Link 1790 

Shads 

Alosa Link, Mag. Physik. Naturg., Gotha, 6(3), 1790: 35; type species not specified; Cuvier, Regne Anim., 
ed. 2, 2, 1829: 319; type species Clupea alosa Linnaeus, same as Alosa Link. 

Generic Synonym: 

Alausa Cuvier and Valenciennes, Hist. Nat. Poiss., 20, 1847: 389; type species Alausa vulgaris Cuvier and 
Valenciennes equals Clupea alosa, apparently a variant spelling of Alosa Link. 

Characters. Body compressed, rather deep to elongate. Bony scutes rather promi- 
nent, with sharp points; the chest and abdomen with a sharp edge. Cheek (bone 
below eye) notably deeper than long (Fig. 69). Mouth large. Maxillary generally 
extending beyond middle of eye. Upper jaw with a definite median notch. Lower jaw 
largely included In upper one, its upper margin within the mouth rising gradually 
(Fig. 69). Teeth in jaws minute or absent (absent in American adults); none on vomer 
or tongue in adults. Dorsal with 16—22 rays, its last ray not produced, its origin 
nearer to end of snout than to base of caudal. Anal with 18—27 ''^ys (18—22 in Amer- 
ican species), the last ones not enlarged. Pelvic usually with 9 rays, inserted under 
base of dorsal. Vertebrae about 52-59. 

Remarks. This genus is very close to Pomolobus., with some of the species being 
difficult to separate generically. The species herein assigned to Alosa have a deep 
median notch in the upper jaw; the lower jaw shorter than the upper, with the upper 
margin (within mouth) not rising abruptly; the cheek deeper than long; and the adults 
without teeth. A combination of these characters is required to identify the species of 
Alosa. ''^ 

25. The two "wing-like plicae of skin, covered with elongate scales [alae)" on the base of the caudal, designated the 
leading diagnostic character by Borodin in his revision of the genera Alosa and Pomolobus {18: 175), seem unusable 
because too often they are indistinct on preserved specimens. 



2 94 Memoir Sears Foundation for Marine Research 




Figure 69. Heads showing shape of mandible and cheek, left, Alosa sapidissima, 93 mmTL, Clifton Beach, 
Maryland; right, Pomolobus aestivalis, 82 mm TL, Smith Point, Maryland. Both drawn by Louise Nash. 



Range. Two species occur off Europe, another off northern Africa, and two off 
North America, where the recorded range extends from southeastern Newfoundland, 
the St. Lawrence River, Quebec, and northern Nova Scotia southward to Indian River 
City, Florida, and the Gulf of Mexico westward to the Mississippi River. 



Key to North American Species 

I a. Lower limb of first gill arch with 59-73 rakers (26-43 i" young under 125 mm). 

sapidissima (Wilson) 181 1, p. 295. 

lb. Lower limb of first gill arch with 42-48 rakers in adults (27-33 •" young 48— 

90 mm long). alabamae Jordan and Evermann 1896, p. 308. 





^ 



Figure 70. First branchial arch, showing numerous gill rakers: left, Jlosa sapidissima, 550 mmTL, USNM 
20969; RIGHT, A. alabamae (with fewer rakers), 390 mmTL, type, USNM 47689. Both drawn by Ann 
S. Green. 



Fishes of the Western North Atlantic 295 

Alosa sapidissima (Wilson) 1 8 1 1 

American Shad, Atlantic Shad, Shad, Common Shad, White Shad 

Figures 69—74 

Study Material. At least 75 specimens, 35-580 mm TL, 29-475 "^"^ ^L, from 
St. Andrews Bay, New Brunswick; Flag Island, Eastport, and Portland, Maine; 
Holyoke and Woods Hole, Massachusetts; Noank, Connecticut; Long Island, and 
Gravesend Bay, New York; Great Egg Harbor, New Jersey; Delaware City, Delaware; 




Figure 71. Alosa sapidissima, adult male, 480mm TL, 19.13 inches, Norfolk, Virginia, USNM 25099. 
Drawn by H. L. Todd. 

many places in Chesapeake Bay and tributary streams; Avoca and Beaufort, North 
Carolina; and New Berlin, Florida. 

Distinctive Characters. For the features that distinguish the true shads {Alosa) 
from their close relatives, and for the difference between sapidissima and alabamae, see 
Key to Species, p. 294, as well as pp. 293, 304, 309, 310. 

Unlike the adult with its toothless jaws and a median indentation in the upper jaw, 
young of 1 50 mm and less lack the indentation on the upper jaw and have minute teeth 
in the jaws. Therefore the jaw and tooth characters of the adult cannot be used in 
identifying the young. However, the deep narrow cheek as well as the slender mandible 
with its upper margin not rising abruptly to form an angle in the mouth will aid in 
separating young sapidissima from young and adult alewlves of the genus Pomolobus 
(Fig. 69). 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens 29-475 mm SL. 



Body: depth 30.2-36.8. 
Caudal peduncle: depth 7.7—10. 



Head: length 23—28. 
Snout: length 4.85—7.1. 



296 Memoir Sears Foundation for Marine Research 



Eye: diameter 3.8-5.7. 
Inter orbital: width 4-55— 5-95- 
Maxillary: length 11— 14. 
Pelvic fin: length 9.9— 11. 9. 
Pectoral fin: length 1 5.3-1 7.7. 



Scales: 52—62. 

Ventral scutes: 35-38. 

Gill rakers: 59-73 in adults. 

Fin rays: dorsal 17-20; anal 20-23; P^'^" 

toral 15-17. 
Vertebrae: 55-57 (4 specimens), 54-59 

given in literature. 



Body slender, compressed, its greatest thickness generally about half of its depth; 
the depth very variable, increasing with age, 2.75—4.0 in SL. 

Scales moderately adherent, with crenulate membranous borders, 15 or 16 rows 
between base of pelvic fin and anterior dorsal rays. Ventral scutes well developed, 
20-22 in front of pelvics, 14—17 behind them. 

Head 3.6-4.4 in SL. Snout 3.6-4.6 in head. Eye 4.0-5.9. Interorbital 4,2— 
5.8. Mouth moderate, oblique; mandible included. Maxillary broad, its width about 
half of length of snout, with a broad supplemental bone, rounded posteriorly; reaching 
to middle of eye in young, generally to below posterior margin of eye in adults, 1.9—2.2 
in head. Gill rakers increasing in number and proportionate length with age, those at 
angle shorter than snout in young but longer than snout in large specimens, 26-43 
on lower limb in young 29—125 mm SL, 59—73 in specimens 300 mm SL and upward. 
Teeth missing in adults though present on jaws and median line of tongue in young. 

Dorsal fin somewhat elevated anteriorly, its margin slightly concave, Its last ray 
a little longer than the preceding one, its origin an eye's diameter or less in advance 
of pelvic fin and about equidistant between tip of snout and vertical from base of last 
anal ray. Caudal rather deeply forked, the lobes of nearly equal length and somewhat 
shorter than head. Anal low, Its origin about an eye's diameter nearer to base of 
pelvic than to base of caudal. Pelvic near ventral edge, about equidistant between 
base of pectoral and origin of anal, 2.1—2.6 in head; the axillary scale equal to, or a 
little more than, half of length of fin. Pectoral a little higher on side than pelvic, 
1.4— 1.65 in head. 

Color. Fresh specimens greenish to bluish with metallic luster above, gradually 
shading into the bright silver on sides. Dark spot on shoulder, sometimes followed 
by smaller spots or dots, rarely with a second parallel row of dots below the first. No 
dark lines along rows of scales. Fins pale to greenish, the dorsal and caudal somewhat 
dusky in large examples, and tips of lobes of caudal dark in some specimens. Peritoneum 
mostly pale to silvery. 

Size. This species is the largest of the herrings. Females attain a maximum 
length of about 600 mm (30 in.) and a weight of 12 pounds In the Atlantic. On the 
Pacific coast, where the species was introduced, the maximum weight is even greater; 
females of 14 pounds have been reported. In the Chesapeake Bay area the average 
weight (and length) of females (roe shad) is probably somewhat under six pounds 
(about 57,5 cm; 22.75 i"-)? '^'^^'^ °^ males (buck shad) about 3.5 pounds (about 50 cm; 
20 In.). 



Fishes of the Western North Atlantic 



297 



Development and Growth. The eggs, about -t^.^ mm in diameter after they are im- 
pregnated, are demersal, but they are only slightly heavier than the water in which 
they are deposited. As they are not adhesive, they drop loosely and singly to the bottom. 
According to Ryder they hatch in 52 hours at an average temperature of 57.2° F, and 
in less than 36 hours at an average of 74° F {ll2: 796). However, a longer incubation 
period was reported by Leim(5j: 31), who said that eggs held under artificial conditions 
hatched in 12-15 <^^ys ^^ I2°C (53.6° F) and in 6-8 days at I7°C (62.6° F). 

The newly hatched fish, transparent and extremely slender, are about 9-10 mm 




Figure 72. Alosa safldissima egg development, left to right: newly laid egg; embryo in natural position 
within enveloping membrane of egg (photograph); egg with embryo nearly ready to hatch. After Leach. 



long. The yolk is absorbed in four or five days at a water temperature of 62.5° F and 
in about seven days at 53.5° F; the fish have then attained a length of about 12— 
15.5 mm {81: 33). In some preserved specimens that show the yolk almost absorbed, 
the depth is only about 0.05 of the length, the large dark eyes protrude prominently, 
the dorsal fin fold is somewhat developed, and the alimentary canal is discernible, 
ending under the beginning of the posterior sixth of the body. During development, 
the vent moves forward as in other Clupeidae. The fins are fully developed in 21- 
28 days at a length of about 20 mm, when metamorphosis is nearly complete {16: 1 18). 
For a more detailed account of the development of the larvae, see Leim {81: 31-43). 

The young remain more slender than the adults until they are nearly grown 
fish, but even among sexually mature fish there is great variation in depth, seemingly 
not related to sex and only partly related to spawning conditions. In 25 young, 35- 
125 mm long, the depth varied between 26.5—30 "/o of SL, and in a lot of 12, 460— 
505 mm long, the depth was 32-48 "/o of SL. 

A gradual increase in the number of gill rakers, as already indicated, also takes 
place with age and growth. For example, the number on the lower limb of the first 
arch varied as follows: in 18 young 35 to 65 mm long, 26 to 31; in 21 examples 



298 Memoir Sears Foundation for Marine Research 




Figure 73. Alosa sapidissima, larval development, top, immediately after hatching; upper center, third day 
after hatching; lower center, five days after hatching; bottom, 17 days after hatching. After Leach. 



70 to 125 mm, 34 to 43; in 16 immature fish 190 to 270mm, 48 to 62; and in 
32 sexually mature examples 465 to 580 mm, 59 to 73. 

The young grow rapidly during their first summer while in fresh water. The 
total lengths of 769 young taken in the Chesapeake Bay area — mostly from the 
Potomac River (in 1912) — ranged between 20-69 "^^ during June, 30-74 mm during 
July, 45-79 mm during August, 50-95 mm during September, 45-99 mm during 
October, and 60-119 mm during November {^9'- 98)- 

Larger young of 1 50, 1 75, and even 225 mm, at an age of seven months, have been 
reported. Such fish, in at least one instance, are known to have lived under specially 
favorable conditions in a pond well supplied with food (at Washington). Furthermore, 
according to Bean : "Nets set offshore in Gravesend bay [New York] in the fall frequently 
inclose large quantities of young shad, sometimes a ton and a half at one time, 
during the migration seaward . . . The fish are usually about 6 to 8 inches long" {8 : 208). 
One may question whether fish of such a large size, taken in such large quantities, 
were only six to seven months old. That the young grow faster north of Chesapeake 
Bay seems unlikely in the light of a statement by Bigelow and Welsh {16 : 1 18) con- 
cerning this Shad in the Gulf of Maine: "The young shad remain in the rivers until 
fall when, at a length of 1^/2 to 4^/2 inches (37 to 112 mm) and resembling their parents 



Fishes of the IVestern North Atlantic 



299 



in appearance, they move down to salt water." This range in length is close to that 
of fish of the same age from Chesapeake Bay and does not indicate more rapid growth 
in the more northern part of the range. However, Bigelow and Welsh added: "In the 
rivers of Maine the fry may be as long as 5 to 7 inches (125 to 175 mm) by the first 
week of November." Included in Hildebrand and Schroeder's data there were three 
specimens from the November collection that ranged somewhere between 145 and 
244 mm in length {5g)\ these fish were so much larger than the other 137 specimens 
taken in the same month that one is led again to suspect that they may have belonged 
to the one-vear class. 








m 



Figure 74. Alosa sapidissima, young, 47 mm TL, 1.88 inches, Beaufort, North Carohna. Drawn by Louella 
E. Cable. 



The growth rate after the young have entered salt water is not well known because 
of the lack of specimens available for examination. Hildebrand and Schroeder listed 
22 specimens that were spending their first winter in the deeper places of Chesapeake 
Bay instead of continuing their migration to sea {^g: 98). Eight of these, taken in 
January, ranged between 95-124 mm and averaged 108 mm, 3 taken in February 
were 95, 105, and 135 mm long, 10 taken in March ranged between 105-159 mm and 
averaged 142 mm, and 2 taken in May were 145 and 154 mm long. Specimens of a 
somewhat similar size were dredged during the winter in the estuary of Newport River, 
Beaufort, North Carolina; 18 specimens caught in January ranged between 73-140 mm 
and averaged 117 mm; 14 caught in February were 100-160 mm and averaged 
121 mm; and 3 caught in March were 84, 102, and 120 mm long. 

Using mostly specimens from the Connecticut River, Borodin-^ {ly. 183) deter- 
mined the age of older fish from the annuli and transverse grooves on the scales, a 
method confirmed as practicable in Barney's study of otoliths (5: 168). Borodin 
provided a list of 10 size groups to which he assigned ages as follows: Group i — 58 

26. Borodin, in determining the age of Shad from the scales, used the transverse grooves in addition to the annuli, 
having discovered that two complete transverse grooves are formed each year. Barney determined their age from 
the "limy deposits" in the otoliths, which form annual rings; he found that these annuli correspond to the annuli 
and crossgrooves on the scales used by Borodin. 



300 Memoir Sears Foundation for Marine Research 

to 1 13 mm long, 26 specimens from the river, all less than a year old; Group 2 — 145 
to 180 mm long, 4 from salt water, in their second year; Group 3 — 210 to 290 mm, 
7 from salt water, in their third year; Group 4 — 300 to 350 mm, 10 young males 
entering the river, in their fourth year; Group 5 — 360 to 390 mm, 22 young males 
entering the river, in their fifth year; Group 6 — 400 to 430 mm, 13 males and 9 
females, in their fifth, sixth, and seventh years; Group 7 — 440 to 480 mm, 34 males 
and 9 females, in their sixth, seventh, and eighth years; Group 8 — 490 to 520 mm, 
28 males and 13 females, in their seventh, eighth, and ninth years; Group 9 — 530 to 
570 mm, 6 males and 43 females, in their seventh, eighth, ninth, and tenth years; and 
Group 10 — 580 to 660 mm long, i male and 23 females, in their eighth, ninth, tenth, 
and eleventh years. 

Young fish from the Bay of Fundy reach a length of 41-45 mm in nine to ten 
weeks {81 : 74-84). Leim's calculations, based on scale readings of adult fish, in general 
indicate a rather faster rate of growth for his fish than for those of Borodin. 

Some of the males, according to Borodin, enter the Connecticut River from the 
sea when they are 300-350 mm long and in their fourth year, as shown in the preceding 
data. Although the author did not say so, these males undoubtedly were sexually 
mature, as migration to the rivers seems to be carried out solely for the purpose of spawn- 
ing. The smallest females that returned to the Connecticut to spawn, according to the 
same author, were between 400 and 430 mm long and in their fifth, sixth, and seventh 
years. Therefore, the males seem to become sexually mature at an earlier age than the 
females. On the other hand, females may reach a greater age; among the older fish ex- 
amined by Borodin, determined by him to be in their seventh, eighth, ninth, tenth, and 
eleventh years, there were only 7 males compared to 86 females. Concerning the Bay 
of Fundy fish, Leim said: "Most of the shad present on the spawning grounds were 
five years old or over and eight or nine years seems to have been the maximum age 
represented. Fish under four years of age apparently do not spawn at all" {81). The 
discrepancy in the rate of growth and the age of spawning fish, as determined by 
Borodin and by Leim, may have resulted from a difference in the interpretation of the 
scale markings, or from a difference in the behavior of two rather widely separated 
races that may respond differently to the climatic or general environmental condi- 
tions in which they live. 

Reproduction. The average number of eggs produced by a single fish varies between 
25,000-30,000 {lig: 127). In th^ Manual of Fish Culture {2: 125) it is stated that the 
average number is not more than 30,000 but that a single fish has been known to yield 
between 60,000 and 115,000 eggs. In 1885, one from the Delaware River yielded 
156,000. In this species, as among others, large fish as a rule produce many more 
eggs than small ones. 

If all of the large number of eggs produced by a female hatched and reached ma- 
turity, the progeny would be exceedingly great. However, it is quite certain that nor- 
mally only a small proportion of the eggs produces fish that live to maturity. The most 
critical time in the life cycle apparently comes when the larvae have absorbed the yolk 



Fishes of the Western North Atlantic 301 

and must find their own food. If food is found and if other conditions are favorable for 
survival, then the spawning season is a successful one. The success or failure of a spawn- 
ing season in a river is usually evident from the size of subsequent catches when the 
fish hatched in a given year return later to spawn in the same stream. The comparatively 
large fluctuations in catches over a period of years no doubt result in large part from 
the measure of success of the particular spawnings from which the fish composing the 
catches were derived. 

Spawning and Migrations. American Shad spawn in fresh water. Some spawn 
as soon as they arrive from the sea while others do not spawn until they reach the 
upper tributaries of rivers. Virtually all spawning, in this species as in many others, 
takes place in the evening between five and ten o'clock. While spawning, the fish are 
paired, swimming close together side by side. Since no nests are made, the eggs fall 
loosely in the water, and doubtless many are smothered when they fall on muddy bot- 
tom. An individual returns to spawn year after year. 

The water temperature governs the date when mature American Shad approach 
the mouths of the rivers to spawn, about 50-55° being required. Accordingly, they 
appear off the Florida coast (St. Johns River) in November, to remain until March, 
in the sounds along the Carolina coast in February and March, in the Chesapeake 
Bay area in March and April, and farther north in May and June. But the precise 
date of arrival at any particular locality along the coast may vary by a couple of weeks 
from year to year, depending on the variation from year to year in the vernal warming 
of the water. 

In large rivers, they run far upstream if their passage is not blocked. Thus they 
still run up 375 miles (or did recently) in the St. Johns River in Florida, 300 miles in 
the Altamaha in Georgia, about 200 miles in the St. John in New Brunswick, and at 
least 200 miles in the St. Lawrence {122 '. 11; 12'j: 129, 130). American Shad as- 
cending the estuary of the St. Lawrence may travel as much as 25-50 miles in 24 
hours (J27: 129). 

The spent fish, very thin after spawning, begin their return journey to the sea 
immediately afterward; i.e. in March to May in the southern part of the range and 
in June to August in the northern part. But the young fry (product of that season's 
spawning) remain in the parent stream until autumn, when the temperature falls to 
60° or lower. In the Chesapeake Bay region, their migration out into salt water takes 
place principally in October and November; but it is not until late November or De- 
cember that all of them are in salt water. 

Southward from Cape Cod the spent fish vanish upon re-entering salt water. While 
a few young fry in the Chesapeake Bay area spend their first winter in the deeper 
parts of the Bay, those from the Hudson may linger nearby for an entire year, for 
catches of 6- to 8-inch fish in Gravesend Bay in autumn have been reported (5: 208). 
The larger sizes of immature fish vanish as completely from the southern part of the 
coast as do the spent fish, until they reappear there for their first spawning. 

Spent fish up to ten pounds and averaging about five pounds have been reported 



302 Memoir Sears Foundation for Marine Research 

as taken in the Gulf of Maine, and schools of them have been seined from time to time 
in summer and autumn, even in December, at various places along the Maine coast, 
where they have been the object of a frozen-fish industry in some years. Immature 
fish (0.5-2.5 pounds) sometimes congregate also along the Maine coast and are more 
or less common in the Massachusetts Bay region. 

Clearly these Gulf of Maine fish are far too numerous to be derived only from 
the streams of the Bay of Fundy, as was formerly considered likely {16: 1 16). Recently, 
extensive tagging experiments (75: iii, ftn. 18—20; 124) have shown that, while 
some of them come from as far south as the Altamaha River, Georgia, others come from 
as far north as the St. Lawrence estuary, both of these contingents to fatten in the 
Gulf. By the end of December, however, they have vanished, not to reappear anywhere 
on the coast until the onset of the next spawning season, when the maturing fish return 
to the rivers of their native watersheds. 

Like salmon and alewives, American Shad spend most of their life in the sea and 
make most of their growth there, but where they winter remains to be discovered. 
While in the sea, Shad are schooling fish, often in the thousands, and they never re- 
enter fresh water until they return to spawn, although they may appear in brackish 
estuaries. They have been taken 25-90 miles out off southern New England, 1 10 miles 
out on Georges Bank, 40-50 miles out off the coast of Maine, and 50-60 miles off 
eastern Nova Scotia, and as deep as 50-68 fms. 

Food. The diet of 14 young fish, 43-60 mm long, from fresh water of the 
Potomac River, consisted chiefly of adult and larval insects and ostracods, and in 
one instance a small fish {^g: 95). Larger juveniles, 100-150 mm long, caught in 
Chesapeake Bay, presumably while en route to the sea, had fed mostly on Mysis, although 
one stomach contained fragments of a small fish, and another fragments of plant tissue. 

Working with young specimens from the Shubenacadie River, tributary to the 
Bay of Fundy and its estuary, Leim found that the first food taken by larvae 1 1 mm 
long consisted of midge larvae (Chironomidae), while the somewhat larger larvae had 
fed principally on mature and immature copepods (81: 16). In fact, these organisms 
constituted the chief food of the young up to the time of transformation, with the 
relative abundance of these forms in a particular locality determining which food pre- 
dominated. Dr. Leim's data show also that young adults taken in the same general 
vicinity continued to subsist principally on these same organisms. Other foods ingested 
consisted of ostracods, amphipods, insects, and fish. 

Little or no food has been found in the stomachs of those caught while in fresh 
water en route to their spawning grounds, indicating that these fish, like salmon, do 
not ordinarily feed then. However, there are some records showing that adults occasion- 
ally do take food while in fresh water, at least late during the spawning season. They 
will often take a live minnow or an artificial fly when working upstream on their 
spawning run. 

Many kinds of food have been found in the stomachs of those taken in salt water 
in estuaries and bays north of Cape Cod. But southward, examples are rarely obtainable 



Fishes of the Western North Atlantic 303 

in salt water, except during the migration to and from tlie spawning grounds. There- 
fore it is of interest that an adult female, caught in Chesapeake Bay in December, 
was gorged with parts of plants, among which were fragments of a mollusk shell. It 
has been reported also that adults taken in the vicinity of Savannah, Georgia, had fed 
on plants. Others caught near Fort Lee, New Jersey, had fed liberally on young shrimp, 
and still others, taken in Delaware Bay, had eaten small fish. 

From an examination of about 350 stomachs of both mature and immature fish 
caught in the salt water of Scotsman Bay, Bay of Fundy, Leim found that, while cope- 
pods constituted the chief food of the smaller ones, as in fresh water, these crustaceans 
were unimportant in fish 400 mm and more in length (^J: 68). Mysids, which were 
sparingly eaten by small fish, were the chief food of adult fish. In general, about 90 "/o 
of the specimens of all sizes from that area had eaten copepods and mysids, with 
ostracods, amphipods, isopods, decapod larvae, insects, mollusks, algae, fish eggs, and 
fish making up the remainder. After examining many stomachs of specimens taken in 
the Bay of Fundy, Willey also concluded that the chief foods consisted of copepods and 
mysids, to which he added shrimp and larval stages of barnacles (JJJ: 313-320). 
Stomachs of fish from the Gulf of Maine have been found full of copepods (chiefly 
Calanus). American Shad also feed greedily on pelagic euphausiid shrimps. 

Enemies. No doubt the larger predatory fishes are the principal enemies of the 
young while the latter are in the rivers. Although no records at hand show definitely 
that this is so, it can scarcely be otherwise, for the young are not rapid swimmers and do 
not ordinarily seek protection among plants or other cover. Furthermore, they are small 
enough, at least during the early part of summer, to be swallowed by bass and other 
carnivorous fishes only an inch or two long. Water birds, turtles, and water snakes 
no doubt feed on them too. The toll taken by predatory fishes while the young are en 
route to the sea may not be very great, for this migration takes place in the fall when 
many marine fishes (at least in Chesapeake Bay) have left the shallow waters. At sea 
they doubtless fall prey to all the larger predaceous fishes, and perhaps to finback whales 
as well. 

The chief enemy of the adults is man, who has not only depleted the stock in many 
rivers through overfishing but has cut them off from spawning areas through the 
construction of barriers and by polluting the waters with deleterious wastes. 

Parasites. The parasites of yf. sapidissima have been made known principally through 
several papers by Edwin Linton and by Charles B. Wilson, based mostly on specimens 
obtained from fish caught in the vicinity of Woods Hole, Massachusetts. The findings 
were summarized by Sumner, et al. {123: 742) as follows: "Acanthocephala (Linton) 
Echinorhynchus acus. Nematodes (Linton) — Ascaris adunca, Ascaris sp. (immature). 
Copepods (C. B. Wilson) — Caligus rapax." To this list was added the copepod Ler- 
naeenicus radiatus (135: 59). Working with both young and adults from the Shuben- 
acadie River and Scotsman Bay, Leim found no external parasites and only three 
internal ones, namely "distomes, nematodes, and Acanthocephali" (^J : 74). It is not 
stated whether any of these parasites were considered seriously detrimental to the host. 



3 04 Memoir Sears Foundation for Marine Research 

Relationships. There are two well-marked species of this genus in Europe, A. alosa 
and A.finta, both of which have the deep cheek and notched upper jaw as in American 
species. However, A. alosa has smaller scales (72-80 lateral series) and more numerous 
gill rakers (72-86 on lower limb in adults) than the American species. A.finta, while 
agreeing with the American species as to scales, has only about 24-32 gill rakers on 
the lower limb, in grown examples. 

The two American species recognized here, namely A. sapidissima and A. ala- 
bamae, differ principally in the number of gill rakers; mature examples of sapidissima 
have 59-73 on the lower limb whereas specimens of alabamae of similar size have 
only 41-48. The relationship of these species is discussed further in the account of 
alabamae (pp. 309, 310). 

Variations. The specimens examined, from various localities within the range 
(p. 295) and used in preparing the Description, do not differ specifically or subspecifically. 
However, racial differences have been recognized. Vladykov and Wallace have found 
slight average differences in the number of vertebrae, pectoral rays, and ventral scutes 
between examples from the Bay of Fundy, Delaware Bay, and Chesapeake Bay; these 
they regarded as different "populations" {130: 52—66). Investigators of the U.S. Fish 
and Wildlife Service, in a special study of A. sapidissima, found that virtually every 
important river system within the range of this fish is inhabited by a different race 
(unpublished data). 

Commercial Importance. This species has been an important foodfish throughout 
its entire range since the earliest settlers arrived, and no doubt these fish were taken 
by the Indians long before. Unfortunately, pollution, overfishing, and dams constructed 
across the streams (preventing the fish from reaching their spawning grounds) have 
caused partial and sometimes serious depletion, or even virtual extermination. So 
many streams have been dammed, particularly in New England, that, in the words 
of Bigelow and Welsh: "the local stock of shad has diminished until now the Gulf of 
Maine stock is but a shadow of its former abundance . . ." {16: 1 14). The great decline 
in abundance in New England is evident from the published statistics of the former 
United States Fish Commission, U. S. Bureau of Fisheries, and the present U. S. 
Fish and Wildlife Service;" the annual New England catch between 18 89-1 908 
seems to have been well above 1.25 million pounds, by 1919 it had fallen somewhat 
below a million pounds, and between 1 924-1 940 it has fluctuated around a half 
million pounds. 

The situation with respect to the American Shad fishery In the Hudson is much 
more encouraging, although depletion occurred in that river also; statistics from the 
sources previously mentioned show that the catch in that river declined from 3.75 
million pounds in 1901 to only 40,000 pounds in 191 6 and to 43,000 in 1917; there- 
after, between 191 8— 1935, *^he catch varied from 94,000 pounds (1924) to 847,000 

27. No adjustment for the differences in the number of persons employed, the type and number of vessels, and the 
different types of gear used from year to year has been made in the statistics given here and elsewhere. However, 
it is highly probable that the "fishing effort" for looo pounds of fish, for example, has increased. 



Fishes of the JVestern North Atlantic 305 

pounds (1935); and in 1 936-1 941 it varied from 2,467,000 (1938) to 3,270,700 
(1939). Although the catch in 1959 was only 1,171,212 pounds, very definite recovery 
has been indicated. According to investigators of the U. S. Fish and Wildlife Service, 
this recovery is ascribed partially to the elimination of pollution and partially to the 
great decrease in fishing effort at the time when depletion made it unprofitable to 
fish. The curtailment in fishing permitted the depleted population to reach the 
spawning grounds, to return to the sea, and to spawn again and again. Ultimately 
this resulted in the rehabilitation of the American Shad in this river system. It is 
decidely encouraging to have evidence that a depleted fishery can be restored. 

The American Shad fishery in the Potomac, though it has fluctuated greatly, has 
also undergone a general decline in recent years, without evident recovery. The catches 
in 1896 and 1901 were above 2.5 million pounds whereas in 1909 and 1915 they were 
only three-quarters of a million pounds. Then from 191 9 to 1923 they ranged above a 
million pounds, reaching the impressive totals of somewhat more than two million 
pounds in 1919 and over three million pounds in 1922; the latter is the greatest catch 
on record. In 1924 the catch dropped to a little over half a million pounds, and since 
then it exceeded a million pounds only in 1926, 1928, 1929, 1931, and 1933; in fact, 
the catch exceeded two million pounds in 1928, 1931, and 1932. However, between 
1 934-1 941 it has fluctuated only from 631,000 pounds in 1935 t° 322,800 pounds in 
1940, the smallest catch recorded. Although great fluctuations in the catches, and no 
doubt in abundance, have occurred in the Potomac River since 1896, a definite decline 
was not indicated by the statistics until 1934. Thereafter, a general decrease in the 
catches has reached alarming proportions. In 1959 the catch was only 188,000 pounds. 

A decline similar to that in the Potomac River has taken place in the entire Chesa- 
peake Bay area, where, exclusive of annual fluctuations in the catch, the general trend 
has been downward since 1930. According to statistics, this same situation also applies 
to the South Atlantic states, and perhaps to a somewhat smaller extent to the fisheries 
of the Middle Atlantic states. 

The over-all yield of late has ranged between catches of 8,222,000 and 14,700,000 
pounds for the United States (1940, 1945, 1950— 1953), and between 1,545,000 and 
2,444,000 pounds for Canada (1940, 1945, 1 950-1 952). The American Shad thus 
remains very important economically, yielding the fishermen an annual return of over 
a million dollars. 

Formerly, when they were abundant, the market for them more limited, transporta- 
tion slower, and refrigeration virtually unknown, a considerable portion of the catch 
was salted. However, now nearly all of it is marketed in the fresh state. Johnson has 
stated, "During recent years the outstanding problem of wholesale dealers, insofar 
as shad are concerned usually has been one of obtaining adequate supplies rather than of 
disposing of the catch" {68'. 15). In fact, the demand on the Atlantic coast has been 
so great that the supply is augmented annually by shipments from the Pacific coast. 
The chief markets on the Atlantic coast are New York, Philadelphia, Baltimore, and 
Washington. 



3o6 Memoir Sears Foundation for Marine Research 

Shad are caught principally with pound nets, gill nets, and seines. Small quantities 
are taken with fyke nets, otter trawls, purse seines, traps, and dip nets. 

Introductions. In the late iSoo's, many introductions of the fry were attempted 
by the former U. S. Fish Commission and U. S. Bureau of Fisheries, but all failed; 
large shipments were liberated in the streams tributary to the Gulf of Mexico, in Great 
Salt Lake, Utah Lake, and Bear Lake, Utah, and in the Colorado River at the Needles, 
Arizona. 

In the 1870's, the New York State Fish Commission introduced numerous fry 
in streams tributary to Lake Ontario and in some of the smaller lakes connected to 
it. According to Smith {118: 193), 658,000 fry were introduced, and "the appearance 
of adult shad promptly followed the introduction of young fish in the Genesee River, 
and from that time on for a number of years the indications pointed to the probable 
successful acclimatization of the species." These fish distributed themselves over the 
entire lake and were taken at irregular intervals from 1875 ^"^ about 1885; but few, 
if any, were seen after 1890. Hubbs and Lagler reported: "In the Great Lakes rare, 
and confined to the Lake Ontario Basin" [^2: I'f). 

Remarkable success has resulted from stocking waters of the Pacific coast north 
of Monterey, California. The first fry, about 12,000, were planted in 1871 in the 
Sacramento River by the California Fish Commission, and additional shipments were 
sent there by the United States Fish Commission between 1 873-1 880. On the northern 
part of the Pacific coast, fry were introduced in the Willamette, Snake, and Columbia 
rivers in 1885 and 1886. From these plantings the species has spread from California 
to Alaska, and it has become important commercially from San Francisco northward; 
the catch during recent years generally has exceeded two million pounds. ^^ 

Artificial Propagation. This species received the attention of fish culturists as early 
as 1848 when eggs were taken artificially and hatched. However, it was not until 1867 
that a really successful apparatus was perfected. This contrivance was known as the 
"Seth Green box," a floating box modified after one used for hatching salmon eggs. 
However, when used in tidal waters It was subject to accidents. Thereafter various devices 
were brought forward in rapid succession, and finally (in 1882) the "universal jar" or 
"McDonald jar," a modification of the "Chase jar" used for hatching whitefish eggs, 
was adopted by the U. S. Fish Commission and is still In use. Many millions of fry, 
or newly hatched larvae, have been liberated In the streams along the Atlantic coast. 

Range. The regular range of the American Shad extends from the lower St. Law- 
rence River and Nova Scotlan Banks as far south as Indian River City, Florida. The 
most northerly record is for one taken in Bulls Bay near St. John's, Newfoundland 
{l: 125). On the Pacific coast It now ranges from about San Pedro, California, to south- 
eastern Alaska (xjj: 221). 

Synonyms and References: 

Clupea sapidissima Wilson in Rees's New Cyclopedia, g,ca. i8i i : no pagination (orig. descr.; no local.; probably 
Philadelphia, Pennsylvania); Jordan and Gilbert, Bull. U. S. nat. Mus., 16, 1882: 267 (descr., range) ; 

28. For further details concerning their artificial propagation and distribution, see A Manual of fish Culture {2: 133-158). 



Fishes of the JVestern North Atlantic 307 

Smith, Bull. U. S. Fish. Comm. (1890), 10, 1892: 193, pi. 30 (deposited in small lakes and streams 
connected to Lake Ontario); Cheney, Rep. Forest Comm. N. Y. (1885), 1896: 125-134, color pi. 
(commerc. import., statist., artif. culture, local spawn., migr.). 

Clupea alosa Mitchill (not of Linnaeus), Trans. Lit. philos. Soc. N. Y., J, 181 5 : 449 (descr., habits, size). 

Alosa freastabilis DeKay, New York Fauna, Ft. 4, Fishes, 1842: 255, pi. 15, fig. 41 (orig. descr.; type local. 
New York; cf. European shad). 

Alausa sapidissima Gill, Proc. Acad. nat. Sci. Philad., 1861 : 54. 

Alosa sapidissima Uhler and Lugger in Rep. Comm. Fish. Md., 1876: 157; 1876: 133 (descr., synon., food 
quahties, Maryland); Ryder, Rep. U. S. Comm. Fish. (1881), 1884: 795 (temp, effects on incub. of 
eggs); Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 427; 47 (4), 1900: fig. 191 (descr., 
range, synon.); Anonymous, Rep. U. S. Comm. Fish. (1897), 1898: 133, pis. 36-39 (artif. propag. 
and develop.); Bean, T. H., 7th Rep. Forest Comm. N. Y. (1901), 1902 : 305, fig. (synon., range, occur., 
size, food of young in rivers, spawn.); Bean, T. H., Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 204 
(synon., refs., decline due to obstructions, food, migr., growth); Fowler, Rep. N. J. St. Mus. (1905), 
1906: 99 (descr., import, as food); Smith, N. C. geol. econ. Surv., 2, 1907: 125, pi. 5 in color (synon., 
descr., range, foodfish; rel. size of male and female, migr., statist., etc.); Kendall, Occ. Pap. Boston Soc. 
nat. Hist., 7, 1908: 39 (refs.. New England); Sumner, Osburn, and Cole, Bull. U. S. Bur. Fish. (191 1), 
5/(2), 1913: 742 (refs.. Woods Hole, Massachusetts, spawn., parasites); Regan, Ann. Mag. nat. 
Hist., (8) 18, 1916: 8 (diagn.); Wilson, Proc. U. S. nat. Mus., JJ, 1917: 59 (parasites); Leim, Contr. 
Canad. Biol, N. S. 2(1), 1924: 163-284 (hfe hist., bibliog.); Barney, Trans. Amer. Fish. Soc, ^4, 
1924: 168, figs. 1-4 (age from otoliths); Borodin, Trans. Amer. Fish. Soc, 5./, 1924: 178 (age from 
scales); Bigelow and Welsh, Bull. U. S. Bur. Fish. (1924), 40 [i), 1925: 113, fig. 46 (descr., size, 
range. Gulf of Maine, migr., food, develop., etc.); Jordan, Fishes, 1925: 276 (relation., qualit)' as food, 
spawn.); Nichols and Breder, Zoologica, N. Y., 9 (i), 1927: 40, fig. (distr.. New York and New Eng- 
land, food, life hist., size); Hildebrand and Schroeder, Bull. U. S. Bur. Fish. (1927), 43 (i), 1928: 93, 
fig. 5 3 (descr., growth of young, food, migr., artif propag., distr., commerc. import., marketing, size, range) ; 
Breder, Field Bk. Mar. Fish. Atl. Cst., 1929: 66, fig. (abund., decline, migr., food, spawn.); Jordan, 
Manual Vert. Anim. NE U. S., ed. 13, 1929: 40 (diagn., range); Truitt, Bean, and Fowler, Bull. Md. 
Conserv. Dep., 3, 1929: 50, fig. 16 (descr., Maryland); Jordan, Evermann, and Clark, Rep. U. S. Comm. 
Fish. (1928), 2, 1930: 42 (range, synon.); Walford, Fish. BuU., Sacramento, 28, 1931: 48 (diagn., 
distr. Pacific coast, commerc. import.) ; Greeley and Bishop, 21st Rep. N. Y. Conserv. Dep. (1931), SuppL, 
6, 1932: 78 (rare, Oswegatchie and Black R. systems); Greeley and Bishop, 22nd Rep. N. Y. Conserv. 
Dep. (1932), SuppL, 7, 1933: 91 (extinct, upper Hudson); Greeley, 24th Rep. N. Y. Conserv. Dep. 
(1934), Suppl., 9, 1935: 90 (Hudson R.); Greeley, 25th Rep. N. Y. Conserv. Dep. (1935), SuppL, 

10, 1936: 78 (Delaware R. within New York); Greeley, 26th Rep. N. Y. Conserv. Dep. (1936), SuppL, 

11, 1937: 91, pi. 3 in color (abund. Hudson R., spawn.); Vladykov and Wallace, Trans, .'^mer. 
Fish. Soc. (1937), ()^ (2), 1938: 52-66; also Contr. 17, Chesapeake Biol. Lab. (commerc. import., Chesa- 
peake Bay; var. in populations); Greeley, 28th Rep. N. Y. Conserv. Dep. (1938), SuppL, 15(1), i939'- 
39; 15 (2), 1939: 82 (fresh water. Long Island, New York); Greeley, 29th Rep. N. Y. Conserv. Dep. 
(1939), SuppL, 16, 1940: 68 (rare. Lake Ontario); Welander, Copeia, 1940: 221 (introd., distr. Pacific 
coast); Hubbs and Lagler, Bull. Cranbrook Inst. Sci., 18, 1941: 27 (range); Fowler, Monogr. Acad, 
nat. Sci. Philad., 7, 1945: 45 (synon., refs.); Anonymous, Fish. Resourc. U. S., Senate Doc. No. 51, 
1945: iii, 67-68, fig. (migr., fish, management; decline, causes and remedies; range); Warfel and Olsen, 
Copeia, 1947: 177-183 (var. in no. of vert, and signif.); Bigelow and Schroeder, Fish. Bull. (74) 
U.S. Fish Wildl.Serv., 53, 1953: 108-112 (descr., life hist.. Gulf of Maine). 

Shad, Clark, Rep. U. S. Comm. Fish. (188 1), 1884: 783 (amount of water needed to keep eggs and young 
alive); Rice, Rep. U.S. Comm. Fish. (1881), 1884: 787 (retard, develop, of eggs); Willey, Contr. Canad. 
Biol., N. S.J (16), 1923: 313-320 (food); Johnson, Invest. Rep. U. S. Bur. Fish., 2 (38), 1938: 1-42, 
figs. 1—9 (nat. hist., sources of fishery, capture, wholesale trade, consump. in homes and public eating 
places, nutr. value, recipes, etc.); Hollis, Science, 108 (2804), 1948: 332 (tagging young; return of 
mature fish to place of hatch.). 

Doubtful References: 

Clupea indigena Mitchill, Rep. in part on Fishes of New York, 1814: 22 (orig. descr.; type local, presumably 
New York; type lost); Mitchill, Trans, lit. philos. Soc N. Y., J, 181 5: 454 (descr. inadequate forident.). 



3o8 Memoir Sears Foundation for Marine Research 

Negative Reference: 

Clupea sapidissima Jordan and Gilbert (not of Wilson), Proc. U. S. nat. Mus., J, 1882: 247 (proportions and 
counts, showing specimens were A.alabamae, not recognized until 1896; Pensacola, Florida). 



Alosa alabamae Jordan and Evermann 1896 

Alabama Shad, Gulf Shad, Ohio Shad 

Figures 70, 75 

Study Material. A total of 23 specimens, 175-505 mm TL, 145-395 mm SL, 
from the Choctawhatchee River, Pensacola, Florida (including two cotypes of A. ala- 




FiGURE 75. Alosa alabamae, 380 mm TL, 15 inches. Black Warrior River, Tuscaloosa, Alabama, tj'pe, USNM 
47689. Drawn by A. H. Baldwin. 

bamae)\ tributary of the Flint River, Georgia; Black Warrior River, Tuscaloosa, Ala- 
bama (male and female types); Coosa River, and the Alabama River, tributary of the 
Tombigbee River, Alabama; Washita River, near Hot Springs, Arkansas; Ohio River 
at the falls, Louisville, Kentucky (including type and several cotypes of A. ohiensis) ; 
and Falrport, Iowa. 

Distinctive Characters. See Key to Species, p. 294, and pp. 304, 309, 310. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on the 23 study specimens, 145-395 mm SL. 



Body: depth 25.5-36. 
Caudal peduncle: depth 8.0—9.0. 
Head: length 23.5-26.5. 
Snout: length 5.5—6.9. 
Eye: diameter 4.2—6.5. 
Interorbital: width 4.5-6.0. 



Maxillary: length 11 — 13.5. 
Pelvic fin: length 10— 11. 5. 
Pectoral fin: length 15.3— 17.5- 

Scales: ca. 55—60, often lost in preserved 
specimens. 



Fishes of the Western North Atlantic 309 

Ventral scutes: 35-38. anal 19-22; pectoral 15-18, usu- 

Gill rakers: 42-48 in adults. ally 16 or 17. 

Fin rays: dorsal 16-20, usually 17-19; Vertebrae: 55 (i specimen). 

Body shaped as in A. sapidissima, the depth in general increasing with age and 
growth, 2.8-3.9 '" SL in adults. Head 3.7-4.25 in SL. Snout 3.8-4.4 in head. Eye 
4.2-6.5. Interorbital 4.1-6.0. Maxillary 1.95-2. 15. Scales only moderately ad- 
herent, about 15 or 16 rows between base of pelvic and anterior rays of dorsal fin. 
Pelvic fin 2.2-2.5 in head. Pectoral fin 1.4-1.5. 

A. alabamae differs from J. sapidissima in only the few respects shown in the 
following comparison: 

A. sapidissima A. alabamae 

Gill rakers close-set, those at angle of Gill rakers less crowded, those at angle 

first arch longer than snout, 59-73 of first arch about as long as snout, 

on lower limb in adults 300 mm and 42-48 on lower limb in adults 300 

upward. mrn and upward in length. 

Surface of cheek and opercular bones Surface of cheek and opercular bones not 

glossy, generally without conspicu- glossy, covered with conspicuous mu- 

ous mucous pores and canals. cous pores and canals. 

Axillary scale of pelvic fin generally only Axillary scale of pelvic fin generally about 

a little more than half of length of fin. 75 "/o of length of fin. 

Size. The Alabama Shad seems to run smaller in size than the American shad; the 
largest female seen was 510 mm (20.4 in.) and the largest male 435 mm (17.4 in.) 
long, both from the Ohio River at Louisville, Kentucky. 

Development. Apparently the eggs and young have not been described. Ten small 
specimens, 48-90 mm TL, 36-73 mm SL, all previously misidentified as Pomolobus 
chrysochloris (USNM 36424, 36620, 36835, and 62225), are no doubt in their first 
year; they were collected from: the Stone River at Nashville and the Clinch River at 
Clinton, Tennessee; the Washita River, and the Mulberry River at Mulberry, Arkan- 
sas. There is also at hand a specimen 105 mm TL (93 mm SL) taken in brackish water 
at Grand Isle, Louisiana. The young differ from their adults principally in having a 
more slender body and in possessing notably fewer gill rakers; the depth is 3.6-4.1 in 
SL and the gill rakers on the lower limb vary between 27-33. The increase in the 
number of gill rakers seems to be rapid, as five half-grown examples 175-215 mm TL 
(145-17 1 mm SL) from Pensacola already have 38-41 gill rakers. 

"Numerous young, 8 or 9 inches in length" were reported for Pensacola, Florida, 
by Jordan and Gilbert (75: 247) as A. sapidissima, before A. alabamae had been recog- 
nized; five of these (USNM 30809) may be part of that material; they are typical 
of the Alabama Shad and are the only half-grown samples at hand. They, like the smal- 
ler American shad, are more slender than their adults and have fewer gill rakers; but the 
raker count in these is higher than in the smallest examples of alabamae reported above. 



3 I o Memoir Sears Foundation for Marine Research 

Spawning and Migrations. A. alabamae, like its close relative A. sapidissima, ascends 
streams in spring and spawns in fresh water. Sexually mature Alabama Shad, ob- 
served at Keokuk Dam, Iowa, in 19 15 and 1916 from the early part of May to the 
middle of July, were in upstream migration to spawn but were stopped by the dam, 
except for a few that might have passed through the lock (27: 169). Coker believed 
there were enough there "to support a substantial fishery." Other runs in the spring 
of the year have been reported for the Black Warrior River (Tuscaloosa) and at Mobile, 
Alabama {32: 127); the Kanawha River at Montgomery, West Virginia; and the 
Ohio River at the falls, Louisville, Kentucky (jj: 275—276). An apparently reliable 
source reported still other runs to Dr. Evermann — for the Ohio River about 30 
miles below Louisville, Kentucky, and for the Mississippi River below Memphis, 
Tennessee. 

In reply to a letter of inquiry addressed to the State Department of Conser- 
vation, Montgomery, Alabama, Mr. Ben. C. Morgan, Director of Conservation, replied 
(June 22, 1946): 

During the months of May and June, and sometimes earlier, we still have enormous quantities of fresh 
water shad which run or attempt to run up our main streams. I say attempt to run because of the fact that the 
Coosa, Tallapoosa, Tennessee, and Tombigbee Rivers all have locks and dams which prevent the shad from going 
further than these dams. Only last year on the Tennessee they banked below the dams by the millions and could 
be scooped with nets, which we do not allow, of course. 

It seems highly probable that the young of this species descend the rivers to salt 
water at the end of their first summer, and it may be assumed, therefore, that the habits 
of the Alabama and American shads are very similar. 

Food. It is well known that anadromous fishes, while migrating to their spawning 
grounds, do not feed regularly, if at all. The Alabama or Ohio Shad is no exception 
to this rule, for Mr. Stringham found that 105 of 163 stomachs examined at Keokuk, 
Iowa, were empty (27: 171). Others contained small quantities of insect and plant 
remains as well as some unrecognizable debris. 

Variations. Although adults of alabamae and sapidissima differ widely in the number 
of gill rakers, the number in the very young is so similar in this respect that overlapping 
definitely occurs (see Description). 

Examples taken from the Ohio River at the falls, Louisville, Kentucky, differed 
sufficiently from alabamae to be recognized by Evermann as a distinct species, which 
he named Alosa ohiensis (ji: 277). He described ohiensis as differing from alabamae 
principally in a more elongated body and in the greater number of gill rakers on the 
first arch. However, the seven large specimens from the Ohio now at hand (including 
the type), though representing the extremes in the two characters mentioned, intergrade 
with the other 1 1 sexually mature examples (including the type material of alabamae 
from Tuscaloosa) from the several other localities in the Gulf drainage already listed. 
Therefore, ohiensis at most is only subspecifically distinct. It is herein regarded as a 
synonym of alabamae. 

Commercial Importance. This species has never been an important foodfish; only 



Fishes of the Western North Atlantic 311 

small catches have been recorded a couple of times in the statistical reports of the former 
U. S. Fish Commission: for 1889, 6,955 pounds from the Ohio River in Indiana and 
Kentucky (1902: 667); and for 1902, 150 pounds from Alabama (1905: 416). No 
later catches seem to have been reported. By correspondence with Herman O. Hessen, 
U. S. Fish and Wildlife Service, Louisville, Kentucky, it was learned that no one now 
(1946) fishes for the "Ohio Shad" at that place, though a few are taken incidentally 
each spring. That these Shad were, and perhaps still are, numerous enough in some 
places to be taken in considerable quantities is evident, but no special effort is apparently 
made to catch them. In fact, the fish seem to be regarded in some localities as too bony 
for human use. At least one fisherman Interviewed by Mr. Hessen at Louisville, Ken- 
tucky, stated that the fish were so bony that fishermen threw them on the bank and made 
no efi-brt to sell them. Nevertheless, the food qualities of this Shad, tested by those 
who are familiar with the delicious American shad, have been found not at all inferior 
(jj: 275; 27: 169). 

It seems probable that Alabama Shad constitute at least a small unutilized food 
supply, but their actual abundance remains unknown. Evermann indicated that if proper 
equipment were used, considerable catches might be taken (jj: 279); for example, a 
great Increase in the catch resulted at Louisville, Kentucky, when surface-fishing seines 
were substituted for bottom-fishing seines. However, until people living on the shad 
streams of the Gulf drainage recognize its delicious food qualities, there will be little 
demand; hence no special effort to catch them will be made. Because of the many dams 
constructed across rivers, barring them from the spawning grounds, a decrease In 
abundance Is no doubt taking place. 

Despite an unutilized supply of Alabama Shad in streams tributary to the Gulf 
of Mexico, large shipments of American shad fry, A. sapidissima, were liberated in those 
streams (2). However, no American shad are Included among the rather numerous 
specimens examined, nor has any authentic record of their occurrence in those streams 
been found. Indeed, Alabama Shad seem to be referred to as "Potomac shad" by fisher- 
men who believe they are the progeny of American shad planted by the former U. S. 
Fish Commission. 29 But it seems quite certain that the Introduction failed. 

Range. The Alabama Shad is represented by specimens in the National Museum 
and has been recorded for all the principal streams tributary to the Gulf of Mexico 
(exclusive of the Pearl River), from the Mississippi eastward to the Choctawhatchee 
River, about 50 miles east of Pensacola, Florida. A 105-mm specimen from Grand 
Isle, Louisiana, is also at hand. In the Mississippi Valley it is known from as far west 
as Hot Springs, Arkansas, and northward to Fairport, Iowa, and Montgomery, West 
Virginia. 

Synonyms and References: 

Clufea sapidissima Jordan and Gilbert (not of Wilson), Proc. U. S. nat. Mus., 5, 1882: 247 (meas., counts 

29. Fishermen generally do not seem to distinguish Alabama Shad from skipjack herring, Pomolobus chrysockloris; ac- 
cording to Coker (27:165) they are referred to as "Government shad" by the fishermen at Keokuk, Iowa, apparently 
in the belief that they are the result of these plantings. 



312 Memoir Sears Foundation for Marine Research 

of half-grown examples from Pensacola, Florida); Bollman, Proc. U. S. nat. Mus., 9, 1886: 462 (two large 
ones, Escambia R., Alabama). 

Clupea chrysochlorh Meek (not of Rafinesque), Bull. U. S. Fish Coram., 14, 1894: 85 (Mulberry, Arkansas; 
USNM 62225 A.alabamae). 

Alosa alabamae Jordan and Evermann in Evermann, Rep. U. S. Comm. Fish. (1895), 21, 1896: 203 (orig. 
descr.; type local. Black Warrior R., Tuscaloosa, Alabama; types USNM 47689 [9] and 47690 [^]; 
cf. A.sapidissima); Evermann and Kendall, Bull. U. S. Fish Coram. (1897), IJ, 1898: 127 (distr.); Jor- 
dan and Evermann, Bull. U. S. nat. Mus., 47 (3), 1898: 2810, figs. 192, 192a (descr., refs.); Regan, 
Ann. Mag. nat. Hist., (8) 18, 1916: 9 (refs., diagn.); Jordan, Evermann, and Clark, Rep. U. S. Corara. 
Fish. (1928), 2, 1930: 43 (range, ref.). 

Alosa ohiensis Everraann, Rep. U. S. Comm. Fish. (1901), 27, 1902: figs, i and 2, $ and (^ ohiensis; figs. 3 and 
4, (J and $ alabamae (orig. descr.; tj'pe local. Ohio R., Louisville, Kentucky; types USNM 50469; 
di. alabamae and sapidUsima; distr.); Jordan, Evermann, and Clark, Rep. U. S. Comra. Fish. (1928), 
2, 1930: 42 (range, ref.) ; Coker, et al., Bull. U. S. Bur. Fish (1929), 45, 1930: 169 (Keokuk, Iowa 
191 5, 1916; influence of dam on migr.; an unused food). 

Alosa alabamae ohiensis Regan, Ann. Mag. nat. Hist., (8) 18, 1916: 9 (ref., diagn.). 



Genus Pomolobus Rafinesque 1820 

Alewives, River Herring 

Pomolobus Rafinesque, West. Rev. Misc. Mag., 2, 1820: 170; Ichthyol. Ohiensis, 1820: 38; genot)'pe by mono- 
typy, Pomolobus chrysochloris Rafinesque. 

Generic Synonym: 

Alausella Gill, Proc. Acad. nat. Sci. Philad., Suppl., i86r : 54; Proc. Acad. nat. Sci. Philad., 1861 : 36; genotype 
by monotypy, Clupea parvula Mitchill equals probably young of Clupea pseudoharengus Wilson. 

Characters. Body oblong, compressed. Bony scutes strong, with sharp points; a 
sharp edge on median line of chest and abdomen. Cheek generally longer than deep, 
about as long as deep in P. mediocris. Mouth large. Maxillary extending to or 
beyond middle of eye. Upper jaw with a more or less definite median notch in adults. 
Lower jaw projecting or included, its upper margin within the mouth rising more or 
less abruptly. Teeth in jaws not entirely wanting in adults; present at least anteriorly 
in lower jaw, though rudimentary in some species; present on median line of tongue; 
wanting on vomer. Dorsal with 1 5-2 1 rays, the last one not produced, its origin 
nearer to tip of snout than to base of caudal. Anal with 16—23 rays, the last one not 
especially enlarged. Pelvic with 9 rays. Vertebr.'\e 46—55. 

Remarks. Two members of this genus, mediocris and chrysochloris, are close to the 
American species oi Alosa in the general structures about the head, except that the lower 
jaw projects beyond the upper one; they diflfer strongly, however, in having fewer gill 
rakers, which do not increase in number with age (18—24 on lower limb). The other 
two species, aestivalis and -pseudoharengus., herein recognized, agree more or less with the 
species oi Alosa in the number of gill rakers (38—51 on lower limb) and in having the 
lower jaw included ; they differ, however, in having much shallower cheeks, which are 
much longer than deep, and in having the upper margin of the lower jaw rising abruptly 
within the mouth to form a more or less obtuse angle. 



Fishes of the IFestern North Atlantic 



313 




Figure 76. Pomolobus heads, showing shape of mandible and cheek, a, chrysoch/oris, 54 mm TL, USNM 9424; 
B, mediocris, 122 mm TL (after Hildebrand and Schroeder); c, aestivalis, 45 mm TL, USNM 69834; 
D, pseudoharengus, 48 mm TL, USNM 69833. a drawn by Ann S. Green, b by Louise Nash, c, d by Nancy 
D. Patton. 

Range. Three species occur in the Atlantic, from Nova Scotia to northern Florida, 
and one occurs in the Gulf of Mexico and tributary streams as far west as Corpus 
Christi, Texas, ranging northward to Minnesota (see Key). 



Key to Species 

Lower jaw strongly projecting, extending nearly or quite into dorsal outline of 
head (Fig. 76 a, b); only 18—24 gill rakers on lower limb of first arch, not increas- 
ing in number with age. 

2 a. Greatest depth of body usually less than length of head, 3.5-4.1 (usually 
3.8-4.0) in SL; teeth present in jaws at all ages, those in lower jaw prominent; 
bluish green color of back ceasing abruptly, not blending into the silvery 
color of side; no spot at shoulder. chrysochloris Rafinesque 1820, p. 315. 



314 Memoir Sears Foundation for Marine Research 

2 b. Greatest depth of body usually greater than length of head, 3.0-3.75 (usually 

3.2-3.6) in SL; teeth missing in upper jaw of adults, those in lower jaw very 
small (rudimentary); the grayish green color of back blending rather gradually 
into silvery color of the side; a dark spot at shoulder just behind head, often 
followed by several smaller spots. mediocris (Mitchill) 18 14, p. 319. 

I b. Lower jaw included or projecting but little, definitely not extending into dorsal 
outline of head (Fig. 76 c, d); gill rakers 38-41 on lower limb in adults, numer- 
ous at all ages. 

3a. Diameter of eye generally about equal to length of snout in adults; 41-52 
(most frequently 44-50) gill rakers on lower limb in adults; peritoneum 
usually black, sometimes pale with dusky dots; back in fresh examples bluish. 

aestivalis (Mitchill) 18 14, p. 324. 

3 b. Diameter of eye longer than snout at all ages; 38-44 (most frequently 39-41) 

gill rakers on lower limb in adults; peritoneum pale, with dusky dots; back 
in fresh examples grayish green. pseudoharengus (Wilson) 181 1, p. 332. 






Figure 77. Pomoloius gill rakers on first branchial arch, top, aestivalis, 300 mm TL, USNM 12 5701; 
LEFT, mediocris, 310 mm TL, USNM 118156; kight, pseudoharengus, 292 mm TL, USNM 117808. All 
drawn by Nancy D. Patton. 



Fishes of the Western North Atlantic 



315 



Pomolobus chrysochloris Rafinesque^" 1820 
Skipjack Herring, Skipjack, Golden Shad, Blue or Green Herring, River Herring 

Figures 76, 78 

Study Material. A total of 32 or more specimens, 1 15-420 mm TL, 95-341 mm 
SL; from Pensacola, Florida; Wetumpka, Montgomery, and Wheeler Reservoir, 
Alabama; Pascagula, Chickasawhay River, Pearl River, Eagle Lake, and Vicksburg, 
Mississippi; shrimp grounds (trawl) in Louisiana waters. Pass Manchac, and borrow 




Figure 78. Pomolobus chrysochloris, 320mm TL, Pensacola, Florida, USNM 30159. Drawn by S.Stearns. 

pits in Mississippi River at Lake Providence, Louisiana; Mustang Island (Gulf side), 
Texas; St. Francis River, Mississippi River, Miles County, Arkansas; Lexington, 
Missouri; Mississippi River, Keokuk, and Fairport, Iowa; Cairo and Mississippi River, 
Victory, Wisconsin; and Fort Snelling, Minnesota; mostly in USNM, but some 
loaned by Miss Fannye A. Cook, Mississippi Game and Fish Commission, Percy 
Viosca, Southern Biological Supply Co., and Reeve M. Bailey, University of Michigan. 
Distinctive Characters. P. chrysochloris is closely related to P. mediocris, the hickory 
shad, of the Atlantic coast; the principal differences are indicated in the account 

30. Although this species has long been considered the genotype of Pomolobus, it may be questioned with good reason 
whether Rafinesque's descriptions of the genus and species are based on the Skipjack Herring or the Alabama shad, 
Alosa alabamae, or on both. If Rafinesque had both species, as suggested by his accounts, he confused them. In his 
generic account he said "Jaws without teeth." This is true of adult Alabama shad but not of the Skipjack. In fact, 
the latter has such prominent teeth in the lower jaw that they would scarcely be overlooked. Furthermore, in the 
specific account Rafinesque used the local name, Ohio shad, together withj^Gold shad and Green herring. Present- 
day fishermen certainly distinguish the two species and refer to A. alabamae as the shad and to P. chrysochloris as 
the herring (generally with a modifier, such as blue, green, or river). 

When listing local names, Rafinesque remarked : "A fine fish from twelve to eighteen inches long. Flesh esteemed, 
white and less bony than the shad." The comparison evidently is with the American shad. The remark concerning 
the flesh seems to suit the Alabama shad, but not the Skipjack. On the other hand, the size given is more nearly that 
of the Skipjack than that of the shad; yet an i8-inch Skipjack would be an e.\ceptionally large one. It is evident, 
then, that one cannot be certain as to the species on which Rafinesque's descriptions were based. Because of the 
doubt involved, it seems advisable to continue the use of Pomolobus chrysochloris for the Skipjack Herring. To do 
otherwise would lead to great confusion of both generic and specific names. 



Body, depth 24.2-28.4. 
Caudal peduncle: depth 7.8 — 11. 
Head: length 24.6-27.8. 
Snout: length 4.8—6.3. 
Eye: diameter 4.4—7.0. 
Interorhital: width 3.8—4.8. 
Maxillary: length 10.5— 12.5. 
Pelvic fin: length 10— 12.6. 
Pectoral fin: length 16—19. 



316 Memoir Sears Foundation for Marine Research 

of mediocris (p. 322). Both of these species are at once distinguishable from the other 
species of Pomolobus and from Alosa by the much smaller number of gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 95—341 mm SL. 

Scales: 51—60, often difficult to count 

accurately. 
Ventral scutes: 33-37, most frequently 

34 or 35. 
Gill rakers: on lower limb of first arch 

20—24, most frequently 20—23. 
Fin rays: dorsal 16—21, most frequently 

18— 20; anal 18—21, most frequently 

1 9 or 2 o ; pectoral 1 6 or 1 7, rarely 1 5 . 
Vertebrae: 53—55 (4 specimens). 

Body slender, compressed, its greatest thickness generally somewhat less than 
half of its depth, its depth increasing with age and growth, usually equal to or less than 
length of head, 3.5—4.1 in SL. 

Scales moderately adherent, with slightly irregular membranous margins, not 
preceded by striae; 15 or 16 longitudinal rows of scales on body between base of pelvic 
fin and anterior rays of dorsal. Ventral scutes well developed, 17—21 (most frequently 
19 or 20) in advance of pelvics and 14-17 behind them. 

Head 3.6—4.0 in SL. Snout 4.2—4.8 in head. Eye 3.8—5.9. Interorbital 5.4- 
7.5. Cheek, from lower rim of orbit, somewhat longer than deep. Mouth moderately 
oblique. Upper jaw with a definite median notch in adults. Lower jaw projecting 
strongly and extending into dorsal profile of head, its upper margin within mouth 
rising gradually, not forming an angle (Fig. 76 a). Maxillary broad, its width 
about 0.67 of length of snout, rounded posteriorly, reaching nearly or quite below 
posterior margin of pupil, 2.25—2.4 in head. Gill rakers not increasing in number 
with age. Teeth in anterior part of both jaws at all ages, those of lower jaw rather 
prominent; sockets present if teeth are broken; teeth on tongue prominent, in a 
rather large elongated patch. 

Dorsal fin with slightly concave margin, its origin scarcely an eye's diameter in 
advance of pelvic and usually equidistant from margin of upper jaw and a vertical from 
last ray of anal; a scaly sheath on dorsal base, composed of a single row of scales and 
extending to 4th or 5th ray from end of fin. Caudal deeply forked, the lower lobe 
rather longer than the upper, about as long as head. Anal low, its margin nearly straight, 
its origin usually about equidistant between base of pelvic and base of caudal ; a scaly 
sheath present on base. Pelvic fin inserted near ventral edge and usually nearer to 
base of pectoral than to origin of anal, 2.15-2.4 in head. Pectoral fin placed some- 
what higher on side, 1.4— 1.6 in head. 

Color. Fresh specimens deep bluish green above, this color ceasing abruptly at 



Fishes of the Western North Atlantic 317 

level of upper margin of gill cover. Sides light green, with strong tinge of gold, espe- 
cially on head, shading into the white of belly. No dark shoulder spot. Tips of jaws 
blackish. Dorsal fin yellowish, more or less dusky at base; caudal yellowish, tips of lobes 
dusky; anal and pelvic pale; pectoral pale, a dusky streak on inner side behind first 
ray. Peritoneum pale. 

Sixe. Skipjacks are smaller in size than Alabama shad, with which they are more 
or less associated in the rivers; their length usually ranges between about 300-450 mm 
(12-18 in.). 

Development. Although ovarian eggs 0.08-1. 11 mm in diameter were observed, 
apparently no young under 107 mm (about 4.33 in.) were secured during Dr. Coker's 
extensive investigations at Keokuk, Iowa {26 ; 2y^. Each of the smallest specimens at 
hand (UMMZ 122510, 128570), from the mouth of Elk River, Wheeler Reservoir, 
Alabama, and from the Mississippi River, Miles County, Arkansas, is about 40 mm 
(31 mm SL), apparently the smallest known. They differ from larger examples princi- 
pally in being more slender, the depth of each being contained 4.75 times in SL. 
Furthermore, they have minute teeth on the margin of the maxillary, these persisting 
in examples up to 150 mm or so in length. The gill rakers, unlike those in some other 
species of the genus Pomolobus, do not increase in number with age and growth. 

Spawning and Migrations. The spawning place and length of spawning are not 
fully known. For such information as is available we are indebted to Robert E. Coker, 
who, with assistants, carried on an extensive investigation at Keokuk, Iowa (26: 25, 26; 
2^: 94, 165-169). They reported fish with large roe taken there on: April 29 and 
May 23 in 19 14; April 29 and 30, May 24, June 5 and 23, July i and 2, and 
"various other dates" in 191 5; and on May 23, June 3 and 17 in 191 6. Fully mature 
fish, ripe for stripping, apparently were not included. Coker concluded his remarks on 
spawning as follows: 

Many attempts made during several years to find a particular place and time where and when river herring 
were clearly spawning met with no success, although various methods were tried. The problem proved unexpec- 
tedly baffling. Our data suggest only that the spawning season is prolonged, possibly beginning early in May; 
that the fish do not spawn in large aggregations; that during spawning operations they are not readily captured 
by ordinary methods of fishery; and that spawning is ended soon after the first of Juh'. 

It has not been proven that Skipjack are anadromous like most of the near rela- 
tives, even though Jordan said categorically, ". . . migrating from the Gulf of Mexico" 
(jl: 40). However, they do enter salt and brackish water. Such a great "run" as was 
reported at Keokuk, Iowa, in the spring of 19 14 as well as smaller runs in succes- 
sive years and their apparent disappearance in September of each year do suggest 
extensive migration. However, Coker expressed the opinion that the migrations do not 
necessarily extend to extreme northern or extreme southern waters. This is in contrast 
to the strictly anadromous relatives, Alosa sapidissima, Pomolobus aestivalis, and P.pseudo- 
harengus (except where the last species is landlocked), whose young leave the fresh- 
water streams at the end of their first summer and do not return until they are ready 
to spawn. 



3 1 8 Memoir Sears Foundation for Marine Research 

Evidence has been set forth that chrysochloris reproduced in Lake Pepin after 
the closing of Keolcuk Dam, an almost complete barrier to them. Dr. Coker said that 
without question there was a decided decline in the number of "River Herring" in this 
Lake during the three years immediately following the dam's construction, but some 
commercial catches there in 1926, 12 years after the dam was closed, convinced him 
that these fish were not rare; and several commercial fishermen stated that "the 
herring were coming back." On the basis of Dr. Coker's observations, it seems 
proper to conclude that P. chrysochloris, although migratory, does not need to have 
access to salt water to reproduce, and the species surely is not strictly anadromous. 
The presence of immature examples in a collection from brackish water as well as from 
the upper Mississippi also may be of some significance. 

Food. Skipjack Herring are carnivorous (^5: 49); the young feed on insects, the 
adults on fish. Coker stated that, of about 150 examples examined for food: 

Approximately one-third were empty; a little more than one-third contained fish, chiefly minnows, with some 
mooneyes, gizzard shad, and other fish not determinable; and less than one-third contained insects and larvae, 
principally May flies, with some caddis flies and others. 

Habits. These are more active fish than most other clupeoids ; they frequent rivers 
with swift currents and often leap from the water, a habit from which they have derived 
the name Skipjack. They leap In part for sport and in part for the pursuit of prey (j5: 
49). Coker remarked that they have, "a striking habit of leaping from the surface 
when feeding upon schools of minnows or in mere play." 

Parasites. We know only that the larvae (glochidia) of the freshwater mussel Fus- 
conaia ehena are parasitic on the Skipjack (see below). 

Commercial Importance. The value of this species as a foodfish Is negligible, for 
these fish are bony and not especially well flavored. No catches have been listed in the 
statistical reports of the U. S. government. However, they represent a very distinct 
economic asset as a host for the larvae of Fusconaia ebena, which has been regarded 
as the most valuable of all the pearly mussels of the Mississippi Basin because of its 
abundance in all the larger waters of strong current and because it yields a shell of the 
best quality for buttons {2y. 166). 

Range. Skipjack Herring range in the Gulf of Mexico from Pensacola, Florida, 
to Corpus Christi, Texas, and sometimes out into the Gulf, as off Breton Island, 
Louisiana. They also inhabit larger streams, occasionally lakes, or even borrow pits, of 
the Gulf drainage. They commonly range as far north in the Mississippi River as Lake 
Pepin, and occasionally to Fort Snelling, Minnesota, to Hudson, Wisconsin, and in 
the Ohio River to Pittsburgh, Pennsylvania. Apparently they are landlocked above 
Keokuk Dam, Iowa. Although Smith {118: 214) said that they were abundant in Lake 
Erie, and Jordan and Evermann (72, 1896: 425) as well as several others have stated 
that they reached Lake Michigan and Lake Erie through canals, Hubbs and Lagler 
{62: 27, ftn.) do not admit them to their list of fishes of the Great Lakes and tributary 
waters. The species apparently Is most common In swift currents of large streams. 



Fishes of the IVestern North Atlantic 319 

Synonyms and References: 

Pomolobus chrysoch/oris Rafinesque, West. Rev. Misc. Mag., 2, 1820: 171; and Ichthyol. Ohiensis, 1820: 39 
(orig. descr.; type local. Ohio R.; type lost; length 12-18 in. attained; seldom ascend to Pittsburgh) ;^'^ 
Jordan and Evermann, Bull. U. S. nat. Mus., 47(1), 1896: 425; 47(4), 1900: fig. 187 (descr., size, 
distr., habits, value as food); Jordan and Evermann, Amer. Food Game Fishes, 1902: loi (descr., range; 
Lakes Erie and Michigan erroneously included); Bean, T. H., 7th Rep. Forest Comm. N. Y. (1901), 
1902: 300, fig. (synon., range, habitat; reported to have reached Great Lakes through canals, but occur, 
there seems doubtful); Bean, T. H., Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 195 (synon., descr., value 
as food, habits, distr.); Forbes and Richardson, Nat. Hist. Surv. 111., Ichthyol., J, 1908: 48 (refs., descr., 
range, occur, in Illinois, size, habits); Coker, Rep. U. S. Comm. Fish. (1913), Append. 8, 1914: 25-28 
(large schools below Keokuk Dam, Iowa; discus, migr. and ascent up spillways of dam; occur, above 
dam; spawn.); Regan, Ann. Mag. nat. Hist., (8)79(112), 1917: 299 (refs., descr., range); Jordan, 
Manual Vert. Anim. NE U. S., 1929: 40 (descr., range, habitat); Coker, Bull. U. S. Bur. Fish. (1929), 
45> 1930: 165, fig- II (below and above Keokuk Dam, Iowa, migr., spawn., econ. import.); Hubbs 
and Lagler, Bull. Cranbrook Inst. Sci., 18, 1941: 27 (not in list of Great Lakes fishes though reported 
there earlier); Gunter, Publ. Inst. mar. Sci. Texas, I (l), 1945 : 25 (Texas coast, Colorado R. at Austin). 

Meletta suoerii Cuvier and Valenciennes, Hist. Nat. Poiss., 20, 1847: 375 (orig. descr.; type local. Wabash 
R., tributary of Mississippi R.); Bertin, BuU. Soc. zool. Fr., 66, 1941 : 23 (ident. of type). 

Clufea chrysochloris Jordan and Gilbert, Proc. U. S. nat. Mus., 5, 1882: 247 (descr.; records two specimens; 
from Galveston, Texas; and Pensacola, Florida); Jordan and Gilbert, Bull. U. S. nat. Mus., 16, 1882: 
266 (descr., range; Lakes Erie and Michigan apparently includ. erroneously); Jordan and Gilbert, 
Proc. U. S. nat. Mus., 9, 1886: 11, 14 (Washita R., Arkadelphia, and Saline R., Benton, Arkansas); 
BoUman, Proc. U. S. nat. Mus., 9, 1886: 462 (few from Escambia R., Alabama); Gilbert, Bull. U. S. 
Fish Comm. (1889), 9, 1891: 158 (near mouth of Escambia R., Alabama); Smith, Bull. U. S. Fish 
Comm. (1890), JO, 1892: 214, pi. 31 (Lakes Ontario and Erie); Evermann and Kendall, Bull. U. S. 
Fish Comm., 12, 1894: 105 (Red R., Fulton, Arkansas; Galveston, Texas); Meek, Bull. U. S. Fish 
Comm., 14, 1894: 78, 82 (White R. at BatesviUe, Little Red R. at Judsonia, Arkansas). 

River herring, Coker, Bull. U.S. Bur. Fish., ^5, 1929: 91-104 (discus, abund. below Keokuk Dam, Iowa, 
1914-1916; dam a barrier to fish). 

Negative References: 

Clupea chrysochloris Meek (not of Rafinesque), Bull. U. S. Fish Comm., 14, 1894: 85 (Mulberry, Arkansas, 
but specimen USNM 62225 is A.alabamae). 



Pomolobus mediocris (Mitchill) 18 14 

Hickory Shad, Hickory Jack, Bonejack, Fall Herring, 
Freshwater Taylor, Shad Herring 

Figures 76, 77, 79 

Study Material. Forty-three specimens, 105-460 mm TL, 82-354 mm SL, from 
Provincetown, Massachusetts; Noank, Connecticut; Long Island, New York; Beesley's 
Point, New Jersey; Annapolis, Love Point, Benedict, Barren Island, Lewistown, and 
Smith Point, all on or near Chesapeake Bay, Maryland; Potomac River, Washington, 
D. C; Buckroe Beach and Lynnhaven Roads, on Chesapeake Bay, Virginia; Beaufort, 
North Carolina; and Brunswick, Georgia. 

31. The data are from Richard E. Call's Ichthyologia Ohiensis (1899), "A verbatim et literatim reprint of the original, 
with a Sketch of the Life, the Ichthyologic Work, and the Ichthyologic Bibliography of Rafinesque," the original 
pubhcations not being available to the author. 



3 2 o Memoir Sears Foundation for Marine Research 

Distinctive Characters. See Relationships for differences between mediocris and chry- 
sochloris (p. 322). From species oi Alosa and other Pomolobus species, excepting chryso- 
chloris, it is marked by the noticeably smaller number of gill rakers. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 43 specimens, 82—354 mm SL. 




Figure 79. Pomolobus mediocris, 309 mm TL, Brunswick, Georgia, USNM 1 181 56. Drawn by Nancy D. 
Patton. 



Body, depth 26.7-32.8. 
Caudal peduncle: depth 9.7— i' 
Head: length 24.5-29.6. 
Snout: length 5.9-7.5. 
Eye: diameter 4.9—7.1. 
Interorbital: width 3.9-5.4. 
Maxillary: length 10.4— 13.4. 
Pelvic fin: length 10.6— 12.4. 
Pectoral fin : length 17—21. 



Scales: 48-57. 

Ventral scutes: 33—38, most frequently 
36 or 37. 

Gill rakers: on lower limb 18—23, usually 
20 or 21. 

Fin rays: dorsal 15—20, most frequently 
17 or 18; anal 19—23, most fre- 
quently 20 or 2 1 ; pectoral 1 5 or 16. 

Vertebrae: 54 or 55 (2 specimens). 



Body moderately slender, well compressed, its greatest thickness notably less than 
half of its depth, its depth rather greater in large than in small specimens, usually 
exceeding length of head, 3.0—3.75 in SL. 

Scales only moderately adherent, with definitely crenulate membranous margin, 
preceded by longitudinal striae; about 16 longitudinal rows of scales on body between 
base of pelvic and anterior dorsal rays. Ventral scutes moderately developed, 19—23 
(usually 20—22) in front of pelvic fins, 12—17 (usually 13—16) behind them. 

Head 3.4—4.1 in SL. Snout 3.7— 5-0 in head. Eye 3.9—5.3. Interorbital ^.c^ — 
7.0. Cheek generally about as deep as long (Fig. 76 b). Mouth oblique. Upper jaw 
with a definite median notch. Lower jaw projecting strongly and extending into dorsal 
profile of head (Fig. 76 b), its upper margin within the mouth rising gradually, not 
forming an angle (Fig. 76 b). Maxillary broad, its width somewhat more than half 
of length of snout, rounded posteriorly, reaching about to vertical from posterior 



Fishes of the Western North Atlantic 321 

margin of pupil, 2.0—2.4 in head. Gill rakers apparently not increasing in number 
with age. Teeth in the upper jaw absent in large examples 230 mm SL and upward, 
but present in smaller ones; present in lower jaw, becoming exceedingly small in large 
examples; those on tongue minute, in a small elongated patch. 

Dorsal with slightly concave margin, its origin rather less than an eye's diameter 
in advance of pelvic, usually nearer to margin of upper jaw than to vertical from last 
anal ray; a sheath of scales on its base, consisting for the most part of more than one 
row of scales and extending nearly or quite to base of last ray. Caudal deeply forked, 
the lower lobe somewhat longer than the upper, fully as long as head. Anal low, its 
margin nearly straight, its origin about equidistant between base of pelvic and base 
of caudal; a sheath of scales on its base. Pelvic fin inserted near ventral edge and 
about equidistant between base of pectoral and origin of anal, 2.0-2.6 in head. Pec- 
toral placed somewhat higher on side, 1.3— 1.7 in head. 

Color. Fresh specimen from Chesapeake Bay grayish green above, shading some- 
what gradually into the iridescent silver of sides. Nape green, side of head brassy. 
Tip of lower jaw dusky. Dorsal and caudal dusky. Anal and pelvic plain, translucent. 
Pectoral slightly dusky. Narrow dark lines along rows of scales on upper part of side; 
missing on specimens less than 1 50 mm SL or so and most distinct in large examples 
that have lost their scales, the dark pigment being in the skin underneath the scales. 
A dark spot on shoulder; several obscure dark spots behind it, but missing in some 
examples. Peritoneum pale, but with scattered dusky punctulations. 

Size. Hickory Shad definitely run larger in size than blueback and grayback 
herrings but smaller than the shad. A maximum length of 600 mm (24 in.) has been 
reported, but the usual size is much smaller. The commercial catches observed by me 
at Chesapeake Bay and at Beaufort, North Carolina, consisted mostly of fish varying 
between about 300—375 mm (12 — 15 ^'^O '" length. 

Re-production, Development, and Migrations. Next to nothing is known about the 
breeding habits of Hickory Shad. Although it has been stated that they spawn in 
North Carolina in the spring, in the headwaters of streams (jxp: 121), spawners have 
never been seen in either fresh or salt water. In fact, many subsequent authors (.^5: 
607; 16: 106; 20: 65) have accepted without reservation the opinion that Hickory 
Shad, like other local members of that genus, are anadromous. However, the only 
supporting evidence is that there are definite spring and autumn runs; in Chesapeake 
Bay, only stragglers are seen in the summer; in the spring they are first taken in 
the lower part of the Bay in March and progressively later farther up the Bay in late 
March and early April. Also, I have observed adults with well-developed roe in the 
Potomac above Washington during the spring, but these were not ripe and may have 
returned to salt water to spawn. The view that this fish does not enter salt water to 
spawn {8: 197; Jj: 95) is supported by Hildebrand and Schroeder {^g: 84); among 
thousands of young clupeids collected during many years in Chesapeake Bay and its 
tributaries, in freshwater streams and estuaries, and in salt water out to 1 5 fms., they 
failed to find any Hickory Shad less than 15 mm (6.12 in.) long. Also, in the vicinity 



32 2 Memoir Sears Foundation for Marine Research 

of Beaufort, North Carolina, they were absent from extensive collections made by 
Hildebrand and his associates with gear of various types, though adults were taken 
in the estuaries there in winter and early spring. 

In short, Hickory Shad probably spawn in the spring, but it remains to be 
learned whether they spawn in fresh or salt water. However this may be, to the north 
of New York they are the most plentiful during autumn. 

Food. Hickory Shad feed chiefly on small fish of various kinds. Their fish diet 
is supplemented by squid, small crabs, various small crustaceans, and by fish eggs 

{59- 84)- 

Enemies and Parasites. Little is known about the enemies of this species ; but no 
doubt they are eaten by larger fish and are caught in limited quantities by man. Linton 
(1^5: 438) found individuals parasitized with nematodes {Ascaris sp.), cestodes (larval 
Scolex polymorphus)., and trematodes (Distomum appendiculatum). Wilson {136 : 1 2) found 
the Hickory Shad to be a host of the copepod Lernanthropus hrevoortiae. 

Relationships. Though closely related to P. chrysochloris, Hickory Shad have a 
heavier and less shapely (less streamlined) body; the depth in adults is greater than 
the length of the head (measured from margin of upper jaw) ; in chrysochloris, exclusive 
of very large or very fat specimens, the depth is equal to or less than the head. In 
mediocris the shield of scales on the dorsal fin base consists of more than one row of 
scales anteriorly and extends nearly or quite to the last ray of the fin, whereas in chryso- 
chloris it is composed of only one row of scales throughout and is missing on the last 
four or five rays. The scales on the side of the Hickory Shad have a pronounced crenulate 
membranous border preceded by longitudinal striations, while in chrysochloris the 
border is only slightly irregular and the striations not evident. In mediocris there is 
only a slight ridge above the base of the pectoral, which ends at the tip of the axillary 
process; in chrysochloris a more pronounced depression (pocket) is provided for the 
pectoral fin, and the more prominent ridge above its base extends well beyond the 
axillary process of the pectoral. 

In large Hickory Shad, 300 mm and upward, teeth in the upper jaw are missing, 
but in specimens at least up to 1 50 mm long they are present there ; however, those 
in the lower jaw persist, though very small; the patch on the tongue is very small, and 
the teeth minute. In chrysochloris the teeth are consistently larger, persist in the upper 
jaw among adults, are large in the lower jaw, and the patch on the tongue is notably 
larger. Although the proportions and counts overlap, it is evident that Hickory Shad 
have a proportionately deeper body, deeper caudal peduncle, more anal rays, and 
fewer gill rakers than P. chrysochloris. 

There are differences in color also, evident in preserved as well as fresh examples. 
Hickory Shad have a large dark spot behind the upper part of the opercle, often 
followed by several smaller dark spots, and the adults frequently have dark lines along 
the rows of scales on the upper part of the side. In P. chrysochloris these markings are 
missing, and the dark blue or grayish color of the back ends very abruptly, not tending 
to shade into the silvery color of the side, as in Hickory Shad. 



Fishes of the Western North Atlantic 323 

Commercial Importance. Hickory Shad are of minor importance as foodfish, mainly 
because the meat is bony and is considered inferior in flavor. Their value is somewhat 
higher in North Carolina than elsewhere, because they are taken there during the 
winter when other fish are scarce. In 1940 or 1941, the following catches were reported: 
New England (all reports for Rhode Island and Connecticut), 25,800 pounds; the 
Middle Atlantic States (all reports for New Jersey), 4,400 pounds; the Chesapeake 
Bay states, 196,200 pounds, of which 22,000 pounds were taken in Maryland and 
174,200 pounds in Virginia; and the South Atlantic states (all of which contributed), 
371,500 pounds, of which 335,900 pounds were listed for North Carolina. It is 
evident, then, that the fishery is centered in Virginia and North Carolina. In the 
United States (1940, 1945, 1950-1953), catches have ranged between o and 25,800 
pounds for southern New England, 200 and 4,400 pounds for the Middle Atlantic 
states, 86,000 and 264,000 pounds for Chesapeake Bay, and 273,000 and 970,000 
pounds for the South Atlantic states. 

Method of Fishing. The commercial fish are caught principally with seines and pound 
nets, the smaller quantities with gill and fyke nets. The fishery is not intense enough 
to affect greatly the abundance of these fish in nature. They take the hook occasionally 
and off"er some sport in the spring, at least at Little Falls on the Potomac River. Anglers 
trolling for striped bass (Roccus saxatilis) and mackerel {Scomber) landed a number of 
Hickory Shad off the Merrimack River, Massachusetts, in 1932. 

Range. The range of Hickory Shad extends from Florida to Maine. They are re- 
corded for Campobello Island, New Brunswick (52 : 47), but no record of their oc- 
currence south of the St. Johns River in Florida has been brought to my attention. 
Though rare north of Cape Cod, they are apparently more numerous in southern New 
England than in the Middle Atlantic States and are most abundant in Virginia and 
North Carolina. ^2 

Synonyms and References: 

Clupea mediocris Mitchill, Rep. in part on Fishes of New York, 1814: 20 (orig. descr. ; type local, presumably 

New York City; type lost); Mitchill, Trans. Lit. philos. Soc. N. Y., I, 181 5: 450 (descr.. New York 

Bay along Staten Island); McDonald in Goode, et al.. Fish. Fish. Industr. U. S., i (3), 1884: 607, 

pis. 216A, 216B (discus., names, distr., abund., econ. import., migr.). 
Clupea mattowaca MitchiU, Trans. Lit. philos. Soc. N. Y., J, 1815:451 (orig. descr. ; type local. Long Island, 

New York; commonly caught in autumn, size). 
Alosa matto-Lcacca DeKay, N. Y. Fauna, Pt. 4, Fishes, 1842: 260, pi. 40, fig. 127 (descr., New York; fig. 

shows no projecting lower jaw, though descr. says "lower jaw longest"). 
Alosa lineata Storer, Proc. Boston Soc. nat. Hist., 2, 1848: 242 (orig. descr.; tj'pe local. Massachusetts); Storer, 

Fish. Mass., 1867: 162, pi. 27, fig. 2 (descr., when and how taken, food value). 
Clupea mattoa'occa Gunther, Cat. Fish. Brit. Mus., 7, 1869: 438 (refs., descr.). 
Pomoloius mediocris Uhler and Lugger in Rep. Comm. Fish. Md., ed. i, 1876: 159; ed. 2, 1876: 136 (descr., 

synon., Potomac R., size); Jordan and Evermann, BuU. U. S. nat. Mus., 47(1), 1896: 425; 47(4), 

1900: fig. 188 (descr., range); 47(3), 1898: 2810 (correc. of earlier acct.); Smith, Bull. U. S. Fish. 

32. A listing of this herring among the clupeoid fishes of Brazil (25: 190) was based on a specimen 170 mm long from 
an "ancient collection," without mention of the place of collection. The description offered suits P. mediocris quite 
well. However, as this herring does not seem to occur on the Adantic coast in semitropical waters, as in southern 
Florida, one is justified in questioning whether it actually was taken in Brazil. 



324 Memoir Sears Foundation for Marine Research 

Comm. (1897), J7, 1898: 91 (Woods Hole, Massachusetts); Smith and Bean, Bull. U S. Fish Coram. 
(1898), 18, 1899: 183 (Potomac R., weight); Evermann and Kendall, Rep. U. S. Comm. Fish. (1899), 
1900: 55 (refs., Florida); Linton, Bull. U. S. Fish Comm. (1899), ig, 1901: 438 (parasites); Bean, 
T. H., 7th Rep. Forest Comm. N. Y. (1901), 1902: 301, fig., also color pi. (synon., names, size, New 
York); Jordan and Evermann, Amer. Food Game Fish., 1902: 102, fig. (names, habits, size, descr.) ; Bean, 
T. H., Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 197 (synon., descr., local and techn. names, size, habits, 
range); Fowler, Rep. N. J. St. Mus. (1905), 2, 1906: 95, 2 figs, (descr., habits, size, refs.); Smith, N. C. 
geol. econ. Surv., 2, 1907: 121, fig. 40 (refs., descr., range. North Carolina, names, econ. import.); Ken- 
dall, Occ. Pap. Boston Soc. nat. Hist., 7, 1908: 37 (refs.. New Engl, locals.); Halkett, Check List Fish. 
Canada, 191 3: 47 (Campobello L, New Brunswick); Regan, Ann. Mag. nat. Hist., (8) jp (112), 19 17, 
300 (descr.); Bigelow and Welsh, Bull. U. S. Bur. Fish. (1924), 40 (i), 1925 : 105 (descr., size, range, 
food. Gulf of Maine); Wilson, Proc. U. S. nat. Mus., 64 (17), 1925 : 12 (host of parasitic copepod); 
Hildebrand and Schroeder, Bull. U. S. Bur. Fish. (1927), 4^ (i), 1928: 83, figs. 42, 43 (synon., descr., 
distinct, char., food, migr., other habits, Chesapeake Bay); Jordan, Manual Vert. Anim. NE U.S., 
ed. 13, 1929: 40 (diagn., range); Truitt, Bean, and Fowler, Bull. Md. Conserv. Dep., 3, 1929: 49 
(descr., Maryland) ; Breder, Field Bk. Mar. Fish. Ad. Cst., 1929: 65 (relation., range, habits, size); 
Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 42 (range, synon.). 

Doubtful References: 

Clupea fusUla Mitchill, Rep. in part on Fishes of New York, 1814: 20 (orig. descr.; type local, presumably 
New York; type lost); Mitchill, Trans. Lit. philos. Soc. N. Y., j, 181 5: 452 (descr.). 

Negative Reference: 

Pomolobus mediocris Evermann and Hildebrand, young, not of Mitchill, but P. pseudoharengus (Wilson), Proc. 
biol. Soc. Wash., 2J, 1910: 158 (Chesapeake Bay locals.). 



Pomolobus aestivalis (Mitchill) 18 14 

Blueback Herring, Blueback, Glut Herring, Blue Herring, Summer Herring, 
Blackbelly, Alewife, Kyak 

Figures 69, 76, 77, 80, 81 

Study Material. A total of 75 or more specimens, 33—295 mm TL, 27—240 mm 
SL; from off Liverpool, Nova Scotia; Freeport and Portland, Maine; Gloucester, 
Cape Cod, and Woods Hole, Massachusetts; Noank, Connecticut; New York Bay, 
N.Y. ; Washington, D. C; Chain Bridge, Bryans Point, Plum Point, Havre de Grace, 
Patuxent River, Barren Island, and Smith Point, Maryland; Rappahannock River and 
Buckroe Beach, Virginia; Avoca and Beaufort, North Carolina; and St. Johns River, 
Florida; in USNM collections. 

Distinctive Characters. See Relationships, p. 330. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on study specimens, 27—240 mm SL. 

Body: depth 21-33.5. Interorbital: width 3.7-6.0. 

Caudal peduncle: depth 8.0—10.5. Maxillary: length 9.0—13.2. 

Head: length 21.5-30. Pelvic fin: length 9.4-13.6. 

Snout: length 3.8-5.8. Pectoral fin: length 15.5-2 1.5. 
Eye: diameter 5.0-9.5. 



Fishes of the Western North Atlantic 



325 



Scales: 46—54. Fin rays: dorsal 15—20; anal 16-21; 

Ventral scutes: 31-36. pectoral 14-16. 

Gill rakers: on lower limb 41-51 in Vertebrae: 49-53 {jiS specimens), 

adults, most frequently 45-50. 

Body rather slender, moderately compressed, its greatest thickness about 2.5 in 
the depth, its depth 3.0—4.6 in SL. 

Scales only moderately adherent, generally lost in young; with crenulate membra- 
nous edge preceded by scarcely visible striae: about 13 or 14 longitudinal rows of 




Figure 80. Pomolobus aestivalis, 265 mm TL, 10.6 inches, USNM 32639. 



scales on body between base of pelvic and base of dorsal. Ventral scutes rather 
strong, 18—21 in advance of pelvic fins and 12—16 behind them. 

Head frequently equal to depth, 3.33-4.6 in SL. Snout 3.6-5.8 in head. Eye 
small, equal to or shorter than snout in adults, 3.0-6.0. Interorbital 4.0-7.0. Cheek 
notably longer than deep (Fig. 76 c). Mouth oblique. Upper jaw with a definite 
notch in adults. Lower jaw not projecting strongly, partly included, not entering into 
dorsal profile, its upper margin within the mouth rising abruptly and forming a 
more or less definite obtuse angle (Fig. 76 c). Maxillary broad, its greatest width about 
75 "/o of length of snout, rounded posteriorly, reaching to, or more often somewhat 
beyond, vertical from middle of eye, 2.0-2.7 in head. Gill rakers at angle of first arch 
fully 0.75 of diameter of eye; increasing in number with age: 28—36 on lower limb in 
examples 30-49 mm SL, 30-39 in specimens 50-69 mm, 35-41 in specimens 70- 
89 mm, 38-44 in specimens 90-109 mm, 42-48 in specimens 110-129 mm, 42-50 
in specimens 130-149 mm, and 42-52 in examples 190-260 mm (adults). Teeth on 
premaxillary exceedingly minute in young, missing in adults; those on free rim of 
maxillary rather prominent in young and often slightly evident posteriorly in adults; 



326 Memoir Sears Foundation for Marine Research 

those present on lower jaw persistent anteriorly, though not with free points in adults; 
a permanent narrow band of granular teeth on median line of tongue. 

Dorsal fin with a definitely concave margin, its origin scarcely an eye's diameter 
in advance of the pelvic fin and usually a little nearer to margin of upper jaw than to 
vertical from last ray of anal; a narrow sheath of scales at base, incomplete, leaving 
base of last 4 or 5 rays exposed. Caudal deeply forked, the lower lobe the longer, a 
little longer than head. Anal low, with straight margin, its origin about equidistant 
between base of pelvic and base of caudal; a narrow sheath of scales at base. Pelvic 
fin inserted near ventral edge and about equidistant between base of pectoral and 
origin of anal, 1.8—2.6 in head. Pectoral somewhat higher on side, reaching about 
0.66 of the distance to base of pelvic, 1.2— 1.7 in head. 

Color. Fresh fish bluish above and sides silvery. Rows of scales on back and upper 
part of side with more or less distinct dark lines in adults. Shoulder with a dark spot in 
the larger examples, generally missing in individuals less than 100 mm long. Fins plain, 
generally greenish or yellowish in life. Peritoneum usually, though not always, dark. 

Size. Bluebacks are known to attain a length of about 380 mm (15 in.) and a 
weight of approximately 1 3 ounces. However, the usual size of these fish in commercial 
catches as observed in the Chesapeake Bay area is under 300 mm (12 in.) and a half 
pound or less in weight. 

Development and Growth. The eggs and early young were described by Kuntz 
and RadclifFe from material taken at Woods Hole, Massachusetts {y6: 123-126). 
The eggs are demersal, somewhat adhesive, semitransparent and yellowish, round, and 
approximately a millimeter in diameter. The development is rapid; hatching occurs 
at ordinary laboratory temperature about 50 hours after fertilization. 

The newly hatched larvae (Fig. 8 i a) are close to 3.5 mm long and quite slender. 
In about four days the yolk is all absorbed and the larvae have grown to a length of 
about 5 mm. The vent in these larvae, as in other clupeids, is situated tar back, about 
at the beginning of the posterior fifth of the body. 

The smallest specimen at hand, 20.5 mm long (16.5 mm SL), taken June 17 (1873) 
at Washington, D. C, is still very slender (Fig. 8 i b), its depth about 7.5 times in 
SL. Its eye is rather longer than the snout, the maxillary extends nearly to the pupil, 
and the fins are all developed, the pelvic being inserted about under the origin of the 
dorsal. In a 25-mm specimen (20 mm SL), the depth is contained 5.0 times in the length, 
and the pelvic fin is well behind the origin of the dorsal (Fig. 8 i c), as in adults. 
The 30-mm fish illustrated by Kuntz and Radcliffe {y6: fig. 100) is much deeper than 
specimens of somewhat similar size now at hand, its depth being about 3.6 times in 
SL; in specimens from the Rappahannock River, Virginia, 34, 35, and 38 mm (28, 
29, and 30 mm SL), the depth is respectively 4.6, 4.65, and 4.3 times in SL, and 
specimens even up to 45 and 50 mm long are more slender than the 30-mm one illus- 
trated. If the proportions are correctly shown, it seems highly probable that the 
illustrated specimen is not this species. 

When the young fish reach a length of about 80-90 mm they are usually pro- 



Fishes of the Western North Atlantic 



327 




,.<^5->. 




Figure 81. Pomolobus aestivalis growth stages; all from Potomac River, a, newly hatched larva, 3.5 mm, 
USNM 12830; B, young, stiU without scales, 20.5 mm TL, 16.5 mm SL, USNM 12830; c, young with 
scales developing, 25 mm TL, 20 mm SL, USNM 12830; d, young adult, 45 mm TL, USNM 69834. 
A drawn by Homer Wheelon; b, c by Ann S. Green; d by Nancy D. Patton. 



328 Memoir Sears Foundation for Marine Research 

portionately about as deep as large examples, though the diameter of the eye continues 
to exceed the length of snout until they are nearly full grown. The gill rakers (see 
Description) increase in number with age, quite rapidly in the young up to about 100 mm 
SL. It has been indicated also that the teeth are more prominent in the young than in 
the adult, and that they disappear entirely from the premaxillary with age. 

In waters tributary to Chesapeake Bay, in the Potomac River in particular, where 
most of the young fish were collected and measured, the rate of growth of the young 
during their first summer and subsequently is shown in a table by Hildebrand and 
Schroeder i^g: 88): 26 taken in June, 30-37 mm; 31 in July, 30-59 mm; 93 in 
August, 35— 64 mm; loi in September, 40— 69 mm; 232 in October, 40— 74 mm; 
and 1 8 I in November, 50—74 mm. Collections of young of the same year-class from 
Chesapeake Bay consisted of: 276 taken in January, 65-119 mm long; 280 in Feb- 
ruary, 65-1 19 mm; 773 in March, 65-1 15 mm; and 8 in May, 95-120 mm in 
length. These data suggest that the young grow to a length of about 50-75 mm 
(2 or 3 in.) while living in fresh water and to a length of 65-120 mm (2.6-4.8 in.) 
when a year old. 

Some older fish were taken in Chesapeake Bay, often in the same otter trawl hauls 
with the younger ones, but they are not separable into year-classes by the lengths 
given in the table referred to above. It is highly probable that, with few exceptions, 
these older and larger fish are approaching an age of two years. The collections con- 
sisted of: 8 taken in January, 160-194 mm in length; 15 in February, 150-209 mm; 
and 8 in March, 140-200 mm. If the six largest examples, which may be about 
three years old, are excluded, the range for the fish that apparently are two years 
old is 140-184 mm. If a length of around 200 mm is attained at three years of age, 
it is reasonable to assume that many individuals will attain a length of 250 mm (10 in.) 
when four years old. As adult gravid fish often are 250 mm and even less, it is highly 
probable that many individuals become sexually mature and perform their first migra- 
tion to the spawning grounds at an age of four years. 

Spawning, Migrations, and Habitat. Bluebacks are anadromous, but apparently 
they do not ascend rivers as far as their near relative, P. pseudoharengus. In the vicinity 
of Woods Hole, spawning takes place in fresh to slightly brackish ponds having an 
outlet to the sea; also, Bluebacks enter at least the mouths of the larger rivers in 
New England. 

This species usually arrives in the coastal waters a month or so later than pseudo- 
harengus — for example, in the lower parts of Chesapeake Bay during the first half of 
April and in the upper sections of the Bay about two weeks later. Hildebrand and 
Schroeder {Sg), in speaking of the migration in Chesapeake Bay, stated: "The peak 
of the catch of bluebacks is usually taken between April i and 20 in the lower bay. 
The numbers decrease throughout May, until after June i only stragglers are caught." 
The time of arrival varies from year to year by as much as a couple of weeks, depending 
upon the temperature of the water, which apparently must reach about 70° F before 
Bluebacks will enter. Northward, of course, the runs take place correspondingly later. 



Fishes of the Western North Atlantic 329 

Spawning apparently begins very soon after the fish enter fresh or brackish water. 
Judging from the size and development of the young and from what is known concern- 
ing their migration locally, spawning in the Potomac seems to be confined principally 
to the last half of April and the first half of May. Ripe fish were taken in July (19 15) 
at Woods Hole, Massachusetts {^6: 123); while most of the females were spent, 
considerable numbers of each sex were still unripe. 

The spent fish, in the Chesapeake Bay area at least, return to the ocean soon after 
spawning (^$g: 86). In the Gulf of Maine, "the spent fish, like alewives, return to the 
sea shortly after spawning. Probably these are the bluebacks taken at Woods Hole 
and north of Cape Cod in September and October. The winter home of our blue- 
backs is unknown; probably like their relatives the sea herring, they move out from 
land and pass the cold season near the bottom" (Bigelow and Welsh, 16: 113). The 
capture of seven large Bluebacks, up to 375 mm (15 in.) in length, on March 5 
(1931) about yomilesofFBarnegat, New Jersey (j^: 377), and the acquisition of nine fish 
by me January 18 (1926) in the market and reported by the dealer to have been taken 
off Beaufort Inlet, North Carolina, with "sink nets" (often operated in fairly deep 
water), tend to substantiate the view that they winter near the bottom. The latter 
fish, one female and eight males, all contained well-developed gonads. Bigelow and 
Welsh reported that large schools of Bluebacks are often seined off the coast of Maine 
"outside the islands" during summer and fall and that they consist mostly of "2-year-old 
fish not yet sexually mature" {16). 

In the Potomac, where the young are abundant below Little Falls throughout 
the summer, fish 20 mm long were taken there as early as June 17, and none less than 
30 mm was secured in July. Upon the approach of cool water, October and November 
in the Chesapeake Bay area, the young leave the fresh waters; they are then about 
five to six months old, and the greater part of them passes through Chesapeake Bay and 
out to sea in company with graybacks and shad of a similar age. However, some of them 
stop to spend their first winter in the deeper parts of the Bay, and a few seem to spend 
their second winter there. However, no young have been listed as occurring off- 
shore, from Chesapeake Bay southward. Pearson did list "Pomolobus sp." as having 
been "observed" in trawl catches off the coast of Virginia and North Carolina {103: 18); 
whether these fish were young or adults was not stated. 

Food. The food (16: 113) consists of plankton, copepods, pelagic shrimp, and 
young launce, and no doubt other fish fry. 

Enemies. Undoubtedly the principal enemies of young Bluebacks are the predatory 
fishes inhabiting the fresh and brackish water they occupy the first several months 
of life. As they school and move into salt water en route to the sea, marine shore 
fishes undoubtedly take their toll. Welsh and Breder found that weakfish {Cynoscion 
regalis) taken at the head of the Acushnet River, Massachusetts, were gorged with 
the young of this species (JJ2: 159). While the schools migrate from the sea to fresh 
water, large predatory fishes again must take a toll. Upon arrival in inshore waters 
and in bays and streams, large numbers are captured by man. 



3 3 o Memoir Sears Foundation for Marine Research 

Parasites. The parasitic acanthocephalan Echinorhynchus acus and the nematode 
Heterakis foveolata have been listed (JSj: 742). A specimen with a large colonial 
hydroid, Obelia commensuralis, attached to its back has been reported for Beaufort, 
N.C. (50: 1-6). 

Relationships. This species and pseudoharengus are closely related, but they differ 
prominently from chrysochloris and mediocris in the more numerous gill rakers. In general, 
aestivalis has a more slender body and a smaller eye than pseudoharengus. However, 
extensive overlapping occurs in a series of measurements of specimens from various 
localities, even though specimens of equal size are measured. But in any one area, as 
in Chesapeake Bay, the separation of the two species by a combination of the two 
characters mentioned generally is not difficult. 

Northward, as in New England, the species run somewhat more slender, and often 
some northern graybacks are quite as slender as some of the Bluebacks from the southern 
part of the range. Although the proportionate size of the eye is a somewhat more reliable 
character than depth of body, some specimens are difficult to distinguish by this char- 
acter also. Fresh examples generally are distinguishable by the color, the Bluebacks 
being bluish above, the graybacks grayish green. However, this difference in color 
disappears in preserved specimens. The Bluebacks usually have more gill rakers and 
vertebrae than the graybacks, but again intergradation occurs in each instance; and the 
use of the number of gill rakers is made more complex because of their increase in 
number with the age of the fish. The range and intergradation of all the proportions 
and enumerations mentioned are shown in the Description. 

The peritoneum (lining of abdominal cavity) is said to be black in the Bluebacks 
and pale or pale gray in the graybacks. Because this character was supposed to be diag- 
nostic, it has been used in keys many times, but in the numerous preserved specimens 
at hand, it was found unreliable; Herbert E. Warfel, who has worked with Pomolobus 
in Connecticut, has informed me (verbally) that this character is unreliable in fresh 
material also. It cannot be trusted, because the peritoneum in the Bluebacks varies all 
the way from black to pale or pearly gray. 

It is evident from the foregoing discussion that aestivalis and pseudoharengus often 
can be distinguished only by a combination of characters, and that, even then, some 
troublesome examples occur, with small specimens more difficult to identify than 
large ones. 

Abundance and Commercial Importance. The relative abundance of the Bluebacks 
and graybacks is not definitely known. Although many fishermen recognize the two 
species, they are not separated in the market, and the catches are listed merely as 
"alewives" in the government's statistical reports. It has been stated that the two species 
are about equally abundant in Chesapeake Bay and that each contributes an equal 
share to the commercial catch {^g: 91); also, the Bluebacks are more abundant than 
the graybacks in the southern states and less numerous northward {y2, 1896: 426). 
For the Gulf of Maine, Bigelow and Welsh found it difficult to arrive at a fair idea 
of the relative abundance of the two {16 : 112); although it is generally agreed that the 



Fishes of the Western North Atlantic 331 

graybacks outnumber the Bluebacks in the northern part of their range, these authors 
were certain that large catches of Bluebacks were taken in the Gulf of Maine from time 
to time. In Nova Scotia the Bluebacks are rather uncommon and of slight value whereas 
the graybacks are common and of considerable economic importance (l2g: ^^). Some 
statistics showing the combined catches of the two species, designated "alewives," are 
given in the account of the grayback (p. 338). 

Fluctuations, Methods of Fishing, and Artificial Propagation. The remarks in the ac- 
count of the grayback concerning fluctuations, effects of fishing, methods of fishing, 
and artificial culture, also apply in general to the Blueback. 

Range. The range of this species extends all the way from Cape Breton, Nova 
Scotia {15: 107, ftn. 92), to the St. Johns River, Florida; it is the most numerous 
southward. 

Synonyms and References: 

Chipea aestivalis Mitchill, Rep. in part on Fishes of New York, 1814: 21 (orig. descr.; type local, presumably 
New York; type lost); Mitchill, Trans. Lit. philos. Soc. N. Y., I, 181 5: 456 (descr.); Jordan and Gil- 
bert, Bull. U. S. nat. Mus., 16, 1882: 267 (cf. C.verna/is Mitchill =-■ C.pseudoharengus Wilson; distinct. 
char., range, synon.); Bean, T. H., Proc. U. S. nat. Mus., 6, 1883: 366 (migr. in Chesapeake Bay); 
McDonald in Goode et a/., Fish. Fish. Industr. U.S., i, 1884: 579-588, pis. 209, 210 (hist, of 
nomencl., abund., geogr. distr., food, reprod., size, uses); Goode, Amer. Fishes, 1888: 393 (discus, 
relation, with P . -pseudoharengus; names, abund., spawn., etc.). 

Pornolobus aestivalis Jordan and Evermann Bull. U. S. nat. Mus. 47 (i), 1896: 426; 47 (4), 1900: fig. 190 
(cf P. pseudoharengus; descr., synon.); Smith, Bull. U. S. Fish. Comm. (1897), ij, 1898: 91 (Woods 
Hole, Massachusetts); Evermann and Kendall, Rep. U. S. Comm. Fish. (1899), 1900: 55 (St. Johns 
R., Welaka, Volusia Bar at Lake George, and Lake Monroe, Florida); Smith, N. C. geol. econ. 
Surv., 2, 1907: 124, fig. 42, pi. 4 in color (synon., descr., range, habits, distinct, char.); Kendall, 
Occ. Pap. Boston Soc. nat. Hist., 7, 1908: 38 (refs., New England); Evermann and Hildebrand, Proc. 
biol. Soc. Wash., 23, 1910: 158 (Bryans Pt., Maryland); Sumner, et al.. Bull. U. S. Bur. Fish. (191 1), 
31 (2), 1913: 742 (refs., Woods Hole, Massachusetts, parasites); Regan, Ann. Mag. nat. Hist., (8) ig, 
1917: 300 (descr.); Kuntz and Radcliffe, Bull. U. S. Bur. Fish., J5, 1918: 123, figs. 87-100 (develop. 
of egg and young; fig. too apparently not this species); Welsh and Breder, Bull. U. S. Bur. Fish. (1923), 
jp, 1924: 159 (eaten by Cynoscion regalis); Bigelow and Welsh, Bull. U. S. Bur. Fish. (1924), 40 (l), 
1925: no, fig. 45 (descr., size, range, Gulf of Maine); Nichols and Breder, Zoologica, N. Y., 9(1), 
1927: 39, fig. (distr. New York, s. New England, life hist., develop., size); Hildebrand and Schroeder, 
Bull. U.S. Bur. Fish. (1927), ./j (i), 1928: 85, figs. 44-50 (synon., descr., cf P. pseudoharengus, 
habits, growth rate of young, commerc. import., range, Chesapeake Bay); Truitt, Bean, and Fowler, Bull. 
Md. Conserv. Dep., 3, 1929: 50 (descr., Maryland); Breder, Field Bk. Mar. Fish. Atl. Cst., 1929: 66, 
fig. (range, migr., spawn., size); Jordan, Evermann, and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 
42 (names, synon., range); Greeley, 24th Rep. N. Y. Conserv. Dep. (1934), Suppl., g, 1935: 89, pi. i 
color (midsection of Hudson R.; cf P. pseudoharengus); Vladykov and McKenzie, Proc. N. S. Inst. 
Sci., ig (i), 1935: 55, fig. 29 (Nova Scotia); Gudger, Amer. Mus. Novit., 945, 1937: 1-6 (colonial 
hydroid on back); Greeley, 26th Rep. N. Y. Conserv. Dep. (1936), Suppl., 11, 1937: 73 ("summer 
herring"), 91 (abund. Hudson R., migr. of adults and young); Greeley, 28th Rep. N. Y. Conserv. Dep. 
(1938), Suppl., 15 (2), 1939: 82 (Long Island, New York); Bigelow and Schroeder, Bull. U. S. Bur. 
Fish., 48 {id), 1936: 327 (adults captured off Barnegat, New Jersey); Anonymous, Fish. Resourc. 
U.S., Senate Doc. 51, 1945: iii, 66, fig. (migr., range, commerc. import., methods of capture, causes for 
decline, remedies; consid. with P. pseudoharengus, not separately); Bigelow and Schroeder, Fish. Bull. 
(74) U. S. Fish Wildlife Serv., 5J, 1953: 106-107 (descr., habits. Gulf of Maine). 

Pornolobus cyanonoton Bean, T. H., 7th Rep. Forest Comm. N. Y. (1901), 1902: 305, fig. (synon., migr., time 
and place of spawn.); Bean, T. H., Bull. N. Y. St. Mus., 60, Zool. 9, 1903: 202 (synon., migr., local 



332 Memoir Sears Foundation for Marine Research 

occur., common names); Fowler, Rep. N. J. St. Mus., 2, 1905: 98, figs, (male and female fig., dist. char., 
local occur., synon.). 
Pomolobus sp. Pearson, Invest. Reps. U. S. Bur. Fish. (1932), I (10), 1936: 18 (in trawl catch). 

Doubtful References: 

Clupea fasclata LeSueur, J. .'\cad. nat. Sci. Philad., J (2), 1818: 233 (orig. descr.; type local. Sandwich, 
Massachusetts). 

Alosa tyranmis De Kay (not of Latrobe), N. Y. Fauna, Pt. 4: Fishes, 1842: 258, pi. 13, fig. 38 (descr., occur, 
along coast; plainly not menhaden; descr. and fig. suggest P. aestivalis). 

Alosa cyanonoton Storer, Proc. Boston Soc. nat. Hist., 2, 1848: 242 (orig. descr. ; type local, presumably Massachu- 
setts); Storer, Fish. Mass., 1867: 161, pi. 27, fig. i (descr., names, occur.). 



Pomolobus pseudoharengus (Wilson) i 8 1 1 

Grayback,='=' Gray Herring, White Herring, Brancli Herring, Bigeye Herring, 
Freshwater Herring, Sawbelly, Alewife, Kyak, Gaspereau 

Figures 76, 77, 82, 83 

Study Material. A total of 65 or more specimens, 30-330 mm TL, 23.5-268 mm 
SL, from Cape Breton, Nova Scotia; St. Andrews Bay, New Brunswick; Freeport and 
Portland, Maine; Falmouth and Woods Hole, Massachusetts; Noank, Connecticut; 
Long Island, New York; off Wildwood, New Jersey; and various localities in the 
Chesapeake Bay area. 

Distinctive Characters. See Relationships under aestivalis, p. 330. 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on 65 or more specimens, 23.5-268 mm SL. 



Scales: 42—50. 

Ventral scutes: 30—36. 

Gill rakers: on lower limb 38—43 in 

adults (occasionally 44). 
Fin rays: dorsal 15—19; anal 17-21; 

pectoral 13-16. 
Vertebrae: 46—50 (36 specimens). 



Body: depth 23.5-35.5. 
Caudal peduncle: depth 8.5-12. 
Head: length 22.6-34.8. 
Snout: length 4.55-7.6. 
Eye: diameter 5.0—12. 
Interorbital: width 3.45—7.3. 
Maxillary: length 10.5— 15. 
Pelvic fin: length 10.4- 15.8. 
Pectoral fin: length 14.6—22.6. 

Body compressed, its greatest thickness usually about 2.75 in its depth, the depth 
2.8-4.25 in SL. 

35. Gray back has been chosen as the leading name for P . pseudoharengus and blueback for P. aestivalis because these 
names in a measure are descriptive, distinctive, and are perhaps understood by more people within the range of 
the species than any others. Alewife certainly is not distinctive, as it sometimes includes both species mentioned, 
and in some places it is applied to the menhaden, Bret'oortia. Branch herring and glut herring are distinctive in 
the Chesapeake Bay area, but these names are not well known northward. However, alewife is used in subsequent 
pages to include both of these species in citing commercial statistics, for the species are not separated in the market, 
where they are sold either as alewives or as river herring. [The American Fisheries Society recommends "Ale- 
wife" for pseudoharengus. — Y. H. o.] 



Fishes of the Western North Atlantic 



333 



Scales only moderately adherent, often lost in preserved specimens; apparently 
not different from those of P. aestivalis-^ 14 longitudinal rows between base of pelvic 
and base of dorsal. Ventral scutes strong, 18—21 in advance of pelvics, 12 — 16 
behind them. 

Head less than depth, often fully by diameter of eye, 2.8-4.2. Snout 4.1-6.0. 
Eye large, longer than snout at all ages, 2.5—4.0. Interorbital 4.0—7.3. Mouth 
oblique, as in P. aestivalis; margin of lower jaw within mouth with a somewhat 
broader angle (Fig. 760); other characters about the mouth, including Teeth, as in 




Figure 82. Pomolobus pseudoharengus, 300 mm TL, from a market in Washington, D. C, USNM 25197. 



P. aestivalis. Maxillary 1.8—2.8 in head. Gill rakers at angle of first arch about 0.66 
of diameter of eye, increasing in number with age; 25—33 on lower limb in examples 
30-49 mm SL, 32-36 in specimens 50—69 mm, 30—39 in specimens 70— 89 mm, 35—38 
in specimens 90— 109 mm, 36—40 in specimens 1 10—129 mm, 36—40 in specimens 
130—149 mm, and 38—44 in examples 210— 270 mm (adults). 

Dorsal fin with a straight to slightly concave margin, its origin definitely less than 
an eye's diameter in advance of pelvic fin and usually nearer to margin of upper jaw 
than to vertical from last ray of anal. Caudal deeply forked, the lower lobe a little longer 
than upper lobe, often exceeding head length by nearly the diameter of eye. Anal low, 
with an almost straight margin, its origin about equidistant between base of pelvic 
and base of caudal. Pelvic fin inserted near ventral edge and usually equidistant be- 
tween base of pectoral and origin of anal, 1.75—2.5 in head. Pectoral fin somewhat 
higher on side, reaching about 0.66 of the distance to pelvic in adults and nearly to 
pelvic in some small examples, 1. 1—2.0 in head. 

Color. Fresh specimens, grayish green above. Sides silvery. Rows of scales on 
upper part of side sometimes with more or less definite dusky lines in adults. A small 



334 Memoir Sears Foundation for Marine Research 

dark spot at shoulder, often missing in specimens less than lOO mm long. Fins all 
plain, slightly yellowish to green in life. Peritoneum generally pale or silvery, often 
with dark punctulations, and sometimes quite dusky. 

Size. The Graybacks, like the bluebacks {aestivalis), attain a length of about 
380 mm (15 in.) and a weight of about 14 ounces; the usual length of market fish, as 
observed in the Chesapeake Bay area, averaged only about 275 mm (i i in.) in length 
and about a half pound in weight. 

Although Graybacks generally grow to a length of 275-300 mm (11-12 in.) in 
salt water along the Atlantic coast, their usual length in Lake Ontario {118: 188) 
is less than 150 mm (6 in.). The same writer mentioned fish only 100-125 mm 
(4-5 in.) long with roe as further evidence of stunted growth resulting from the 
local conditions under which the fish were living. 

Development and Growth. The eggs are 1.25 mm in diameter and very glutinous 
when first laid, adhering to brush, rope, stones, piling, and other objects {lig: 123). 
The incubation period extends over six days at a mean water temperature of 60° F. 

Recently hatched larvae are very slender and extremely transparent, and have 
only a row of pigment spots on the side of the tail (JJJ: 506, pi. i, fig. 8). A two- 
day-old fish is about 5.0 mm long, and Ryder's figure shows the greatest depth, in- 
cluding the fin folds, to be about 10 times in TL, with the vent (as usual in clupeoid 
fishes) situated very far back at about the beginning of the posterior fifth of the body. 
This agrees very well with a larva of similar age and size described as P. aestivalis 
(j6: 1 26). 3* The appendages (pyloric caeca), connected to the intestine near the 
stomach, are very small or wanting in the young but become greatly developed with age 
{59 •• 90). 

The young reach a length of i 5 mm when a month old {16: 109). On the smallest 
specimen at hand, 29 mm (22 mm SL), taken on June 18 (1873) at Washington, 
D. C, the depth is 4.0 times in SL; in other respects it already resembles the adult; 
the eye is notably longer than the snout and remains so throughout life. When the 
fish reach a length of about 100 mm, the proportionate depth is more or less the 
same as in adult fish. 

The gill rakers (see Description) increase in number with age, this increase being 
especially rapid in the young up to about 100 mm SL. In this species, as in P. aestivalis, 
the teeth are more prominent in the young than in the adults and disappear from the 
premaxillary with age. 

The young reach a length of 50-100 mm (2-4 in.) by autumn in streams trib- 
utary to the Gulf of Maine {16). This rate of growth is about the same as that 
shown by Hildebrand and Schroeder {59: 91) in a table based on collections made in 
streams tributary to Chesapeake Bay (especially the Potomac River). This table in- 
cludes: 26 young for June, 45-74 mm; 105 for August, 50-84 mm; 134 for Sep- 

34. Early larvae and some young adults, supposedly of P. pseudoharengus, were described and figured by Prince 
(/05: 103-108, pis. 8, 10, in part). It seems doubtful whether his identifications were correct, for his drawings of 
young adults, compared with young adults from the Potomac, now before me, show the body to be much deeper 
and the mouth nearly vertical. 



Fishes of the Western North Atlantic 



335 




Figure 83. Pomolobus pseudoharengus, from Potomac River, top, young, showing scales developing, 29 mm TL, 
22 mm SL, USNM 85668; bottom, young adult, 47.5 mm TL, USNM 69833. upper drawn by Ann S. 
Green; lower by Nancy D. Patton. 



tember, 50—79 mm;^* 138 for October, 55—94 mm; 221 for November, 55—109 mm; 
and 19 for early in December, 60— 79 mm TL. In Chesapeake Bay, young that ap- 
peared to be in their first winter, as judged from their small size, were 75—140 mm in 
January, 80—1 14 mm in February, and 80—145 mm in March. The young fry taken 
in the Bay during the winter varied so widely in length that it is clear they represented 
more than one year-class. The 262 older fish taken during the winter ranged between 
1 15—224 mm, or 1 15— 194 mm, if the three largest examples are excluded. 

The foregoing data seem to show that the young of this species grow to a length 
of 55—109 mm (2.20—4.36 in.) while in fresh water, to about 65—145 mm (2.6— 
5.8 in.) at one year of age, and to about i 15-195 mm (4.6—7.8 in.) at two years of 

35. The largest young of the season evidently are not represented in this collection. 



336 Memoir Sears Foundation for Marine Research 

age. Mature gravid fish often are only 250 mm (10 in.), or even less, and if a length 
of 200 mm (8 in.) or more is attained at three years of age, it is probable that many 
become sexually mature and perform their first migration to the spawning grounds 
when four years old. However, Bigelow and Welsh {16: no) have reasoned that at 
least some of the fish in the Gulf of Maine area reach maturity at three years of age. 
This conclusion was reached because the "progeny" of adults used in restocking streams 
returned in the third year after the streams were restocked. 

Spawning. This species is very prolific; a total of 644 females from the Potomac 
River yielded an average of 102,800 eggs per fish (lig: 123). The time of spawning 
varies from some time in March in Chesapeake Bay to late April or May in Maine. 
They may spawn in running water or in ponds. Ripe fish have been taken many 
times in the Potomac, near Washington, where the eggs have been found adhering 
to fish traps and where young occur throughout the summer. Greeley has described 
their spawning in dense schools in fast water approximately two feet deep over a 
bottom of coarse stones, some sand, and gravel in the Poestenkill, at Troy, New York 
(47: 89). Around the Gulf of Maine, however, they spawn more commonly in ponds 
(including those behind barrier beaches), though probably not so exclusively as Bigelow 
and Welsh believed (16: 109). 

According to Greeley (47: 89): 

The act of spawning is characterized by the whirling of a pair of fish in a close spiral, usually ending as 
they reach the surface with a splash and separate. The whirling is done rapidly and often only a flash of silver 
is to be seen. A number of spawning pairs were observed here [presumably the Hudson] by Dr. C. W. Greene 
and the writer. The sexes of the fish cannot be determined from the appearance of the fish in the water, but it 
was assumed that the fish seen whirling were mated pairs. It is probable that the male is on the outside of the 
spiral during the act of spawning and that the rapid circling is adopted for maintenance of contact as the eggs 
are shed and fertilized. Many eggs were scattered over the bottom and had drifted downstream where they 
were deposited by the current in slack water areas behind obstructions. The temperatures here May 16(11 A.M.) 
were: air, 57 degrees; water 52 degrees. 

Migrations and Habitat. This species is strictly anadromous. It has been common 
knowledge since colonial times that they run 

indifferently up rivers as large as the St. John (New Brunswick), the Merrimac and Potomac, or streams so small 
one can almost leap across, and only a few inches deep. In large rivers they may run far upstream ... or their 
journey may be only a few yards, as it is in the artificial cuts that are kept open through barrier beaches to allow 
the fish access to fresh water ponds behind the latter (jj: 103). 

The Graybacks arrive in inshore waters a month or so ahead of the bluebacks and 
a week or two ahead of the shad. In Chesapeake Bay the first catches are made in late 
February or early March. In southern New England (Woods Hole, Massachusetts) 
they arrive sometime during March and are caught through March and April 
{120: 91). However, they do not reach the rivers of Maine until late April or early 
May and, as shown by the commercial catches, they are most numerous there during 
the last half of May (16: 109). 

That the young return to the stream of their birth (the so-called "parent stream 



Fis/ies of the Western North Atlantic 337 

theory") was confirmed when the progeny of gravid adults planted in depopulated 
streams returned to these rivers three years after hatching (z6: iio). This incident 
is also cited in a subsequent paragraph as a successful method of restocking streams. 
But a much more intensive study of this interesting question Is needed before any 
categorical statement can be made as to how generally the parent stream theory applies 
in the case of the Graybacks (75: 103). 

As in the case of bluebacks and shad, relatively little is known about the home 
of Graybacks in the sea. The capture of 18 adult Graybacks 70 odd miles off Barnegat, 
New Jersey, on March 5, 1931 {14: 327), suggests that the home may be some distance 
offshore. Since these adults were taken in March, they may have been en route to some 
nearby stream, hence it should not be assumed that the fish actually were taken from their 
usual abode in the sea. It has been surmised for some time that the Graybacks, as well 
as blueback herring and American shad, live quite a distance offshore from the rivers of 
their origin and that their migrations consist of Inshore and offshore movements. This 
theory finds slight support in the discovery that different races spawn In different 
river systems. However, further study is needed. 

The young, often with bluebacks and American shad of similar age, have been 
seen descending as early as June 15 in the more southerly streams fringing the Gulf 
of Maine. Successive companies of fry move out of the places where they were hatched, 
run down with the current throughout the summer and are at sea by autumn. In the 
Potomac, which the fish cannot ascend beyond Little Falls, the young remain abundant 
in the vicinity of Washington, D. C. Here they may be taken in the same seine hauls 
with bluebacks and American shad until October, when the main seaward migration 
takes place; however, some young remain in the river until November and a few 
linger until December. Most of the young apparently pass directly through Chesapeake 
Bay and out to sea, but some stop to spend their first winter in the deeper parts of the 
Bay, a few seem to remain for their second winter, and very few apparently stay there 
for their third winter. 

The fish In the freshwater lakes endeavor to carry out the same spawning migra- 
tion as those that inhabit the Atlantic coast, for the fish enter the shallow parts of 
the lakes and ascend the tributary streams to spawn. 

Food. The food of the young while in fresh water apparently consists principally 
of diatoms, copepods, and ostracods. After they enter salt water, their chief diet seems 
to be shrimps and fish. Young 50-70 mm long, taken in fresh water, had fed principally 
on copepods and ostracods; and larger young, 83-178 mm long, caught in Chesapeake 
Bay, had continued to feed on copepods, though the larger ones had ingested prin- 
cipally mysids and fish {^g : 90). Copepods, shrimps, and young squid were listed as 
the food of the young by Sumner, et al. {123: 72). Bigelow and Welsh (16: 1 10) have 
described their chief food as copepods, amphlpods, shrimps, and appendicularians, small 
fish (such as herring, eel, launce, and cunner) as well as fish eggs and young of their 
own. Alewives often contain diatoms, even when adult. It is conceivable that diatoms 
could be screened from the water by the rather numerous, close-set gill rakers of adults. 



338 Memoir Sears Foundation for Marine Research 

It has long been known that anadromous fishes, as a rule, feed sparingly or not 
at all while ascending streams, and Graybacks are no exception to this rule. However, 
the spent fish feed ravenously on shrimp on entering brackish water along the coast, 
where anglers often hook them on an artificial fly il6). 

Enemies. The young are undoubtedly preyed upon by various predatory fishes, 
turtles, snakes, and birds while they live in fresh water. When they begin to school and 
migrate out to sea, marine fishes and numerous other sea inhabitants very probably 
destroy many. When they re-enter shallow shore water as adults in great schools, large 
marine fishes of many kinds, finback whales, birds, and turtles, take a further toll. 
Upon arriving inshore and in the rivers, man — who captures great quantities for food 
— must be their chief enemy. 

Parasites. The following have been listed by Sumner, et al. {12 j: 742): Acanthoce- 
phala {Echinorhynchus acus) \ cestodes {Rhynchobothrium imparispine) \ trematodes {Distomum 
appendiculatum, D. bothryophoron., D. vitellosum., and Monostomum sp.); and copepods {Ar- 
gulus alosae, Caligus rapax, and Lepeophtheirus edwardsi). 

Variations. Specimens from various localities within the range of this species 
along the coast {Study Material, p. 332) show a tendency toward a more slender body 
and a proportionately larger head toward the northern part of the range; and a limited 
number of specimens from fresh water, Cayuga Lake and Lake Ontario (proportions 
not used in Description), show this tendency to be even more pronounced there. In 
general, the fish from the northern part of the range (especially from Casco Bay, Maine), 
and more particularly those from fresh water, have the appearance of being more or 
less emaciated. However, insufficient specimens have been studied to determine the 
significance of the apparent differences mentioned. These differences may be of sub- 
specific value, but more likely they are of only racial importance. The decrease in 
body depth in northern specimens makes their separation from P. aestivalis more 
difficult than identification of specimens from the southern part of the range. 

Abundance and Commercial Importance. The relative abundance of pseudoharengus 
and aestivalis is discussed in the account of the latter (p. 330). It has been claimed 
by various writers that Graybacks are far less numerous now than during colonial days, 
or even a half century ago. The matter was summed up for the Gulf of Maine by 
Bigelow and Welsh {16: 108) in these words: 

During the past two centuries ... its numbers have declined, and its range has been restricted, both by 
actual extirpation from certain streams by overfishing, by the pollution of river waters by manufacturing wastes, 
and by the. erection of dams that it cannot pass. However, the alewife is still a familiar fish all along our coast, 
and yields an abundant catch in many of our streams. 

These remarks also apply in a measure to fish in the more southern part of the range, 
though the decline apparently has been less pronounced southward. 

The following statistics give the catches of Graybacks and bluebacks combined, 
under the name "Alewives." These records show in part the decline in their abundance 
during the past 50 years or so, but they also show that these fish are still abundant 
and of great economic importance. There has been no compensation for any variations 



Fishes of the Western North Atlantic 339 

in the "fishing effort" — the number of men, vessels, and amount and efficiency of the 
gear used — but it may be assumed that there was no decrease in this respect. Federal 
Government statistics, which have been taken at irregular intervals, report total catches 
in round numbers for the Atlantic coast of the United States as follows: 62,067,000 
pounds In 1896, 52,061,000 in 1905, 35,809,000 in 1921, 35,290,000 in 1929, 
and 29,542,000 in 1941. The greatest proportionate decline took place in New 
England, for which the following catches are listed: 12,116,000 pounds in 1896, 
8,429,000 in 1905, and 3,975,000 in 1924; however, from 1924 to 1941 the catch 
there remained fairly even. 

South of New England the decline in abundance began later and was more gradual 
and less pronounced. For example, the Chesapeake Bay states (Maryland and Virginia), 
which have furnished a larger part of the total catch of "Alewives" than any other section 
of the coast since the beginning of this century, yielded a fairly even catch until about 
1925. Thereafter a decline has been indicated by the statistics: 25,611,000 pounds in 
1925, 21,129,000 in 1930, 18,884,000 in 1937, and 17,012,000 in 1941. The catch 
in the South Atlantic states, from North Carolina to Florida, is also very large, has 
fluctuated greatly, and has shown a slight downward trend: 15,857,000 pounds for 
1897, 11,601,000 for 1902, 15,186,000 for 1918, 7,571,000 for 1923, 11,176,000 
for 1929, 6,218,000 for 1937, 11,611,000 for 1938, and 9,116,000 for 1940. The 
over-all catch of the two species combined has ranged between 47,300,000 and 
57,700,000 pounds for the United States (1940, 1945, 1950— 1953), and between 
6,250,000 and 41,000,000 pounds for Canada (1940, 1945, 1950-1952). 

The Graybacks and blueback herring are still abundant. In fact, every state 
along the Atlantic coast exclusive of Pennsylvania, South Carolina, and Georgia, make 
substantial contributions to the catch, as do Nova Scotia and New Brunswick, Canada. 
Although these species are among the cheapest fish on the market, they furnish a sub- 
stantial income because of their abundance. During recent years the annual yield to 
the fishermen from this fishery alone has been % 400,000-800,000 in the United States, 
and $ 33,000-350,000 in Canada. 

Most of the "Alewives" caught are salted, but an increasingly large part of the 
catch is sold in the fresh state. Although the flesh is quite bony, as in other herrings, 
it is delicious and is in demand whether fresh, salted, or smoked. At the salting 
houses the roe generally is separated for canning and the offal is made into fish meal 
and oil. 

Although these fish have become common to abundant in some freshwater lakes, 
as in Lakes Ontario and Erie, they are of little economic importance there because of 
their small size. However, in 1892 Smith said that they were of more economic impor- 
tance in Lake Ontario than is generally realized {118: 189). He pointed out that they 
had replaced more important foodfishes (young ciscoes and suckers) as bait for sturgeon 
and lake trout in the trawl-line fishery, that the young were used as bait in extensive 
sport fishing, and that many tons of those that wash ashore annually are used as 
fertilizer. 



340 Memoir Sears Foundation for Marine Research 

Fluctuations. Overfishing, pollution, and impassable dams in streams have been the 
chief factors in diminishing the abundance of "Alewives" in many localities, but all of 
these causes can be remedied at least in some measure by restocking suitable streams 
with gravid fish {l6: \\o\et ah). This has been done in Massachusetts localities, where, 
in the third year after the ripe fish were introduced, the progeny appeared in those 
streams. 

It is not always easy to determine whether an increase or decrease in catch is due 
to the amount of fishing, to special conditions created by man, to successful or unsuc- 
cessful spawning, or to general conditions obtaining in nature. However, if the increase 
or decrease extends over a period of several years, reasonable conclusions may be reached. 
For example, the tremendous abundance of "Alewives" in the Chesapeake Bay area in 
1908 and 1909, when the catches were about twice those in preceding and successive 
censuses (66,690,000 and 51,425,000 pounds), may be ascribed to especially favorable 
conditions during and probably after the spawning season. On the other hand, the 
general decline in the fishery in New England (discussed above), despite some 
fluctuations, quite surely was the result of adverse conditions created by man. There- 
fore, frequent censuses are necessary to show the status of this as well as other fisheries 
for the guidance of those entrusted with the promulgation of regulations for the con- 
servation of the fishery. 

Artificial Cultivation. "Alewives" have not been artificially propagated regularly be- 
cause it has not been deemed necessary. However, in 1882 two million fry hatched 
by fish culturists were transported to the Colorado River, Texas (^5: 586), an in- 
troduction that apparently was not successful. As already stated, another method 
of "planting" was successfully used in Massachusetts. Some depopulated streams in 
that state, which had been reconditioned and made suitable, were successfully restocked 
merely by releasing gravid fish. Graybacks apparently were unintentionally introduced 
into Lake Ontario when shad fry were planted there (45: 588; 118: 188). 

Methods of Fishing. "Alewives" are easily caught, chiefly with dip nets in New Eng- 
land. Pools are constructed about the mouths of rivers and the gravid fish are led into 
them by means of nets and stone diversions. Southward, as in the Chesapeake Bay 
area and in North Carolina, they are caught chiefly with pound nets. However, they 
are taken also with weirs, seines, gill nets, fyke nets, and a few with otter trawls. 

Range and Distribution. The range oi pseudoharengus extends from the Gulf of St. 
Lawrence and Nova Scotia, where it is common {l2g: 54), to North Carolina. It has 
been recorded for the St. Johns River, Florida, but the first report (^5: 336), as 
Clupea vernalis, was apparently based on a P. aestivalis (USNM 17929, collected by 
Spencer F. Baird), now before me; and the second report, so far as known, by Lonn- 
berg {88: 114), may be disregarded, for this species is too difficult to distinguish 
from P. aestivalis by a simple "observation." According to Smith, "extensive collec- 
tions of fishes in the St. Johns River, in Florida, have failed to disclose its presence" 
(lig: 122). Indeed, no specimens of this species from south of the Albemarle Sound 
area (where aestivalis is plentiful) have been found in the National Museum, and 



Fishes of the Western North Atlantic 341 

it was only sparingly represented among many specimens of P. aestivalis collected over 
a period of years in the Newport River and its estuary near Beaufort, N. C. Possibly 
this species does not occur south of North Carolina. 

This species, as stated above under Artificial Cultivation^ apparently was uninten- 
tionally introduced into Lake Ontario, whence it has spread to Lake Erie and to Lake 
Huron. It also inhabits Lakes Seneca and Cayuga, which it probably reached by natural 
means from the ocean (.^5 : 116), and the Oswego River system in New York State 
(^6: 95). It has remained common in Lake Ontario, has become common in Lake 
Erie, and has penetrated Lake Huron, where it is still rare {62: ii). 

Synonyms and References: 

Clupea pseudoharengus Wilson, in Rees' New Cyclopedia (Art. Cluped), g, undated, ca. 1811: no pages (orig. 
descr.; no local.; probably Philadelphia, Pennsylvania); Smith, Bull. U.S. Fish Comm. (1890), JO, 1892: 
187, pi. 29 (unintent. introd. in Lake Ontario, abund. there, diseases, mortal., uses). 

Clupea vernalis Mitchill, Trans. Lit. philos. Soc. N. Y., I, 1815: 454 (orig. descr.; type local. New York; 
effects of mill dams on migr. to spawn); Jordan and Gilbert, Bull. U. S. nat. Mus., 16, 1882: 267 (descr., 
range, synon.); McDonald in Goode, et al.. Fish. Fish. Industr. U. S., i, 1884: 579-588, pis. 207, 
208 (hist, of nomencl., abund., geogr. distr., migr., food, reprod., size, uses); Bean in Goode, et al.. Fish. 
Fish. Industr. U. S., i, 1884: 588-593 (Lakes Cayuga, Seneca, and Ontario, origin in lakes, movements, 
enemies, fatal., capture); Ryder, Rep. U. S. Comm. Fish. (1885), 1887: 506, pi. I, fig. 8 (descr. of egg 
and newly hatched young); Goode, Amer. Fishes, 1888: 393 (discus, with P. aestivalis, names, abund., 
spawn.). 

Clupea megalops Rafinesque, Amer. Mon. Mag. (i 817-18 18), 2(3), 1818: 206 (orig. descr.; type local. 
Delaware R.). 

Meletta venosa Cuvier and \'alenciennes. Hist. Nat. Poiss., 20, 1847: 374 (orig. descr.; type local. New York); 
Benin, Bull. Soc. zool. Fr., 66, 1941 : 23 (ident. of types). 

Pomolobus pseudoharengus Gill, Rep. U. S. Comm. Fish. (1871-72), 1873: 811 (names, range); Uhler and 
Lugger, Rep. Comm. Fish. Md., ed. l, 1876: 158; ed. 2, 1876: 135 (descr., synon., habits, Maryland; 
pseudoharengus and aestivalis not distingu.); Jordan and Evermann, Bull. U. S. nat. Mus., 47 (i), 1896: 
426; 4-7 {4), 1900: fig. i8g (descr., range, habits, synon.); Smith, Bull. U. S. Fish Comm. (1897), ly, 
1898: 91 (Woods Hole, Massachusetts); Bean, T. H., 7th Rep. Forest Comm. N. Y. (1901), 1902: 
302, with fig. (synon., names, occur, in lakes, migr., size, methods of capture); Bean, T. H., Bull. N. Y. 
St. Mus., 60, Zool. 9, 1903 : 199 (synon., descr., names, range, migr., size, growth of young, econ. value, 
methods of capture); Fowler, Rep. N. J. St. Mus., 2, 1905: 95 (figs, of male and female, descr., migr., 
synon.); Smith, N. C. geol. econ. Surv., 2, 1907: 122, fig. 44, pi. 3 in color (synon., descr., range, migr., 
reprod., commerc. import.. North CaroUna, methods of capture) ; Kendall, Occ. Pap. Boston Soc. nat. Hist., 
7, 1908: 38 (refs., New England); Sumner, Osburn, and Cole, Bull. U.S. Bur. Fish. (191 1), JI, 191 3: 
742 (refs., Woods Hole, Mass., food, parasites); Regan, Ann. Mag. nat. Hist., (8) ig, 1917: 300 (descr.); 
Bigelow and Welsh, Bull. U. S. Bur. Fish. (1924), 40, 1925: 107, fig. 44 (descr., size, range. Gulf of 
Maine, habits, food, commerc. import.); Nichols and Breder, Zoologica, N. Y., 9 (l), 1927 : 38, fig. (distr. 
New York, s. New England, food, life hist., size); Hildebrand and Schroeder, Bull. U. S. Fish Comm. 
(1927), 4J (i), 1928: 89, figs. 51-52 (synon., descr., relation., growth of young, food, habits, commerc. 
import., range, Chesapeake Bay); Greeley, 17th Rep. N. Y. St. Conserv. Dep. (1927), Suppl., 1928: 95 
pi. 3 in color (Oswego R. watershed); Jordan, Manual Vert. Anim. NE U. S., ed. 13, 1929: 40 (diagn., 
range); Truitt, Bean, and Fowler, Bull. Md. Conserv. Dep., 3, 1929: 49, fig. 15 (descr., Maryland); 
Breder, Field Bk. Mar. Fish. .'\tl. Cst., 1929: 65 (food, spawn., size); Greeley, i8th Rep. N. Y. 
Conserv. Dep. (1928), Suppl., 1929: 167 (common Lake Ontario, enters Niagara R.); Jordan, Evermann, 
and Clark, Rep. U. S. Comm. Fish. (1928), 2, 1930: 42 (names, synon., range); Greeley and Greene, 
20th Rep. N. Y. Conserv. Dep. (1930), Suppl., 5, 1931: 82 (St. Lawrence watershed); Greeley and 
Bishop, 2ist Rep. N. Y. Conserv. Dep. (1931), Suppl., 6, 1932 : 78 (common Lake Ontario, St. Lawrence 
R., annu. mortal.); McKay, Copeia, 1934: 97 (first rec. for Lake Huron); Vladykov and McKenzie, 
Proc. N. S. Inst. Sci., ig (i), 1935: 54, fig. 28 (Nova Scotia); Greeley, 24th Rep. N. Y. Conserv. Dep. 



342 Memoir Sears Foundation for Marine Research 

(1934), Suppl., 9, 1935: 89 (Mohawk-Hudson watershed, descr. of spawn.); Bigelow and Schroeder, 
Bull. U. S. Bur. Fish., 48 (20), 1936: 327 (adults off Barnegat, New Jersey); Greeley, 26th Rep. N. Y. 
Conserv. Dep. (1936), Suppl., 11, 1937: 90 (abund. Hudson R.; landlocked in reservoir and lake); 
Greeley, 26th Rep. N. Y. Conserv. Dep. (1937), Suppl., 12, 1938: 62 (Allegheny and Chemung water- 
sheds); Greeley, 28th Rep. N. Y. Conserv. Dep. (1938), 13, 1939: 39 (in fresh water, Long Island, 
New York); Greeley, 28th Rep. N. Y. Conserv. Dep. (1938), Suppl., 1 5 (2), 1939: 82 (salt water. 
Long Island, New York); Greeley, 29th Rep. N. Y. Conserv. Dep. (1939), Suppl., 16, 1940: 68 (notes 
on alewife), 59-61 (abund. Lake Ontario; stunted growth in fresh water; annu. mortal., spawn.); Hubbs 
and Lagler, Bull. Cranbrook Inst. Sci., 18, 1941 : 27 (range, Lakes Ontario, Erie, and Huron, where 
introd.); Anonymous, Fish. Resourc., U. S. Senate Doc., 51, 1945: iii, 66, fig. (migr. range, commerc. 
import., methods of capture, causes for decline, remedies, consid. with P. aestivalis); Bigelow and Schroe- 
der, Fish. Bull. (74) U. S. Fish Wildl. Serv., 55, 1953: loi (descr., habits. Gulf of Maine). 

Pomolobus pseudoharengus lacustris Jordan, Manual Vert. Anim. NE U.S., ed. I, 1876: 265 (orig. descr.; 
type local. Cayuga Lake, New York; landlocked form subspecif distinct from anadromous one). 

Pomolobus mediocris Evermann and Hildebrand (not of Mitchill), Proc. bid. Soc. Wash., 2J, 1910: 158 (notes 
on young; found later to be fseudoharengus). 

Pomolobus sp., Pearson, Invest. Rep. U. S. Bur. Fish., j (10), 1932: 18 (in trawl catch, off Virginia capes). 

Doubtful References: 

C/upea fiarvula MhchiW, Rep. in part on Fishes of New York, 1814:21 (orig. descr.; type local, presumably New 

York; type lost); Mitchill, Trans. Lit. philos. Soc. N. Y., j, 181 5: 452 (descr.). 
Clupea virescens De Kay, New York Fauna, Pt. 4, Fishes, 1842: 252, pi. 13, fig. 37 (orig. descr.; type local. 

New York Bay; type lost). 
Clupea pseudoharengus Lonnberg, Ofvers. Svensk. Vet. Akad. Forh., 5r (3), 1894: 114 (observed enter. St. 

Johns R., Florida in spring; probably P. aestivalis'). 
Pomolobus pseudoharengus Evermann and Kendall, Rep. U. S. Comm. Fish. (1899), 1900: 55 (St. Johns R., 

after Bean 1883 and Lonnberg 1894). 



Genus Brevoortia Gill 1861 

Menhaden 

Brevoortia Gill, Proc. Acad. nat. Sci. Philad., 1861 : 37; genotype by original designation, Brevoortia menhaden 
Gill equals B. tyrannus (Latrobe). 

Characters. Body oblong, compressed. Bony scutes present, the median line of 
chest and abdomen with a sharp edge. Scales adherent, the exposed part much deeper 
than long, the margin serrate or pectinate in adults; a series of modified scales pres- 
ent next to median line on back in front of dorsal fin. Cheek deeper than long. 
Mouth large. Maxillary extending to or beyond middle of eye. Upper jaw with a 
distinct median notch. Lower jaw included in the upper one, not projecting, its up- 
per margin (within mouth) nearly straight. Teeth wanting in adults. Lower limb of 
first Gill arch with an obtuse angle. Gill rakers long and numerous, increasing in 
number with age, those on upper limb of first arch extending downward and over 
those on upper part of lower limb. Dorsal with 17—22 rays, the last one not greatly 
produced; its origin about equidistant between rim of snout and base of caudal. Anal 
with 17—25 rays, the last one little enlarged. Pelvic small, with 7 rays. Vertebrae 
about 42—50. Intestine very long. Peritoneum black. 

Remarks. The species of this genus generally are recognizable by the strongly 



Fishes of the Western North Atlantic 343 

serrate or pectinate adherent scales, the exposed part of which is much deeper than 
long; the modified series of scales next to the median line of the back in front of the 
dorsal fin; the very long gill rakers on the upper limb, which project downward 
over the equally long rakers of the lower limb; the very deep cheek; and the rather 
small pelvic, composed of only seven rays. 

The young, about 60 mm TL and less, of North American representatives at least, 
have minute teeth on the margin of the maxillary, but these disappear with age. The 
gill rakers on the upper limb do not yet extend downward over those on the lower limb, 
as in larger fish; the number, as already indicated, increases greatly with age. Further- 
more, until the fish reach about 100-125 mm TL, the two series of scales, one on 
each side of the median line of the back in front of the dorsal fin, do not become fully 
difTerentiated or modified. Though the exposed margin of the scales on young may 
be somewhat uneven, the scales are not definitely serrate; the serrae become somewhat 
developed when the fish attain a length of about 100 mm and continue development 
until they are comparatively large, but these serrations are not developed uniformly 
on all parts of the fish. For some species the length and shape of these serrations or 
pectinations are diagnostic characters, but fish of nearly equal size must be compared 
and scales from the same part of the body must be used. 

The species of Menhaden included in this work fall into two closely related pairs: 
tyrannus from the Atlantic and patronus from the Gulf; smithi from the Atlantic and 
gunteri from the Gulf — and one odd species, brevicaudata, from Noank, Connecticut, 
related to tyrannus. Such a close relationship between species exists in other groups 
as well.^^ 

In each pair of fishes named, the range probably was once continuous but became 
discontinuous when the last passageway for marine fishes across the Florida peninsula 
was closed. At least these fishes do not now occur on the coasts of southern Florida, 
indicating that suitable conditions do not exist there. Therefore, the separation between 
Atlantic and Gulf representatives is complete. Under this separation, and apparently 
under the influence of differences in environment, these fishes have become sufficiently 
differentiated to constitute distinct species. 

Range. The Atlantic and Gulf coasts of the United States northward to Nova 
Scotia; southern Brazil to Argentina; and western Africa. 

Measurements and Counts for Brevoortia. For measurements and counts not dis- 
cussed below, see p. 258. 

Head depth : for this genus, measured from the slight crossgroove at occiput to 
the keel of the first ventral scute. Although this is not an exact vertical, it does 
provide definite and uniform points for measuring. 

Scales; the number in a lateral series represents the number of oblique series 
(running upward and slightly backward) across the middle of the side ; these should 

36. Among the shad, Alosa lapidissima of the Atlantic and A. alabamae of the Gulf; also the hickory shad, Pomolobus 
mediocris, of the Atlantic and the skipjack herring, P. chrysockloris, of the Gulf and Mississippi Valley. In the 
family Sciaenidae, the weakfish, Cynoscion regalis, of the Atlantic and the sand seatrout, C. arenarius, of the Gulf 
(Ginsburg, 41: 83); and another pair occurs in the genus Menticirrhus (Ginsburg, unpublished). 



344 



Memoir Sears Foundation for Marine Research 



be counted uniformly, since they tend to become reduced in size on the back and 
larger on the lower part of the side. The comparisons and figures given on these pages 
are based on scales from the middle of the side below the anterior dorsal rays. 

Caudal fin: lower lobe measured from middle of caudal base. 

Pectoral fin: axillary appendage measured from base of upper pectoral ray. 






Figure 84. Brevoortta pelvic fins: a, patronus, 215 mm TL, off Galveston, Texas, USNM 1298 10, see Fig. 
91 ; B, tyrannus, 320 mm TL, Chesapeake Bay, USNM 129809, see Fig. 86; c, gunteri, 270 mm TL, Rock- 
port, Texas, USNM 129798, see Fig. 93; d, smhhi, 295 mm TL, USNM 1 18723, see Fig. 92; all equally 
enlarged. Drawn by Ann S. Green. 

Species. Five species fall within the scope of this work (see Key to Species), and 
two others, aurea iLud pectinata, occur in southern Brazil, Uruguay, and Argentina. Still 
another one, sometimes identified with tyrannus, has been reported for western Africa. 
I have seen no specimens of the latter and cannot add any information as regards its 
relationship to American forms. In general, the scales of South American representa- 
tives are scarcely reduced in size on the back and on the base of caudal, whereas in 
North American species they are greatly reduced there. 



Key to Species of the Western Atlantic 

I a. Scales fairly regularly placed, 35-36 oblique series crossing middle of side; great- 
est depth of body 30-40 "/o of SL. 
2 a. Scales on back and at base of caudal much smaller than those along middle 



Fishes of the Western North Atlantic 345 

of side; pelvic fin with a definitely convex margin, the innermost ray more 
than 67 «/o of the length of the outermost one, the fin not pointed when 
folded (Fig. 84 a, b); upper part of opercle with prominent radiating striae; 
shoulder spot followed by a variable number of smaller dark spots in adults. 
3 a. Ventral outline of body only moderately convex anteriorly, usually only 
about half of greatest depth below a straight line extending through 
lower margin of eye to middle of base of caudal; sheath of scales at base 
of dorsal fin composed for the most part of a single row of scales covering 
only the basal third of the longest rays when standing erect; pectoral 
fin falling far short of base of pelvic fin, 3-7 vertical series of scales 
between its tip and base of pelvic, the fin with 16-18 (rarely 15) rays. 
4 a. Maxillary reaching well beyond vertical from posterior margin of 
pupil, 13-16 «/o of SL; mandible 16-190/0 of SL; pectoral fin 
failing to reach base of pelvic fin by less than diameter of eye, 3 or 
4 vertical rows of scales between tip of pectoral and base of pelvic, 
its length 17-21 «/o of SL; caudal fin with lower lobe about as 
long as head, 25-35.5 "/o of SL. tyrannus (Latrobe) 1802, p. 346. 
4 b. Maxillary reaching only to a vertical from posterior margin of pupil, 
12-13 Vo of SL; mandible 1 5.5-1 6.5 «/„ of SL; pectoral fin failing 
to reach base of pelvic fin by a distance exceeding diameter of eye, 
5-7 vertical series of scales between its tip and base of pelvic, its 
length 1 5. 5- 1 6 "/o of SL; caudal fin with lower lobe shorter than 
head, 22.5-25 "/o of SL. brevkaudata Goode 1879, P- Z^2,- 

3 b. Ventral outline of body strongly convex anteriorly, much more than half 
of greatest depth below a straight line extending through lower margin 
of eye to middle of base of caudal ; sheath of scales at base of dorsal fin 
composed for the most part of two rows of scales covering basal two- 
thirds of shortest rays when standing erect; pectoral fin often reaching 
nearly to base of pelvic fin, seldom more than i or 2 vertical series of 
scales between its tip and base of pelvic, the fin with 14-17 (usually 15 
or 16) rays. patronus Goode 1879, P- 2>^S- 

2 b. Scales on back and at base of caudal not much smaller than those along middle 
of side; pelvic fin with a nearly straight oblique margin, the innermost ray 
not more than 67 »/o of the length of the outermost one, the fin pointed when 
folded (Fig. 84 c, d); upper part of opercle with feeble radiating striae, if any; 
shoulder spot not followed by smaller dark spots. 

5 a. Scales in only 35-46 oblique series crossing middle of side ; five longitudinal 
rows on side of caudal peduncle; pectoral fin sometimes reaching base 
of pelvic fin, occasionally falling short of this point by diameter of pupil, 
its length 1 7.5-21 "/o of SL, 0-3 vertical series of scales between its tip 
and base of pelvic. pectinata Jenyns 1842. 

Rio Grande of Brazil to Bahia 
Blanca, Argentina. 



346 Memoir Sears Foundation for Marine Research 

5b. Scales in 48—56 oblique series crossing middle of side: seven longitudinal 
rows on side of caudal peduncle; pectoral fin failing to reach base of 
pelvic fin by a space varying from half to a full diameter of eye, its length 
16.7— 18 "/o of SL, 3-6 vertical series of scales between its tip and base 
of pelvic. aurea Agzs?,\z 1829. 

Bahia (Bai'a), Brazil, prob- 
ably to the Rio de la Plata. 
lb. Scales small, irregularly placed, about 60—75 oblique series crossing middle of 
side (difficult to count); greatest depth of body 36-45.5 "/o of SL. 
6a. Length of head 29-31.5 "/o of SL; maxillary reaching from below middle of 
eye to vertical from posterior margin of pupil, 2.0—2.4 in head; pectoral fin 
generally failing to reach base of pelvic fin by somewhat more than half of 
diameter of eye, 5—8 vertical series of scales between its tip and base of pelvic, 
its length 1 8.5-21 "/o of SL; total number of ventral scutes 30—32 (usually 
30); vertebrae 45-47. smithi Hildebrand 1941, p. 372. 

6b. Length of head 3 1-35-5 "/o (usually 32-34 "/o) of SL; maxillary reaching to, 
or a little beyond, vertical from posterior margin of pupil, 1.8-2.2 in head; 
pectoral fin generally failing to reach base of pelvic fin by less than diameter 
of pupil, 2—4 vertical series of scales between its tip and base of pelvic, its 
length 19-23.570 (usually 20-22 "/„) of SL; total number of ventral scutes 
27—30 (usually 28 or 29); vertebrae 42-44. 

gunteri Hildebrand 1948, p. 376. 



Brevoortia tyrannus (Latrobe) 1802 
Atlantic Menhaden, Menhaden, Mossbunker, Bunker, Fatback, Shad, Pogy, Bugfish 

Figures 84—89 

Study Material. At least 1 50 specimens, sometimes many more, 70-430 mm TL, 
55—2S5 "^"^ SL, collected at many localities from Massachusetts Bay to Fernandina, 
Florida. Some of the data used were accumulated by me over a period of 33 years; in 
addition, unpublished data from Dr. A. Bascom Hardcastle and some gathered by the 
late William W. Welsh; also juveniles from Woods Hole, Massachusetts; Chesapeake 
Bay; and Beaufort, North Carolina. 

Distinctive Characters. The close relationship of Brevoortia tyrannus to B. patronus 
of the Gulf of Mexico is shown in the account of the latter (p. 370). From B. smithi, 
which occupies the same range as the Atlantic Menhaden from North Carolina 
southward, tyrannus differs prominently in having larger scales with notably longer pec- 
tinations; these species also differ in several other respects as shown in the account of 
B. smithi (p. 374). 

Description. Proportional dimensions in per cent of standard length, and counts, 
based on at least 150 study specimens, ^^—JiSS "^"^ S^- 



Fishes of the Wester?! Nort/i Atlantic 



347 






Figure 85. Brevoortia scales from middle of side below anterior dorsal rays, a, fatronus, 21 5 mm TL, see Fig. 
91; B, tyrannus, 320 mm TL, see Fig. 86; c, gunteri, 270 mm TL, see Fig. 93; d, smithi, 295 mm TL, see 
Fig. 92; all equally enlarged. 



Body, depth 30-40, usually 32-37. 

Caudal -peduncle: depth 8.5— 11. 

Head: length 29—36, most frequently 

31-34- 
Snout: length 5.5-9.0, usually 6.0-8.0. 
Eye: diameter ca. 5.0—8.0. 
Interorbital: width 6.2—8.2. 
Maxillary: length 13—16. 
Mandible: length 16—19. 
Anal fin: length of base 14—19, usually 

15-18. 
Pelvic fin: length 8.0-10. 
Pectoral fin: length 17-21; usually 

18-20. 



Scales: oblique series along middle of 
side, 41—55, most frequently 45— 

52. 
Modified scales: in a series in front of 
dorsal 31—43, most frequently 33— 

39- 

Ventral scutes: 30—35, usually 32 or 

Fin rays: dorsal 18—22, usually 19—21; 
anal 18-24, usually 20—23; pec- 
toral 16-18, rarely 15. 

Vertebrae: 47—49, very rarely 45, 46, or 
50 (194 specimens). 



348 Memoir Sears Foundation for Marine Research 

Body elongate, compressed, its greatest thickness varying greatly, from nearly 
half of depth in large fat fish in spawning condition to only about a third of depth in 
lean half-grown examples; greatest depth generally at, or slightly in advance of, vertical 
from tip of pectoral, 2.5—3.3 in SL in specimens 100 mm and upward; ventral outline 
moderately convex anteriorly, usually only half of greatest depth below a straight line 
through lower margin of eye to middle of base of caudal. Caudal peduncle rather 
slender, 2.9-4.0, usually 3.2-3.8, in head, and 3.2-4.1, but usually 3.5-4.0, In greatest 
depth of body. 







Figure 86. Brevoortia tyrannus, 320 mm TL, 247 mm SL, Chesapeake Bay, USNM 129809. Drawn by 
Ann S. Green. 

Scales adherent, the exposed part notably deeper than long, the scale itself some- 
what deeper than long (Fig. 85 b); margin irregular in young about 60 mm long; 
short blunt serrae on margin in examples of about 100 mm, the serrae increasing in 
length with age, becoming slender hair-like appendages, often extending nearly across 
the exposed portion of the next scale in large specimens; scales greatly reduced in size 
on upper part of side, on back, and at base of caudal; in rather regular series on lower 
half or so of side; most frequently 3 or 4 vertical rows of scales exposed between tip 
of pectoral and base of pelvic. A row of large modified scales on each side of median 
line of back in front of dorsal fin, these scales not fully modified in specimens under 
about 125 mm TL, rather variable in number, most frequently 34-38 in each series. 
Ventral scutes rather strong, 17-22, usually 18-21, in front of pelvic fins; 10-15, 
usually 12—14, behind these fins. 

Head 2.8-3.5, usually 2.9-3.2, in SL, its depth 3.1-4.0, usually 3.3-3.8. 
Snout only moderately blunt, with a prominent median notch in adults, the length 
3.9-5.5, usually 4.0-4.7, in head. Eye difficult to measure accurately in adults 
because of much adipose tissue, about 4.4-6.2. Interorbital 4.2-5.3, usually 4.4- 
5.0. Cheek deeper than long except in young. Maxillary rounded, generally scarcely 
reaching below posterior rim of eye, the length 1.9-2.7, usually 2.1-2.5, i" head. 
Mandible included in upper jaw, its upper margin (within mouth) nearly straight. 



Fishes of the Western North Atlantic 



349 



its length to joint i .7—2.0 in head. Upper section of Opercle with prominent radiating 
ridges. Gill rakers very numerous; those on upper limb extending down and over 
those on lower limb, the longest about as long as snout in adults; increasing in number 
with age and growth, the greatest increase occurring before a loo-mm length is at- 
tained: about 60 on lower limb in 60-mm specimens, about 100 in lOO-mm fish, about 
140 in 200— 250-mm examples, and 150—160 in large adults 330— 360 mm long. 
Teeth absent except for minute ones on margin of maxillary in young about 60 mm 
and less. 

Dorsal fin moderately elevated anteriorly, its margin definitely concave, its longest 
rays as long as snout and fully half of eye, the last ray somewhat longer than the pre- 
ceding ones; origin of fin generally slightly nearer to base of caudal than to margin of 




Figure 87. Brevoortia tyrannus, modified scales in front of dorsal fin, from same specimen shown in Fig. 
Drawn by Ann S. Green. 



snout; a low sheath on base of fin, composed for the most part of a single series of 
scales not extending above the basal third of shortest rays when standing erect. Caudal 
rather deeply forked, the middle rays about as long as eye; the lobes rather short, the 
lower one the longer, about as long as head, 2.8—4.0, usually 3.0—3.8, in SL. Anal 
low, little elevated anteriorly, the margin nearly straight ; its origin under or somewhat 
behind vertical from tip of last dorsal ray, its base 5.2—7.0, usually ^.^—6.6, in SL; 
a very narrow sheath of scales at base. Pelvic fin small, with a very gently convex 
margin, the outermost ray only a little longer than the innermost (Fig. 84 b), the fin in- 
serted slightly behind vertical from origin of dorsal, the length 3.1—3.8 in head. Pec- 
toral fin slightly falcate, its length 5.0—5.9 in SL in adults, 1.6— 1.9 in head, the 
longest ray about four times the length of the shortest one, the fin generally failing to 
reach base of pelvic in adults by a distance somewhat greater than diameter of pupil. 
Axillary appendage of pectoral variable in length, about 75 "/g of the length of fin 
in large examples, only about half of length of fin in examples about 100 mm long, 
and little developed in young of 50 mm. 

Color. In fresh specimens, back dark green to bluish. Sides generally brassy and 
sides of head bright silvery to slightly brassy. A large black spot at shoulder a very 
short distance behind margin of opercle and well above middle of side, variable in size 
and shape among specimens, generally roundish but sometimes vertically elongate, and 
about as large as pupil. This spot develops first and is generally evident when the 
fish reach a length of about 75 mm (3 in.). The shoulder spot usually followed among 
halt-grown and large examples, from about 150 mm (6 in.) upward, by a variable 



3 50 Memoir Sears Foundation for Marine Research 





Figure 88. Brevoortia tyrannus. a, newly laid egg in advanced stage of cleavage; b, egg with advanced stage 
of embryo; c, newly hatched larva, 4.5 mm long. After Kuntz and RadclifFe. 

number of smaller dark spots, sometimes only a few small ones and sometimes many, 
differing greatly in number, size, and intensity of darkness, these spots sometimes 
arranged in two, three, or more indefinite horizontal series on anterior part of side, 
seldom extending far beyond vertical from origin of anal; a profusely spotted example 
illustrated in Fig. 86. Fins pale yellow to brassy, sometimes with dusky punctulations; 
caudal generally dusky at base and often with a dusky margin. Peritoneum black. 

Variation in color, especially in the dark spots on the side, is very great. 

Size. The usual size of adult fish at Beaufort, North Carolina, is about 300- 
350 mm (12-14 in.) TL. It is generally accepted that the fish grow rather larger in the 
northern part of the range than in the southern part. The largest specimen in the U. S. 
National Museum, from Long Branch, New Jersey, is 430 mm (17 in.) long, but a 
larger one, 480 mm (18 in.), has been reported as taken at Woods Hole, Massa- 
chusetts {120: 91), and one of 500 mm (20 in.) was reported to the writer by Dr. 
H. F. Prytherch verbally. The last mentioned specimen was taken at Beaufort, North 
Carolina, and weighed 3.5 pounds. 



Fishes of the Western North Atlantic 351 

Development and Growth. The eggs and larvae have been described by Kuntz and 
RadcHffe from material collected at Woods Hole, Massachusetts {^6 : 119). The eggs 
are buoyant and spherical, have a diameter of 1.4-1.6 mm, and are highly transparent; 
they contain a small oil globule, and are covered with a thin horny membrane; the 
perivitelline space is broad. Incubation occupied "not more than 48 hours." 

The newly hatched fish are "relatively slender" and about 4.5 mm long. The 
vent, as usual in clupeoid fishes, is "less than one-fifth the length of the body from the 
posterior end." At four days of age the larvae were 5.7 mm long, and black chromato- 
phores were present in a series along the entire digestive tract. At 9.0 mm the dorsal 
and caudal fins were at least partly developed, and posteriorly the intestine was convo- 
luted. At 23 mm all the fins had become well differentiated, and the body remained 
"relatively slender"; black chromatophores were present on the nape, along the margin 
of the opercle, near the base of the dorsal, anal, and caudal fins, and in groups posterior 
to the dorsal fin and ventral to the pectoral fin; internal black areas could be seen 
along the dorsal wall of the abdominal cavity, and a series of dark spots was present 
above the notochord. 

Compared to the Woods Hole larvae, specimens from Beaufort, North Carolina, 
of comparable size agree with them fairly well in development, but in each phase they 
are notably more slender. Beaufort examples 9.0 mm long are little compressed and 
scarcely thicker than a number-60 sewing thread; the depth is contained about 
30 times in TL. Specimens 18 mm long are fully as well developed as those of 23 mm 
described and illustrated by Kuntz and Radcliffe, and the 23-mm examples are rather 
more advanced than the northern ones of equal length; the body continues to be more 
slender; the depth is contained about 21 times in TL. In 25-mm preserved examples 
from Beaufort, the convoluted intestine noted in younger fish is scarcely discernible, 
and the vent Is now situated slightly behind a vertical from the base of the last dorsal 
ray, about as in the adult. Some examples, at a length of 30 mm, or even up to '^1^ mm, 
remain very slender, their depth being contained about 13 times in TL; and in these 
slender examples the ventral surface of the chest and abdomen remains round to slightly 
flat, with indications of a row of bony points along each lateral edge. Juvenile pigment 
spots remain about as described for examples of 23 mm or so, but other specimens of 
the same length are much more compressed and notably deeper; their greatest depth 
is contained only about 7.7 times in TL, or only about 6.7 in SL. In these deeper 
and older fish, the two series of lateral bony points noted in the slender fish have 
become coalesced to form definite ventral scutes on the median line of the chest and ab- 
domen ; also, the fish are rather fully pigmented, the sides of the head are bright silvery, 
a silvery lateral band is evident, and a sheen of the same color is present on the sides of 
the abdomen. It seems, then, that when the fish reach a length of about 30 mm the 
increase in length is retarded during metamorphosis from the somewhat roundish 
slender larvae to the deeper and more compressed young adults. 

Schools of young Atlantic Menhaden were observed repeatedly at Beaufort in 
winter and spring, and specimens 24-35 '""^' caught from a school on April 23, 



3 52 Memoir Sears Foundation for Marine Research 










Figure 89. Brevoortia tyrannus. a, larva four days after hatching, 5.7 mm; b, young, 9 mm; c, 23 mm; d, 33 
mm. A, B, c drawn by Albert Kuntz, d by E. B. Decker. 

1926, differed little in development; all had remained slender and were more or less in 
the larval stage. As seen in the water they were almost transparent, but when removed 
from the water, dark pigment spots were evident along the middle of the side in a more 
or less definite lateral band. 

Specimens 40-45 mm long are shaped very much like large fish, although they 
remain rather more slender; their greatest depth of body is contained about 3.6-4.0 in 



Fishes of the Western North Atlantic 353 

SL. The obtuse angle in the lower limb of the first gill arch is evident, and the gill rakers 
on the upper limb of this arch now definitely extend down and over the upper ones on 
the lower limb. Striations on the upper plate of the gill covers are evident, but the cheek 
remains about as long as deep. Scalation is about complete. The silvery lateral band of 
smaller specimens has blended with the silvery abdomen ; the dark shoulder spot, always 
present in adults, is evident though small; and in some specimens suggestions of sup- 
plementary spots are already present. 

The gill rakers increase in number and in proportionate length with age (see 
Description). Therefore, if their number is to be usable at all for differentiating species, 
specimens of equal size must be compared. 

In the Gulf of Maine, the fish hatched during the summer reach a length of 
60— 80 mm during their first winter and average slightly more than 160 mm the second 
winter, and those hatched in the fall are 30 mm long the first winter and 1 30 mm the 
second winter, with every gradation in size between the two {16: 122). 

In Chesapeake Bay, where the fish apparently spawn only during the autumn, 
the young are only about 27 mm long the following January and 46 mm the following 
April {^g: 103). These data, though not complete, are in general agreement with more 
extensive data collected by me at Beaufort, where spawning quite certainly occurs only 
during late fall and early winter. Seventy-nine larvae, collected there from the last half 
of November to and including the first half of March, were between 4.0—20 mm, and 
the average increase in length in each month was about 1 2 mm, excluding December, 
when only three larvae were taken (5.0, 5.0, and 7.0 mm). Thereafter, more rapid 
growth was indicated: seven specimens taken the last half of March were between 
22—31 mm and averaged 27 mm; 1 10 collected in April were 30—40 mm and averaged 
about 'iiS irin^; 50 collected during May were 30—41 mm and averaged about 35 mm; 
and 48 examples taken in June ranged between 38—56 mm and had an average 
length of 45 mm. Lengths of 85-150 mm (3.4-6.0 in.) were attained during their 
second winter; the average was about iiomm (4.4 in.); measurements were based 
on 167 fresh examples. 

"Sexual maturity is apparently attained in the season following the third winter, 
and a few of the older fish Welsh examined showed as many as 9—10 winter rings on 
their scales" (Bigelow and Schroeder, i^\ 116). 

Spawning. The Atlantic Menhaden spawn at sea in saline waters. Many large 
fat fish with well-developed roe were observed by me off Beaufort, North Carolina, during 
autumn only. Such fish were seen also in the fall in the Chesapeake Bay area i^^g'. 
103) and in December off Fernandina, Florida (from unpublished notes by William 
W.Welsh). However, off New England, spawning takes place in late spring and through- 
out the summer, the height of the spawning season occurring in June; and in 19 15 
off Woods Hole, Massachusetts, eggs were taken as late as the last week of October 

(76: 119)- 

At Beaufort, spawning apparently takes place only off Beaufort Inlet, for all of 
the small ones, 1 6 mm and less, were taken at sea. This would be expected, for the large 



3 54 Memoir Sears Foundation for Marine Research 

mature fish in that vicinity seldom enter the harbor or other inshore waters. Northward, 
however, as at Woods Hole, Massachusetts, the eggs and larvae were collected in the 
harbor {j6'. 1 19). The place and season of spawning, then, seem to vary according to 
locality. 

Habits. Both young and adults, acccording to Bigelow and Schroeder, feed and 
travel in compact schools at the surface, swimming close together side by side and 
tier above tier (75 : 114): 

In calm weather they often come to the surface when their identity can be recognized by the ripple they 
make, for pogies, like herring, make a much more compact disturbance than mackerel do. . . Also, pogies as 
they feed frequently lift their snouts out of water, which we have never seen herring do, while they break the 
water with their dorsal fins, also with their tails. And the brassy hue of their sides catches the eye. 

The larvae appear to be more or less solitary, for only one specimen or at most a 
few were generally taken in a haul, off Beaufort. They do not stay at or near the surface 
as strictly as older ones do, for most of the larvae at hand were taken in nets towed 
along the bottom. However, schooling and surface swimming begin early in life, for 
many schools of young 25—35 mm long (some just reaching the adult stage) were 
observed many times on windy days near shore on the leeward side of Pivers Island 
at Beaufort. 

Migrations. These fish are present principally in the Gulf of Maine from July to 
September, off Woods Hole, Massachusetts, from June to September, and off the coasts 
of New York, New Jersey, and Delaware from May to September, being most numer- 
ous during midsummer; in Chesapeake Bay small to medium-sized fish are numerous 
throughout the summer, and large ones are taken off the mouth of the Bay (or off the 
Capes) during October and November; in North Carolina, small and medium-sized 
fish are present all summer, the large ones from October to December; and on the 
east coast of Florida the fish are numerous from early spring to late fall. 

Small to medium-sized specimens are present in Chesapeake Bay throughout the 
year but in greatly reduced numbers during the winter. Adults off Beaufort, though 
most numerous during late fall, sometimes remain in the nearby offshore waters until 
January if the winter is mild. Schools composed of fish that are generally 1 50 mm 
(6 in.) and less are present in Beaufort Harbor and connecting estuaries all winter, 
and schools of young, generally under 50 mm in length, ascend freshwater creeks or 
enter small drainage ditches in the spring. Unlike adults in northern waters that come 
close inshore, into the bays and harbors {16: 122), large fish in North Carolina very 
rarely cross Beaufort Inlet to enter the Harbor and adjacent waters. 

It is still much of a mystery where the schools of adults go after the fish leave 
the waters within 10—15 ™^^^ of shore, the usual operating range of the fishing 
schooners. Sufficient taxonomic work has been done, however, to indicate that each 
section of the coast, that is the New England, Middle Atlantic, and South Atlantic 
states, has a more or less distinct population. Therefore, if a north-south migration 
does take place, then the fish from each section return to the general vicinity from 
which they migrated originally. It seems more probable now that the principal migra- 



Fishes of the Western North Atlantic 355 

tion is offshore and inshore." No doubt temperature is an important factor in the 
movements of the fish; it certainly affects the supply of plankton upon which they 
feed. In years when inshore runs fail, it is even probable that the fish remain so far 
offshore that they are beyond the usual range of the fishing vessels. In slight con- 
firmation of this theory, I recall that one autumn, when the usual inshore runs failed 
at Beaufort, a progressive operator, upon my suggestion, induced one of his boat 
captains to go farther offshore than usual in search of a catch. Fish, indeed, were 
found, at a considerable distance offshore, apparently too far off for practical fishing 
with the crews and schooners available. 

Food. Menhaden feed on small organisms strained from the water by their numer- 
ous long, slender, close-set gill rakers, which form an effective strainer. While feeding, 
the fish generally swim near the surface and often "break water"; they whirl around, 
sound a short distance, come out of the whirl, and swim up and straight ahead at a 
considerable speed for a rather short distance. During this time the mouth is wide 
open and the gill covers are lifted, thus making it possible for a fish to filter a great 
amount of water with minimum effort. The food that is ingested depends in large 
measure upon the organisms that are present where the fish are feeding. Even a 
considerable amount of mud and general debris is often swallowed. Included in the 
stomach contents examined by various investigators were: numerous small crustaceans, 
especially copepods; small annelid worms; rotifers; and unicellular plants, particularly 
diatoms and peridinians. The plant organisms, as a rule, constitute the chief food. 

This species has a very thick-walled stomach, and a long intestine, as is usual among 
fish that feed on vegetable matter. Among 64 fish taken at Beaufort, ranging between 
65—214 mm, the intestinal tract increased more or less proportionately with the length 
of the fish; generally it remained between 400 and 450 "/o of the standard length, 
according to A. Bascom Hardcastle (unpublished data). 

Enemies. Most predatory animals associated with the sea feed on Atlantic Men- 
haden — an easy prey because of their habit of schooling. Their fiercest enemy prob- 
ably is the bluefish {Pomatomus)^ which, it is said, kills many more than it eats. Among 
the other fish that feed on them extensively are the cod, pollock, hakes, weakfish, 
swordfish, tuna, dolphin, amberjacks, and sharks. Whales and porpoises, as well as 
birds, also devour many of them. 

Parasites. Atlantic Menhaden so often have a large crustacean parasite in the 
mouth that they are sometimes referred to as "bugfish." It is said that in some schools 
nearly every fish has a parasite in its mouth. This "bug," an isopod listed as Olencira 
praegustator (Latrobe) (J07 : 231), has a robust body, generally fully half an inch long 
and occasionally approaching an inch in length. Sometimes the mouth of the fish is 
deformed because of its presence ;''thus the amount of water screened for food must be 
considerably reduced in fish so infected. 

37. According to Rose {tio: 104, 105), the recent investigations by the U. S. Fish and Wildlife Service make it likely 
that the north-south migrations of the Menhaden are more regular and extensive than Dr. Hildebrand supposed. 
— H. B. B. 

23* 



3 5^ Memoir Sears Foundation for Marine Research 

Another large parasite, the copepod Lernaeenicus radiatus (LeSueur), buries its 
horns in the flesh on various parts of the fish's body, most commonly along the back 
{JST- 480). This parasite sometimes approaches a length of two inches and has a pre- 
dominant reddish color, which makes it very conspicuous. Other copepods listed 
by Wilson as parasitic on B.tyrannus are: Bomolochus teres Wilson, Caligus schistonyx 
Wilson, Caligus cheliferWWson, Pandarus sinuatus Say, Lernanthropus brevoortiae Rathbun, 
and Clavell'isa spinosa Wilson. 

The following have been listed by Linton {85: 440): Scolex polymorphus'D\i]2.rd\n, 
a cestode, small and free in the intestine; cysts and blastocysts of Synbothrium on the 
viscera; and the trematodes Z)«/ow«;« appendiculatum Rudolphi and I), vitellosum Linton; 
later he added Dactylocotyle sp., and Distomum pyriforme Linton {86: 352). 

Hardcastle has reported a sporozoan parasite of the genus Eimeria for the testes 
of specimens taken at Beaufort; he stated that a "study of over a hundred individuals 
[Menhaden] showed an infection of nearly 50 per cent" (54: 214). 

Diseases. No report of a specific disease has been found in the literature, but 
there are records of many fish having become stranded along the shore. It has been 
suggested that their enemies at times drive them ashore, but my observation is that the 
fish make no special effort to flee from enemies. In fact, when caught in a net they 
make little effort to escape. It seems more probable, therefore, that the stranded fish 
died of a disease. An epidemic has indeed been reported (J25 : 742). 

Variations. Large fish generally are proportionately deeper than small ones, and 
those from the southern part of the range are proportionately deeper than northern 
ones. However, the variation in depth among fish from any one general region is great 
enough to obscure any definite distinctions. Southern fish have a somewhat larger head 
and longer fins (especially the pectoral) than those from the northern part of the range, 
but the gap is fully bridged by examples from intermediate areas. Contrary to expecta- 
tion, data at hand show that the vertebrae do not average fewer in southern specimens. 
But further detailed studies are needed to determine the exact characteristics of popula- 
tions inhabiting the different general sections of the coast ; even now specimens from 
the Gulf of Maine are too few in the collections at hand to determine their finer charac- 
teristics. 

As shown in the Description^ the number of gill rakers increases greatly with age, 
with the principal increase occurring before the fish reach a length of about 100 mm 
(4 in.). Also, the scales in small specimens have only irregular margins while those in 
larger fish are pectinate. In using scales to differentiate species (their development in 
all species being similar), it is necessary to compare scales from examples of about equal 
size; and it is preferable to use large specimens, for in them the differences are most 
pronounced. 

See Color for variations in color. 

Commercial Importance. These fish, though exceedingly valuable, are not used very 
extensively as food by man, mainly because of their bony nature and oiliness. How- 
ever, some find the flesh delicious, and those living along the coast, especially the 



Fishes of the Western Nortli Atlantic 357 

fishermen, eat them in season as a common article of diet. Considerable quantities are 
often "corned" (salted) for home use during winter, and the writer knows from first- 
hand experience that these fish are delicious when smoked. They were canned to a 
limited extent for export during the last war, and a small quantity is still canned for 
home consumption. 

Because of their abundance, large numbers are available for reduction to oil, fish 
meal, and fish scrap; the Atlantic Menhaden fishery is one of the most valuable of 
our fish industries. The catch for 1945 amounted to 759,073,820 pounds, valued at 
$ 11,202,127 (Canned Fish and Byproducts, U. S. Fish Wildl. Serv., 1945: 15); of 
this amount, 120,493,920 pounds were credited to Florida, 142,209,510 to North 
Carolina, 77,232,910 to Virginia, 280,463,340 to Delaware, and 138,584,140 to New 
Jersey. For 1950 through 1953, the approximate catches have been as follows: between 
9,304,000 and 39,900,000 pounds for New England, mostly taken south and west 
of Cape Cod; between 372,950,000 and 857,584,000 pounds for the Middle Atlantic 
states; 92,374,000 for Chesapeake Bay; and between 147,125,000 and 314,841,000 
pounds for the South Atlantic states. In 1953 the total catch was 1,259,031,000 
pounds. 

Because of the great fluctuations in catch from year to year, the catch for a single 
year, or even for many successive years, may not indicate their actual abundance. 
Furthermore, since the fish are not equally fat every year, or at all times when 
available, equal quantities of fish (pounds) do not always yield equal quantities of oil 
(gallons), the most valuable product obtained from these fish. Thus the value of a given 
catch depends to a large extent upon the amount of oil it yields. In general, large fish 
yield more oil than small ones. For example, at Beaufort, North Carolina, the summer 
"runs," consisting of small to medium-sized fish, yield so little oil that they are con- 
sidered scarcely worth handling, but the fall runs, consisting of large and fat gravid 
fish, make the industry profitable. 

In discussing this subject, Greer (^48: 24)^* wrote as follows in 191 5: 

The yield of oil varies greatly on different parts of the coast, and also from year to year and at different 
seasons. The fish are always larger and fatter during the fall than at any other season, consequently there is a 
greater yield of oil during that period. The fish taken from northern waters produce more than those taken from 
southern waters; the writer visited one of the northern factories during August 191 3, and was told that a short 
time previous to his visit they had made 20 gallons, and that the average yield for July had been 16.5 gallons 
per 1,000 fish.^^ The maximum for the Middle Atlantic States is about 15 gallons, though it usually is less; 
in the early part of the season it is very much less than that amount. The South Atlantic fish rarely produce more 
than 8 gallons per 1,000 fish, and frequently it is less than i gallon during the early spring fishing. The average 
number of gallons of oil made from 1,000 fish during the season of 1912 was as follows: Connecticut 11.73; 
New York 11.36; New Jersey 5.08; Delaware 7.72; Virginia 3.72; and North Carolina 1.98. 

38. This paper, entitled "The Menhaden Industry on the Atlantic Coast," although written before the manufacture 
of fish meal for stock and poultry feed, contains much valuable information concerning the history of the industry, 
the construction of factories, the machinery used in these reduction plants, the vessels employed, the methods of 
fishing, their reduction, men employed, etc. 

39. The fish generally are measured, and the measure used is said to contain 1,000 fish, no account being taken of the 
size of fish. Thus, the measure arbitrarily contains 1,000 fish regardless of whether they are 6 or 12 inches long. 
In converting the number of fish to pounds, the factor 0.67 is used in the statistical division of the U. S. Fish and 
Wildlife Service. That is, the number of fish is multiplied by o.ffj. 



358 Memoir Sears Foundation for Marine Research 

As further evidence of the great value of this extensive fishery, additional data 
from Greer show that the total catch for 1912, a near banner year, was 1,061,483,750 
fish (711,435,312 pounds); this catch yielded 6,651,203 gallons of oil valued at 
$ Ij55I)990 and 88,520 tons of scrap worth $2,138,165; 48 factories, valued at 
^ 3)625,983, were in operation; the vessels engaged in the fishery numbered 147, 
valued at % 3,456,792; and there were 2,159 employees in the factories and 3,735 
fishermen, drawing a combined wage of $ 1,579,984. 

The value of the fish scrap, now that most of it is used as fish meal for poultry 
and livestock feed instead of as fertilizer, is considerably greater than it was ; this change 
in turn produced other changes, for this processing requires quicker and more sanitary 
handling of the fish, including cooking before decomposition sets in; some changes at 
the factories were also required, for apparatus had to be installed for drying the residue 
quickly after the oil had been extracted. 

Fluctuations in Abundance. No species has been more erratic in its appearances along 
the coast than B. tyrannus. For example, in 1 9 1 1 and in 1 9 1 2 large catches were made 
at Beaufort, the center of this fluctuating industry in North Carolina, but then these 
near banner years were followed by several exceptionally lean years, which forced 
some of the local operators into bankruptcy. However, in 191 8 the fish were back 
again. Even though the annual catches (without adjustment for variations in fishing 
effort) have fluctuated since that time from 54,476,000 pounds in 1932 to 181,968,000 
pounds in 1939, there has been no complete failure in any year since the period 
1913-1916. 

Although at least some of the reasons for the fluctuations in abundance of certain 
species of fish are known, no explanations have been given for the fluctuations in At- 
lantic Menhaden. For the Gulf of Maine, Bigelow and others have stated that these fluc- 
tuations date back to the early history of the fishery and have continued since then (15 : 
116, 117; j6: 120). Thus, in 1845 they were abundant and in 1847 scarce. Then, for 
some years prior to 1875, they were tremendously abundant off the coasts of Massa- 
chusetts and Maine every summer, and a considerable fishery for them grew up on the 
Maine coast. However, during the cold summer of 1877 few were taken in the Gulf un- 
til September and October, when they were reported as "about as abundant as normal." 
So few of these fish appeared north of Cape Cod from 1879 through 1885 that the cap- 
ture of one caused comment, and many people thought that the fish would never return. 
However, in 1886 the fish reappeared and in 1888 they were so abundant that the 
fishery was revived. The period of abundance was short, however, for in 1891 the 
catch was less than half of that taken the year before, and in 1 892 few were seen north 
of Cape Cod. Similar periods of scarcity and abundance followed each other until 1 904, 
when a 20-year period of scarcity began. Then in 1922 the fish returned in such abund- 
ance that 18 steamers fished successfully in Massachusetts Bay; further, "upward of 
1,500,000 pounds were landed by the larger fishing vessels besides what the small 
boats brought in." From the middle 1920's until the middle 1940's there were not 
enough of them in the Gulf of Maine to be of any commercial importance there. 



Fishes of the Western North Atlantic 359 

Although they were plentiful again in the Gulf from 1946 through 1949, fewer were 
present from 1950 through 1956. 

Comparison of statistics for the past 20 years or so shows no correlation as to the 
quantity taken along different sections of the coast. If North Carolina, for example, has 
a good run in a given year, it does not follow that Florida, New York, or New Jersey 
will have a good run that same year. For example, in 1929 the fishery in North Caro- 
lina yielded 173,490,000 pounds of Atlantic Menhaden — considered an unusually large 
catch. But in that same year the catch credited to Florida was rather less than average, 
50,532,000 pounds, and the combined catch for New York and New Jersey was a 
near failure with only 11,092,000 pounds. Again in 1937, when the catch in North 
Carolina amounted to only 61,706,000 pounds (next to the smallest during about 20 
years), the catch for Florida was 139,788,000 pounds and the combined catch for New 
York and New Jersey was 99,684,000 pounds, both well above the average for the 
20-year period. 

It seems improbable that the commercial catches taken by the fisheries in the 
western Atlantic have made any serious inroads on Atlantic Menhaden populations in 
any general area. If it is true that the handiwork of man has had no important effect 
on the abundance of this species, then the fluctuations must be attributed to natural 
phenomena that occur in their enviroment. 

Successful and unsuccessful spawning seasons are known to profoundly affect the 
abundance of some species and this is probably so in the case of this species also. 
However, it does not explain why a fish that may attain an age of 9-10 years 
(as concluded from the number of winter rings on the scales) may be abundant one 
year in a certain locality but very scarce or even absent there the next year. Furthermore, 
since the fish caught during a fishing season in any one area are not all of the same 
age and size, the failure of one or even several successive spawning seasons would not 
cause an abrupt diminution from one year to the next. The theory that these fish follow 
their food and therefore remain in water having an abundant growth of plankton of 
suitable size has been advanced. It would follow then that if sufficient food is not 
available near the shore, the fish will remain offshore, perhaps beyond the range 
of the vessels employed in the fishery. This theory seems to be the most plausible one 
advanced to date. It has been suggested also that the temperature of the water is a 
factor, which no doubt is true, particularly insofar as it affects the growth of the plank- 
tonic food required by these fish. It is obvious that further study is needed to determine 
more specifically the reason or reasons for the exceedingly great fluctuations in the runs 
of Atlantic Menhaden. 

Methods of Fishing. These fish are caught chiefly with purse seines, though smaller 
quantities are taken with long haul seines and in pound nets; and minor quantities are 
caught with ordinary seines or drag nets and with gill nets. 

The vessels employed in the operation of this large industry are a very familiar 
sight to those living near the fishing centers, and in some localities, for example 
Beaufort and Moorehead City, North Carolina, t