Tu (^ ¥:c^ HARVARD UNIVERSITY Library of the Museum of Comparative Zoology nio4'^ MUS. COMP. ZOOLi LIBRARY SEP 8 1978 HARVARO UNIVERSITY VOLUME 20 1978 TULANE UNIVERSITY NEW ORLEANS TULANE STUDIES IN ZOOLOGY AND BOTANY, a publication of the .Biology Department of Tulane University, is devoted primarily to the biology of the waters and adjacent land areas of the Gulf of Mexico and the Caribbean Sea, but manuscripts on organisms outside this geographic area will be considered. Each number is issued separately and contains an individual monographic study, or several minor studies. As volumes are completed, usually on an annual basis, title pages and tables of contents are distributed to recipients receiving the entire series. Manuscripts submitted for publication are evaluated by the editor or associate editor and by an editorial committee selected for each paper. Contributors need not be members of the Tulane University faculty. When citing this series authors are requested to use the following abbreviations: Tulane Stud. Zool. and Bot. INFORMATION FOR AUTHORS: The editors of Tulane Studies in Zoology and Botany recommend conformance with the principles stated in CBE Style Manual, 3rd ed., published in 1972 by the American Institute of Biological Sciences, Washington, D.C. Manuscripts should be submitted on good paper, as original typewritten copy, double-spaced, and carefully corrected. Two copies, carbon or other suitable reproduction, must accompany the original to expedite editing and assure more rapid publication. Legends for figures should be prepared on a separate page. Illustrations should be proportioned for one or two column width reproductions and should allow for insertion of legend if occupying a whole page. Photographs should be on glossy paper. Many tables, if carefully prepared with a carbon ribbon and electric typewriter, can be photographically reproduced, thus helping to reduce publication costs. Letter- ing in any illustrative or tabular material should be of such a size that it will be no less than 1 V^ mm high when reduced for publication. An abstract not exceeding three percent of the length of the original article must accompany each manuscript submitted. This will be transmitted to Biological Abstracts and any other abstracting journal specified by the writer. Authors of contributions will receive a Statement of Page Charges, calculated at $45/page. Partial or complete payment of these charges is solicited from authors who have funds available for this purpose through their institutions or grants. Acceptance of papers is not dependent on ability to underwrite costs. Illustrations and tabular matter in excess of 20 percent of the total number of pages may be charged to the author; this charge is subject to negotiation. EXCHANGES. SUBSCRIPTIONS, ORDERS FOR INDIVIDUAL COPIES: Exchanges are invited from institutions publishing comparable series but subscriptions are available if no exchange agreement can be effected. A price list of back issues is available on request. Remittance, preferably money order, should accompany orders from individuals. Make remittances payable to "Tulane University." Authors may obtain separates of their articles at cost. Copies of Tulane Studies in Zoology and Botany sent to regular recipients, if lost in the mails, will be replaced if the editorial offices are notified before the second subsequent issue is released. COMMUNICATIONS: Address all queries and orders to: Editor, Tulane Studies in Zoology and Botany, Department of Biology. Tulane University, New Orleans, Louisiana 70118, U.S.A. Harold A. Dundee, Editor Arthur L. Welden, Associate Editor David C. Heins, Assistant to the Editors TULANE STUDIES IN ZOOLOGY AND BOTANY VOLUME 20 INDEX TO AUTHORS AND SCIENTIFIC NAMES (New taxonomic entries in boldface) A cue an da fitzp atricki, 59 Alligator, 50 Arnbloplites rupestris, 105 Amia, 50 Artemia, 138 Barbourofelis, 53 Beckham, Eugene C, III, article, 137 Branson, Branley A., article, 99 Bullatifrom, 2-3,6-9, 14, 17, 19-20, 23-24 Cambarus, 57 advena, 58, 60 carolinus, 5 7-58, 60 hage1^iaf^us, 57-58, 60, 69 hagenianus carri, 83 hagenianus evansi, 64, 69 hagenianus forrestae, 69 hagenianus hagenianus, 60 hagenianus X vesticeps, 60 Campbell, Roger D., ^ticle, 99 Carassius auratus, 104 Cephalanthus, 50 Chaenohryttus gulosus, 132 C/ie/ns, 49 fimbriatus, 36 Chelydra, 49 serpentina, 36 C/jry'remj5, 35, 38-39, 43, 45-49 cae/a/c2, 44-45,50,53-54 concinna, 38, 50 floridana, 50 inflata, 45 nelsoni, 40, 50 ornata, 47 platymarginata, 54 script a, 50 williamsi, 45, 54 C/em3'5, 49 Cranioceras, 53 Cyc/ops bicuspidatus, 27 bicuspidatus thomasi, 28, 32 crassicaudis brachycercus, 28, 32-33 varicans rubellus, 28, 32 vemalis, 28, 32 Cyprinodon variegatus, 137-148 Cyprinus carpio, 114 Deirochelys, 35-39, 41^4, 46-50, 5 5 carri, 37, 43-48 floridana^ 35 reticularia, 35-43, 45-50, 5 5 reticularia chrysea, 49 reticularia reticularia, 49 Diapfomui bogalusensis, 27, 30 bogalusensis marii, 29-31 dorsalis, 28-29 m ississipp ie nsis, 2 7-3 0 pallidus, 27-29, 33-34 reighardi, 28-29,33 sanguinensis, 28 silicoides, 28-29 sinuatus, 30 sp., 30-31 virginiensis, 27-29 Emydoidea, 35, 42, 49 blandingi, 36 , 49 Emys, 49 Epischura fluviatilis, 27 £Mcyc/op5 a^i/is, 28, 32-34 agilis montanus, 28, 32-33 prionophorus, 28, 32 speratus, 28, 32-33 EwrjcercMX, 1-2, 8-9, 14, 17, 19-20, 22, 24 glacialis, 5, 14, 16, 19, 20-21, 23 lamellatus, 1-2, 5, 9, 11-14, 16 macracanthus, 9, 14-17, 21-23 microdontus, 1-25 pompholygodes, 5, 9, 11-14, 16-17, 21 Ewr^^fewora af finis, 27-28, 33 F alii cambarus, 58 macneesei, 59 Fitzpatrick, J.F., Jr., article, 57 Frey, David G., article, 1 Geochelone, 42 alleni, 54 tedwhitei, 53 Girardiella, 5 7-58,93-94 barbiger, 65,67-68, 76-77 come to, 65, 67-68, 74-76, 78-79 connus, 65, 67-68, 76-83 gracilis, 93 hagenianus, 60, 64, 69, 74, 83 hageniatms carri, 83 hagenianus evansi, 64 hagenianus forrestae, 69 hagenianus hagenianus, 60-65, 67 hagenianus vesticeps, 64-65, 67-68, 70-71 hagenianus hagenianus X vesticeps, 65, 67 pogiim, 65, 67-68, 83-87 sp.B,69 sp.D,83 sp.E, 74 subsp. A, 64 tulanei, 93 Graptemys, 42 Harris, Martin J., article, 27 Hipparion plicatile, 53 Hirochelys, 37 reticulata, 38 Hyalella, 101, 115 Hypothalmichthys molitrix, 129 Ictalurus melas melas, 99-133 natalis, 104 nebulosus, 102-103 punctatus, 100-101 Jackson, Dale R., article, 35 Lemna, 50 Lepisosteus, 50 Lepomis macrochirus, 114 Macroclemys auffenbergi, 54 Macrocyclops albidus, 2 7-28,33 ater, 28, 33 fuscus, 28, 32 edax, 27-28, 32 Mesocyclops leuckarti, 27 tenuis, 28, 32-33 Mettee, Maurice F., Jr., article, 137 Micropterus dolomieui, 105, 132 salmoides, 114 Nannippus ingenuus, 53 No tr opts sp., 114 Nymphozanthiis advena, 101 Oedogonium, 106 Ortmannicus hagenicmus, 60 Osphranticum labronectum, 28 Osteoborus, 53 fimbriatus, 28, 32-33 fimbriatus poppei, 28,32-33 Perca flavescens, 105 Pimep/ja/e5 notatus, 114 promelas, 114-1 15 Pleuroxus denticulatus, 24 Potamogeton, 101, 106 Procambarus, 57-59 barbiger, 61, 65, 67-74, 76-77, 87-96 cometes, 65, 67-68, 74-76, 78-79, 87-96 connus, 65, 67-68, 76-83, 87-96 fitzpatricki, 59 gracilis, 93 Gracilis Group, 93 hagenianus, 58-61, 64, 69, 74, 76, 83, 91 Hagenianus Group, 58-59, 61-62, 65,92-95 hagenianus forrestae, 73 hagenianus hagenianus, 60-65, 67, 69, 73, 82, 87-96 hagenianus vesticeps, 64-71, 83, 87-96 hagenianus hagenianus X vesticeps, 65-67 pogum, 65, 67-68, 83-92, 94-96 sp.B, 69 sp. C, 76 sp. D, 83 sp. E, 74 subsp. A, 64 tulanei, 59, 93 Salvelinus sp., 104 Spirogyra, 106 Synthetoceras, 53 Terrapene, 42-43, 49 Carolina putnami, 50 Testudo reticularia, 37 Tilapiaheudiloti, 128 Trachemys jarmani, 35 Trionyx, 42, 50, 55 ferox, 50 TropocycZops prasinus, 28, 32-34 prasinus mexicanus, 28, 32-34 Wolffiella, 50 Zygnema, 106 CONTENTS OF VOLUME 20 NUMBER PAGE 1-2. A NEW SPECIES OF EURYCERCUS (CLADOCERA, CHYDORIDAE) FROM THE SOUTHERN UNITED STATES David G. Frey 1 COPEPODA OF NORTHERN MISSISSIPPI WITH A DESCRIPTION OF A NEW SUBSPECIES Martin J. Harris 27 EVOLUTION AND FOSSIL RECORD OF THE CHICKEN TURTLE DEIRO- CHELYS, WITH A RE-EVALUATION OF THE GENUS Dale R. Jackson 35 3-4. SYSTEMATICS OF THE CRAWFISHES OF THE HAGENIANUS GROUP OF THE GENUS PROCAMBARUS, SUBGENUS GIRARDIELLA (DECAPODA, CAMBARIDAE) J. F. Fitzpatrick, Jr. 57 ECOLOGY AND POPULATION DYNAMICS OF THE BLACK BULLHEAD, ICTALURUS MELAS (RAFINESQUE), IN CENTRAL KENTUCKY. Roger D. Campbell and Branley A. Branson 99 NOTES ON THE BREEDING BEHAVIOR, EMBRYOLOGY AND LARVAL DEVELOPMENT OF CYPRINODON VARIEGATUS LACEPEDE IN AQUARIA. Maurice F. Mettee, Jr. and Eugene C. Beckham, III 137 TUl ®®IL®(g^ ^S!5ID ....1(1).,^ ,^P^ uanAMY JAN30t978 Volume 20 Nos. 1-2 $3.00 MAf*VAl4l9uary 9, 1978 ONt^^ — "^tTY A NEW SPECIES OF EURYCERCUS (CLADOCERA, CHYDORIDAE) FROM THE SOUTHERN UNITED STATES DAVID G. FREY p. 1 COPEPODA OF NORTHERN MISSISSIPPI WITH A DESCRIPTION OF A NEW SUBSPECIES MARTIN J. HARRIS p. 27 EVOLUTION AND FOSSIL RECORD OF THE CHICKEN TURTLE DEIROCHELYS, WITH A RE-EVALUATION OF THE GENUS DALE R.JACKSON p. 35 TULANE UNIVERSITY NEW ORLEANS TULANE STUDIES IN ZOOLOGY AND BOTANY, a publication of the Biology Department of Tulane University, is devoted primarily to the biology of the waters and adjacent land areas of the Gulf of Mexico and the Caribbean Sea, but manuscripts on organisms outside this geographic area will be considered. Each number is issued separately and contains an individual monographic study, or several minor studies. As volumes are completed, usually on an annual basis, title pages and tables of contents are distributed to recipients receiving the entire series. Manuscripts submitted for publication are evaluated by the editor or associate editor and by an editorial committee selected for each paper. Contributors need not be members of the Tulane University faculty. When citing this series authors are requested to use the following abbreviations: Tulane Stud. Zool. and Bot. INFORMATION FOR AUTHORS: The editors of Tulane Studies in Zoology and Botany recommend conformance with the principles stated in CBE Style Manual, 3rd ed., published in 1972 by the American Institute of Biological Sciences, Washington, D.C. Manuscripts should be submitted on good paper, as original typewritten copy, double-spaced, and carefully corrected. Two copies, carbon or other suitable reproduction, must accompany the original to expedite editing and assure more rapid publication. Legends for figures should be prepared on a separate page. Illustrations should be proportioned for one or two column width reproductions and should allow for insertion of legend if occupying a whole page. Photographs should be on glossy paper. Many tables, if carefully prepared with a carbon ribbon and electric typewriter, can be photographically reproduced, thus helping to reduce publication costs. Letter- ing in any illustrative or tabular material should be of such a size that it will be no less than 1 14 mm high when reduced for publication. An abstract not exceeding three percent of the length of the original article must accompany each manuscript submitted. This will be transmitted to Biological Abstracts and any other abstracting journal specified by the writer. Authors of contributions will receive a Statement of Page Charges, calculated at $45/page. Partial or complete payment of these charges is solicited from authors who have funds available for this purpose through their institutions or grants. Acceptance of papers is not dependent on ability to underwrite costs. Illustrations and tabular matter in excess of 20 percent of the total number of pages may be charged to the author; this charge is subject to negotiation. EXCHANGES, SUBSCRIPTIONS, ORDERS FOR INDIVIDUAL COPIES: Exchanges are invited from institutions publishing comparable series but subscriptions are available if no exchange agreement can be effected. A price list of back issues is available on request. Remittance, preferably money order, should accompany orders from individuals. Make remittances payable to "Tulane University." Authors may obtain separates of their articles at cost. Copies of Tulane Studies in Zoology and Botany sent to regular recipients, if lost in the mails, will be replaced if the editorial offices are notified before the second subsequent issue is released. COMMUNICATIONS: Address all queries and orders to: Editor, Tulane Studies in Zoology and Botany, Department of Biology, Tulane University, New Orleans, Louisiana 70118. U.S.A. Harold A. Dundee, Editor Arthur L. Welden, Associate Editor David C. Heins, Assistant to the Editors TULANE STUDIES IN ZOOLOGY AND BOTANY Vol.20 Nos. 1-2 $3.00 January 9, 1978 A NEW SPECIES OF EURYCERCUS (CLADOCERA, CHYDORIDAE) FROM THE SOUTHERN UNITED STATESl DAVID G. FREY DEPARTMENT OF ZOOLOGY, INDIANA UNIVERSITY, BLOOMINGTON, INDIANA 47401 ABSTRACT Two species of Eurycercus occur in the south- ern United States, of which E. microdontus in the subgenus Enrycercus is described here. Its out- standing characteristic is that the large number of minute teeth on the postabdomen exhibits a sig- moid relationship to size of animal, rather than linear as in the other known species. Moreover, although the two major populations from North Carolina and Florida agree very closely in most morphological details, they differ significantly in this regression and in the body size at which repro- duction begins. The meaning of these differences in terms of evolution within the genus is not yet understood. Other characters that are size-depend- ent, particularly details of setation and spinulation on trunk limbs I and II, are described by regres- sions and compared with the other species. Rela- tively little can be deduced concerning the biology of tlie species because of the small size samples available. INTRODUCTION Two species of the highly distinctive genus Eurycercus occur in the south and southeastern region of the United States and south at least to Mexico City. Neither is the species E. lamellatus, which is the only one presently reported as occurring in the United States, excluding Alaska (Brooks 1959). Until recently (Fray 1971) E. lamellatus was considered to be widespread over North America and Eurasia, with disjunct popula- tions in Argentina and Soutii Africa. Now (Frey 1973, 1974) this taxon is known to be made up of a number of species, each with a much more restricted distribution. E. lamel- latus sensu strictu is the dominant, and probably in most water bodies the only, species in Western Europe and for an un- known distance eastward through the north- ern half of Asia. Evidence to date suggests it does not occur in North America at all. The most outstanding characteristics of the nominate subgenus to which E. lamel- latus sensu strictu belongs are a sharp keel on the shell in all instars and a pronounced transverse fold or notch immediately behind the median head pore (Frey 1974). More- over, the keel actually begins on the head shield behind the median pore, which results in a triangular posterior extension of the head shield where it contacts the keel on the shell. The original paper on head shields and head pores of the chydorid Cladocera (Frey 1959) contains a sketch o( a Eurycerus head shield from White Lake, N. C, having the same triangular extension posteriorly as in the European taxon and hence at that time judged to be the same. This was an unfortu- nate coincidence that may have delayed the recognition of the pluraUty of Eurycercus for some years, because from almost any other region of North America, particularly from the more northern lake districts in the 1 Contribution No. 982, Department of Zoology, Indiana University. Editorial committee for this paper: CLYDE E. GOULDEN, Chairman, Department of Limnology, Academy of Natural Sciences, Philadelphia, Pennsylvania 19103 DR. ALFRED E. SMALLEY, Professor of Biology, Tulane University, New Orleans, Louisiana 70118 Tulane Studies in Zoology and Botany Vol. 20 United States and Canada where Eurycercus is a common component of the littoral fauna, the head shields are rounded behind. Close examination of intact specimens of these populations would have revealed an unkeeled shell and the head pore located on a rounded protuberance — features that are characteristic of the subgenus Bullatifrons (Frey 1974) to which the second species in the South and Southeast belongs. Studies have not yet been made to determine if this latter species is the same as the widespread Bullatifrons sp. to the north, although super- ficially they seem different. Resolution of this matter will have to be delayed until the description of the northern taxon has been completed. Except for matters of distribu- tion, the present paper is concerned only with Eurycercus microdontus sp. nov. DISTRIBUTION The region under consideration comprises the 11 states from North Carolina west to Oklahoma and south to Florida and the Gulf of Mexico plus Mexico. Here Eurycercus, although obviously widespread, is very sporadic in occurrence and seldom abun- dant, except perhaps in winter to late spring. The subgenus can be determined only from actual specimens. None of the literature records, including the lengthy description of Chien (1969), provides any basis for decid- ing the subgenus. Specimens derive chiefly from my own collections in North Carolina, Tennessee, Florida, Mississippi and Louisiana and from the sediment samples collected by DeCosta (1964). Scattered specimens were discovered in the U. S. National Museum and the E. A. Birge collection. Collections made in various states by R. V. Harmsworth, W. M. Lewis, Jr., P. R. Becker, Dewey Bunting, D. L. Dycus, D. C. Wade, and F. N. Young yielded occasional specimens or exuvial frag- ments. All known occurrences of the genus in this region are summarized in Tables 1 and 2 and Figure 1. DeCosta (1964) found the remains of Eurycercus in the sediments of 7 out of 22 lakes sampled from Reelfoot Lake, Tennes- see, to New Orleans, generally in very low abundance. Chien (1969) recovered furycer- cus from only 3 samples out of 80 collected along the entire length of the Pearl River in Mississippi and always in his lowest category of abundance. The only record from Texas is one intact specimen from Smithville in the E.A. Birge Collection. Intensive faunistic effort, documented by a number of master's theses from several universities in the State, have not disclosed any additional records (Becker and Sissom 1971). For Eurycercus microdontus described in this paper (Table 1), the original material consisted of head shields and postabdomens from Pages Lake and White Lake in North Carolina, plus a few intact specimens and exuvial fragments from four lakes in High- lands Co., Florida, near the southern tip of the interior lake district. The Florida lakes were revisited in early December 1972 for the express purpose of obtaining more speci- mens and hopefully gamogenetic individuals. Intensive sampling yielded a total of only 70 specimens from Lake June in Winter, a few specimens each from two of the other lakes, and none at all from the fourth. No gamo- genetic individuals were collected. More than 30 samples in the D. G. Frey Collection from various lakes in North Carolina, South Carolina, Georgia, Florida, Louisiana, Texas, and Mexico were searched for Eurycercus with completely negative results, except for an offshore sample of benthic algae from White Lake, N. C, which yielded the second largest collection of this taxon. The fairly extensive collection of D. L. Dycus from Georgia was not used in the morphological analysis because all specimens are badly dis- torted from preservation. Other records are based on scattered specimens and exuviae. Fortunately, the raw sediment samples from all the lakes studied by DeCosta (1964) and from a few additional ones not used in his study have been stored in glass jars in a 5° room at Indiana University. Samples, from his seven positive lakes in this region were processed, plus Sardis Reservoir, Miss., and Lake Charles, La., known to be positive on other evidence, plus three negative lakes that were significant to reexamine for one reason or another, and also Johnson's Lake, Tenn. Processing consisted of de flocculating 30-50 Nos. 1-2 New Eurycercus cc of raw sediment in 10% NaOH on a mag- netic stirrer-hotplate (temperature kept below boiling) for about 1 1/2 hours. The sediment was then screened through a Clarke-Bumpus bucket provided with a No. 18 metal screen, which is fine enough to retain headshields and postabdomens of even the first-instar individuals of the two species of Eurycercus. Fractions of the residue were scanned systematically in a petri dish at a magnification of 15 X. Remains oi Eurycer- cus were picked out as encountered and sub- sequently mounted in polyvinyl-lactophenol stained with lignin pink for study at higher magnifications. The results are given in Table 3. Remains of Eurycercus were recovered from all of DeCosta's positive lakes, except Horseshoe Lake, La., from which he had recovered only a single fragment. Remains were also recovered from Sardis Reservoir and Lake Charles. All head shields, except three from Bennett's Bay in Lake Cocodrie, La., were from the subgenus Bullatifrons. The single Bullatifrons head shield from Lake Charles is interesting, because the specimen from this locality in the E. A. Birge collection is E. microdontus. Accord- ingly, another 50 cc of sediment was proces- sed without disclosing any more remains of either species, demonstrating how very tenu- ous even some of the positive records can be. No remains of the genus were recovered from Alligator Lake and Lake Concordia in Louisiana, from the lagoon in Audubon Park, New Orleans, or from Johnson's Lake, Tenn. A trip to some of the southern states was made in May 1974. Bullatifrons sp. was found at two of Chien's (1969) three Eurycercus stations in the Pearl River System, Miss., at Reelfoot Lake, Tenn., and at Courtebleau Bayou, La. (see Table 2). In addition, collections from various TVA reservoirs were examined at Muscle Shoals, Ala. No new populations of E. microdontus were discovered. Aside from Lake Charles, which may be doubtful because of the 60 years between the records for the two species, both species Southern PInas Whita LaK* Pagaa LaKa Par Pond Swampllka Pond Laka Juna in WIntar Laha Jackaon Laka Annia Buck Laka Smithvilla KiLOMerens Figure 1. Known distribution by counties or parishes of the two species oi Eurycercus in the southern and southeastern part of the United States. In addition there is a record from Mexico City (Juday 1915). Localities from which £. microdontus are known are indicated by a white dot and specifically labelled as to lake or locality. The other records are either for the species in the subgenus Bullatifrons or those for which the specific identification is uncertain (cf. Table 2). Most of the records from Tennessee and northern Alabama are from various T.V. A. reservoirs. Tuhne Studies in Zoology and Botany Vol. 20 u to c CLi -o en (U 'G a en C <^ s c o O )^ u s o S O 3 a CO (J C 3 u u o c o c a! H C it u X) 3 z -a "o 2 ^ c/5 f^ + rt .5 u u C/5 II 0 U + + + 3 ^ 00 00 IT) lD > > 00 00 D3 \0 00 O O Q Q > > (N --I X 2 r^ O rt- -T3 03 C "o i-i U O o 2 .-I rj O 00 *-» ^ 3 O u OJ 3 OJ C T3 C o a. ^*- « O o. oi "o J 00 a! u. rt 00 o a. 3 3 Q Q J J Q Q > > o o o H =a 3 3 rt rt 00 (^ r-< ON O ^ Ol 01 U .2 rs ^ 5c 73. 3 o o O >< ON rn 00 o ^ o ^ ON .-I (M O O Q Q > X C w w T! c c < a; 3 oJ J 3 pa oa < < W PJ 9 --^ o ^ O x£, >< >< ^ r<1 (NJ ON c £ 3 E .2i S tJ OJ ^ 00 U »i ur, m U o ON o _ U OJ o c 00 J CQ rt C A •IFt to Vi • •-t rt 3 X o 00 3 D ^ JLQ S^J S Q c/l • -. OJ I ^ „ H c^ >- C 3 c « S2 Z 2 ■^^ E^ 3 V OJ C {/3 l- D n! S^ >. ■^ ^ C o O ^ ■n CTJ rt J z ^ aj II bO O! X a. 5 E ^ 2 c c Z ^ ■' (U c c s 3 o o JD <-• ^ rt t30 *-» 'si, t >- o ^-j o .^ C .S -' OJ ;-■ tn u O >^'0 — ^ C O -T3 « 0) + « o tA ' OJ tj c s ^ O o rt 'C <-> J i^ . ^ .2 D 1L> oa ~ "O c ■? o ^ bO u< §.s CL llecti Wash 11 o „ rt u .£ '5c ^? i-H o £ s O E o Q _ rt VI II c D p- .2 D. o « E Q Z «5 Nos. 1-2 New Eurycercus occur together in Bennett's Bay and in Par Pond, S. C. In Par Pond only Bullatifrons was present — abundantly so — on 5 May 1973 and again on 30 January 1974 in much smaller numbers, and only E. microdontus (in very low abundance) in November 1973, suggesting a seasonal replacement of one species by the other. In other parts of the world where the ranges of two species of Eurycercus overlap, although any given body of water generally has only one of the species, both can occur in the same water body and at the same time. This is true, for example, of E. lamellatus sensu strictu and E. glacialis in Bislet Dam, Denmark (Kaiser 1959) and in Iceland (D. G. Fray observa- tion in 1972) and of E. lamellatus and E. pompholygodes in several lakes in northern Sweden (Frey 1974) All two-species popula- tions known to date involve species from different subgenera. At least in the lakes in Iceland and northern Sweden, both species of the pairs were present and reproductive at the same time, demonstrating that temporal displacement is not a requirement for oc- cupying the same water body. METHODS Personal samples collected prior to 1965 were obtained with a short conical plankton net of 70 /im Nitex 15 cm in diameter mounted on a rod about 1 meter long. Sub- sequent samples were collected with a 110 jUm mesh Nitex net of about the same size, which could be screwed into a sectional metal rod 3 meters long. The mouth of the latter net was protected by a hinged metal screen with meshes 5 mm bar measure to keep out leaves, macrophytes, and other coarse material. The entire contents were preserved in rough tly 5% commercial forma- lin. In the laboratory subsamples were spread in a petri dish, systematically scanned at 15 X stereo magnification, and all specimens, exuviae, and exuvial fragments were re- moved as encountered, all of which were gradually transferred to glycerol for facili- tating measuring and dissecting. Because of the small size of this species, all specimens were measured in temporary uncovered glycerol mounts with a Wild (M20) com- pound microscope using lOX eyepieces and either 3X, lOX, or 20X objectives, depend- ing on the size of the specimen. By this pro- cedure length of the specimen and length of the postabdomen and postabdominal claws could be measured conveniently, and the number of denticles on the postabdomen and of spinules on the postabdominal claw could be counted except in the smallest specimens. Details of the smaller exuvial components were studied in permanent mounts with 40X or lOOX objectives. In the previous studies (Frey 1973, 1974) the regressions of denticle counts on trunk limbs I and II as a function of length of the toothed zone of the postabdomen, which in turn is a function of overall size of the animal, were determined from complete exuviae recovered from the collections. In the present study virtually no intact exuviae were available, and hence these regressions had to be based on specimens sacrificed for dissection. Twelve specimens each from Lake June in Winter and White Lake, evenly distributed over the size ranges of the popu- lations, except for the largest specimens which were too valuable for sacrifice, were used for this purpose. Dissections were made in glycerol at a magnification of 45X with a stereoscope, using tungsten needles. After removal of the head with its attached appendages and then the shell and post- abdomen, the trunk limbs were separated longitudinally into right and left groups. In the smallest specimens only trunk limbs I and II were separated from their respective groups, whereas in larger specimens all trunk limbs were separated from one another. These were then mounted in polyvinyl lacto- phenol and eventually sealed wath glyceel (available from ESBE, Toronto) for conveni- ence in studying at whatever magnification was necessary, up to oil immersion. All measurements and counts were punched on IBM cards for calculation of re- gression statistics and correlation coeffici- ents, using the SPSS programs (Table 4). Although these simple techniques are not entirely appropriate because of the nature of the variables, they do identify which char- Tulane Studies in Zoology and Botany Vol. 20 a. o c 1) d -a s ™ 2 C OQ JC ^ O J3 CO o ^ '^ c "« — 3 s-^ 3 -6 <« O §^ ^ 2 :3 2 3 a. C J3 4) 3 « -£ u -c c *^ o <*< •« o % c E Si X ^ u 3 I. u '^ 11 3 C V 3 no O U o 3 « o ■£ a o -3 J 3 3 na j: c c « a. S £ Ai > o a. O 5:> o J c n - CO < 00 o ^> _0 "o 11 -^ o -o ^ C ■J Ml 1- H c/:i _o < < ■§ > > on 3 "a 3 CQ CQ cQ *- ^ lA OJ .— '^ a> c — _ ^ ■•= H CQ ~ :3 :3 OQ CQ CQ u • ' n I- fl " 3 OO J5 -L c 13 i'i D.. ^ ^ 3 ^ ^ a. 2 j2 = T3' c ■c 1= rt 4J 5 ° (U -^ a. j: oc c rt ^ n — rt ■f 00 oa a. u o Z a o on C 4> E a. :§ E S5 3 w 00 c 3 := < u. o o >> Eu > o U O >> O T < — ■* O ^ >> « < OS ^ in _ Tj- 4 ^ > E c«: ji O. O. ffl «> flj <1> > > k. nj U a: u: UJ UJ o O 00 c o "3) ™ „ ^ « •^ " -^'' .- -'" A ^ »; >- c s ^ " S 3i u « * n " j: .D 3 Nos. 1-2 New Eurycercus 3 O fV. (V. o.. & 2 2 3 aj OQ D5 V) c ij - C c 60 o o qj ^5 E ■^ u c D. rt IH cJo E > u E 0 c o a. ^ 2 a. c C ^ c 4J "rt I- 0 C E § z 15 E < 1^ ^ U (« n rt ^ > D.04 aJ >J H o E E ^ c 3 OQ < < < < JS \0 in < > o 1 C w i .§ ■« E T) i; (U U QJ T3 T3 TS -T3 -^3 cfl (B rt rt (< 2o -^ ;:! ^ ^ it ^ ;^ l-l< ■— • ■4 > 6 1—1 ■<»• ro c o ■5 O V C E s o u ic O c o c .^ •« o c u !3 o c?) 5 flQ < C o oa z H 2 c c u H Tulane Studies in Zoology and Botany Vol. 20 Table 3. Reexamination of sediment samples from lakes along the lower Mississippi River collected by DeCosta (1964) in 1960 and 1961. All the head shields and shells (and hence presumably the other remains as well) were oi Eurycercus ( Bullatifrons ) sp. except in Bennett's Bay, where 3 of 24 head shields were from E. (Eurycercus) microdontus (see Table 1). DeCosta' s data Remains recovered in present study % total No. chydorid Head Post- No. Lake and State Eu rycercus remains shields 19 abdomens 6 Shells 1 Ephippia 24 Reelfoot Lake, Tenn. 4 1.9 26 Horseshoe Lake, Ark. 4 1.0 46 14 11 36 Horseshoe Lake, La.^ 1 0.5 37 Bennett's Bay, La. 3 1.2 24 4 38 Miller's Lake, La. 25 6.6 47 46 2 39 Chicot Lake, La. 15 1.6 35 26 3 2 41 Cazan's Lake, La. 21 9.7 30 28 1 — Sardis Reservoir, Miss. — 2 1 — Lake Charles, La.l — — 1 lAbout 75 cc of sediment from Horseshoe Lake, La., and 100 cc from Lake Charles were processed. Hence, the taxon at these localities, if present, must be in extremely low abundance. acters are size dependent, and they help to quantify differences between various species and populations. A product-moment correla- tion analysis of one set of data yielded no better insight than a simple regression analysis. In an attempt to locate other specimens and records for this part of the continent, many individuals who had worked in this region were contacted by letter, mostly with negative results. Remains recovered from sediments may be one of the best ways of working out geographic distributions, once the species have been characterized suffici- ently for their remains to be identified to species. Certainly DeCosta's (1964) sediment samples from the region of the Lower Missis- sippi River were invaluable in demonstrating that Eurycercus microdontus apparently has a smaller frequency of occurrence than Bullatifrons sp. Eurycercus (Eurycercus) microdontus new species Type locality. Lake June in Winter, High- lands Co., Florida; approximate geographic coordinates 27.3°N and 81.4»W. The speci- mens were collected along the point of land close to U.S. Highway 27 at the northeast corner of the lake in SE 1/4, S25, T36S, R29E. This is where the species had been collected previously in 1960. At this locality the bottom drops off quite steeply, and the wave action can be strong. The specimens were collected chiefly around the emergent stems of cane at water depths up to 80 cm and up to 4 m offshore. Type material. All 70 specimens collected on 3 December 1972 plus the scattered exuvial fragments recovered. Holotype: A parthenogenetic female 1.67 mm long in alcohol has been deposited in the National Museum of Natural History, Washington, D. C, their catalog number USNM 1.51210. Paratypes: 1. National Museum of Natural History: two reproductive females in alcohol and one mounted in glycerine jelly, catalog number USNM 151210. 2. British Museum (Nat. Hist.) London: three reproductive females, two in alcohol Nos. 1-2 New Eurycercus (registration numbers 1974. 717-718) and one mounted in glycerine jelly (registration number 1974. 716). 3. All other specimens and exuvial frag- ments are in the D. G. Frey Collection in Bloomington. Brief diagnosis. Small, seldom as large as 2 mm. Shell strongly and sharply keeled in all instars. Seen from the side, dorsal margin of head flattened or even slightly concave between compound eye and median pore. Rostrum short, not extending much beyond base of antennules. Many minute teeth on postabdomen, increasing to 140 or more in largest specimens. Anal groove broad and shallow. Middle seta on inner distal lobe of trunk limb I well developed, rather strongly and evenly curved, heavily chitinized. Etymology. From Gr. micro small, and Gr. odons, odontos tooth. Named for the many extremely minute teeth on the post- abdomen. General comments: E. microdontus is the second described species in the subgenus Eurycercus, E. lamellatus sensu strictu being the first. These two species are readily sepa- rated from the other known species of the genus by the sharp dorsal keel in partheno- genetic females and males beginning immedi- ately behind the median head pore and by the sharp notch behind the pore which inter- rupts the dorsal contour. Ephippial females of the subgenus Bullatifrons also have a keeled shell and hence might be confused with parthenogentic females of E. micro- dontus, but the keel does not involve the head. As a result the head shields even of these keeled ephippial females are rounded behind, rather than pointed as in the sub- genus Eurycercus. Moreover, the keel of ephippial females is rounded, rather than sharp as in parthenogenetic females of the subgenus Eurycercus, as illustrated for E. pompholygodes in Frey (1974). Although the populations from White Lake and Lake June in Winter are indistin- guishable in most respects, they do differ sig- nificantly in number and shape of teeth on the postabdomen, in length of the postabdo- men relative to body length, and in size at which reproduction begins. These differ- ences and others will be pointed out where appropriate. At least for the present, the two populations are considered to belong to the same species. Size. The sample from Lake June in Winter has a size range of 0.58-1.67 mm, with the smallest reproductive female (em- bryos in brood pouch) measuring 1.41 mm^ The specimens from White Lake range in size from 0.58 to 1.61 mm, with individuals as small as 1.19 mm having embryos in the brood pouch. One isolated postabdomen 701 nm long from the sediments of Pages Lake corresponds to a body length of 2.02 mm, according to regression 1 in Table 4. The four specimens from Southern Pines, N. C, measured 1.75-2.23 mm, all larger than the largest specimens from Lake June in Winter and White Lake. Of the other miscel- laneous specimens available, the largest was a parthenogenetic female about 1.7 mm long from Slidell, La. Hence, the species is small, seldom attaining a length as great as 2 mm. Shape (Figs. 2-4). The shape of £. micro- dontus changes during ontogeny. The early pre-reproductive instars have the shell only weakly arched dorsally and the head relative- ly high (Fig. 4), resulting in an overall appearance rather similar to E. macracanthus although more elongate. Beginning immedi- ately prior to maturity, the back becomes progressively more highly arched and more sharply keeled (Fig. 2), both characters being considerably more extreme than in E. lamellatus. The high arch to the shell can be regarded as one mechanism for increasing the brood pouch capacity and hence the re- productive potential of this small species. In lateral contour the head is flattened or even slightly concave between the com- pound eye and the median pore (Figs. 2, 5). Because the rostrum is short, the compound eye is located closer to the tip of the ros- trum than in E. lamellatus, more resembling E. macracanthus in this respect. The pos- terior margin of the shell slopes posteriorly resulting in the dorsal angle being located considerably farther forward than the ven- tral angle (Figs. 2, 4). This condition also tends to occur in E. lamellatus and E. pompholygodes, although not so extreme. 10 Tulane Studies in Zoology and Botany Vol. 20 M h '^ o <^ i- 4j <-' -C 4-. -T3 z - = 1 3 O o i-t (N -^^ Ui O c p :3 V3 Uh c C 3 r 3 •— > "a. O 5 "J - on 3 ^^-^ I— ' .13 O *-' - CL,— U £ c« — !> U 4-1 ^ vi3 to .2 O S O ^. o s> crt ;s o o V S ^ ^ u O O U " >- = a u u a. o s ^ ^ s • ^ ° ? a! u JO (1> OS E c o r; V o o 4-> u c/) a> c J= 4-1 a) o 4-' hn « 4-> 2i ON <->-i -rf c cx, o T— 1 Cl, -> 4-» O c c a: 0^0^00^0^000^0^0^0^00 O^OO OOOO OnO^ ldO r^OO inOO 0000 00[~~ uD o m i/^ OS tN OS >-< 00 Ov OS 00 OS SO Tl- oor-^Of^^osr^-m^i" ■to \oos t^ orn iTiTj- inr^ MLn (sjsoossor^'-Hsooomcsi'^ i^co Ti-o rno f^ osin in-* -^m so-^ ro ■^ r~- r^ "—I O"— I (sjtN OO .— (i— i OS o OS Tl- OS in O IT) so 00 ^ o so OS o o sD t^ sD O o en f— < 1 1 0^ \0 OS T— t r~- Lo r^ un Csl en Ol 1—i 1 1 1 1 00 so O 00 LO r^ o LO en so (N so en en en en t en 00 in en OS r-< .—1 00 OS ■^ so 00 en in CS) 00 t-H en o ^ SO 00 00 sO ^ ^ Cs^) in en Cs) so O rn Os 00 in en r-l o OS 00 t— 1 00 00 T— 1 00 00 so o T— I Cs) CM CS) 00 SO 1—1 so 00 so t— I o OS 1—1 T— I r- t^ 1—1 CS) 1— 1 so in CM 1—1 O 00 C^) en i-( en so 00 o 1—1 o en 1—1 C nl en so 00 OS OS OS (M o 00 en OS o in en 00 rn 00 en o en en OS rsi in en 00 en en en in 00 en o en (N en en en SO O 1— ( en en in en en en o en OS in 00 in u C o so o so o in o sO o SO in 'I- in o so so in un o so so in in in so in in in u C c u -o c u a u ■a c c Hi -a -~, a u Q 00 sOCsl i^Os CMsO CMsO OSi— ( ■^t^ OsCM ■^r^ mm ^^00 00'^ ■^ inoo^^i— 100 cn'^cMTf ^H^M 1— ii— losso i— im m'^ en^n Oi— icMinmeno CS) -I- 00 en C^ 00 CS) "^ 1-1 O 1— I rs) en •— I OS OS in o 1—1 "^ 00 OS 1—1 OS en en 00 o so OS o CM CM OS Cs) CM ^^ ^-^ cs)-^ cn^ en^ rtcs) ^ Cs) c^) en en en mcM oom 00'^ oo'^ >— 't^ o^so mm mo mso t^so ^^""J <^ 1-iod mo CMsO csjsD oo"oo" odos osr-i sor-^ som som t~^os i— i CTso oen rr^i ^r\ cnen r- 1-1 1-^ 1-1 CM in o ^ so in so m O o ^ Csl CM 00 00 m so o in i^t- CM en ^ 1— 1 1— I 00 1— ' o o OS -^ in m r~~ so m 00 -H Tj- in t^ OS -^ 1— 1 1— 1 T— ( m T^ m en so <-< so 00 r-- 00 so t^ CS) CM OS -rf en ^ CM r~ CM CM so 00 CM O CM CM 1-1 1-1 m ^H in r~~ in ■* m en in so en rs] in so m CM 1—1 rs) (N CM ^ 00 (N CM en rU en en CM CS] m »— ' CM en O — I -^ 1— 1 ^H OS 00 1^ •4 -4 in -4 ■* ^ m •-< c^) en r-~ OS 1—1 1—1 CM r^ CS) CM en CS] en r^ m CM (^) in r^ T}- o cs] (N 1-1 1-1 r^ ^ -> 5 c bC c en O G. bO C nl rt 4-J 4-> o o Xi n Vi O cx .= -C J= bC bC c c bO C O bO c 4-) o ^ rt o bC c bO C -t3 o bC C 4-> o bD C -T3 O -a M 01 4>> o a. -C -C so c bO c -o o a. c bo C nl 4-t o a. bc c u bO C pv4 ^" « <-' -^ -s C o Ji a. _i; 1) 5 c C c CS) C/5 en C/5 C/3 c ■■J -^ 00 00 3 -5 3 c o^ ^ rt lU 1— 'cercM5 (Chydor- idae, Cladocera) with a description of Eury- cercus macracanthus sp. nov. /nf. Rev. ges. Hvdrobiol. 68:221-267. 1975. Subgeneric differentiation within Eurycercus (Cladocera, Chydoridae) and a new species from northern Sweden. Hydrobiologia 46:263-300. FRYER, G. 1963. The functional morphology and feeding mechanism of the chydorid cladoceran Eurycercus lamellatus (O. F. Miiller). Trans. Roy. Soc. Edi7jburgli 66:335-381. HANEBRINK, E. L. 1965. Summer net plankton in Sardis Reservoir, Mississippi. Southwest Nat. 10:261-277. HOFF, C. C. 1943. The Cladocera and Ostracoda of Reelfoot Lake. J. Tennessee Acad. Set. 18:49-107. JONES, W. H. 1958. Cladocera of Oklahoma. Trans. Amer. Micros. Soc. 77:243-257. JUDAY, C. 1915. Limnological studies on some lakes in Central America. Trans. Wisconsin Acad. Sci. Arts Lett. 18:214-250. KAISER, E. W. 1959. Biologiske og jf^kologiske unders^gelser over dafnierne Eurycercus glaci- alis Lilljeborg og Eu. lamellatus (O. F. Muller). Flora og Fauyia 65:17-34. PENN, G. H. 1947. Branchiopoda and Copepoda of the New Orleans area as recorded by Ed. Foster in the early 1900's. Proc. Louisiana Acad. Sci. 10:189-193. SHAN, R. KUO-CHENG. 1974. Reproduction in laboratory stocks of Pleuroxus (Chydoridae, Cladocera) under the influence of photoperiod and light intensity. Int. Rev. ges. Hydrobiol. 59:643-666. TURNER, C. H. 1910. Ecological notes on the Cladocera and Copepoda of Augusta, Georgia, with descriptions of new or little known species. Trans. Acad. Sci. St. Louis 19:151:176. 26 Tulane Studies in Zoology and Botany Vol. 20 COPEPODA OF NORTHERN MISSISSIPPI WITH A DESCRIPTION OF A NEW SUBSPECIES MARTIN J. HARRIS DEPARTMENT OF ZOOLOGY, WASHINGTON STATE UNIVERSITY PULLMAN , WASHINGTON 99163 ABSTRACT Variable types of Mississippi surface waters north of 33° north latitude were examined qualita- tively for the occurrence of calanoid and cyclopid copepods. A total of 283 collections was made from 59 sites from August 1972 through July 1973. Samples were taken once each season. Twenty-five species in four families, including three which were represented by two subspecies, were taken in the study area. A new subspecies of Diaptomus bogalusensis is described. Diaptomus pallidus and Tropocydops prasinus and its sub- species T. p. mexicanus were the most numerous and widespread calanoid and cyclopoid copepods taken. Seventeen species are new records for Missis- sippi. Despite their importance, the free-living, freshwater copepods have been given fela tively little attention in the United States. In Mississippi, in particular, there is a lack of systematic work on this group. Marsh (1907, 1912, 1929) noted the presence in Mississip- pi of Macrocyclops albidus, Eurytemora affinis, Diaptomus mississippiensis and D. pallidus. Eddy & Simer (1928) also recorded Cyclops bicuspidatus and Mesocyclops leuckarti from Mississippi. Wilson & Yeat- man (1959) specifically mention the pres- ence in Mississippi of D. virginiensis, and Grantham (1958) noted the presence of Mesocyclops edax and D. mississippiensis. One other rare calanoid, Epischura fluvi- atilis, was recently taken from Pickwick Reservoir on the Mississippi-Alabama border by D. L. Bunting (personal communication). The current investigation surveys and up- dates the knowledge of the cyclopoid and calanoid copepods in surface waters of Mississippi north of 33° north latitude. MATERIALS AND METHODS Samples were taken with a 30.5 cm plankton net using number 25 silk bolting cloth. Collections were made at 59 sites once per season for a total of 283 collections (some sites collected in several locations). Samples were taken from 13 lakes, nine ponds, nine streams, 10 rivers, 10 ditches and other intermittent waters, and nine swamps, sloughs or small oxbow lakes. Six Mississippi river drainages were sampled: Tombigbee, Big Black, Pearl, Tennessee, Yazoo and Upper Mississippi including the Horn, Hatchie, and Wolf systems. The study area covered approximately 53,000 km^ and 40 counties. Samples were taken from the surface to a depth of approximately 2.0 m or less depending upon the depth of the water, except that in the five large reservoirs, samples were taken to depths of approxi- mately 5.0 m. Specimens were killed in 5% formaUn and preserved in 70% ethanol. Dissected and whole specimens were mount- ed in glycerin jelly. Species identification and terminology follows that of Wilson and Yeatman (1959). At each site dissolved oxygen content, temperature and pH measurements were taken using a YSI oxygen meter, model 51A and Coleman Metrion IV pH meter. EDITORIAL COMMITTEE FOR THIS PAPER: DR. WALTER G. MOORE, Professor of Biology, Loyola University, New Orleans, Louisiana 70118 DR. ALFRED E. SMALLEY, Professor of Biology, Tulane University, New Orleans, Louisiana 70118 27 28 Tulane Studies in Zoology and Botany Vol. 20 RESULTS Twenty-five species, three of which were represented by two subspecies, were taken in the study area. One undescribed species and a new subspecies, Diaptomus bogalusensis marii, were included in the collections. The order Calanoida was represented by three families: Centropagidae (one genus with one species), Temoridae (one genus with one species) and Diaptomidae (one genus with nine species) The order Cyclopoida was represenred by only one family, the Cyclopi- dae, with six genera and 14 species. Seven species, including Diaptomus sanguineus, D. virginiensis, Osphranticum lahronectum, Eurytemora affinis, Cyclops bicuspidatus thomasi, C. crassicaudis brachy- cercus and Macrocyclops ater were taken only during the winter and/or early spring sampling period when temperatures were considerably lower and water levels were higher. Four species, Diaptomus dorsalis, Macrocyclops fuscus, Mesocyclops tenuis and Eucyclops agilis montanus were found only in the summer collections. All other species were present throughout the year with the exception of two species which appeared in three of four collections. Oxygen levels ranged from a low concen- tration of 0.4 ppm to a high of more than 15.0 ppm without any evident effect on the number or diversity of copepod species. The same was true of the pH values for the vari- ous bodies of water which ranged from a low of pH 5.4 to a high of pH 9.7. Temperatures ranged from 19.5'C to 34.0'X:: in the sum- mer, O.O'C to Q.O'X: in winter and 8.5'<: to 23.5'C in the early spring. LIST OF SPECIES Order Calanoida Family Centropagidae 1. Osphranticum labronectum Forbes, 1882 Family Temoridae 2. Eurytemora affinis (Poppe), 1880 Family Diaptomidae 3. Diaptomus dorsalis Marsh, 1907 4. D. siciloides Lilljeborg, 1889 5. D. sanguineus Forbes, 1876 6. D. virginiensis Marsh, 1915 7. D. pallidus Herrick, 1897 8. D. mississippiensis Marsh, 1894 9. D. reighardi Marsh, 1895 10. D. bogalusensis marii new subspecies 11. D. sp. Order Cyclopoida Family Cyclopidae 12. Paracyclops fimbriatus (Fischer), 1853 13. P. fimbriatus poppei (Rehberg), 1880 14. Eucyclops prionophorus Kiefer, 1931 15. £. speratus (Lilljeborg), 1901 16. E. agilis (Koch), 1838 17. E. agilis montanus (Brady) 18. Tropo cy clops prasinus (Fischer), 1860 19. T. prasinus mexicanus Kiefer, 1938 20. Cyclops vernalis Fischer, 1853 21. C. bicuspidatus thomasi Forbes, 1882 22. C. crassicaudis brachycercus Kiefer, 1929 23. C. varicans rubellus Lilljeborg, 1901 24. Mesocyclops edax (Forbes), 1891 25. M. tenuis (Marsh), 1909 26. Macrocyclops fuscus (Jurine), 1820 27. M. albidus (Jurine), 1820 28. M.a^er (Herrick), 1882 ECOLOGICAL AND DISTRIBUTIONAL NOTES Genus Osphranticum Forbes Osphranticum labronectum was collected during the winter and early spring periods when water temperatures ranged from 2.5*^ to 14.5'C. It occurred in shallow sloughs with little or no current, connected to or immediately adjacent to rivers and creeks. The species was taken from the Yazoo, Big Black and Pearl river drainages. Genus Eurytemora Giesbrecht Eurytemora affinis was collected during early spring in waters that ranged from Nos. 1-2 Mississippi Copepods 29 H.O'C to 16.5°C It was found in flooded oxbow lakes adjacent to and continuous with the Mississippi River. Genus Diaptomiis Westwood Diaptomus dorsalis was found once in considerable numbers in the late summer collection. It was taken from a large creek with little or no current, in the Yazoo drainage. D. siciloides was collected during all seasons, except late summer, in large lakes, streams and rivers of the Yazoo and Missis- sippi drainages. D. sanguineus was found during the winter and early spring periods in the Yazoo and Pearl River drainages. It occurred in slow moving streams and sloughs and in a flooded field adjacent to a farm pond, but was not taken from the pond itself. D. virginiensis was found in sloughs ad- jacent to rivers of the Yazoo drainage during the winter collection period. Water tempera- tures at the collection sites ranged from 3.5°C to 7.5°C. D. palUdus occurred in all habitats sampled, and within a wide range of pH, temperature and dissolved oxygen values. D. mississippiensis was collected during all seasons in the Yazoo, Tombigbee, Ten- nessee and Big Black river drainages. It was predominantly lentic, appearing most often in lakes and ponds, but was not taken from creeks. D. reighardi was abundant and wide- spread and was found during all seasons in all drainages and in all types of habitats. Two-thirds of the collections of this species were made from lakes and ponds. Description of New Subspecies D. bogalusensis marii new subspecies (Figs. 1-6) Occurrence: This species was taken from all drainages and during all seasons in north- ern Mississippi. It was found most often in lakes, ponds and other standing waters, but was collected on rare occasions from lotic habitats. Type lot: Holotype female, USNM 151227, paratypes, male and female, USNM 151228 and 151229, from Tishomingo St. Pk. Lake, Tishomingo Co., Miss. Diagnosis: Female. Average length of 15 preserved specimens 1.11 mm (0.93 mm-1.30 mm). Metasome (Fig. 1) with head rounded; greatest width at posterior portion of first metasomal segment and second meta- somal segment; width about 37% of total length of metasome. Metasome tapering gradually, especially distal to metasomal segment 3. Metasomal segments 5 and 6 not separated dorsally, "wings" of segment 6 directed posteriorly with small sensillum on lateral posterior corners. Urosome (Fig. 2) 3-segmented, approxi- mately 35% of length of metasome. Genital segment asymmetrical to nearly symmetri- cal, left side rounded, symmetric, right side more flattened on posterior margin with sensillum located on anterior margin. Seg- ment 2 shorter than segment 3, caudal rami equal to or slightly longer than segment 3. Inner margins of caudal rami bearing hairs. Antennules reaching beyond distal ends of caudal rami setae, one seta on segments 11 and 13-19. Maxillipeds (Fig. 3) with distal lobe of basal segment bearing three setae. Leg 5 (Fig. 6) basipod 1 subglobular and about as long as wide. Basipod 2 subtri- angular and slightly longer than wide with lateral seta on external margin reaching along outside margin. Basipods 1 and 2 — 1.05; exopod 1 - 1.00; and exopod 2 - 1.08 (proportional lengths). Exopod 2 base slightly more than one-third of length. Exo- pod 2 bearing no latteral seta. Exopod 3 represented by two lateral setae on proximal lateral margin of exopod 2. Length of setae variable, but outer usually at least twice length of inner. Endopod reaching beyond base of exopod 2 with distal end bearing numerous fine setae and two longer setae, outer longer and its length slightly greater than greatest width of endopod. Endopod ends in pointed protuberance. Male. Average length of six preserved specimens 1.03 mm (0.9 mm - 1.11 mm). Metasome similar to female. Urosome symmetrical. Left antennule as in female. Right anten- 30 Tulane Studies in Zoology and Botany Vol. 20 nule (Fig. 4) with spines of segments 8 and the proximal margin of segment 15. In D. b. 12 shorter than width of their respective seg- marii this spine almost reaches the proximal ments; spine 12 shorter than 8. Spine of margin of segment 15 but in D. bo^a/uien^is segment 10 nearly as long as width of seg- it reaches just beyond the middle of segment ment and spine of segment 11 longer than 14, The fifth leg of D. b. marii presents a width of segment 11. Spine on segment 13 number of similarities to D. sinuatus and D. almost reaching proximal margin of segment bogalusensis. D. b. marii and D. sinuatus 15. Spinous process present on segments 15 have a triangular lamella on the distal inner and 16 with 15 the larger. margin of right exopod l.D. b. marii and D. Leg 5 (Fig. 5) with right posterior face of sinuatus both have a small bifid spine on the basipod 2 bearing blunt, bifid spine on distal distal margin of right basipod 2. The claw of margin. Exopod 1, inner distal margin with ^ y marii is unlike that of the other two small rounded lamella. Exopod 2 inflated species in being sickle-shaped and robust, anteriorly, spine about as long as width of -pj^g j^fj gxopod 2 is quite similar in all three segment at spine. Claw about as long as exo- groups, with D. b. marii and D. bogalusensis, pod, sickle-shaped, very robust. having a somewhat longer short spine than is Leg 5 left (Fig. 5) (excluding process) present in D. sinuatus. Exopods 1 and 2 of reaching to end of right exopod 1. Basipod jhe left leg are not completely separated in longer and wider than exopod. Exopod £> ^ marii and D. bogalusensis while they segments 1 and 2 incompletely separated. are completely separated in D. sinuatus. The Exopod 1 rounded inner pad and proximal left endopod is similar in D. b. marii and D. pad of the second exopod segment with long bogalusensis but that in D. sinuatus is much hair. Exopod 2 with two prominent proces- longer and narrower. These three groups ses; the outer process stouter and longer presently are found only in the southeastern than exopod. Inner, slender process about United States: D. bogalusensis in Louisiana, 80% of length of outer. Endopod very large D. b. marii in northern Mississippi and D. and slightly longer and as wide as exopod. sinuatus in northwestern Florida. The mor- Distal end of endopod covered with many phological and geographical nature of D. b. short, stout hairs. marii indicates there may be a very close Systematic position: This subspecies relationship between these three groups, appears to be an intermediate form between With further collecting and a better under- D. sinuatus ¥Ar\c2t.\A, \9S2> and. Ubogalusen- standing of their ranges, this relationship sis Wilson & Moore, 1953, but more closely should be clarified, related to the latter. In the female this rela- tionship is indicated by the nature of the Descriptive Notes on D/cjpfomu5 sp. genital segment of the urosome. The right V^^g^- '' °) side of this segment in D. sinuatus forms a Male right first antennae without spinous small conical protuberance while in D. process on segment 14 or 15. Exopod 2 of bogalusensis the right side is formed into a right fifth leg (Fig. 7) with outer margin large laterally directed protuberance. This rounded and inner margin with small stout protuberance in D. b. marii is larger than in spine. Claw of exopod 2 angled in two places D. sinuatus but much more symmetrical and and much longer than exopod. Endopod of shorter than in D. bogalusensis. The fifth leg right leg 5 enlarged as in D. mississippiensis. of D. b. marii differs from the other two Left leg 5 terminal segment (Fig. 8) with species in that the spines representing the two processes, outer more anterior process vestigal third exopod segment are unequal in nearly as long as entire exopod with distal length, the outer being much longer. two-thirds of inner margin flattened. Inner In the male the major areas of difference and more posterior process triangular and lie in the nature of the right antennule and less than one-half length of outer. Endopod the fifth legs. In D. sinuatus the spine on reaching almost to distal end of short inner segment 13 of the right antennule reaches process. Average length of male 0.80 mm. Nos. 1-2 Mississippi Copepods 31 ,AJJ^ JlSjL 15 I 132H , AliL 8 90 ti 28 u 106 u Figs. 1-8. Diaptomus bogalusensis marii new subspecies. 1. Female dorsal view. 2. Female urosome. 3. Female maxilliped. 4. Male right first antennae segments 8-16. 5. Male fifth leg posterior view. 6. Female fifth leg. Diaptomus sp. 7. Male fifth leg anterior view. 8. Male fifth leg detail of processes of left exopod two. 32 Tulane Studies in Zoology and Botany Vol. 20 Occurrence: This copepod was found dur- ing all seasons and in all drainages except the Tombigbee. It was collected from all types of habitats with the exception of swamps and sloughs. More than 70% of the collec- tions of this species were taken from lakes and ponds. Remarks: According to H. C. Yeatman (personal communication), this species was being described by the late M. S. Wilson of Anchorage, Alaska. It has been collected by several other workers, from southeastern United States, including Dr. Yeatman and Dr. D. L. Bunting of the University of Tennessee at Knoxville. At this time and until Wilson's unpublished papers are studied, its status is uncertain. Genus Paracyclops Glaus Paracyclops fimbriatus was found during all seasons in all types of habitats. It was present in all drainages except that of the Big Black and Pearl Rivers. P. fimbriatus poppei was much less common than P. fimbriatus, appearing only during the winter and early spring collec- tions in creeks, rivers, and intermittent waters of the Yazoo River drainage. Genus Eucyclops Glaus Eucyclops prionophorus was collected only tour times. It was found in late summer and early spring in the Yazoo and Tennessee River drainages. E. prionophorus was taken from one lake, two creeks and one pond. E. speratus was found throughout the year in all types of habitats except rivers. It was present in the Yazoo and Tombigbee drainages. E. agilis was present in all types of habi- tats and in all drainages throughout the year. It was one of the most common and wide- spread cyclopoid copepods encountered in northern Mississippi. E. agilis montanus is much smaller and characterized by shorter caudal rami which are not more than 2.5 times their width. It was taken from sw^amps, sloughs and inter- mittent waters of the Yazoo River drainage. Genus Tropocyclops Kiefer Tropocyclops prasinus was widespread in all habitats and drainage systems. It was taken during the late summer, winter, and early spring periods. T. prasinus mexicanus was found in all drainage systems, all types of habitats, and during all seasons. Genus Cyclops MuUer Cyclops vernalis was taken from all drain- ages, at all seasons and from all types of habitats. It is quite variable, especially in the spination of the fourth swimming legs and the length of the spines of the fifth legs. Although collected during all seasons, C. vernalis was most abundant during the winter and early spring periods. C. bicuspidatus thomasi was one of the most abundant cyclopoid species in the early winter and spring collections, but was not taken during the summer. It occurred in all drainages except the Tombigbee and in all types of habitats. Of the 27 collections made of this species, 17 were from lakes and ponds. C. crassicaudis brachycercus was collected only in offshore plankton tows from two large impoundments in the Yazoo drainage. C. varicans rubellus is a small, obscure species that was widespread but not com- mon. It was taken from all habitats with the exception of ponds, and in all seasons. C. v. rubellus was found in all drainages except the area immediately adjacent to the Missis- sippi River. Genus Mesocyclops Sars Mesocyclops edax was present through- out the year and was widespread during the early spring collection. It was found in all habitats and all drainages with the exception of the Tennessee. M. tenuis is a rare species that was taken once during the late summer of 1972 in a large oxbow lake between the levee and the Mississippi River in northwestern Mississippi. Genus Macrocyclops Glaus Macrocyclops fuscus was found and identified only once from an immature form Nos. 1-2 Mississippi Copepods 33 collected during the late summer of 1972 in a small lake in the Tombigbee River drain- age. M. albidus was a common species wide- spread throughout the year in all habitats and all drainage systems except that of the Mississippi River proper. M. ater is a widespread but uncommon form in the United States. It was collected once during the early spring in the Y-azoo drainage in about 0.3 m of water in a flood- ed field adjacent to a farm pond, but was never taken from the pond itself. DISCUSSION During the course of the investigation, several significant range extensions were un- covered. Eurytomora affinis is a euryhaline species that is usually found in lakes and ponds of coastal areas (Wilson & Yeatman, 1959); however, during the current investiga- tion specimens were recovered 1145 km up- river from the Gulf of Mexico well beyond any saltwater intrusion. Either this species has been missed in previous freshwater col- lections, or it is currently migrating into the inland freshwater environment. Mesocyclops tenuis is a rare species that has been collect- ed primarily in the southwestern United States but was taken once during the sum- mer of 1972 in northwestern Mississippi. Cyclops crassicaudis brachycercus, according to Bunting (1973), has previously been taken only from stagnant, temporary pools and wells and from littoral areas of shallow lakes. In northern Mississippi it was taken only in offshore plankton tows in two large man-made reservoirs. With more thorough collecting, many of these species with re- stricted ranges will undoubtedly prove to be more widely distributed than once thought. A number of problems arose during this investigation concerning the interpretation of several species-subspecies complexes. Females of Diaptomus pallidus and D. reighardi posed one of the most difficult problems of the survey. The chief distin- guishing charateristic between the two is the length of the two lateral setae of the terminal exopod segment of the fifth legs. In D. reighardi the inner seta is longer and reaches beyond the middle of the claw, while in D. pallidus, the two setae are nearly equal in length and do not reach beyond the middle of the claw, A large number of indi- viduals were encountered during this study that were transitional between the two species. Until a detailed investigation of the differences between them can be compele- ted, identification of females of these species should be made with caution. Another problem concerns the Eucyclops agilis-E. speratus relationship. From the material collected in northern Mississippi, there seems to be no easily recognizable dis- tinction between them. Supposedly the main differences between the two are relative length of the caudal rami, length of the inner corner setae of the caudal rami and the prominence of the lateral spinules on the caudal rami. In the study area numerous individuals were found with characters inter- mediate between these species. Many indi- viduals with the long caudal rami of JB. sper- atus also had prominent lateral spinules of the caudal rami characteristic of E. agilis. Tropocyclops prasinus and T. p. mexi- canus presented another area of concern. Distribution notes by Wilson & Yeatman (1959) indicate that T. prasinus is the com- mon form, but in northern Mississippi T. p. mexicanus is prevalent. Although, as ex- pected, many individuals appeared inter- mediate between these two groups, the fact that T. p. mexicanus, T. prasinus and inter- mediates were taken from the same body of water at the same time indicates a need for reviewing the validity of this species/ sub- species complex. Two other species were represented by two subspecies in the study area: Eucyclops agilis and E. agilis montanus and Paracyclops fimbriatus and P. fimbriatus poppei. Neither of these couples were col- lected from any body of water at the same time although they did occur in the same waters at different seasons. Further investi- gations of these subspecies is needed to determine if they are truly different or mere- ly seasonal variations. Of the calanoid copepods D. pallidus was the most common and widespread during all 34 Tulane Studies in Zoology and Botany Vol. 20 seasons and the most common and wide- spread copepod collected during the summer period. T. prasinus and T. p. mexicanuswere by far the most widely distributed cyclopoid copepods in the late summer, winter and early summer collections and were second only to Eucyclops agilis in the early spring collections. T. p. mexicanus was the most widespread copepod during the winter col- lection with D. pallidus being second. ACKNOWLEDGEMENTS I am deeply grateful to my wife, Mary Jo, for her valuable assistance in the field work and in the preparation of this manuscript. My thanks also to Dr. H. C. Yeatman, Uni- versity of the South, for his aid in the species identifications and to Dr. J. W. Crane, Washington State University, for his review of the manuscript. LITERATURE CITED BUNTING, DEWEY L. 1973. The Cladocera and Copepoda of Tennessee II. Cyclopoid copepods. J. Tenn. Acad. Sci. 48:138-144. EDDY, S. and P. K. SIMER. 1928. Notes on the food of the paddlefish and the plankton of its habitat. Illinois Acad. Sci. Trans. 21:59-68. GRANTHAM, B. J. 1958. The seasonal variation of the plankton of Lake Geiger. Unpublished thesis, Univ. of Miss. KINCAID, T. 1953. A contribution to the taxono- my and distribution of the American fresh- water calanoid Crustacea. Privately printed by the author, Calliostoma Co., Seattle, Wash- ington. MARSH, C. C. 1907. A revision of the North American species of Diaptomns. Wise. Acad. Sci. 15:381-516. 1912. Notes on fresh-water copepods in the U. S. National Museum. Proc. U. S. Nat. Mus. 42:245-255. __i 1929. Distribution and key of the North American copepods of the genus Dwpfo- mus, with the description of a new species. Proc. U.S. Nat. Mus. 75:1-27. WILSON, M. S. and W. G. MOORE. 1953. New records of Diaptomns sanguineus and allied species from Louisiana, with a description of a new species (Crustacea:Copepoda). J. Wash. Acad. Sci. 43:121-127. and H. C. YEATMAN. 1959. Free- living Copepods, pp. 735-861. In: W. T. Edmondson (ed.) Fresh-water biology. 2nd. ed. John Wiley & Son, New York. January 9, 1978 EVOLUTION AND FOSSIL RECORD OF THE CHICKEN TURTLE DEIROCHELYS, WITH A RE-EVALUATION OF THE GENUS DALE R.JACKSON FLORIDA STATE MUSEUM AND DEPARTMENT OF ZOOLOGY, UNIVERSITY OF FLORIDA, GAINESVILLE 32611 ABSTRACT Prior evidence of the fossil history of the mono- typic genus Deirochelys is limited to a single upper Pleistocene fragment and a number of sub-Recent elements from Florida. On the basis of several morphological adaptions unusual among emydine turtles (e.g., neural bone width and rib structure), fossils from 20 Florida sites, ranging from Miocene to sub-Recent in age, are referred to the genus Deirochelys. Evidence of the gradual evolution of a specialized suite of characters associated with pharyngeal feeding is presented. The middle Plio- cene representative of the genus is recognized as a distinct species (Deirochelys carri, sp. nov.) inter- mediate between Recent D. reticularia and less specialized emydines such as Chrysemys. Deiro- chelys fossils trom the Thomas Farm Miocene are more primitive than D. carri and further bridge the morphological gap between Chrysemys and Deiro- chelys^ The evolutionary history of the mono- typic genus Deirochelys is one of the more enigmatic chapters in our knowledge of North American emydid turtles. Previous workers (Carr, 1952; Loveridge and Wil- Uams, 1957; McDowell, 1964) have generally agreed that Deirochelys is a highly special- ized derivative of the genus Chrysemys (sensu McDowell, 1964). Furthermore, Baur's (1889) suggestion of a close phylo- genetic relationship between Deirochelys and another North American monotypic emydine genus, Emydoidea, has been sup- ported by most subsequent workers Love- ridge and Williams, 1957; C. Jackson, 1959; McDowell, 1964; Zug and Schwartz, 1971). Recently Waagen (1972) and Bramble (1974) have cast doubt on this idea based on their respective studies of musk glands and shell mechanics. The fossil record has been of no help in these matters to date. Prior knowledge of the fossil history of the genus Deirochelys is hmited to description of one partial nuchal bone from the upper Pleistocene of Florida (C. Jackson, 1964, 1974a) and to mention of the presence of D. reticularia in a sub- Recent Florida site (Hirschfeld, 1968). Crawford Jackson (1964) found that the Pleistocene element represents a turtle con- specific with Recent D. reticularia. All other fossils previously assigned to the genus, i.e., Deirochelys floridana Hay and Trachemys jarmani Hay (Hay, 1908; Weaver and Robertson, 1967), actually represent the genus Chrysemys (C.Jackson, 1964, 1974a). This paper examines material referable to the genus Deirochelys from one Miocene, five Pliocene, 12 Pleistocene, and two sub- Recent sites, all in Florida. The Miocene fossils are the oldest known representatives of the genus. Two species of Deirochelys, one new, are recognized as fossils. As will be shown the major course of evolution within Deirochelys has been the extreme elongation of the head and neck, a condition achieved by only one other emydine genus (Emy- doidea) and presumably developed as a trophic specialization. The accompanying cervical musculature hypertrophy has neces- sitated further structural modifications of the shell and vertebral column. It is for this reason that in tracing the evolution of the genus I dwell primarily upon this cervico- cranial elongation and associated morphol- ogical modifications (e.g., changes in neural bone width and rib and vertebral structures), EDITORIAL COMMITTEE FOR THIS PAPER: DR. CRAWFORD G. JACKSON, JR., Research Associate, Department of Paleontology, San Diego Natural History Museum, San Diego, California 92112 DR. SAMUEL B. McDOWELL, Professor of Zoology, Rutgers University, Newark, New Jersey 07102 35 36 Tulane Studies in Zoology and Botany Vol. 20 to which I collectively refer hereafter as a single "character suite." MATERIALS AND METHODS All fossil specimens except those from Waccasassa River and a few from Thomas Farm are part of the vertebrate paleontology collection of the Florida State Museum (UF); the Waccasassa River I specimens are from the Timberlane Research Organization (TRO), the Thomas Farm fossils are from the collec- tions of the Museum of Comparative Zool- ogy, Harvard University (MCZ). Comparative skeletal material was examined from the herpetology collection of the Florida State Museum (UF), the National Museum of Natural History (USNM), and my personal collection (DRJ). Extant specimens examined were Deirochelys reticularia: DRJ 264, 266, 270, 274, 278-280, 300, UF 1420, 7744, 14244, USNM 11610, 11615, 29477, 29584, 62219, 80965, 95789; Emydoidea blandingii: UF 14249, 18931; Chelydra serpentina: DRJ 253; Chelus fimhriata: UF 21977. A shell thickness index (STI) was deter- mined for most fossils. Thicknesses of fossil shell elements were measured and divided by corresponding measurements of a series (N = 10) of Recent adult D. reticularia of cor- responding size, or by linear extrapolations to approximate such if a Recent specimen of sufficient size were unavailable. As the rela- tionship between shell thickness and body length may not be strictly linear, the STI values given for the largest fossils may actu- ally be underestimates. There was little individual STI variation among the Recent specimens when corrected for differences in body length. More medial elements (neural bones and proximal ends of pleural bones) generally yielded slightly higher STI values than peripheral elements (peripheral, pygal, and nuchal bones), indicating that increase in shell thickness is not necessarily propor- tional for all parts of the same shell. Medial edges of peripheral elements were measured to reduce this discrepancy. An index of free rib length (width of rib canal) was determined by dividing the straight-line distance from the proximal tip of the pleural bone to its union with the rib by the width of the pleural bone at the level of the union. The fragmented condition of most of the fossils necessitated the use of pleural bone width rather than length. In comparing neural and pleural bones of fossil Deirochelys with those of Recent specimens, it is necessary to determine which of the eight neural or pleural bones the fossils represent. The presence and position of scute sulci as well as the relative proportions of the anterolateral and postero- lateral borders of the bones usually make this possible. Because of the relatively great width and frequent anomalies of the posteri- or neural bones of most emydine turtles, these bones are of little taxonomic value. All measurements are maximum and given in millimeters. FOSSIL LOCALITIES The Appendix provides an annotated list of Florida localities that have yielded fossil Deirochelys mentioned in this paper. Refer- ence is made to other publications in which stratigraphy, paleoecology, and correlative age of each of these deposits is described in detail. Figure 1 shows the geographic distri- bution of these sites. SYSTEMATIC DESCRIPTIONS All past descriptions of the genus Deiro- chelys (Agassiz, 1857; Baur, 1889; White, 1929; Schwartz, 1956; C. Jackson, 1959; McDowell, 1964; Zug and Schwartz, 1971) have necessarily been drawn solely from the single extant species, D. reticularia. Hence, many characters which would have been more appropriately designated as specific characters, particularly those involving color pattern, have been incorporated into the definition of the genus. Therefore, in order to accommodate the fossil members of the genus it is necessary to relegate many of the generic characters, including all references to color pattern, to specific level. Additionally, an examination of osteological characters Nos. 1-2 Deirochelys 37 through time reveals phylogenetic changes within the genus that may be used to distin- guish certain allochronic forms. For these reasons I find it necessary to give a brief systematic reevaluation of the genus as a pre- lude to a formal description of the fossil forms. The present chronologically- expanded definition of the genus, like those of Baur (1889), White (1929), C. Jackson (1959) and McDowell (1964), is based solely on osteological characters. As fossil skull material is presently unknown, all skull characters are drawn from Recent D. reticu- laria. Schwartz (1957) gives a brief but adequate account of the taxonomic history of the genus. Family Emydidae Subfamily Emydinae Genus Deirochelys Agassiz To the generic synonymy given by Zug and Schwartz (1971) should be added the following entry: Hirochelys Beyer, 1900: 45. Type. Testudo reticularia Latreille. Referred species. Deirochelys reticularia, the only extant species, at present distribu- ted throughout the southeastern United States and known from the Pleistocene of Florida; Deirochelys carri, new species, middle Pliocene Alachua clays of Florida, Hemphillian age. Definition. Shell elongate to subovate in adults; carapace elliptical or cuneiform in outline and usually sculptured with fine parallel ridges or scales (Fig. 2); anterior edge of nuchal bone generally truncate and acuminate; lateral sulci of nuchal scute usually parallel above and below; nuchal scute usually two to three times longer than wide above, approximately as wide as long below; nuchal bone overlapped by only small corner of first coastal scute or not at all; vertebral scutes as wide as long; neural bones hexagonal, short-sided in front; first neural bone circular to subovate in outline; other neural bones generally as wide or wider than long (Fig. 2); peripheral bones unnotched; pygal bone approximately par- allel-sided with a shallow mesial notch; ribs dorsally free from pleural bones well below proximal ends of pleurals, their free portions slender and bowed ventrally (Fig. 3) accom- modating the enlarged trunk vertebral muscle complex (Shah, 1963). Plastron usually considerably narrower than carapace, akinetic, and firmly united to carapace by a high bony bridge and plastral buttresses; inguinal scutes large (contrary to ^ V_ ® 100 KM J \^ rt-fl iJ (3- Fig. 1. Fossil sites in peninsular Florida containing Deirochelys. Site ages are given in Appendix. 1. McGehee Farm 2. Haile sites 3. Love Bone Bed 4. Wall Company Pit 5. Mixson's Bone Bed 6. Kendrick lA 7. Waccasassa River sites 8. Coleman IIIC 9. Seminole Field 10. Catalina Gardens 1 1. Bradenton 12. Warm Mineral Springs 13. Nichol's Hammock 38 Tulane Studies in Zoology and Botany Vol. 20 Holman's [1967] statement that they are absent); plastron smooth ventrally or with traces of sculpturing similar to but less pro- nounced than that of carapace; entoplastron usually anterior to humcfopectoral sulcus and overlapped by gular scutes for approxi- mately one third of length. B Fig. 2. Third neural bones of Chrysemys floridana (A) and Deirochelys reticularia (B); note greater width and characteristic sculp- turing of latter. Skull and second through seventh cervical vertebrae elongate; neural spines of anterior thoracic vertebrae laterally compressed as vertical sheets (Fig. 4); triturating surfaces of maxilla and mandible narrow, vdthout ridges; beak never hooked; interorbital width very narrow, less than one-half diameter of orbit; palate decidedly flat; posterior pal- atine foramina much larger than foramina orbito-nasale (Gaffney, 1972) [= anterior palatine foramina of Hoffman, 1890] ; temporal arcade complete; quadrate nearly enclosing stapes; coronoid relatively low; hyoid apparatus strongly developed, lateral horn length at least as great as skull wddth; cervical musculature as described by Shah (1963). A specialization of the genus almost certainly related to the elongate neck and hypertrophied vertebral musculature is the modification of the spinal column. The dif- ferences between Deirochelys and more primitive emydines (e.g., Chrysemys), sum- marized in Table 1 and Fig. 4, are most con- spicuous in the first four thoracic vertebrae. In both forms ribs attach intercentrally and the thoracic vertebrae are united by their neural spines to the overlying neural bones. The net effect of these modifications in Deirochelys has been to move the rib attach- ment ventrally (away from the carapace), allowing for the hypertrophied trunk verte- bral musculature without changing the distance of the spinal cord from the ventral surface of the carapace. Deirochelys reticularia (Latreille) Chicken turtle The only addition to the species synony- my listed by Zug and Schwartz (1971) is: Hirochelys reticulata Beyer, 1900:45. Type: The type was formerly in the col- lection of the French Museum National d'Histoire Naturelle but is now considered lost (Schwartz, 1956). Schwartz (1956) described a neotype and neoallotype from the vicinity of the original type locaUty. Type locality. Restricted by Harper (1940) to the vicinity of Charleston, South Carolina. Diagnosis: A Deirochelys characterized by relatively low length: width ratios for third through fifth neural bones (means, 0.6 to 0.7; Fig. 5) and relatively great length of free portions of dorsal ribs (Fig. 6); colora- tion as described by Schwartz (1956) with notation that the yellow forelimb band is Fig. 3. Frontal aspects of third pleural bones of Chrysemys concinna (A) and Deiro- chelys reticularia (B), showing dorsal ribs. Nos. 1-2 Deirochelys 39 Table 1. Comparison of the thoracic verte- brae o( Deirochelys and Chrysemys. Character Chrysemys Deirochelys Neural spines low and robust Centra. Site of rib attachment to veter laterally com- pressed as vertical sheets narrowest dorsally ; dorsoventrally compressed; ventral sur- faces wide and flattened expanded dorsal expanded region of ventral region centra of centra narrowest ventrally ; not compressed usually but not always wide; neck nearly as long as plastron; usual pattern of cervical central articulation (perhaps a generic char- acter): (2( (3( (4) )5) )6) )7( (8) (WiUiams, 1950; C.Jackson, 1974b). Description of fossil material. The follow- ing fossils, listed in reversed chronologic order by site, are here assigned to D. reticu- laria. Nichol's Hammock: contains more D. reticu- laria than any other post-Pliocene site; 75 carapacial elements (UF 20892), a cervical vertebra (UF 20904), and a supraoccipital crest (UF 20905) represent 12 to 20 indi- viduals ranging from 65 mm to 195mm cara- pace length (CL); many additional elements from this deposit, particularly plastral and peripheral bones which lack diagnostic features, probably represent D. reticularia as well; fossils from the site are indistinguish- able from extant D. reticularia, their shallow rugosity probably reflecting their relatively small size; STI 0.95 to 1.05. Warm Mineral Springs: To date, 35 elements — one nuchal, seven neural, one suprapygal, six pleural, 13 peripheral, and five plastral bones, plus a scapula and broken femur ~ all assigned field number WMS 19352 and representing 5 to 10 individuals of CL 138 to 184, have been removed from this site. The bones are similar to those from Nichol's Hammock and have an average STI of 1.15. Vero: A large number of plastral and cara- pacial elements, including at least two nuchal, two neural and two pleural bones (all recently acquired by the Florida State Museum as part of the former Florida Geo- logical Survey .collection and as yet un- catalogued) are virtually indistinguishable from modern D. reticularia; STI 0,85-0.95. Waccasassa River I: Two second neural bones (TRO 101, 102) and a third neural bone (TRO 103), representing three indi- viduals of 130 to 210 CL (Fig. 7); STI 1.1 to 1.3. Waccasassa River V: A lightly -sculptured nuchal bone, UF 16271 (Fig. 7): greatest length 30.5, greatest width 35.5, estimated CL 135; proximal end of a pleural bone, UF 16275; STI 1.1. Waccasassa River VI: A distinctly grooved nuchal bone, UF 21906: greatest length 39.8, greatest width 42.3, estimated CL 170; STI 1.1. Reddick IIC: A first neural bone (UF 21955) from an adult turtle (estimated CL 180) and the proximal end of a fourth pleur- al bone from a juvenile; STI 1.1. Coleman IIIC: Four elements (UF 15186E) representing at least three individuals: a lon- gitudinally rugose, relatively deeply notched pygal bone (length 21.5); a left epiplastron (interepiplastral suture length 13.6); a char- acteristically rugose left xiphiplastron missing its distal portion (hypo-xiphiplastral suture length 40.7); and a distinctly sculp- tured right hypoplastron (interhypoplastral suture length 58.4); STI 1.3. St. Petersburg, Catalina Gardens: Lower two thirds of a right fifth pleural bone (UF 19248): greatest width 30.0, estimated length 60, estimated CL 220; STI 1.3. Seminole Field: A deeply sculptured frag- ment of a right second pleural bone (UF 9927) with rib attachment - width at rib level 28.0, thickness at rib level 5.9, esti- mated CL 210; fragment of a left hypo- plastron (UF 9927) with deep longitudinal grooves on ventral surface, estimated CL 210; STI 1.4. Bradenton 51st Street: A characteristically 40 Tulane Studies in Zoology and Botany Vol. 20 sculptured fragment of a nuchal bone (UF 2482): estimated CL 210, STI 1.25. Kendrick lA: A sixth neural bone (UF 19250) with a pronounced, scale-like sculp- turing and a low, rounded keel — greatest length 19.3, greatest width 33.2, greatest thickness 6.0, estimated CL 250, STI 1.3; a deeply grooved partial nuchal bone (UF 9292) possibly from the same individual and described previously by C. Jackson (1964): estimated CL 250, STI 1.1 to 1.6; (Fig. 8). Wall Company Pit: Proximal halves of two broken pleural bones (UF 5026): a second left (estimated CL 175, STI 1.6) and a deep- ly rugose fourth right (estimated CL 220, STI 1.5) with rib distance: pleural width ratios of 0.84 and 0.80, respectively. Haile XVI: 38 elements representing at least 15 individuals of CL 116 to 240: a nuchal bone (UF 20896), length 40.0, estimated CL 182; a second neural bone contiguous with the second and third right pleural bones (UF 20888; Fig. 9), and the first left and second right peripheral bones (UF 20889) almost certainly from the same individual, esti- mated CL 230; fifteen fragmentary pleural bones (UF 20895; UF 20898) and seven peripheral bones (UF 21970); a hypoplas- tron (UF 21969) and partial hypoplastron (UF 21968); second left pleural bone and first four neural bones (UF 20893) from a turtle of 227 CL; and the third (UF 20897), fourth (UF 21971), two fifth (UF 20894 and UF 20898), and sixth (UF 20898) neural bones from five turtles with CL of 240, 140, 225, 160 and 220, respectively. Neural length: width and rib distance: pleural width ratios are included in Figs. 5 and 6; STI of neural bones 2.0 to 2.2. Haile XVA: A fifth neural bone (UF 19249), the dorsal surface of which is extremely flat but moderately sculptured: greatest length Fig. 4. Lateral (A, C) and frontal (B, D) aspects of third neural bones and associated verte- brae of Deirochelys reticularia (A,B) and Chrysemys nelsoni (C, D); c, centrum; s, neural spine. Nos. 1-2 Deirochelys 41 20.2, greatest width 32.0, estimated CL 210; an anterior fragment of a nuchal bone (UF 19168), estimated CL 230; STI 1.5; (Fig. 10). Discussion of fossil material. All of the Rancholabrean and sub-Recent material is clearly referable to D. reticularia. With the exception of shell thickness, relative dimen- sions of individual fossils show no significant differences from corresponding measure- ments of extant turtles. The Blancan and Irvingtonian material, as well as the Ken- drick nuchal, indicate that this species reached a sUghtly larger maximum size dur- ing the Late Pliocene and Pleistocene than at present. The blunt median keel on the Ken- drick neural, although not typical of most extant D. reticularia, occurs posteriorly in a 1^1 K) I »- O »- O z < UJ Z J> K) 10 * 10 ^ 1 o 10 1 10 10 1 1 1 1 1 2 3 4 5 6 NEURAL NUMSER Fig. 5. Length: width ratios of second through sixth neural bones of Recent (cir- cles), Irvingtonian (stars) and Hemphillian (triangles — Love; square — Haile VI; asterisk — Mixson's) Deirochelys. Dice — Leraas diagrams depict mean, range, arul two stand- ard errors; numbers above and below bars represent sample sizes. 1.0 z o .8 -.1 < UJ \u U Z Q .4 '" # i {] ml fi| 1 A -I 1 1 I 1— 2 3 4 5 6 PLEURAL NUMBER Fig. 6. Rib distance: pleural bone width ratios for Recent, Pleistcicene, and Pliocene (Deirochelys; all symbols as in Fig. 5. Fig. 7. Nuchal bone (UF 16271) and three neural bones (TRO 101-103) of Rancholabrean Deirochelys reticularia from Waccasassa River VA and I, respectively. 42 Tulane Studies in Zoology and Botany Vol. 20 Fig. 8. Distinctly sculptured sixth neural bone (UF 19250) oi Deirochelys reticularia from Kendrick lA. few individuals. Though tending to be more pronounced in the Pleistocene, shell rugosity patterns are within the range of variation of extant D. reticularia. The single consistent difference between Pleistocene and Recent D. reticularia is that of shell thickness. The STI of Pleistocene D. reticularia is 1.1 to 2.2 times that of Recent specimens. The trend towards shell thickness reduction appears roughly chronoclinal since at least the Irvingtonian (Table 2), though the absence of material from some glacial and interglacial periods may conceal unseen fluctuations. Similar trends in post-Pliocene shell thickness reduction have been suggest- ed, though less well supported by a time- transgressive series of fossils, for Chrysemys (Preston, 1966, 1971), Emydoidea (Taylor, 1943), Graptemys (D.Jackson, 1915), Kino- stemon (Fichter, 1969), Trionyx (Wood and Patterson, 1973) and Geochelone (Auffen- berg, 1963b). Although shell thickness alone is inadequate as a basis for taxonomic sepa- ration, it is not a simple function of turtle size as Auffenberg (1958) states for Terra- pene. From middle Pleistocene to the pres- ent, the shell of D. reticularia has become progressively thinner. Gaps in the STI-time curve may reflect our incomplete sampling of the fossil record. Nevertheless, the possi- bility of sexual and ontogenetic polymor- phism in this character, as well as hidden fluctuations in the curve, could complicate the matter. Unfortunately, sample sizes from most fossil sites are inadequate for a thorough treatment of the data. The Irvingtonian (Haile XVI) fossils differ from younger material in two additional ways: a higher length: width ratio tor the second and third neural bones (Fig. 5) and a slightly more proximal site of rib juncture with the second and third pleural bones (Fig. 6). These characters do not exhibit allome- A J':* .idl^ ^' 41 A- B Fig. 9. Dorsal (A) and ventral (B) views of part of carapace (UF 20888) of Irvingtonian Deirochelys reticularia from Haile XVI. Note characteristic sculpturing, neural bone width, and rib junctures. Nos. 1-2 Deirochelys 43 Table 2. Shell thickness index (STI) of fossil Deirochelys from 16 Florida sites listed chronologically by faunal periods. Age and Site STI Hemingfordian Thomas Farm Hemphillian Love Mixson Haile VI Blancan Haile XV Irvingtonian Haile XVI Rancholabrean Wall Co. Pit Bradenton Kendrick Seminole Field Catalina Gardens Coleman IIIC Waccasassa I Waccasassa V Sub-Recent Warm Mineral Spring Nichol's Hammock 1.9 1.6-2.1 1.8 1.9 1.5 2.0-2.2 1.5-1.6 1.25 1.3 1.4 1.3 1.3 1.1-1.3 1.1 1.15 0.95-1.05 try in Recent adult specimens, and there is thus no reason to suspect it in Pleistocene populations. In these respects Irvingtonian Deirochelys are morphologically intermedi- ate between the later Pleistocene and the middle Pliocene turtles discussed below. The near identity of the Blancan neural (UF 19249) with that of an Irvingtonian one (UF 20894) suggests that little shell evolution was experienced between these periods. The differences between Recent and upper Pliocene to middle Pleistocene Deiro- chelys are real and might justify taxonomic distinction were it not for the intermediate P.ancholabrean material. Such time-related changes are, however, to be expected within a chronoclinal lineage, as shown previously by Milstead (1967) with Terrapene. The modern subspecies of D. reticularia are dis- tinguished by coloration and shell shape (Schwartz, 1956) and consequently can not be compared to these fossils. Furthermore, D. reticularia hkely varied geographically during the Pleistocene as it does now. For these reasons I refrain from erecting sub- specific epithets for any of the Pleistocene or upper' Pliocene fossils and simply refer them all to the species Deirochelys reticu- laria. The exigence of certain morphological differences, reflected in carapacial osteology, of specimens referred above to D. reticularia and all earlier representatives of the genus (Figs. 5 and 6) is accentuated by the absence of late Hemphillian fossils. This gap in a gradually evolving lineage creates a conveni- ent (though admittedly artificial) point of division between morphologically distinct forms. I therefore designate the turtle repre- sented by the middle Pliocene fossils as Deirochelys carri, new species Etymology . Named in honor of Archie F. Carr, Jr. for his extensive contributions to our knowledge of Recent turtles and to herpetology in general. Holotype. UF 20908, a fragmented but nearly complete carapace lacking only the nuchal bone, first neural bone, and anterior peripheral bones (Fig. IIA); a partial plas- tron consisting of the left hyoplastron, hypoplastron, and xiphiplastron apparently represents the same individual (Fig. IIB). Type locality and horizon. Alachua Clay, Love Bone Bed, near Archer, Alachua County, Florida; early Hemphillian, middle Pliocene. B Fig. 10. Blancan Deirochelys reticularia from Haile XVA; (A) fifth neural bone (UF 19249) and (B) nuchal bone fragment (UF 19168) showing broad nuchal scute under- lap. 44 Tulane Studies in Zoology and Botany Vol. 20 Referred material. All from four Florida sites producing Hemphillian faunas: MLxson's Bone Bed: a fourth neural bone, UF 20890 (formerly Florida Geological Survey V-2599), assigned previously by Hay (1916) to Chrysemys caelata: estimated CL 290, STI 1,8. McGehee Farm: a complete (UF 19204) and two partial (UF 20891 and UF 20903) nuchal bones — measurements of UF 19204: length 57.1, width 60.8, corresponding to a CL of approximately 263; right hypoplas- tron, ninth right peripheral bone, and left and right xiphiplastral fragments (UF 20899). Haile VI: contiguous second neural bone fragment and proximal portion of left second pleural bone (UF 20887); estimated CL 253, STI 1.9; contiguous pygal bone and eleventh left peripheral bone (UF 6485a); anterior end of third cervical vertebra (lack- ing zygapophyses) (UF 6485b , Fig. 12); five peripheral bones UF (6485c); first neural bone, (UF 6485d); 17 pleural bone frag- ments (UF 6485e); many other elements and fragments from this site may represent either D. carri or Chrysemys caelata (D. Jackson, 1976). Love Bone Bed: Although excavation is incomplete, this deposit is already the rich- est source of fossil Deirochelys known. At the time of this writing over 400 carapacial elements and half as many plastral elements of Deirochelys have been removed. Other than the holotype, only two sets of associ- ated carapacial bones have been found (UF 24100 and UF 20900, Fig. 13). The less water-worn carapacial elements display the distinct scale-like sculpturing characteristic of the genus (Fig. 14). Many elements represent turtles of exceptionally large size for Deirochelys: the largest nuchal bone (UF Fig. \\. Deirochelys carri holotype, UF 20908. (A) Dorsal aspect of carapace; (B) Ventral aspect of plastron. Hatched areas missing from fossil. Nos. 1-2 Deirochelys 45 20906) measures 59.8 (length) x 61.8 (width). STI range is 1.6 to 2.1. Diagfiosis. Deirochelys carri differs from D. reticularia in having relatively narrower neural bones (mean length: width ratio of third through fifth neural bones 0.8 to 0.9; Figs. 5, 11 A) and a more proximal site of emergence of the ribs from the pleural bones (Figs. 6, 13). Elongation of cervical verte- brae and patterns of shell rugosity are similar in these species, but the carapace of D. carri appears to be relatively broader. Shell rugosity and width of first vertebral scute of D. carri are Uke those of Chrysemys caelata and C. williamsi, respectively, also from the Florida Pliocene (D. Jackson, 1976); nevertheless, other generic characters distinguish these species from D. carri. Neural bones of D. carri are similar in shape to those of the Florida Pliocene Chrysemys inflata (Weaver and Robertson, 1967), yet distinguished from them by absence of the pronounced keel and deeply excavated sur- face of the lattei*. Fig. 12. Ventral surfaces of third cervical vertebrae of (A) Deirochelys carri (UF 64856) and (B) D. reticularia (DRJ 300) (x2.2). Description. With the exception of the less developed character suite previously alluded to, D. carri is, in most respects, simi- lar to D. reticularia. Nevertheless, many of the fossils indicate that the former reached a greater size than D. reticularia, perhaps as large as 320 mm CL compared to approxi- mately 250 mm CL today (Carr, 1952). The shell of D. carri is about twice as thick as that of extant D. reticularia but not unlike that of Blancan and Irvingtonian repre- sentatives of the modern species (Table 2). Additionally, the reconstructed holotype shell is relatively broad and flat compared to Recent chicken turtles. In this respect, as well as in the flaring of the posterior periph- eral bones, D. carri is reminiscent of some members of the genus Chrysemys and appears to have been more streamlined than D. reticularia. One fairly constant difference between D. reticularia and D. carri is that the anterior edge of the fourth vertebral scute (incised at the fifth neural bone) of D. reticularia projects forward to form a sharp anteriorly-directed V, whereas that of D. carri projects forward only slightly (and more bluntly) or not at all (Fig. IIA). The plastron of D. carri, like that of D. reticu- laria, is narrow. The anal notch in the plastron associated with the holotype of D. carri is twice as deep as that of D, reticularia. There is no significant morphological varia- tion among D. carri from the four sites. Measurements and qualitative observations of all material from Haile VI, McGehee Farm, and Mixson's Bone Bed fall within the range of variation of elements from the Love Bone Bed. Comment. Deirochelys carri is similar in most respects to its presumed descendant D. reticularia. The major differences are modifi- cations associated with the further develop- ment of the specialized elongate neck and head in D. reticularia. In this respect both D. carri and D. reticularia surely represent seg- ments of a single chronoclinal lineage. The neural spines and dorsal rib heads of D. carri are typical of the genus and only slightly more robust than those of D. reticularia. The single cervical vertebra (UF 6485b) referable 46 Tulane Studies in Zoology attd Botany Vol. 20 to D. carri is likewise slightly more robust than the corresponding vertebra of D. reticu- laria (Fig. 12). Although it is impossible to determine accurately the length of the Plio- cene vertebra from the Haile VI fragment, it appears that the characteristic cervical elongation and development of associated modifications in Deirochelys had already approached present levels by middle Plio- cene. Nevertheless, the narrower neural bones and more proximal rib union with the pleural bones in D. carri, compared with D. reticularia, imply a shorter free rib between the pleurals and vertebral column and a cor- respondingly less developed set of cervical extensor muscles in the former. A slightly shorter or less powerful neck in the Pliocene species therefore seems likely. Certainly any future finds oi Deirochelys skull and cervical material in the Love Bone Bed would be particularly valuable. Although the Love Bone Bed provides an exceptionally fine series of Deirochelys fossils, far older than any previously known for the genus, one can trace the evolutionary record back one step further — to the Mio- cene. THE THOMAS FARM DEIROCHELYS The only emydine turtle previously recog- nized from the Florida Miocene (Thomas Farm) is a Chrysemys species of uncertain status (Williams, 1953; Rose and Weaver, 1966). In an effort to determine the rela- tionships of this turtle, I examined the hold- ings of the Florida State Museum for additional material. Among the elements retrieved were a faintly sculptured neural bone (UF 21949) only slightly narrower than those of Deirochelys carri, and the proximal fragment of a pleural bone (UF 21950) with a rib juncture scar too low for that of Chrysemys (Fig. 15). Comparisons with Recent and fossil Deirochelys and Chrysemys, including "typical" Chrysemys elements from Thomas Farm, leave no doubt that the two fossils represent Deirochelys. Curvature of the scute sulcus, relative length of the anterolateral borders, and extreme lowness of the neural spine all indicate that the neural bone is a fifth, while the relative proportions of the medial borders of the pleural bone in addition to the position of the sulcus indicate that it probably is the second pleural bone from the left side. As -I k 1?^^^: *- Fig. 13. Ventral surface of posterior region o( Deirochelys carri carapace, UF 20900, show- ing rib juncture with pleural bones. Arrows indicate where junctures would occur in a D. reticularia of equivalent size. Nos. 1-2 Deirochelys 47 with the Pliocene Deirochelys, the shell is relatively thick (STI 1.9). In addition to the two fossils described above, I tentatively refer to Deirochelys the followdng elements from Thomas Farm: one complete epiplastron (UF 21932) and the medial half of another (UF 21939), the pos- terior part of a right xiphiplastron (UF 21946), the major part of an entoplastron (UF 21942), and the proximal end of a pleural bone (UF 21951). Additionally, one complete and two fragmentary nuchal bones (MCZ 3432; see Fig. 4 in Williams, 1953, and Fig. 2B in Rose and Weaver, 1966), al- though probably representing Chrysemys, may be Deirochelys. The width of the first vertebral scute and shape of the nuchal scute are Uke those of both Deirochelys and Chrysemys ornata. Both the shape of the neural bone (length: width ratio, 0.94, Fig. 5) and the point of juncture of the rib with the pleural bone (rib distance: pleural bone width ratio , 0.41, Fig. 6) indicate that, in terms of cervi- cal hypertrophy, the Thomas Farm Deiro- chelys was even more primitive (less special- ized) than D. carri. Remains of the very low neural spine fused to the neural bone con- firm this. Hence, I beUeve that the limited Thomas Farm material represents a turtle distinct from D. carri. However, any taxo- nomic assignment of the Thomas Farm fossils other than to genus must await additional and preferably associated materi- B Fig. 14. A distinctly sculptured nuchal bone (A) and posterior peripheral bone (B) of Deirochelys carri from the Love Bone Bed. B Fig. 15. Deirochelys fossils from the Thomas Farm Miocene (x 1.1); (A) neural bone, (B) visceral surface of pleural bone fragment showing rib juncture scar. al. More important at present is that in the Thomas Farm Miocene we find an important Unk in the gradual evolutionary sequence from a generalized emydine ancestor (cf. Chrysemys) into the more specialized D. carri and its highly specialized descendant, D. reticularia. DISCUSSION The material now available shows that the genus Deirochelys, instead of being an evolu- tionary enigma, possesses one of the most complete evolutionary records of any Recent turtle. Evolution oi Deirochelys has been by specialization of a generalized emy- dine stock (presumably Chrysemys). The earliest fossils are, in fact, difficult to dis- tinguish from Chrysemys. We may estimate by extrapolation at what point the two genera would be no longer distinct — i.e., the time at which a generalized turtle began its initial shift to a new adaptive zone in re- sponse to selective pressure. The elongated neck (and presumably skull) as well as associated muscular (Shah, 1963) and osteo- logical modifications of Deirochelys had already developed by middle Phocene. This character suite is already conspicuous in hatchHng D. reticularia, so that phylogenetic recapitulation must occur very early during ontogenetic development if it occurs at all. The divergence from a more generalized aquatic emydine stock (moderately short 48 Tulane Studies in Zoology and Botany Vol. 20 neck, long neural bones, weak hyoid appara- tus, robust ribs emerging from very near the proximal ends of pleural bones, limited trunk vertebral musculature, and a relatively broad shell, as in the genus Chrysemys) had certainly begun by the Miocene. Extrapo- lations based on an average rate of evolution from such a generalized ancestor suggests an Oligocene origin of the genus (Fig. 16). This character suite almost certainly evolved as a peculiar trophic structure; Deirochelys uti- lizes a "pharyngeal" method of feeding (Bramble, 1973) for capturing prey capable of quick movements (primarily aquatic arthropods). Arguments such as those of G. J. W. Webb and Johnson (1972), in which cervical elongation is held to represent a thermoregulatory device, seem at most of secondary significance in this case, particu- larly in light of the hypertrophied hyoid skeleton. The thick shell of D. carri and the Thomas Farm Deirochelys, as well as of Blancan and Irvingtonian D. reticularia (Table 2), suggests that until Late Pleisto- cene Deirochelys was a moderately thick- shelled turtle. Pleistocene reduction in weight and volume of the shell may have allowed faster pursuit and increased maneu- verability necessary for capturing fast- moving prey (author's unpublished data) on which Deirochelys had come to specialize. Loss of armor (if the thick shell served this purpose) may have been offset by crypsis and behavioral immobility (unpublished observations). In addition to changes in shell thickness, general reduction in body size, accompanied by relative elongation and heightening of the shell, seems to have occurred from at least Hemphillian to Rancholabrean times. 1.3H .9- < a: .5- •o Ol .? igocene Miocene CHRYSEMYS DEIROCHELYS Pliocene 24 "T" 16 T" 8 MILLIONS OF YEARS B. P. PIst. Recent Fig. 16. Extrapolation through time of fifth neural bone length: width ratio oi Deircfc^ielys, indicating hypothetical point of divergence from a more generalized emydine line. Nos. 1-2 Deirochelys 49 Relationships. Baur (1889) was the first to hypothesize a close relationship between Emydoidea and Deirochelys on the basis of similar skull and rib specialization. Although Carr (1952) beheved the similarity between Emys [= Emydoidea] hlandingii and D. reticularia to be "purely fortuitous," most subsequent workers supported Baur's idea. Bramble (1974) summarizes the situation: Williams (in Loveridge and Williams, 1957) presented a forceful case for a relationship between Emydoidea and Deirochelys. Although Deirochelys possesses no plastral hinge and on many points of shell morphology closely approaches certain members of the genus Chrysemys (McDowell, 1964), it does, as Wil- liams noted, share with Emydoidea a number of specializations of the skull, cervical vertebrae and neck musculature. On these grounds Wil- liams suggested that Emydoidea was a deriva- tive of Deirochelys and only convergent with Emys. This view has been widely adopted by later workers (Tinkle, 1962; McDowell, 1964; Zug, 1966; Pritchard, 1967; Milstead, 1969; Ernst and Barbour, 1972), some of whom (Tinkle, 1962; Zug, 1966) have presented additional evidence in support of it. McDowell (1964: 275) found no 'significant cranial differ- ences between Deirochelys and Emydoidea' and accordingly placed both genera in a Deirochelys Complex within the Emydinae. However, Bramble's (1974) study of shell kinesis and other osteological and myo- logical characters indicates instead that Emydoidea is a "close phyletic associate of Emys and Terrapene" as well as oi Clemmys (the four genera comprising the Clemmys Complex), and that these genera may be distinguished as a group from Deirochelys and McDowell's (1964) Chrysemys Com- plex. Waagen (1972) formed an identical opinion from his analysis of musk glands in Recent turtles. On the basis of fossils dis- cussed in this paper I agree with the con- clusions of Waagen (1972) and Bramble (i974) that Deirochelys shares a close rela- tionship with the genus Chrysemys, and that similarities between Emydoidea and Deiro- chelys are "undoubtedly the result of con- vergent feeding system" (Bramble, 1974). In fact, most of the modifications used to sub- stantiate a close relationship between Deiro- chelys and Emydoidea (elongated ventrally- bowed free ribs, widened neural bones, elongated cervical vertebrae, and a greatly hypertrophied cervical musculature) are also present in the totally unrelated (at least at the familial level) cryptodire genus Chelydra as well as the pleurodire genus Chelus. They are, moreover, all modifications associated with the pharyngeal method of feeding (Bramble, 1973) employed by these turtles. Hence, the taxonomic use of this particular character suite, so clearly convergent among members of three distinct families, should be treated cautiously in attempts to determine intrafamilial relationships. This paper has presented evidence of the gradual develop- ment of these adaptations as a unit of func- tional morphology (Wilson, 1975) within one of these phyletic lines. Pleistocene and late Pliocene fossils of Emydoidea, which are clearly referable to the extant species E. blandingii (Taylor, 1943; Preston and McCoy, 1971), show no special resemblances to late Tertiary Deiro- chelys, other than the convergent character set already discussed, and hence do not support a theory of their divergent evolu- tion. The fossil records and present distri- butions (Carr, 1952; Preston and McCoy, 1971; Zug and Schwartz, 1971; McCoy, 1973; C. Jackson and Kaye, 1974) indicate that the two genera have remained essential- ly allopatric, although the southern exten- sion of the range of Emydoidea in the late Pleistocene (C.Jackson and Kaye, 1974) just touches the present limit of Deirochelys in northeastern Mississippi (C. Jackson and Fortman, 1976). Further ecological studies might help to determine if this allopatric relationship reflects a Gause-type competi- tive relationship or a difference in thermal requirements. Distribution and Paleoecology. The genus Deirochelys is endemic to the southeastern United States, and it is therefore not surpris- ing that the first extensive evidence of its fossil history should be from Florida. All vertebrate fossil sites known to contain Deirochelys (Fig. 1) occur within the range of the modern subspecies D. reticularia chrysea or its zone of intergradation with D. r. reticularia (Schwartz, 1956; Zug and Schwartz, 1971). 50 Tulane Studies in Zoology and Botany Vol. 20 Deirochelys reticularia usually inhabits quiet, shallow bodies of freshwater through- out its range although it occasionally enters the quieter portions of streams (Pope, 1939; R. Webb, 1950; Carr, 1952; Schwartz, 1956; Campbell, 1969) and perhaps rarely salt- water (Neill, 1948; Martof, 1963). Personal observations in north-central Florida indi- cate that the densest populations of Deiro- chelys occur in shallow (less than one meter) ponds with abundant basking logs, emergent bushes (e.g., Cephalanthus) and an extensive Lemna-Wolffiella surface mat. From a struc- tural standpoint, the relatively short limbs, long nuchal scute underlap, and absence of streamlining (as compared to a lotic from such as Chrysemys concinna) reflect its evolution as a quiet-water form. The turtle also shows a proclivity for overland wander- ing (Neili, 1948; Carr, 1952; Gibbons, 1969, 1970). Its typical association with the south- eastern Coastal Plain (Mount and Folkerts, 1968) implies adaptation to a warm temper- ate climate. The presence oi Deirochelys and associated fauna (Lepisosteus, Amia, Alli- gator, Chrysemys caelata [D. Jackson, 1976] , Trionyx cf. T. ferox) in Hemphillian sites thus indicates the existence of quiet freshwater (e.g., sinkhole ponds or sluggish streams) and a warm, equable qlimate in the Florida middle PUocene. Even in the most favorable habitats Deirochelys today rarely reaches densities comparable to those of sympatric emydine turtles (e.g., Chrysemys nelsoni, C. flori- dana, C. scripta). This relationship appears to hold also in the PUocene; in the -only Pliocene deposit containing large numbers of Deirochelys (Love Bone Bed), Chrysemys caelata elements outnumber those of D. carri approximately four to one. In what presum- ably was a suboptimal habitat for Deiro- chelys at McGehee Farm the ratio is even more disparate. This indicates that popula- tions of Deirochelys may be more restricted by Hmiting factors than are other emydines. All fossil records for Deirochelys from sites near the present coastline of Florida (Fig. 1) are either sub-Recent or late Rancholabrean. All other sites except those in the Waccassassa River are in presently well- p. ft ai en a> ■H o 0) ft CQ m 0) ■H O • ft-p 03 U-l CD C -H 03 M 0 0) 0) +^ fH CD 3 aJ a ctf -P ' ft 03 03 s ■3 43 ft •H rH 03 X) U -p w) o C Cd 03 p J -p C 03 S -P cd -H 03 43 +^ •H -p c 03 (rt .cl rH -P o bo 03 ti U 03 < 1-:) 03 O Cd ft cd u cd ■H > Xi to a 03 tEl -P •H 03 VD 1- CO r- U &> •s Cd ,0 ^o ON • • o CO 1— ^ o J- CTs CM T— O CTs PhI 03 o rH o OO CM CJN • O CE> J- Vi3 -d- T- »*> 03 O OS t— C\J • • • lA O ■^ MD O UN • • • ctn ^ r~- •I- (^ ■Lp» • • • 1- D— OO * C\J C\J o OO CJN Cvl CM M3 • c^ • • o • CM CM UN CM 1- o CM * o CM T-TN • • • CJN CO t— * CO o « • • o ON ON '" * ON a • • ■LA CM CM CM ON UN -J- CO • • • \A f- UN CJN 1-fN t^ • • • CM CO VO C3N V£3 O • • • CO CTN t-- o CM J- CO UN OO CJN CO O j rA J- UN UN CTN ^ d CM CM CM -J- -3- CM CM -d- t-- UN • • • • • J- J- f^ 1- \D NO CO 1- • • • \A -d- UN CM O O • • • C3N CO C— o • J- lA J- • -J- • NO J- • NO NO • UN UN t— t^ • • • CO rr\ ^ '" ■^ ■^^ ^ r^ C\J • • • f- c^ r^ T- OO r- • • • J- r^ T" '" "" ^ J- T— • • • r— \A -J- CO c— -d- • • • .J^ UN J- UN UN _J • • • CO CO UN UN 1- ■<- J- t-^ J- T- CTn t~- • • • J- UN '- NO NO f»^ rr-\ UN UN CM 1^ r^ O t- OO o CJN ON OO NO CJN 1- •t- CM NO UN UN NO J- t— c^ UN '" '" NO o • • CM UN m 03 -p 03 a C3 C3 03 ^ • +J (J O 03 ■P C f^l o m 03 O & O f^l 03 O <-l O m 03 O OO NO OO o UN CM f»N CM 03 m ft 03 o •H -P CD 0) > r^ r^ 03 0) s • 0) 03 ^ S -p O ^ S +3 +^ j3 • HJ +> o o & tjj O <-< O rH o a 3 o . o rH «»1 OO • J- o • J- NO CO J- • • CM J- UN CO UN NO rr\ PhI 03 & O •a o Nos. 3-4 Systematics of Crawfishes 69 (avg. 33.70), for females was 23.9-40.0 (avg. 33.23) and 23.2-37.9 (avg. 33.28) for Form II males. Form I males had a rostrum length of 4.44-5.56 (avg. 4.82), females 4.66-6.33 (avg. 5.23) and males, Form II, 4.57-4.93 (avg. 4.73); there was significant sexual di- morphism in this characteristic. The rostrum was 1.13-1.54 (avg. 1.36) times longer than wide in Form I males; 1.10-1.36 (avg. 1.27) in females and 1.28-1.46 (avg. 1.36) in Form II males. /u"eola length values for Form I males were 2.24-2.50 (avg. 2.40), 2.25-2.86 (avg. 2.45) for females and 2.34-2.86 (avg. 2.52) for Form II males. The antennal scale length for Form I males was 6.30-9.11 (avg. 8.03), for females 7.47-9.41 (avg. 8.35) and 7.73-9.00 (avg. 8.31) for Form II males. As expected, sexual dimorphism existed in the characteristics of the chela. Chela length in Form I males was 1.36-1.53 (avg. 1.41), in Form II males 1.38-2.16 (avg. 1.65) and in females 1.50-2.01 (avg. 1.76); chela width was 2.18-2.65 (avg. 2.31), 1.46-2.40 (avg. 2.08) and 2.07-2.88 (avg. 2.26), respectively. Form I males had an inner palm length of 2.48-2.82 (avg. 2.60) and a dactyl length of 1.62-1.83 (avg. 1.73). In Form II males the same values were, respectively, 1.67-2.84 (avg. 2.46) and 1.07-1.72 (avg. 1.46); and in females they were 2.42-2.73 (avg. 2.57) and 1.61-1.83 (avg. 1.74). Pleopod length in Form I males was 2.18-3.67 (avg. 3.35) and 3.27-3.67 (avg. 3.48) in Form II males. Most morphological variations are within the limits set by the types, but some discus- sion of the populations around Muldon, Monroe County, Mississippi, seems neces- sary. This is the principal population on which I base my conclusion that vesticeps and hagenianus are conspecific, despite some striking morphological differences, most obvious of which are setiferous ornamenta- tions. In one collection (USNM 57278, 2 661, 5 99)i all of the characteristics are of vesticeps except that the chelae are not bearded in one male, and the lateral margin of the ischium of the third maxilliped in all specimens is poorly provided with setae. In the females the postannular sternite is flat and trapezoidal in four and has a conical protuberance in the other. In one male the central projection is enveloped basally by a fold as in hagenianus on one pleopod, while on the other pleopod the composition is as in vesticeps. In USNM 44746, of 90+ speci- mens, there is one female with a postannular sternite and maxilliped setation intermediate between hagenianus and vesticeps; in two the sternite is hagenianus-like and the maxil- liped is vesticeps-hke; in nine the sternite- maxilliped characterizations are the reverse. Among the Form I males, seven have the setation of the ischium of the maxilliped and of the mesial margin of the hand much less dense than typical vesticeps. The remaining specimens are within the limits of variability typical of vesticeps. For morphometric com- parisons a total of 130 specimens (55 661, 4 6611, 71 99) of the intergrade population were used. Specimens examined. - MISSISSIPPI: "NE Mississippi," 10 April 1911 (4 66l, 3 99); Chicka- saw Co., Egypt, no date (6 66l, 3 66ll, 6 99), Okalona (7 66l, 1 6ll): Monroe Co., Slough at jet. St. Rtes. 8 and 25, 1 April 1966 (1 dimm.); Ponto- toc Co., "Nr. Hurracine [= Hurricane] nr. Thax- ton," 1 May 1937(1 9). Intergrades (P. h. hagenianus x P. h. vesticeps). - MISSISSIPPI: Monroe Co., Muldon, July, 1912 (3 6<^, 6 99), Fall, 1922 (1 (5ll, 6 99), no date (1 dl), (6 66l, 1 9), (1 dl, 1 9), (1 61, 2 99); no locality, no date {66[, 66[\, 99, 90+ specimens). Remarks. - Lyle (1937:44-45) reported: "The wn-iter attempted to cross this crawfish [C. h. evansi {=P. h. vesticeps)] with C. hagenianus but was unable to get either species to mate in .captivity." One must, however, consider this in light of the fact that none of his attempts at laboratory mating o( any species was successful. Etymology. — From vesticeps (L.): bearded, arrived at puberty; an allusion to the setiferous bearding characteristic of the hand and maxilliped in this subspecies. Procambarus (Girardiella) barbiger n. sp. Figures 37-52 Camharus (Girardiella) hagenianus forrestae Lyle, 1938:76, nom. nud. Procambarus hagenianus. — Fitzpatrick, 1968:37 (part). Procambarus (Girardiella) hagenianus. — Hobbs, 1972b:47 (part), 151 (part), 154 (part) (by implication). Procambarus (Girardiella) sp. B. - Fitzpatrick, 1975:385,386,387, 388. 70 Tulane Studies in Zoology and Botany Vol. 20 Diagnosis. — Body pigmented, eyes small but well -developed. Rostrum with gently converging margins, lacking marginal spines; acumen short, indistinctly delimited basally. Areola 38.32-43.87% of entire length of carapace (avg. 41.81); areola linear. Carapace devoid of cervical spines or tubercles. Sub- orbital angle obsolete. Postorbital ridges lacking spines or tubercles cephalically. Antennal scale 2.40-2.71 (avg. 2.69) times longer than wnde, widest near midlength, thickened lateral portion terminating in strong, acute spine. Mesial margin of palm provided with dense setiferous beard, tuber- cles obscure; movable finger with small tuft of stout setae and strong tubercle distal to it on proximal fourth. Opposable margin of immovable finger with row of five promi- nent tubercles in proximal half, two in distal half, distal half with crowded minute den- ticles; opposable margin of movable finger with three stout tubercles, distal third with crowded minute denticles. Ischia of third pereiopods only with stout spine; no promi- nence or bosses on coxae of third through fifth pereiopods. Inner ramus of uropod with two conspicuous spines protruding beyond distal margin. First pleopods sym- metrical, shoulder present at base of central projection; pleopods reaching caudal margin of coxae of third pereiopods when abdomen flexed; distal extremity bearing (1) promi- nent, subacute mesial process directed distal- ly and sHghtly mesially, extending distally beyond other terminal elements; (2) well- developed central projection with apical portion bent slightly mesially, centro- cephalic process enveloped at base by fold continuous with centro-caudal process; and (3) conspicuous, well-developed, subtri- angular (in lateral aspect) caudal process extending distally subparallel to central projection for 80% of length of central projection, caudodistal margin lamelliform. Annulus ventralis of female very deeply ex- cavate cephaUcally with strong cephalo- lateral ridges terminating (ventrally) in three to four tubercles and set off cephalically by groove; deep sinus originating in sulcus near center of annulus and describing gentle arc to caudal margin; postannular sternite in shape of triangular prism. Holotypic male, Form I. — Body sub- ovate, distinctly compressed. Abdomen narrower than thorax (11.6 and 16.1 mm). Width of carapace at caudodorsal margin of cervical groove less than height (16.1, 19.0 mm). Areola 42.38% of total length of cara- pace and linear; cephalic section of carapace 1.4 times as long as areola (Fig. 39). Ros- trum sharply depressed (Fig. 43), deeply excavate dorsally, vAth thickened lateral margins slightly converging cephalically, marginal spines lacking, acumen indistinctly dehmited basally; upper surface sparsely punctate, usual submarginal row of setifer- ous punctations present. Subrostral ridges moderately well- developed to level of sub- orbital angle and visible in dorsal aspect to midlength of rostrum. Branchiostegal spine small but acute. Carapace punctate dorsally and laterally; granulate cephalolaterally with granules best developed just posterior to sub- orbital angle. Cervical spines or tubercles absent. Cephalic section of telson with two spines in each caudolateral corner (Fig. 51). Cephalic portion of epistome subtriangular in outline, about as broad as long, lacking Figures 19-36. Procambarus (Girardiella) hagenianus vesticeps n. subsp.: 19, mesial view of first pleopod of holotypic male, Form I; 20, mesial view of first pleopod of morphotypic male, Form II; 21, terminal elements of first pleopod of holotype, mesial view; 22, terminal elements of first pleopod of holotype, cephalic view; 23, dorsal view of carapace of holotype ; 24, terminal elements of first pleopod of holotype, lateral view; 25, terminal elements of first pleopod of holotype, caudal view; 26, lateral view of first pleopod of morphotype; 27, lateral view of first pleopod of holotype; 28, lateral view of carapace of holotype; 29, caudal view of first pleopods of paratypic Form I male; 30, mesial margin of palm of allotype; 31, proximal podomeres of third through fifth pereiopods of holotype; 32, distal podomeres of cheliped of topoparatypic Form I male; 33, cephalic portion of epistome of holotype; 34, antennal scale of holotype; 35, telson and left uropod of holotype; 36, annulus ventralis and postannular sternite of allotype. Nos. 3-4 Systematics of Crawfishes 71 19 24 ^:m 27 36 72 Tulane Studies in Zoology and Botany Vol. 20 cephalomedian tubercle (Fig. 41). Anten- nules of usual form with well-developed spine on ventral surface of basal segment slightly distal to midlength. Antennae broken, but seemingly reaching to about caudal margin of carapace; antennal scale (Fig. 50) 2.4 times longer than wide, broad- est approximately at midlength; thickened lateral portion terminating cephalically in strong acute spine about 18% of total length of scale. Ischium of third maxilliped with long, stiff, dense setae arising from mesial and ventrolateral margins. Right chela (Fig. 49) vdth palm inflated, moderately depressed; lateral margin nearly straight, not costate; entire palmar area covered with setiferous punctations; mesial margin of palm with mat of dense long setae obscuring tubercles; row of four squamous setiferous tubercles median to setal promi- nences; mesial margin of movable finger with tuft of setae in proximal fourth, prominent tubercle in proximal third distal to tuft of setae. Opposable margin of movable finger slightly excavate in basal third; with three stout tubercles in basal half, finger excavate in same region; three small tubercles distal to prominent tubercles and crowded minute denticles in distal fourth. Opposable margin of immovable finger with row of two small, one large, two small tubercles in basal half, one only in basal fourth; distal half with crowded minute denticles and with two small tubercles, one at origin of denticles, second halfway to tip. Both fingers with sub- median longitudinal ridge above and below flanked by setiferous punctations. Lower surface of palm sparsely punctate with few tubercles in distal portion. Carpus of right chelipcd longer than broad with mesial margin bearing two strong spines near midlength and two spinose tuber- cles in mesiodistal fourth; punctate dorsally with longitudinal furrow slightly mesial to midline; lower distal margin with spine in each corner, third spine halfway between mesiodistal spine and aforementioned stout mesial spines. Merus of right cheliped with dorso- median row of 14 spines, distal three acute. remainder low and rounded, almost tuber- cular; ventral surface with mesial row of 11 irregularly spaced acute spines, stoutest distalmost, and lateral row of seven irregular- ly spaced spines, distalmost stoutest. Ischium with mesial row of two spines. . Hooks on ischia of third pereiopods only; hooks simple, directed proximally, project- ing proximad to midlength of basis (Fig. 48). Coxae of third, fourth and fifth pereiopods lacking prominences or bosses. First pleopods (Figs. 37, 38, 40, 42, 45, 47) as described in "Diagnosis"; central pro- jection corneous. Uropods (Fig. 51) with mesial ramus bearing two subequal spines projecting beyond caudal margin, one from lateral corner and one from median ridge. Sternites and coxae of all pereiopods, especially i" pereiopodal segments II-V, bearing setae partially concealing pleopods when latter held under thorax. Male, Form II. — Unknown. Allotypic female. — Differing from holo- type in following respects: areola 41.44% of entire carapace length; left antennal scale (Fig. 44) with two stout flat spines originat- ing from middle of mesial margin of lamellar portion and mesiodistal margin of antennal scale. Mesial margin of palm (Fig. 46) with mesial row of seven setiferous squamous tubercles, row of six dorsomedial to them and irregularly spaced squamous setiferous tubercles over much of mesial half of upper surface; chelae proportionately shorter and less inflated; setiferous beard of ischia of third maxillipeds not as dense; pereiopodal sternites essentially lacking concealing setae. Annulus ventralis (Fig. 52) deeply exca- vate cephalically with strong ceph;Jolateral prominences extending (ventrally) one and one-half times depth of remainder of annu- lus, left terminating (ventrally) in two large and two small tubercles, right in three sub- equal tubercles; sinus originating near center of annulus, curving gently sinistrad to caudal margin; annulus delimited cephalically by groove. Postannular sternite as described in "Diagnosis." Types. - USNM nos. 146258 (holotype) and 146259 (allotype); Paratypes: Missis- Nos. 3— 4 Systematics of Crawfishes 73 sippi, Scott County (4 661, 11 99). palm; furthermore the annulus is not typical. Type-locality. — Forrest, Scott County, Neither specimen possesses the lateral row of N4is'sissippi. dense setae on the ischium of the third max- Range. - This species is known only from iUiped, but both have a postannular sternite burrows in and about Forrest, Scott Co., characteristic of barbiger. In the topotypic Mississippi, but probably occurs throughout series (4 661, 9 99), one female has a dextral- the band of Jackson Prairie running south- ly oriented sinus and less prominent tuber- eastjvard from central Mississippi through cles on the annulus than the rest of the Rankin, Scott, Smith, Newton, Jasper, specimens; in another female the tubercles Clarke and Wayne counties and possibly are placed much more cephalomesially than Madison and Hinds counties. All of the the rest; in a third specimen the tubercles are known specimens were collected from the broad, prominent and lateral; and in a fourth environs of Forrest, and I have never been the tubercular prominences seem almost able to collect the species personally despite montane in comparison. The shoulder on the repeated efforts. I have, however, attempted cephalic surface of the first pleopod of one the excavation of burrows which were male is much more sharply angular than in probably made by this species near Forrest, the remaining specimens, and the environmental situation in which Morphometric analysis was based on 15 these burrows occurred was nearly identical specimens (4 661, 11 99). Carapace length with that in which P. hagenianus is found— was 24.0-36.1 (avg. 29.15) in Form I males large open fields relatively remote from and 28.5-36.3 (avg. 32.17) in females. In watercourses. Lyle (1937:48) recorded the males, Form I, rostrum length was 3.87-4.51 species (as P. h. forrestae) from "Leesburg, (avg. 4.27) and 4.16-5.00 (avg. 4.48) in Rankin County," but I have been unable to females. The rostrum was 1.29-1.55 (avg. locate the specimens on which this record is 1.45) times longer than wide in Form I based. The only "Rankin" specimens I have males and 1.16-1.76 (avg. 1.43) in females, seen are USNM no. 93101 ("near Jackson The areola length was 2.36-2.43 (avg. 2.39) [Hinds or Rankin Co.], leg. Carlysle Carr, in Form I males and 1.77-2.61 (avg. 2.42) in May, 1936") which are unquestionably females. Antennal scale values were available assignable to P. h. hagenianus. Mr. Carr, an only for females, and they were 7.68-7.80 employee of the U. S. Biological Survey, (avg. 7.74) for length and 2.67-2.71 (avg. worked intimately with the state Fish and 2.69) for width. Chela length was 1.47-1.50 Wildlife Service, and I beheve that the locali- (avg. 1.48) in Form I males and 1.69-1.84 ty appearing on the label of these specimens (avg. 1.77) in females. Chela width was reflects that the specimens were received 2.17-2.40 (avg. 2.28) in Form I males and from Jackson, on the site of the Fish and 2.03-2.15 (avg. 2.09) in females. In Form I Wildlife Service Museum. I offer in support males length of the mesial margin of the of this thesis the fact that no other speci- palm and dactyl length were, respectively, mens resembling the subspecies have been 2.67-2.80 (avg. 2.75) and 1.66-1.72 (avg. collected or reported from the vicinity of 1.69), while the same two characters in Jackson. females were 2.06-2.68 (avg. 2.46) and Color. - I have never seen fresh speci- 1.61-1.73 (avg. 1.67). Pleopod length in the mens personally, but Lyle (1937:48) gives a porm I males was 3.08-3.84 (avg. 3.44). limited description: "color is rather variable Specimens examined. - MISSISSIPPI: Scott but blue apparently predominates." Co., Forrest, 2 May 1936 (4 66[, 9 99), March, Variation. — Most of the limits of varia- 1934 (2 99). tion are within the Umits set by the primary Remarks. - In the topotypic series, col- types, but in a collecrion of two females lected 2 May 1936, three of the nine females (USNM no. 146273) one specimen is typical had a sperm plug in the annulus ventralis. of barbiger, and the other lacks the hirsute Etymology. - From barbiger (L.): wear- ornamentation of the mesial portion of the ing a beard; so named because of the hirsute 74 Tulane Studies in Zoology and Botany Vol. 20 nature of the mesial margin of the hand in both sexes and the bearding of the ischium of the third maxilliped. Procanibarus (Girardiella) cometes n. sp. Figures 53-71 Procanibarus (Girardiella) liageiiianns. — Hobbs, 1972b:47 (part), 151 (part), 154 (part) (by implication). Procanibarus (GirardicUa) sp. E. - Fitzpatnck, 1975:385,386,387,388. Diagnosis. — Body pigmented, eyes small but well -developed. Rostrum with slightly converging margins, lacking marginal spines; acumen short and indistinctly delimited basally. Areola linear and 38.77-43.24 (avg. 40.01)% of total length of carapace. Cara- pace devoid of cervical spines or tubercles. Suborbital angle absent. Postorbital ridges without spines or tubercles. Antennal scale 2.33-2.87 (avg. 2.51) times longer than wide, broadest near midlength, thickened lateral portion terminating cephalically in long, acute, stout spine. Mesial margin of palm with surface obscured by dense mat of setae in first form male, second form male with numerous tufts of short stiff setiferous bristles, female with single row of five or six tubercles and sparse setiferous squamous tubercles; basal portion of movable finger excavate along opposable margin with row of five or six tubercles in basal two-thirds; opposable margin of immovable finger v^th row of seven or eight tubercles in basal two- thirds and single tubercle at base of distal fourth. Ischia of third pereiopods only bear- ing simple hooks; coxae of third, fourth and fifth pereiopods lacking prominences or bosses. Inner ramus of uropod with two con- spicuous spines projecting distally beyond margin. First pleopods symmetrical, reaching caudal margin of coxa of third pereiopod when abdomen flexed; small shoulder present at base of central projection; distal extremity bearing (1) prominent central projection directed more or less latero- distally (2) subacute mesial process directed laterodistally, projecting only slightly beyond tips of other elements; and (3) sub- rhomboidal (in lateral aspect) caudal process excavate mesially, lamelliform in caudodistal half. Annulus ventralis of female deeply ex- cavate cephalically with prominent cephalo- lateral ridges terminating (ventrally) in tubercles, sulcus relatively narrow (only 20-25% width of annulus), annulus set off cephalically by shallow groove; sinus origi- nating near center of annulus and passing in gentle arc to caudal margin; postannular sternite subconical in shape. Holotypic male, Form I. - Body sub- ovate, distinctly compressed. Abdomen nar- rower than carapace (13.3 and 17.1 mm). Carapace higher than wide at caudodorsal margin of cervical groove (18.5, 17.1 mm). Areola 38.87% of total length of carapace and linear (Fig. 57). Cephalic section of cara- pace 1.6 times as long as areola. Rostrum slightly depressed (Fig. 62), deeply excavate dorsally with thickened lateral margins gently convergent distally, marginal spines lacking, acumen indistinctly delimited basal- ly; upper surface sparsely punctate, usual row of submarginal setiferous punctations present. Subrostral ridges moderately well- developed from suborbital level and visible to about midlength of rostrum in dorsal aspect. Postorbital ridges moderately promi- nent, grooved dorsolaterally, and termina- ting cephalically without spines or tubercles. Suborbital angle lacking. Branchiostegal spine reduced. Carapace very sparsely punc- tate but with squamous tubercules laterally, tubercles most numerous in cephaloventral portion. Cervical spines or tubercles absent. Telson (Fig. 70) divided vdth two spines in right caudolateral corner and three in left. Cephalic portion of epistome (Fig. 68) about as broad as long with tubercular mesioce- phalic projection. Antennules of usual form with well-developed spine on ventral surface of basal segment slightly distal to midlength. Antennae extending about to caudal margin of carapace; antennal scale (Fig. 69) 1.57 times longer than wide, broadest distal to midlength, thickened lateral portion termi- nating cephalically in strong acute spine approximately 23% of total length of anten- nal scale, cephalic margin of lamellar portion provided with relatively few tubercular eminences bearing tufts of setae. Nos. 3-4 Systematics of Crawfishes 75 Ischium of third maxilliped with promi- nent, stiff, simple, unmatted setae running along entire mesial and ventrolateral margin. Right chela (Fig. 67) with palm inflated, not strongly depressed; lateral margin not costate; upper and lower surfaces punctate except along mesial margin; mesial margin of palm almost devoid of tubercles but pro- vided with dense mat of long, stiff setae; setae in distinct tufts arising from large, deep punctations. Opposable margin of fixed finger with row of three small, one large, and three small tubercles in basal two-thirds, single large tubercle directed ventromesially from base of distal tourth, distalmost 10% broken, but with crowded minute denticles distal from tubercle; submedian longitudinal ridge flanked by setiferous punctations above. Opposable margin of dactyl excavate in basal half with row of three small, one large and two small tubercles along basal two-thirds, crowded minute denticles in distal fourth; mesial margin with two evenly spaced setiferous tubercles in basal fourth; submedian longitudinal ridge above flanked by setiferous punctations. Carpus of right cheliped longer than wide, with mesial margin bearing stout acute spine in distal fourth, small acute spine in mesio- distal corner: remainder of upper surface punctate, but with longitudinal furrow in distalmost three-fourths; lower surface with strong acute spines in each distalmost corner. Merus of right cheliped with row of ten tubercles on upper surface and two acute spines terminating row distally; lower sur- face with mesial row of 14 acute spines increasing in size distally, and lateral row of nine smaller acute spines likewise increasing in size distally. Ischium with mesial row of two small tubercles. Hooks on ischia of third pereiopods only (Fig. 65); hooks simple, directed proximally and projecting proximally to distal fourth of basis. Coxae of third, fourth and fifth pereiopods lacking prominences or bosses. First pleopods symmetrical and as de- scribed in "DiagYiosis"; central projection and caudal process corneous (Fig. 53, 55, 56, 58, 59,61,66). Uropods (Fig. 70) with mesial ramus bearing two spines projecting beyond caudal margin, one from lateral corner, and other, slightly longer, from median ridge. Sternum between coxae of second through fifth pereiopodal segments slightly excavate, lateral margins bearing sparse tufts of setae only partially obscuring first pair of pleopods when pleopods held beneath thorax. Morphotypic male, Form II. — Differing from holotype in following respects: dorsal punctation of carapace present; tip of im- movable finger of chela unbroken and crowded minute denticles occurring to level of corneous acute tip; mesial margin of palm (Fig. 63) lacking dense bearding, but pro- vided with numerous tufts of shorter setae clearly arising from deep large setiferous punctations; carpus of cheliped bearing small acute spine between mesial spine and ventro- mesial spine on both chelipeds. First pleo- pods (Figs. 54, 60) with tips non-corneous, proportionally less developed, and central projection with fold at base wrapping around mesial, cephalic and lateral portions, mesial process extending proportionately farther beyond tip of other elements. Only two spines in left caudolateral margin of cephalic portion of telson. Hooks on ischia of third pereiopods much less pronounced than in holotype. Allotypic female. — Differing from holo- type in following respects: cephalic portion of epistome subtrapezoidal and broader than long; areolar region with transverse crease pre-mortem injury and only 33.02% of total length of carapace; mesial margin of chela (Fig. 64) lacking beard, bearing instead five tubercles, some of which provided with quite light setiferous tufts; distalmost por- tion of immovable finger similar to that of morphotype; small acute spine between spine and ventrodistal spine of carpus as in morphotype. Annulus ventralis (Fig. 71) about as broad as long, with deep subtrapezoidal ex- cavation in centrocephalic portion, cephalo- lateral margins markedly raised (ventrally) with three tubercles on left ventralmost margin and four on right; sinus originating near center, forming arc sinistrally and be- 76 Tulane Studies in Zoology and Botany Vol. 20 coming lost in caudal margin near midline; postannular sternite subconical in shape. Types. - USNM nos. 130227, 146260, and 131280 (holo-, alio-, and morphotype, respectively); Paratypes: Oktibbeha (2 661, 9 99) and Lowndes (1 61) counties, Missis- sippi. Type-locality. — Field behind Luxury Mobile Homes (T18N, R14E, SWA, Sec. 3), Starkville, Oktibbeha County, Mississippi. Range. — The species is known from only two localities in Oktibbeha County (the type locality and 8.8 mi S of Starkville, jet. St. Rte, 25, on St. Rte. 12) and one in Lowndes County (6.6 mi E of Old State Route 12 in Starkville). It seems to be associated with the Flatwoods belt in Oktibbeha Co., but the specimen from Lowndes Co. came from the Black Belt. Color. — P. cometes resembles P. hageni- anus in color— dark blue ground color with cream markings— but I have never seen a "red" phase. This species was unknown to Lyle, therefore he left no color notes. Variation. — Most of the Umits of vari- ation are incorporated in the diagnosis and descriptions above. In four female specimens from south of Starkville (USNM no. 146274), the mesial margin of the hands is provided with sparse, widely scattered tufts of setae not markedly different from the condition found in P. barbiger; an associated Form I male is typical of the species. The extremely short areola (comparatively) of the allotype seems to be unique. Morphometric variation analysis is based on 14 specimens (4 661, 10 99). Carapace length in males was 29.6-37.3 (avg. 33.30) and 31.8-40.7 (avg. 35.54) in females. In males the rostrum length was 4.05-4.63 (avg. 4.24) and in females 3.984.69 (avg. 4.35), while rostrum width values were 1.33-1.66 (avg. 1.54) and 1.34-1.67 (avg. 1.46), re- spectively. Areola length was 2.45-2.58 (avg. 2.54) in males and 2.31-3.03 (avg. 2.54) in females. Antennal scale length was 7.65-8.05 (avg. 7.75) in males and 7.39-8.85 (avg. 8.12) in females. Chela length in males was 1.39-1.66 (avg. 1.51) and in females was 1.71-1.92 (avg. 1.81); chela width was 2.23-2.46 (avg. 2.34) in males and 2.05-2.47 (avg. 2.26) in females. Length of the mesial margin of the palm and length of the dactyl were, respectively, in males 2.79-2.97 (avg. 2.88) and 1.52-1.70 (avg. 1.64); in females the same characters in the same sequence were 2.59-3.01 (avg. 2.77) and 1.56-1.90 (avg. 1.66). Pleopod length in males was 3.65-3.89 (avg. 3.72). Specimet2s examined. — MISSISSIPPI: Lowndes Co., 6.6 mi. E of Starkville, 12 July 1968 (1 dl); Oktibbeha Co., Luxury Mobile Homes (T18N, R14E, SWy4 Sec. 3), Starkville, no date (1 9), (1 6l, 1 9), (1 6l), (1 9), (3 99), 8.8 mi. S of Starkville on St. Rte. 12, no date (1 c^, 4 99). Etymology. — From cometes (L.): one with long hair; used with reference to the long setae on the mesial margin of the palm of males and the long setae on the ischium of the third maxilliped. Procambarus (Girardiella) connus n. sp. Figures 72 - 90 Procambarus hageyiiayius. — Fitzpatrick, 1968:37 (part). Procambarus (Girardiella) hagenianus. — Hobbs, 1972b:47 (part), 151 (part), 154 (part) (by implication). Procambarus (Girardiella) sp. C. — Fitzpatrick, 1975:385,386, 387, 388. Diagnosis. — Body pigmented, eyes small but well- developed. Rostrum with gently I Figures 37-52. Procambarus (Girardiella) barbiger n. sp.: 37, mesial view of first pleopod of holotypic male, Form I; 38, teminal elements of first pleopod of holotype, cephalic view; 39, dorsal view of carapace of holotype; 40, terminal elements of first pleopod of holotype, caudal view; 41, cephalic portion of epistome of holotype; 42, lateral view of first pleopod of holotype; 43, lateral view of carapace of holotype; 44, left antennal scale of paratypic female; 45, terminal elements of first pleopod of holotype, mesial view; 46, mesial margin of palm of allotype; 47, terminal elements of first pleopod of holotype, lateral view; 48, proximal podomeres of third through fifth pereiopods of holotype; 49, distal podomeres of cheliped of holotype; 50, antennal scale of holotype; 51, telson and left uropod of holo- type; 52, annulus ventralis and postannular sternite of allotype. Nos. 3-4 Systematics of Crawfishes 77 78 Tulane Studies in Zoology and Botany Vol. 20 converging margins, lacking marginal spines, acumen short and indistinctly delimited basally. Areola constituting 38.39-44.79 (avg. 41.41)% of total length of carapace; areola linear. Carapace devoid of cervical spines or tubercles. Suborbital angle obso- lete. Postorbital ridges without spines or tubercles. Antennal scale 2.00-2.88 (avg. 2.44) times longer than wide, thickened lateral portion terminating cephalically in stout, comparatively short spine, broadest slightly distal to midlength. Mesial margin of palm with setiferous mat of hairs obscuring tubercular ornamentation. Opposable margin of immovable finger with row of eight tubercles in basal three-fourths, third from base largest; opposable margin of dactyl slightly excavated in basal third, with row of seven tubercles, first and fourth larger than remainder. Ischia of third pereiopods only bearing simple hooks; coxae of third, fourth and fifth pereiopods lacking prominences or bosses. Inner ramus of uropod with two con- spicuous spines projecting distally beyond margin. First pleopods very slightly asym- metrical, slight shoulder at base of central projection; distal extremity bearing (1) prominent central projection directed latero- distally, tightly applied fold continuous with centrocaudal process enveloping base of centro-cephalic process; (2) subacute mesial process directed caudal subparallel to main axis of shaft, extending length of central projection beyond other terminal elements; and (3) subrectangular (in lateral aspect) caudal process subparallel to mesial plane of pleopod with caudodistal margin lamel- liform. Female with rather simple (in com- parison with relatives) annulus ventralis; annulus deeply excavate in centrocephalic portion with prominent lateral prominences (ventrally), prominences lacking tubercular ornamentation; annulus set off cephalically by groove; sinus originating slightly anterior to center of structure, curving sinistrally to be lost in caudal fourth; postannular sternite sub rhomboid in outline, not highly elevated. Holotypic male, Form I. — Body sub- ovate, distinctly compressed. Abdomen narrower than carapace (10.3 and 17.1 mm). Width of carapace at caudodorsal margin of cervical groove greater than height (17.1, 14.1 mm). Areola 41.24% of total length of carapace; areola linear (Fig. 76). Cephalic section of carapace 1.4 times as long as are- ola; rostrum depressed, deeply excavated dorsally, with gently converging thickened lateral margins, marginal spines lacking, acumen indistinctly delimited basaUy; upper surface not punctate except for usual sub- marginal row of setiferous punctations. Sub- rostral ridges moderately well -developed from level of suborbital angle and visible dorsally to approximately midlength of rostrum. Postorbital ridges moderately well developed, grooved cephalolaterally, termi- nating cephalically without spines or tubercles. Suborbital angle obsolete (Fig. 81). Branchiostegal spine small, but acute. Carapace very sparsely punctate dorsally but with squamous tubercles on cephalolateral portion. Cervical spines or tubercles absent. Cephalic section of telson (Fig. 88) with three spines in right caudolateral corner and two in left. Cephalic portion of epistome (Fig. 90) subovate in outline, lacking tuber- Figures 53-71. Procambarus (Girardiella) cotnctcs n. sp.: 53, mesial view of first pleopod of holotypic male. Form I; 54, mesial view of first pleopod of morphotypic male, Form II; 55, terminal elements of first pleopod of holotypc, mesial view; 56, terminal elements of first pleopod ofholotype, cephalic view; 57, dorsal view of carapace of holotype; 58, terminal elements of first pleopod ofholotype, lateral view; 59, terminal elements of first pleopod of holotype, caudal view; 60, lateral view of first pleopod of morphotype; 61, lateral view of first pleopod of holotype; 62, lateral view of carapace of holotype; 63, mesial margin of palm of morphotype; 64, mesial margin of palm of allotype; 65, proximal podomeresof third through fifth pereiopods of holotype; 66, caudal view of first pleopods of holotype; 67, distal podomeres of cheiipcd of holotype; 68, cephalic portion of epistome of holotype; 69, antennal scale of holotype; 70, telson and left uropod of holotype; 71, annulus ventralis and postannular sternite of allotype. Nos. 3-4 Systematics of Crawfishes 79 80 Tulane Studies in Zoology and Botany Vol. 20 cular median protrusion. Antennules of usual form with well-developed spine on ventral surface of basal segment slightly distal to midlength. Antennae extending caudad approximately to caudal margin of carapace. Antennal scale (Fig. 83) 2.07 times longer than wide, widest slightly distal to midlength, thickened lateral portion termi- nating cephalically in very stout spine approximately 21% of total length of an- tennal scale; cephalic half of margin of mesial lamellar portion provided with numerous tubercular eminences bearing tufts of setae. Ischium of third maxilliped with promi- nent simple, stiff, unmatted setae along mesial and ventrolateral margins. Right chela (Fig. 85) with palm inflated, not strongly depressed; lateral margin not costate; entire palmar area covered with setiferous punctations; mesial margin with row of five prominent tubercles and second row of four above, both rows nearly ob- scured by beard of setae in tufts arising from deep punctations. Opposable margin of im- movable finger excavate in basal third; with row of three small, one large and three small tubercles in basal two-thirds, another tuber- cle at level of basal fourth, except for afore- mentioned tubercle distal one-third with crowded minute denticles; fmger with sub- median longitudinal ridge above and below, both flanked by setiferous punctations. Opposable margin of dactyl with row of one large, two small, one large and three small tubercles in basal two-thirds, with crowded minute denticles beginning just distal to penultimate tubercle and extending to tip of finger, slightly excavate in proximal third; mesial margin with two tubercles in basal one-fourth (distalmost largest); submedian longitudinal ridge above, and less distinct one below, both flanked by setiferous punc- tations. Carpus of right cheliped longer than broad; v^dth mesial margin bearing stout acute spine at base of distal third, two small but acute spines, equally spaced, proximal to aforementioned spine, upper mesiodistal margin and corner with three small spines; longitudinal groove above; small acute spine in each ventrodistal corner with row of two acute spines between mesiodistal spine and stout spine of medial surface. Merus of right cheliped with two stout spines on upper surface near distal margin; mesioventral margin with row of 12 acute spines, distalmost largest, and ventrolateral margin with row of six, increasing in size distally. Ischium with mesial row of three acute spines. Hooks on ischia of third pereiopods only (Fig. 86); hooks simple, directed proximally and projecting to about midlength of basis. Coxae of third, fourth and fifth pereiopods lacking prominences or bosses. First pleopods (Fig. 72, 74, 75, 77, 78, 80, 82) slightly asymmetrical and as de- scribed in "Diagnosis "; central projection and caudal process corneous. Uropod (Fig. 88) with protopodite with small acute spine; mesial ramus with two spines projecting beyond distal margin, one from lateral angle and second of approxi- mately same size from median ridge. Sternum between third, fourth and fifth Figures 72-90. Procambams (Girardiella) connus n. sp.:72, mesial view of first pleopod of holotypic male. Form I; 73, terminal elcinents of first pleopod of morphotypic male, Form II; 74, terminal elements of ' first pleopod of holotype, mesial view; 75, terminal elements of first pleopod of holotype, cephalic view; 76, dorsal view of carapace of holotype; 77, terminal elements of first pleopod of holotype, lateral view; 78, terminal elements of first pleopod of holotype, caudal view; 79, lateral view of first pleopod of morphotype; 80, lateral view of first pleopod of holotype; 81, lateral view of carapace of holotype; 82, caudal view of first pleopods of holotype; 83, antennal scale of holotype; 84, mesial margin of palm of allotype; 85, distal podomeres of cheliped of holotype; 86, proximal podomeres of third through fifth pereiopods of holotype; 87, annulus ventralis and postannular sternite of allotype; 88, telson and left uropod of holotype; 89, mesial view of aberrant terminal elements of paraty pic male, Form I (c = central projection, m = mesial process); 90, cephalic portion of epistome of holotype. Nos. 3-4 Systematic s of Crawfishes 81 82 Tulane Studies in Zoology and Botany Vol. 20 pereiopods moderately deep and bearing long setae partially obscuring first pleopod when held under thorax. Morphotypic male, Form II. — Differing from holotype in following respects: mesial margin of left cheliped not so densely bearded as holotype; carpus of cheliped with additional spine along mesial margin (three total); right cheliped lacking beard, with only three tubercles in mesial row on margin of palm; latter apparently regenerated; cephalic portion of telson with only two spines in right caudolateral corner. Terminal elements of first pleopod (Fig. 73, 79) not corneous and less prominent than in holo- type. Hooks of ischia of third pereiopods reduced markedly, almost tubercles. Allotypic female. — Differing from holo- type in following respects: mesial margin of palm (Fig. 84) with mesial row of seven spin- iform tubercles and second, more median row of six squamous tubercles, margin pro- vided with several squamous setiferous tubercles but no beard evident. Cephalic section of telson with only two spines in right caudolateral corner; sternites associated with third to fifth pereiopods not so deeply excavate, likewise, setae much reduced. Annulus ventralis (Fig. 87) with broad centrocephalic excavation and set off cephalically by groove; prominent, but smooth cephalolateral ridges rising (ventral- ly) above sulcus, right continuing as lower ridge to midcaudal line; sinus originating cephalolateral to midline and arcing sinistral- ly to be lost in caudal fourth of annulus. Postannular sternite subtrapezoidal in out- line and not highly elevated. Types. - USNM nos. 146261, 146262, and 146263 (holo-, alio, and morphotype, respectively); Paratypes: Carroll County, Mississippi (47 661). Paratypes at USNM. Type -locality. — Carrollton, Carroll County, Mississippi. Range. — P. connus is known only from the environs of Carrollton, Carroll County, Mississippi. This species is the only member of this group found associated with loess, being located in an area of brown loam and thick loess. Although Carrollton is in the Red Hills, remnants of the mantle do occur in the vicinity, too close to be ignored. Color. — No color records for this species exist. Variation. — Morphometric analysis was based on 44 Form I males. Carapace length was 26.6-41.0 (avg. 34.15). Rostrum length was 3.45-6.34 (avg. 5.31) and width was 1.08-1.58 (avg. 1.38). The length of the are- ola was 1.71-2.57 (avg. 2.40) and the length of the antennal scale was 6.05-9.31 (avg. 8.24). Chela length was 0.99-1.47 (avg. 1.37) and chela width was 2.09-2.59 (avg. 2.35). Length of the mesial margin of the palm and length of the dactyl were 2.38-2.82 (avg. 2.63) and 1.40-4.65 (avg. 1.81), respectively. Pleopod length was 2.51-3.84 (avg. 3.35). In the topotypic series, one male of 17 had the mesial margin of the palm of the left cheliped almost devoid of setae; the right cheliped of this specimen and all the re- maining specimens were like the holotype. In a second paratypic collection, one of 14 males had both chelipeds devoid of seti- ferous ornamentation (thus resembling hagenianiis), and the ischia of the third max- illipeds were likewise devoid of dense setifer- ous beards. In the third paratypic series one of 17 males had a clearly regenerated hand which was lacking the beard of the inner margin of the palm; two had a straight mesial process. Otherwise, variation fell within the limits established in the diagnosis and descriptions outlined above. In one male. Form I, the terminal elements of the left first pleopod were markedly aberrant (Fig. 89); whether or not this was the result of an injury could not be ascertained. Specimens examined. - MISSISSIPPI: Carroll Co., Carrollton, no date (17 66\, 1 dll, 1 9), (14 66\), (17 66\). Remarks. — P. connus seems almost as closely related to P. h. hagenianus as is P. h. vesticeps. Indeed, I would say as closely were it not for the fact that apparent inter- grade populations exist between the latter two. I find no suggestion that a zone of intergradation exists between P. hagenianus and P. connus. Therefore, I have taken the more conservative position, designating connus a species. Should intergrade popula- tions be discovered subsequently, a minor Nos. 3-4 Systematics of Crawfishes 83 nomenclatorial change will rectify the situa- tion. Although the degree of relationship existing between P. h. liagenianiis, P. h. vesticeps and P. connus is not as precisely delimited as one might wish, each popula- tion represents a morphologically and geographically distinct entity as indicated more thoroughly in the "Discussion" section following. Etymology . — The name of this species is taken from konnos (Gr.): a beard: this is in reference to the bearding of the mesial margin of the palm in males and the beard- ing of the third maxilliped. Procambarus (Girardiella) pogum n. sp. Figures 91-108 Cambarus (Girardiella) hagcuianus carri Lyle, 1938:76, now. ymd. Procambarus hagenianiis. — Hobbs, 1968:K25 (Fig. 19c): Fitzpatrick, 1968:37 (part). Procambarus (Girardiella) hagenianus. — Hobbs, 1972b:47 (part), 151 (part), 154 (part) (by im- plication). Procambarus (Girardiella) sp. D. — Fitzpatrick, 1975:385, 386,387, 388. Diagnosis. — Body pigmented, eyes small but well- developed. Rostrum with sub- parallel to gently converging margins, lacking marginal spines: acumen short, indistinctly delimited basally. Areola 38.36^1.72 (avg. 39.69)% of entire length of carapace: areola linear. Carapace devoid of cervical spines or tubercles. Suborbital angle lacking. Post- orbital ridges lacking spines or tubercles cephalically. Antennal scale 2.21-3.08 (avg. 2.62) times longer than wide, widest distal to midlength, thickened lateral portion term- inating in strong, acute spine. Mesial margin of palm provided with dense setiferous beard, tubercles obscure: movable finger with relatively dense tuft of setae in basal third, tuft obscuring three or four small tubercles. Opposable margin of immovable finger with row of one large and one small tubercle in basal third and third tubercle at base of distal third, distal third with crowd- ed minute denticles: opposable margin of movable finger with row of two small, one large and one small tubercle in basal half, small tubercle at base of distal third with crowded minute denticles distal to it. Ischia of third pereiopods only with stout spine ; no prominences or bosses on coxae of third through fifth pereiopods. Mesial ramus of uropod with two conspicuous spines pro- truding beyond distal margin. First pleopods asymmetrical, shoulder weak at base of cen- tral projection; pleopods reaching midlength of coxae of third pereiopods when abdomen flexed; distal extremity bearing (1) promi- nent, subacute mesial process directed distal- ly and slightly mesially, curved caudally in distal half, extending beyond other terminal elements; (2) well-developed central pro- jection with apical portion directed latero- distally; and (3) prominent subrhombiform (in lateral aspect) caudal process extending distally not quite so far as central projection, distal margin sloped gently caudad and caudodistal margin lamelliform. Annulus ventralis of female deeply excavate centro- cephalically with strong cephalo lateral ridges terminating (ventrally) without tubercles; sinus originating slightly cephalic to center of annulus, arcing gently to be lost in caudal 10% of annulus; postannular sternite broadly conical in shape. Holotypic male, Form I. — Body subo- vate, distinctly compressed. Abdomen nar- rower than thorax (9.3 and 14.1 mm). Width of carapace at caudodorsal margin of cervical groove less than height (14.1, 14.5 mm). Areola 38.46% of total length cf carapace and linear; cephalic section of carapace 1.6 times as long as areola (Fig. 95). Rostrum sharply depressed (Fig. 101), deeply exca- vate dorsally, with thickened lateral margins slightly converging cephalically, marginal spines basally; upper surface sparsely punctate, usual submarginal row of setiferous puncta- tions present. Subrostral ridges moderately well -developed to level of antennae and visible to approximate distal third of ros- trum in dorsal aspect. Suborbital angle lack- ing. Branchiostegal spine much reduced. Carapace sparsely punctate dorsally and laterally; granulate cephalolaterally with granulations best developed just posterior to orbit and antennal scale. Cervical spines or tubercles absent. Cephalic section of telson with two spines in each caudolateral corner 84 Tulane Studies in Zoology and Botany Vol. 20 (Fig. 107). Cephalic portion of epistome (Fig. 105) subovoid in outline, about as broad as long, with small cephalomedian tubercle. Antennules of usual form with well- developed spine on ventral surface of basal segment slightly distal to midlength. Anten- nae broken, but apparently reaching to about caudal margin of carapace; antennal scale (Fig. 106) 2.73 times longer than wide, widest distal to midlength; thickened lateral portion terminating cephalically in stout , acute spine approximately 24% of total length of antennal scale; cephalic half of mesial margin of lamellar portion provided with numerous tubercular eminences bearing tufts of setae. Ischia of third maxillipeds with long, stiff, unmatted, dense setae arising from mesial and ventrolateral margins. Right chela (Fig. 104) with palm inflated, moderately depressed; lateral margin nearly straight, not costate; entire palmar area covered with setiferous punctations; mesial margin of palm with mat of dense setae obscuring tubercles; row of six squamous tubercles medial to setiferous beard; mesial margin of movable finger with tuft of setae in proximal third partially obscuring row of three low tubercles. Opposable margin of movable finger slightly excavated on basal one-third; with row of two small, one large and one small tubercle in basal one-half; small tubercle at base of distal third with crowded minute denticles between it and tip of finger; finger with submedian longitudinal ridge above and below, both flanked by seti- ferous punctations. Opposable margin of im- movable finger with row of one large, one small and two nearly inconspicuous tuber- cles in basal half; moderate sized tubercle at base of distal third directed ventromedially, crowded minute denticles between it and tip of finger; finger with submedian longitudinal ridge above and below, both flanked by setiferous punctations. Carpus of right cheliped longer than broad with mesial margin bearing single strong acute spine slightly distal to mid- length, two smaller equally spaced acute spines proximal to it, row of four squamous setiferous tubercles along middle third of mesial portion of carpus medial to afore- mentioned spines, small subacute spine in mesiodistal corner; punctate dorsally with longitudinal furrow slightly mesial to mid- line; lower distal margin with acute spines in each corner and row of three low, almost tubercular, spines between distal ventro- mesial corner and stout spine of mesial margin. Merus of right cheliped with dorsal row of six tubercles terminating distally in two acute spines (distalmost strongest); ventro- mesial margin with row of 1 3 acute spines increasing in size distally; ventrolateral margin with row of seven spines, increasing in size distally. Mesial margin of ischium with row of three small but acute spines de- creasing in size distally. Hooks on ischia of third pereiopods only (Fig. 102); hooks simple and directed proxi- mocephalically, extending just beyond distal margin of basis; coxae of third, fourth and fifth pereiopods lacking prominences or bosses. First pleopod (Figs. 91, 93, 94, 96, 97. 99, 100) as described in "Diagnosis"; central projection and caudal process corneous. Figures 91-108. Procambarus (Girardiella) pogum n. sp.: 91, mesial view of first pleopod of holotypic male, Form 1; 92, mesial view of first pleopod of morphotypic male, Form 11; 93, terminal elements of first pleopod of holotypc, mesial view; 94, terminal elements of first pleopod of holotype, cephalic view; 95, dorsal view of carapace of holotypc; 96, terminal elements of first pleopod of holotype, lateral view; 97, terminal elements of first pleopod of holotypc, caudal view; 98, lateral view of first pleopod of morphotype; 99, lateral view of first pleopod of holotype; 100. caudal view of first pleopods of holotype; 101, lateral view of carapace of holotype; 102, proximal podomeres of third through fifth pereiopods of holotype; 103, mesial margin of palm of allotype; 104, distal podomeres of cheliped of holotype; 105, cephalic portion of epistome of holotype; 106, antennal scale of holotypc; 107, telson and left uropod of holotype; 108, annulus ventralis and postannular sternite of allotype. Nos. 3-4 Syste ma tics of Crawfishes 85 '92 96 97 •04 106 102 105 108 86 Tulane Studies in Zoology and Botany Vol. 20 Uropods (Fig. 107) with mesial ramus bearing two spines projecting beyond distal margin, one from lateral corner and one, slightly larger, from median ridge. Sternites and coxae of third through fifth pereiopodal segments bearing setae partially obscuring pleopods when latter held under thorax. Morphotypic male, Form II. — Differing from holotype in following respects: mesial margin of palm with setiferous mat not so dense, but still obscuring margins complete- ly; lacking tuft of setae on mesial margin of dactyl: hooks on ischia of third pereiopods not so strong, almost tubercular. Terminal elements of first pleopod (Fig. 92, 98) non- corneous, proportionately smaller and less developed; base of central projection envel- oped by tightly applied fold on mesial, cephalic and lateral surfaces. Setiferous orna- mentation of pereiopodal sternites and coxae less well -developed. Tip ot rostrum markedly more truncate (broken early in life?). Allotypic female. — Differing from holo- type in following respects: mesial margin of palm (Fig. 103) and dactyl lacking setiferous beard, instead palm with mesial row of six tubercles and with two rows of seven each squamous setiferous tubercles median to it; row of nine spinose tubercles median to stout spine of mesial margin of carpus giving serrated appearance to upper mesial edge; spines between stout mesial spine and distal ventromesial spine of carpus strongly acute, additional row of four spines ventromesial to aforementioned spines and row of two ventromesial to stout spine of mesial margin. Annulus ventralis (Fig. 108) deeply exca- vate centrocephaHcally, sulcus sloping pre- cipitously cephalically; annulus set off cephalically by deep groove; quite large cephalolateral ridges lacking tubercles, ex- tending ventrally approximately three times depth of remainder of annulus; sinus origi- nating slightly cephalic to center of annulus and forming gentle arc sinistrally to be lost in caudal 10% of annulus; postannular ster- nite broadly subconical in shape. Types. - USNM nos. 146270, 146271, and 146272 (holo-, alio-, and morphotypes, respectively); Paratypes: Chickasaw County, Mississippi (2 661, 4 6611, 999). Paratypes are located at the National Museum of Natural History. Type-locality. — Houston, Chickasaw County, Mississippi. Range. — The species is known only from burrows at the type-locality and from 0.4 mi E of Houlka Creek where the animals were taken from burrows in a roadside ditch. The latter, the only precise locality, is associated with the Ripley formation. Apparently P. poguni is associated with the upper coastal plain of the Tibbie Creek drainage. Color. — The basic color pattern of P. pogum is reddish, infused with henna and with lighter areas associated with the lateral carapace, the rostral margins and the post- orbital ridges. Variation. — The limits of variation as known are incorporated into the diagnosis and description sections above. Morpho- metric analysis was based on a total of 15 specimens (2 66\, 4 6611, 9 99). Carapace length in Form I males was 28.6-31.1, for females 26.4-37.4 (avg. 32.51) and for Form II males 24.8-34.0 (avg. 28.8). Rostrum length was 3.36-4.85 in Form I males, 4.24-5.45 (avg. 4.85) in females and 4.20-5.26 (avg. 4.59) in Form II males; width of the rostrum was 1.39-1.77, 1.17-1.51 (avg. 1.37) and 1.23-1.59 (avg. 1.41) in the same three categories, re- spectively. Areola length was in Form I males 2.55-2.60, in females 2.40-2.61 (avg. 2.52) and in Form II males 2.45-2.53 (avg. 2.46). Antennal scale length was 6.98-7.97 in Form I males, 7.33-9.42 (avg. 8.34) in females and 7.18-9.16 (avg. 8.37) in Form II males. Chela length was 1.48-1.51 in Form I males, 1.57-4.03 (avg. 2.01) in females and 1.67-2.03 (avg. 1.80) in Form II males; chela width in Form I males was 1.47-2.53, 0.45-2.44 (avg. 2.11) in females and 2.45-2.60 (avg. 2.53) in Form II males. Length of the mesial margin of the palm and length of the dactyl in Form I males were 2.76-3.44 and 1.58-1.68, respectively. The same two characteristics in the same se- quence were in females 1.10-2.72 (avg. 2.45) and 0.76-1.72 (avg. 1.55); in Form II males Nos. 3-4 Systematics of Crawfishes 87 they were 2.65-2.93 (avg. 2.78) and 1.51-1.76 (avg. 1.61). Pleopod length in Form I males was 2.18-3.21 and in Form II males was 3.64-3.74 (avg. 3.68). Specimens examined. - MISSISSIPPI: Chicka- saw Co., Houston, 11 April 1936 (2 ddl, 3 66ll, 8 99), 0.4 mi E of Houlka Creek, no date (1 6l\, 1 Etymology. — Taken from pogon (Gr.): beard; in reference to the setiferous beard found on the mesial margin of the hands in males and on the third maxilliped. DISCUSSION Each of the taxa described here, except P. h. vesticeps, probably represents a distinct breeding population of crawfish. There are many morphological features which dis- tinguish each of the populations: likewise, there are no, again with the exception of vesticeps, evidences of intergradation. In each instance there seems to be either dif- ferences in ecological habitat preferences and/or physical disjunction of the habitat. Morphometric analyses strongly support the conclusions offered for other kinds of data and reveal that the several species are no more variable than other crawfish species similarly studied, exhibited no greater inter- specific variation within a taxonomically compact grouping and exhibited no evidence of clines. The most taxonomically significant dif- ferences can be found in the details of the terminal elements of the first pleopod of the Form I males. In P. hagenianus hagenianus and P. h. vesticeps the centro-cephalic pro- cess is enveloped at its base by a tight fold which becomes continuous with the cen- tro-caudal process. A similar, but not so tightly applied, fold wraps the lateral base of the centro-cephalic process in P. harbiger. Contrastingly, in P. cormus the anterior base of the cephalic portions of the centro- cephalic process has a tightly applied fold, but the fold is absent from the lateral portions. In P. cometes and P. pogum such a fold is absent. In all populations the caudal process is a brqad, truncate extrusion, later- ally compressed in its caudodistal half. The caudodistal margin of all, except harbiger, forms an angular intersection with the distal margin; in the latter this intersection is curvilinear. Only in P. h. hagenianus, how- ever, is the distal margin subperpendicular to the main axis of the shaft; in all other species the margin is gently sloping, and in no species does the caudal process project distally as far as the central projection. The mesial process in all is a long, subconical, nearly setiferous protrusion extending, ex- cept in cometes, well beyond the distalmost portions of the other terminal elements; in cometes it extends distad approximately the same distance as the central projection. Only in P. connus is the mesial process subparallel to the main shaft of the appendage. In P. cometes it is curved sharply laterad, and in the remaining species it is curved gently laterad. A caudally oriented curve of the mesial process exists, additionally, in both subspecies of hagenianus, in barbiger and in connus, whereas in pogum the process curves gently ^ephalad to recurve gently caudad in its distal half. A pronounced shoulder occurs at the base of the cephalic margin of the central projection in both subspecies of hagefiianus, in barbiger and in connus, although it is not sharply angular in vesticeps and barbiger. In other details the appendages are as outlined in the diagnoses above and are similar. The most conspicuous feature of the pre- viously undescribed taxa is the beard of plumose setae along the entire mesial margin of the palm. Such pubescence is also present in varying degrees (varying according to species) on the dactyl, on Form II males and on females, but is best developed in Form I males. AU of the species bear a dense double row of plumose setae along the surface of all units of the third maxilliped. This char- acteristic is sufficiently common in species of primary burrowing habits to suggest an adaptation to such an habit. In P. barbiger and P. poguyn the ventrolateral surface of the exopodite of the third maxilliped is pro- vided with a dense row of heavy, stout setae forming a mat. In P. cometes, P. connus and P. h. vesticeps the setae of this row are stout but markedly more sparsely distributed. In P. h. hagenianus the row is usually absent. 88 Tulane Studies in Zoology and Botany Vol. 20 o o O to s II QL CO en B ^ en u CM CM cn > cn X o -r in 0. I xl E o U o ■I -I I II Ol pi o o o If) o o Nos. 3-4 Systematics of Crawfishes 89 ^ ^ III o -§!-' I > iMS Ex*" o §'^ ^ a, U I I o I U > no •b To E c ■ o o y u > > I -I-.-' 1 90 Tulane Studies in Zoology and Botany Vol. 20 ■b •o CD U 00 's E o 1^ I > n -) u 3 O u o s o > E o o s o .SP Nos. 3-4 Systematics of Crawfishes 91 but if present, at best consists of a very thin row of delicate setae. The antennal scale reflects the degree of setation found along the hand of Form I males in every species. In P. h. hagenianus the setae are restricted to the short bristles common along the mesial margin in most crawfishes. In P. harhiger they are longer and stouter; in all of the other taxa the mesio- cephalic margin is provided with tufts of long setae arising from tubercular eminences on the margins. The antennal scale is widest approximately at midlength in P. harhiger, P. cometes, P. connus and P. pogum, but it is widest distal to midlength in the subspecies of P. hagenianus. In all taxa the thickened lateral margin terminates distally in a strong acute spine, although the relative length of the spine varies according to the species. In P. harhiger, P. cometes and P. pogum it con- stitutes about 33% of the total length of the scale; in P. connus it is about 25% of the length; and it is about 10-15% in the sub- species of P. hagenianus. The spine is mark- edly more attenuate in the first three men- tioned species than in the others and reaches the other extreme in P. h. vesticeps in which it is nearly conical in shape. Other variations are mentioned in previous discussions. A much more difficult characteristic to evaluate is the annulus ventralis of the female. All have an annulus in which there is an anterior trough (sulcus) which rises caudally; the highest (ventrally) portions are cephalolateral and/or lateral, and the annulus is movable; usually the prominences are decorated with tubercular eminences. In the subspecies of P. hagenianus and in P. pogum the anterior portion of the sulcus slopes precipitously toward a deeply exca- vate sternum associated with the fourth pereiopodal segment. The annulus is poorly delineated from the sternum associated with its anterior margin in the subspecies of P. hagenianus, but in P. pogum is set off by an intervening groove. The annulus of P. harhiger has a less severe slope and is delinea- ted anteriorly from the sternite by a groove. In P. cometes the anterior sulcus is narrow, less sloped and delimited anteriorly by a groove. The postannular sternite (associated with pereiopodal segment V) is subconical in P. h. hagenianus and P. cometes, broadly conical in P. h. vesticeps and P. pogum, but in P. harhiger is in the shape of a triangular prism. In morphometric analysis 10 character- istics were analyzed in males and nine in females. For analysis the males were treated as two separate categories — Form I and Form K. Thus, 29 comparisons could be made between any two populations. When these comparisons were made using a null hypothesis, the following significant (p > 0.05) differences were established (Table 1): P. h. hagenianus differed from P. h. vesticeps in 24 characteristics, from P. harhiger in 14 (of 19), from P. connus in 5 (of 10), from P. pogum in 21 and from P. cometes in 14 (of 19). P. h. vesticeps differed from harhiger m 10 (of 19), from connus in 8 (of 10), from pogum in 21 and from cometes in 18 (of 19). P. harhiger differed from connus in 4 (of 19), from po^wm in 7 (of 10) and from cometes in 6 (of 19). P. connus differed irom pogum in 6 (of 10) and from cometes in 7 (of 10); and P. pogum differed from P. cometes in 12 (of 19). In many of these p^ 0.01. The report of less than all possible combinations results from the fact that no Form II males are known for P. harhiger, and there are insufficient numbers for analy- sis in the category of Form II males for P. cometes and P. connus and for females in P. connus. Procamharus harhiger and P. connus seem to be geographically isolated from the other species, and the latter is the only one associ- ated with loess. The subspecies of P. hageni- anus are apparently restricted to the prairie soils of the Black Belt of Alabama and Mis- sissippi, with the nominate subspecies oc- cupying the area south and east of Tibbie Creek, a tributary to the west bank of the Tombigbee River in east-central Mississippi; an intergrade population occurring just north of the creek; and P. h. vesticeps being confined to the upper Tombigbee River system (Fig. 120). P. cometes SLud P. pogum occupy limited areas in different habitats just outside the periphery of the range of P. hagenianus; both are associated with the in- terior flatwoods. 92 Tulane Studies in Zoology and Botany Vol. 20 LEGEND o P. h. hagenianus • P. h. vesticeps 9 P. hagenianus X vcsticaps a P. barbiger ▲ P. comates ■ P. connus Ci. P. pogum i Figure 120. Distribution of Hagenianus Group crawfishes. Nos. 3-4 Systematic s of Crawfishes 93 The species of the Hagenianus Group probably represent a derivative stock rather than an ancestral stock within the subgenus Girardiella. The loss of the cephalic process, present in all other species of the subgenus, is an excellent indication that this group is unlike the ancestral stock. Four taxa— ^. hagenianus, h. vesticeps, barbiger and con- HU5— possess a highly specialized structure in the fold which envelops the base of the central projection. All Hagenianus Group species have a nearly unique spinose orna- mentation on the uropods. indicating a secondarily acquired characteristic. In the subgenus, the most generalized pleopod is probably found in Pro camb am s (Girardiella) gracilis, but the relatively broad areola found in P. (G.) tulanei is generally considered to be a more primitive condition in crawfishes than the obliterated or linear areolae char- acteristic of species of the subgenus. Bearding of the hands and maxillipeds is a more widespread phenomenon, but general consensus favors the unbearded hand as the more primitive condition in crawfishes. P. tulanei becomes interesting in a study of the Hagenianus Group because it seems to fill certain essential criteria as a form inter- mediate between the two species complexes constituting the Gracilis and Hagenianus Groups. Geographically, it is found in the Red and Ouachita River systems of Arkansas and Louisiana; the other taxa of the Gracilis Group are found in an arc and are northeast to southwest of it. It is the only member of this assemblage with a beard of long setae along the mesial margin of the palm, while P. h. hagenianus is the only member of the Hagenianus Group to lack it. The broad are- ola has already been mentioned. Thus, as I have previously outlined (Fitzpatrick, 1975), a fj 3 0) rH •H (U S P, ^^ (0 at • M o U -H (U O ^ • -p Pi O u S o p s § 0) o •H o •H ^ o -P Vl m a (U fi • > Figure 121. Proposed relationships of Hagenianus Group crawfishes. Nos. 3-4 Systematics of Crawfishes 95 that they do not represent the population from which the Hagenianus Group stock descended, and their prairie soil habitat does not represent a continuum through which invasion could take place. Although P. bar- biger seems to retain several features which associate it with the primitive stock (bearded hand, lamellar portion of antennal scale lack- ing a tubercular margin, and a groove at the anterior margin of the annulus), it also has some unique and probably derived char- acteristics: the beard of the hand occurs in distinct tufts; the distal margin of the caudal process has a curvilinear relationship with the caudal margin rather than angular; the lateral base of the central projection is cover- ed by a loosely applied fold; and the post- annular sternite is uniquely in the shape of a triangular prism. Couple these morphological features with the prairie-habitat difficulty mentioned above, and one is inevitably drawn to the conclusion that this species represent* a secondary, albeit early, invasion of an available habitat. Thus, one is left with one species, P. pogum, having relatively primitive char- acteristics within the group and occupying a habitat compatible with an environmentally plausible invasion route. But pogum does not in itself represent a relict of the ancestral stock. Several features indicate that it, too, has undergone specialization. The shoulder at the base of the central projection is lack- ing in pogum and only one other species in the subgenus, cometes. Only in pogum in the Hagenianus Group is the mesial process recurved distally, and the slope of the sulcus of the annulus ventralis falling precipitously toward the anterior margin is more like the derivative hagenianus populations than the other species. The antennal scale is an excel- lent structure to indicate the mixed combi- nation of characters found in this species: primitive features include a long spine forming the distalmost part of the thickened lateral portion (33% as in barbiger and cometes, compared with 25% in connus and 10-15% in the subspecies o{ hagenianus) and being widest near midlength (as in barbiger, cometes and connus). On the other hand, it shares the apparently derived condition of having setae arising in tufts from tubercular prominences along the cephalic portion of the lamellar margin with cometes, connus and h. vesticeps. The picture presented, then, is one in which a crawfish stock entered the area across the Mississippi Flood Plain from the northwest probably in a single invasion. But shortly after invasion an adaptive radiation occurred resulting in the reproductive isola- tion of five populations, with the eastern- most filling the most expansive habitat and undergoing clinal (?) variation which eventu- ally resulted in geographic races. Suggested relationships are given in Figure 121. A Key to the Hagenianus Group of Crawfishes In Hobbs' (1972b:45) "Key to Species of Subgenus Girardiella," couplet 1 must be modified as follows (italics indicate changes): 1 Mesial surface of palm of chela bearded; areola always with 3 or 4 punctations across narrowest part; cephalic process pres- ent P. (G.) tulanei Penn. Mesial surface of palm of chela never bearded, or if bearded areola linear and cephalic process absent, areola linear or with only 1 or 2 punctations across narrowest part. 2. The keys here offered may then be used following identification of P. hagenianus in couplet 4. Alternately, after eliminating the genus Fallicambarus and the subgenus Acu- cauda from consideration, one may proceed directly to this key if the specimens possess a median spine extending beyond the distal margin of the inner ramus of the uropod. 1 Sex male 2. 1' Sex female 7. 2 (1) Mesial margin of palm with setifer- ous beard 3. 2' Mesial margin of palm lacking setifer- ous beard. P. h. hagenianus (Faxon). 3 (2) One or more tufts of long setae arising along mesial margin in basal half of dactyl 4. 3' Dactyl lacking tufts of long setae along mesial margin 5. 96 Tulane Studies in Zoology and Botany Vol. 20 4 (3) Tufts of setae on dactyl scanty; tight fold enveloping base of central pro- jection in Form I male. P. connus n. sp. 4' Tufts of setae on dactyl well-devel- oped; base of central projection of Form I male lacking enveloping fold. P. pogum n. sp. 5 (3') Mesial process of first pleopod subequal to central projection in length; lacking enveloping fold at base of central projection in Form I males P. cometes n. sp. 5' Mesial process of first pleopod longer than central projection; base of cen- tral projection enveloped by fold in Form I males 6. 6 (5') Distal margin of caudal process of first pleopod subperpendicular to main axis of shaft of pleopod; setae of lamellar portion of antennal scale arising in tufts from tubercular eminences. . P. h. vesticeps n. subsp. 6' Distal margin of caudal process of first pleopod not subperpendicular to main axis of shaft of pleopod; setae of lamellar portion of antennal scale arising individually P. barbiger n. sp. 7 (1') Antennal scale widest near mid- length, distal spine of thickened lateral portion constituting at least one-fourth of total length of scale. 8. 7' Antennal scale widest distal to mid- length, distal spine of thickened lat- eral portion constituting less than one-fourth of total length of scale. 11. 8 (7) Setae of lamellar portion of antennal scale arising in tufts from tubercular eminences; postannular sternite conical 9. 8' Setae of lamellar portion of antennal scale arising individually; postannular sternite in shape of triangular prism. P. barbiger n. sp. 9 (8) Distal spine of thickened lateral portion of antennal scale consti- tuting about one-third of total length of scale 10. 9' Distal spine of thickened lateral portion of antennal scale consti- tuting about one-fourth of total length of scale. ... P. connus n. sp. 10 (9) Trough of annulus broad and de- scending precipitously cephalically toward sternites; postannular sternite broadly conical P. pogum n. sp. 10' 1 rough ot annulus narrow and subparallel to horizontal plane of sternites; postannular sternite sub- conical P. cometes n. sp. 11 (7') Setae of lamellar portion of antennal scale arising in tufts from tubercular eminences; postannular sternite broadly conical. P. li. vesticeps n. subsp. 11' Setae of lamellar portion of antennal scale arising individually; post annu- lar sternite subconical. P.h. hagenianus (Faxon). LITERATURE CITED BLACK, J. B. 1967. A new crawfish of the genus Cambarus from southwest Louisiana. (Deca- poda, Astacidae). Proc. Biol. Soc. Wash. 80:173-178. BROCCHI, M. 1875. Recherches sur les organes genitaux males des crustaces decapodes. Ann. Sci.Natur. 2(3-6): 1-131. CARR, C. 1936. The control of the dry -land craw- fish (Camhariis hagenianus). U. S. Biol. Surv., Control Meth. Res. Lab., Denver, lip. (Mimeo- graphed). DOWELL, V. E. and L. P. WINIER. 1970. A bi- lateral color anomaly in the crayfish Orconectes virilis (Hagen). Proc. Iowa Acad. Sci. 76:487-492. ERICHSON, W. F. 1846. Ubersicht der Arten der Gattung Astacus. Arch. Naturgeschichte 12:86-103, 375-377. FAXON, W. 1884. Descriptions of new species of Cambarus, to which is added a synonymical list of the known species of Cambarus and Astacus. Proc. Amer. Acad. Arts Sci. 20:107-158. 1885. A revision of the Astacidae; Part I: The genera Cambarus and Astacus. Mem. Mus. Comp. Zool., Harvard Coll. 10:vi + 186pp. — — _ 1885. A revision of the Astacidae; Part I: The genera Cambarus and Astacus. Mem. Mus. Comp. Zool., Harvard Coll. 10:vi + 186pp. 1914. Notes on the crayfishes in the United States National Museum and the Museum of Comparative Zoology with descrip- tions of new species and subspecies to which is appended a catalogue of the known sjsecies and Nos. 3-4 Systematics of Crawfishes 97 subspecies. Mem. Mus. Comp. Zool., Harvard Coll. 40:351-427. FISHER, A. K. 1911 [1912]. Crawfish as crop destroyers. U. S.D. A. Yearbook 1911: 321-324. FITZPATRICK, J. F. JR. 1967. The Propinquus Group of the crawfish genus Orconectes. Ohio J. Sci. 67:129-172. 1968. The subspecies of the crawfish Procambarus hagefiianus (Faxon). A. S. B. Bull. 15:37. 1975 [1976]. The taxonomy and biol- ogy of the prairie crawfishes, Procambarus hagenianus (Faxon) and its allies. Pages 381-391 in Avault, J. W. Jr., ed. Freshwater Crayfish. Louisiana State University Division of Continuing Education, Baton Rouge. and H. H. HOBBS III. 1968. The Missis- sippi River as a barrier to crawfish dispersal. Amer.Zool. 8:807. FOWLER, H. W. 1911 [1912]. The Crustacea of New Jersey. Ann. Rept. New Jersey Mus. 1911, Pt. 11:29-650. HAGEN, H. 1870. Monograph of the North Ameri- can Astacidae. Illus. Cat. Mus. Comp. Zool., Harvard Coll. 3:109pp. HARRIS, J. A. 1903. An ecological catalogue of the crayfishes belonging to the genus Cambariis. Kansas Univ. Sci. Bull. 2:51-187. HAY, W. P. 1902. On the proper application of the name Carnbarus carolmus Erichson. Proc. Biol. Soc.Wash. 15:38. HOBBS, H. H. JR. 1938. A new crawfish from Florida. J. Wash. Acad. Sci. 28:61-65. 1942a. A generic revision of the cray- fishes of the subfamily Cambarinae (Decapoda, Astacidae) with the description of a new genus and species. Amer. Midi. Natur. 28:334-357. 1942b. The crayfishes of Florida. Univ. Florida Publ., Biol. Sci. Ser. 3:1-279. 1959. Malacostraca [part]. Pages 883-901 in Edmondson, W. T., ed. Fresh-water Biology. John Wiley and Sons, New York. 1968. Crustacea: Malacostraca. Pages K-1 - K-36 in Parrish, F. K., ed.. Keys to water quality indicative organisms (southeastern United States). U. S. Dept. Interior, Fed. Water Pollution Control Adm., [Washington, D.C.?]. 1971. A new crayfish of the genus Pro- cambarus from Mississippi (Decapoda: Astaci- dae). Proc. Biol. Soc. Wash. 83:459-468. 1972a. The subgenera of the crayfish genus Procambarus (Decapoda: Astacidae). Smithsonian Contrib. Zool. no. 117:1-22. 1972b. Crayfishes (Astacidae) of North and Middle America. Identification Manual no. 9, X + 173pp. in Biota of Freshwater Eco- systems. U. S. Environ. Protection Agency, Water Pollution Res. Control Ser. U. S. Govern- ment Printing Office, Washington, D. C. __i^_«_ 1973. New species and relationships of the members of the genus Fallicambarus. Proc. Biol. Soc.Wash. 86:461-482. 1974. A checklist of the North and Middle American crayfishes (Decapoda: Asta- cidae and Cambaridae). Smithsonian Contrib. Zool. no. 166:iii+ 161pp. and R. W. BOUCHARD. 1973. A new crayfish from the Cumberland River system with notes on Cambarus carolinus (Erichson). Proc. Biol. Soc.Wash. 86:41-68. and A. VILLALOBOS. 1964. Los cambarinos de Cuba. An. Inst. Biol., Univ. Nat Aut. Mexico 84:307-366. LYLE, C. 1937. The crawfishes of Mississippi, with special reference to the biology and control of destructive species. Ph. D. Dissertation. Iowa State College [=University ] , Ames. 1938. The crawfishes of Mississippi, with special reference to the biology and control of destructive species. Iowa St. Coll. J. Sci. 13:75-77. (Abstr.) MARTIN, A. C. and F. M. UHLER. 1938. Food of game ducks in the United States and Canada. U. S. Dept. Agri., Tech. Bull. no. 634:1-156. MOMOT, W. T. and J. E. GALL. 1971. Some eco- logical notes on the blue phase of the crayfish, Orconectes virilis, in two lakes. Ohio J. Sci. 71:363-370. ORTMANN, A. E. 1902. The geographical distribu- tion of fresh water decapods and its bearing upon ancient geography. Proc. Amer. Phil. Soc. 41:267-400. 1905a. The mutual affinities of the species of the genus Cambarus and their dis- persal over the United States. Proc. Amer. Phil. Soc. 44:91-136. 1905b. Procambarus, a new subgenus of the genus Cambarus. Ann. Carnegie Mus. 3:435-442. 1906. Mexican, Central American, and Cuban Cambari. Proc. Wash. Acad. Sci. 8:1-24. PAYNE, J. F. 1972. The life history of Pro- cambarus hayi. Amer. Midi. Natur. 87:25-35. PENN, G. H. 1953. A new burrowing crawfish of the genus Procambarus from Louisiana and Arkansas. J. Wash. Acad. Sci. 43:163-166. PENNAK, R. W. 1953. Fresh-water Invertebrates of the United States. Ronald Press, New York, ix -t- 769pp. REIMER, R. D. 1975. Procambarus (Girardiella) curdi, a new crawfish from Arkansas, Okla- homa, and Texas. Tulane Stud. Zool. and Bot. 19:22-25. SMILEY, J. W. and W. W. MILLER. 1971. The occurrence of blue specimens of the crayfish Procambarus acutus acutus (Girard) (Decapoda, Astacidae). Crustaceana 20:221. SMITH, E. W. 1953. The life history of the craw- fish Orconectes (Faxonella) clypeatus (Hay). Tulane Stud. Zool. 1:79-96. 98 Tulane Studies in Zoology and Botany Vol. 20 ECOLOGY AND POPULATION DYNAMICS OF THE BLACK BULLHEAD, ICTALURUS MELAS (RAFINESQUE), IN CENTRAL KENTUCKY ROGER D. CAMPBELL^ and BRANLEY A. BRANSON Department of Biological Sciences Eastern Kentucky University Richmond, Kentucky 40475 ABSTRACT The black bullhead, Ictalurus melas (Raf- inesque), is a prominent element of the fish fauna in central Kentucky, but relatively little has been reported concerning either the population dynam- ics or physiological ecology in that area. This work reports growth and reproduction in pond, lake and stream populations of the black bullhead. Empha- sis was placed on pond populations for the analysis of reproductive development, longevity and sur- vival. Some aspects of the physiological ecology, including temperature ranges and oxygen consump- tion rates, were measured as functions of prior acclimation histories. Black bullheads were well-represented in each of the habitats surveyed; however, growth in length and in weight, expressed as a function of age at the time of annulus formation, was significantly lower for fish taken from ponds. Pond populations ex- hibited the ability to achieve great numbers, and the effect of crowding was significantly reflected in lower length-weight and age-weight correlations, and in lower mean fecundity. Mortality in ponds was high throughout the year, and highest during spring and early summer. Adult black bullheads sampled from all populations relied heavily on chironomid larvae as a staple source of food, although this item had a selection value compa- rable to several less abundant food sources. Larval feeding was found to be highly selective. Repro- ductive development, and the attainment of peak gonad weight-body weight ratios was retarded in ponds as compared with development in lakes and streams. Temperature thresholds measured at 7°, 12° and 23°C indicated that different life stages responded differently to temperature. Upper criti- cal thresholds for the first four free-swimming stages formed a graded sequence, with lower thresholds graduated upward in direct proportion 1 Present address: Department of Zoology, Idaho State University, Pocatello, Idaho. to upper critical values. Preference ranges differed significantly for postlarval through yearling stages. Oxygen consumption rates, QIO values, and abso- lute consumption curves suggested a relative insen- sitivity to temperature. Black bullheads ranging in weight from 20.15 to 37.00 g had a critical oxygen consumption value per gram of 0.1 ml. 02/g/hr at 0.2 to 0.5 ml/1 ambient oxygen. INTRODUCTION Changes in fish stocking combinations (Swingle, 1950), shifts in angler preferences (Houser and CoUins, 1962) and selective construction of commercial gear to favor larger, if less abundant, fishes have de- emphasized the value of the black bullhead. Although management personnel and anglers in Kansas and Iowa (Hastings and Cross, 1962; Forney, 1955) continue to prize this species for its angling value most states east of the Mississippi River have shown little interest in buUhead propagation and manage- ment for sport or commercial purposes since early in the present century. General disregard for the black bullhead is reflected in the early complete absence of investigations designed to foster a better understanding of its potentials and limita- tions. In short, very little is known con- cerning many important aspects of the life history and physiological ecology of /. melas, aspects that might prove valuable if the species is to be most effectively utilized or managed. No detailed bullhead investiga- tions have been conducted in Kentucky waters. Reports that have appeared (Clay, 1962) have simply acknowledged the EDITORIAL COMMITTEE FOR THIS PAPER: DR. KENNETH D. CARLANDER, Distinguished Professor of Fisheries, Department of Animal Ecology, Iowa State University, Ames, Iowa 50011 DR. FRANK B. CROSS, Curator of Fishes, Museum of Natural History, University of Kansas, Lawrence, Kansas 66045 99 100 Tulane Studies in Zoology and Botany Vol. 20 species' presence and recorded some of its taxonomic features. The purpose of this study is to present pertinent aspects of the black bullhead life cycle in order to assess the species' ecologi- cal interactions, and to analyze some physio- logical phenomena which favor the fish in some habitats and restrict it in others. The study was begun in April 1970 and termi- nated in early July 1971. During this period, our investigation concentrated on the repro- ductive potential, natural mortality, feeding behavior and food, and physiological ecol- ogy associated with the black bullhead's suc- cess in natural and impounded waters, with particular emphasis on the earlier stages of life. However, several aspects of adult biol- ogy not previously investigated in Kentucky waters were studied with significant results. REVIEW OF THE LITERATURE Black bullheads have been found in virtu- ally every type of fresh-water habitat. Houser and ColHns (1962), for example, indicated that no important Oklahoma water was free of this species. Although sampled from reservoirs, lakes and streams, the only significant populations were encountered in smaller bodies of water. This finding is sup- ported by the results of lake and stream investigations conducted throughout the range (Lynch, Buscemi and Lemmons, 1951; Trautman, 1957). According to Fogel (1964), black bullheads comprise an insig- nificant 0.5 to 1.5 per cent of the annual catch in South Dakota waters, and Clark (1960) listed the black and brown bullheads as the least important species encountered in Lake St. Marys, Ohio, as reflected by annual harvest statistics. He added, however, that strong year classes occasionally appear, which assure well-balanced distribution of the species. Important populations of bullheads and channel catfish (I. punctatus) formerly oc- curred in the Mississippi River and its tribu- taries. According to Barnickol and Starrett (1951), this species complex amounted to 50% of the commercial catch reported in 1899. This figure had dwindled to 15.1 per cent by 1946 corresponding to concom- mitant increases in the populations of other commercial species, particularly carp and buffalofishes (Barnickol and Starrett, loc. cit.). Black bullheads have become prominent in small artificial lakes in many areas of the midwest (Jenkins, 1959), sometimes to such an extent that selective eradication projects have been directed toward population reduc- tion (Houser and Grinstead, 1961). The greatest predominance appears to be in farm ponds where the management objective may or may not be bullhead propagation (Ben- nett, 1952). In Douglas County, Kansas, Hastings and Cross (1962) found black bull- heads in 15 of the 22 farm ponds surveyed (68 per cent), and considered them as domi- nants in 14 of these. The ponds studied were aribtrarily classified on the basis of turbidi- ty: 50 ppm or less suspended material con- stituted a clear pond, while one exceeding this limit was considered turbid. Black bull- heads predominated in all turbid ponds for which quantitative estimates were obtained. Trautman (1957) found that black bullheads have increased in Ohio waters during the present century at the expense of other bull- head species. Increased siltation resulting from more intensively applied agricultural practices apparently has favored this replace- ment trend. Differences in ponderal indices from dif- ferent portions of Clear Lake, Iowa, sug- gested that several subpopulations could exist in the same body of water (Forney, 1955). Each of these subpopulations ex- hibited its own vital characteristics based, in part, on differences in food supplies. Various authors have attributed the distributional success of black bullheads to omnivorous feeding habits (Clay, 1962; Viosca; 1931; Smith, 1949), although Forney (1955) indi- cated that selective feeding occurred in Iowa. Rose and Moen (1951) concluded that selective feeding by adults was more appar- ent than real, usually reflecting the prepon- derance of one or more items over others in the food supply. Kutkuhn (1955) listed a variety of food items taken by adult black bullheads in Iowa lakes: insects, ento- Nos. 3-4 Black Bullhead 101 Foods taken by different age groups appear to differ significantly, resulting in part from differences in feeding times between adults and young. Darnell and Meierotto (1965) found that young-of-the- year bullheads school during the day, their social appetites stimulated by tactile and gustatory senses not shared by older indi- viduals (Bowen, 1931). Adults were rarely active by day but commenced feeding with the onset of darkness and remained active throughout the night. Young bullheads (15 - 53 mm TL) in Clear Lake, Iowa, fed almost exclusively on Entomostraca (Forney, 1955). Ewers (1931) found that small bull- heads (36 - 76 mm TL) in Buckeye Lake, Ohio, fed primarily on the amphipod Hyalel- la, Entomostraca and small insect larva, with peaks of feeding activity occurring just before dawn. Feeding preferences appear to change sequentially with different stages of the re- productive cycle. Raney and Webster (1940), working with Cayuga Lake bullheads in New York, noted the following feeding pattern: adults moved into cover during periods of 50° to 55°F temperature (early April) and began feeding avidly on aquatic vegetation (Nymphozanthus advena and Potamogeton) prior to spawning. Through- out the remainder of the year, feeding habits were assessed as truly omnivorous. Forney (1955) noted a similar sequence, and con- cluded that plant feeding was somehow re- lated to the attainment of reproductive readiness. He listed as staples throughout the first year of life, Crustacea (60 per cent) and Diptera larva (25 per cent, chiefly Chriono- midae). The inclusion of adult dipterans and bryozoan statoblasts suggested some surface feeding data, but such activity has not been described. Age and growth statistics, vital to an understanding of any species' population complexities, have been reported from vari- ous states (Carlander, 1968); however, none have appeared representing the bullheads of Kentucky waters. Synthesis of the available material yields the following generalizations: growth tends to be more rapid in clear water, notwithstanding the black bullhead's propensity to dominate turbid waters (Hastings and Cross, 1962). Bullheads in Kansas and Iowa ponds tended to overcrowd turbid waters, rarely maintaining suitable growth rates. Houser and Collins (1962) noted slower growth rates in steeams than in ponds or reservoirs. Variations in growth rates have been so extreme throughout the range that comparisons for different habitat types are all but precluded (Carlander, 1968), as in the case of direct comparisons of Kentucky populations with others. Albaugh (1969) found great intraspecific variability in growth per season. In artificial ponds provided with unlimited food sup- plies, males made greater gains than females, but when food supplies were kept near maintenance levels, growth rates were com- parable. Simco and Cross (1966) reached similar conclusions concerning differential growth rates in /. pnnctatus. Growth is apparently rapid during the first year of life, and subsequently slower coincident with the degree of intraspecific competition. Moen (1959) noted that bull- heads in Silver Lake, Iowa, attained an average total length of 127 mm in Septem- ber at age I, but only 132 mm (41 g) in September at age II. Following a severe winter kill, which significantly reduced com- petition, these same fish reached a length of 201 mm and 122 g at age III. The age IV group increased from 208 mm to only 213 mm at age V. Corthell (1961) assessed rapid growth of bullheads in the Empire Lakes, Oregon, as the result of recent introduction and low population density. Some of the most worthwhile data pertinent to the management of scaled fish has been gained through the evaluation of age -growth relationships based on the scale method of age determination (Creaser, 1926: Buckman, 1929, Van Oosten, 1929). Since members of the genus Ictdlurus lack scales, workers have resorted to annuli on bones and spines for age determination (Lewis, 1949, 1950; Forney, 1955; Carlander and Sprugel, 1948; Sneed, 1951). Houser and Collins (1962) found that annuli usually 102 Tulane Studies in Zoology and Botany Vol. 20 formed on fin spines during the second week in May in Oklahoma waters. The uniform appearance of annuli from all age groups at the time indicated that all ages developed annuli at approximately the same time. Few descriptions of the breeding behavior of /. melas are available. The early ichthyo- logical notes of Fowler (1917) provided the following account. (It should be noted, how- ever, that the author referred to the breeding pair simply as "bullheads", leaving some doubt as to their species identity.) "After a nest was constructed by the female, both sexes took charge of it; both guarded the eggs and the fry after spawning; and both were observed occupying the nest at the same and at alternate times. The deposited eggs numbered 200, were creamy white in color, and took approximately five days to hatch." Wallace (1957), in a more recent account, found that only the female exca- vated the nest. As the nest neared comple- tion, each parent began to pay closer atten- tion to the other. Actual spawning lasted for about one second, and occurred five times in a single hour. The female guarded the eggs during the first day, the male thereafter. Clemens and Sneed (1957) described a simi- lar behavior in the channel catfish. Black bullheads reportedly make no elab- orate preparation for spawning, although gonadal development comprises a relatively long period of time. Dennison (1970) found rcproductively mature males and females during the 1969 season in Clear Lake, Iowa, starting to mobilize material in the gonads for the next spawning season during August and September. Mesenteric fat was assumed to be the principal source of energy for gonadal development. Differences in the availability of mesenteric fat between the 1969 and 1970 seasons were reflected in the success of spawning and in fecundity indices. Because of wide variation in weight at a given length within the rather narrow length range for mature bullheads, the value of the length-weight relationship as a measure of re- productivity is considerably reduced. Never- theless, length has been used as an arbitrary index to reproductive readiness. Mississippi River bullheads with a total length of 154 mm or more, and belonging to age group III were considered reproductively mature by Barnikol and Starrett (1951). A similar range was proposed by Shields (1957a, b) for the bullheads of Gavins Point Reservoir, South Dakota. Dennison (1970) arbitrarily chose 229 mm total length as the size separating mature and immature bullheads in Clear Lake, Iowa. Fecundity appears to be more closely related to length than to age. Clear Lake females, for example, 203 to 226 mm in total length contained and average of 3,283 eggs, whOe females of the same population 229 to 251 mm long contained an average of 3,845 eggs (Forney, 1955). Dennison (1970) noted a good relationship between length and fecundity in bullheads. Adult females with an average length of 250 to 259 mm had a mean fecundity of 3,892 eggs. Limits of fecundity throughout the range could not be determined, since few statistics, other than those listed above, are available (Car- lander, 1968). Few attempts have been made to estab- Ush the periods of reproduction for black bullheads. Chance observations of spawning periods have been the rule rather than the exception. Spawning occurs from late June to late July in South Dakota (Shields, 1957a, b). The earliest spent female was taken on June 26, and the latest ripe female on August 6, 1951 in Clear Lake (Forney, 1955). Dennison (1970) reported that the 1969 spawning season in Iowa extended from June 1 through July 1, while the 1970 season was somewhat shorter (June 10 to 23), each seasonal limit reflecting the in- fluence of prolonged temperature increase. The literature is incomplete with respect to environmental conditions that influence initiation of the spawning response. Buser and Blanc (1949) tested the effect of photo- period on gonadal development in /. nebu- losus. Continuous light in winter produced increase in testicular and ovarian weight, accompanied by increased in the deposition of interstitial tissue. According to Burger (1939) and Medlen (1951) temperature was a more influential factor than light for a number of fishes (see also, Liley, 1969; Har, Nos. 3-4 Black Bullhead 103 1965). Wallace (1967) observed spawning in captive bullheads that had been retained in the laboratory from August through March. Environmental conditioning w^as not in- tended, i.e., photoperiod control was not maintained, and temperature was held at a constant 25'C The absence of control did not inhibit induced spawning. Dennison (1970) stated that spawning was triggered by an abrupt increase in water temperature, ranging from 6° to 8°C over a period of several days. Any fish that lives in farm ponds is sub- jected to seasonal and sudden changes in temperature, but not necessarily in light in- tensity (Wallen, 1955). The Secchi disk visibility in Oklahoma ponds, determined on 132 occasions from 1951 to 1954 averaged only 45.7 cm. Readings of 6 mm, varying slightly throughout the year, were recorded for some turbid waters known to contain bullheads. Accounts of black bullhead embryonic development are scarce. Spawning seems dif- ficult to induce in the laboratory, since even reproductively mature individuals often become refractory under imposed conditions (Sneed and Clemens, 1960). To avoid this difficulty, especially where regulation of spawning is economically important, several methods have been devised to heighten or induce spawning. The dependence of sexual behavior on the gonadal secretions of fish is well-authenticated (Jones and King, 1952; Cohen, 1946; M.C. Tavolga, 1949). There is, however, experimental evidence to indicate that sexual behavior and spawning attempts may be influenced by pituitary hormones in the absence of gonadal mediation (Noble and Kumpf, 1936). Hypophysectomy, on the other hand, can cause the complete ces- sation of the behavioral sequence which normally leads to spawning (W.N. Tavolga, 1955). An authoritative review of the role of pituitary hormones in spawning, and the use of excised pituitary glands in pisciculture has been presented by Pickford and Atz (1957), and specific accounts have appeared period- ically (Robinson and Rinfret, 1957; Schmidt, et al. 1965; Graybill and Horton, 1969). Sneed and Clemens (1957) dealt with induced catfish spawning, a method used successfully to influence the sequence of bullhead reproduction by Wallace (1967). Several investigators have commented on the ability of black bullheads to survive in waters that are uninhabitable for most other species (Black, 1953, 1955; Lewis, 1949; Trautman, 1957; Bennett, 1948). Bullheads have apparently evolved certain adaptive mechanisms which elevate tolerance levels in contrast to less hardy species. Hathaway (1927) found little change in feeding and behavior patterns with markedly increased water temperatures, and Bailey (1955) de- scribed the chance observation of mass- mortality in Bass Lake, Michigan. Death to a complex of species was assessed as the result of high temperature, since none of the char- acteristic signs of anoxia were apparent. Ictalurid fishes were the least affected; a water temperature of 38°C was not limiting to the catfishes so long as free movement was assured. Temperature relation analyses of poikilo- thermic animals are extensive in the litera- ture (Bullock, 1955; Precht, et al. 1955). Brett (1956) measured the median resistance times of several species of fish that were tested at various high temperatures and accli- mated at 20'Xr. The upper incipient lethal temperature (the point at which no further death occurred with time) for /. nebulosus fell between 32°C and 33°C. Morris (1965) reported similar information for the yellow bullhead, a species he found adaptable to extremes of both heat and cold. Black (1953) found that black bullheads accli- mated at 23°C had an upper lethal tempera- ture of 35°C, but she failed to report the effects of cold and associated lethal temper- atures. Cowles and Bogert (1944) introduced the concept of critical thermal maximum (CTMax). According to these authors, upper lethal temperatures have little ecological significance, since species in nature will usually be prevented from reaching ex- tremes. The critical thermal maximum was subsequently defined as the thermal point at which locomotory activity becomes dis- 104 Tulane Studies in Zoology and Botany Vol. 20 organized, and the animal loses the ability to escape conditions that would promptly lead to its death. According to this definition, physiological and ecological death points do not necessarily coincide, and may, in fact, be widely separated. A similar argument could be posed for critical thermal minimum (CTMin). The ecological significance of tempera- ture thresholds in fish has been questioned by Hart (1952) and Brett (1959), who con- tended that temperature tolerances are usually well above habitat extremes. Critical thermal determinations that reflect the ecological rather than physiological death range have also been considered more valid than incipient lethal temperatures by Lowe and Heath (1969). Few CTM determinations have been made for freshwater fish, presum- ably because of the difficulty associated wath determination of the ecological death point (Hutchison, 1961;Norris, 1963). Forbes and Richardson (1909) stated that buUheads cease feeding and become sluggish, sometimes burrowing into the mud, in late fall. Underbill (1952) demonstrated that, although brown bullheads acclimated to lO'Xi: showed reduced rates of feeding, they did continue to feed throughout the year, and Nordlie (1966) noted that black bull- heads, which remained relatively active throughout the winter, had peptic digestive rates that could be correlated to acclimation histories. The extent to which other physio- logical functions depend on temperature, and the limits to which these dependences can be expected to adjust, have been poorly authenticated for ictalurid fishes. Prosser and Brown (1962) reported that fish approximately double their oxygen re- quirements for every lO'C rise in tempera- ture. Thus, it is important from the stand- point of total metabolic expenditure for fishes to enjoy acclimation in gradients of both temperature and dissolved oxygen. According to Brett (1944), however, accli- mation from 3.9° to 13°C took approxi- mately 20 days in the goldfish (Carassius auratus Linnaeus), and may have been even slower for the reciprocal gradient. To com- pensate the need for continuous acclimation. fishes seek optimum temperatures (at which metabolic oxygen demand is probably also optimum) and tend to congregate there (Dendy, 1945, 1946; Buck and Cross, 1952; Brett, 1944; Fry and Hart, 1948a, b). Pre- ferred temperatures have been determined for a relatively few fish species (Fry and Hart, 1948a), and not at all for the black bullhead. Morris (1965), opposed to Brett's (1944) observation, noted that /. natalis ad- justed rapidly to different acclimation tem- peratures by invoking compensatory meta- bolic adjustments. When moved to water at 27'C;, a fish acclimated at 12°C, lowered its metabolic level from 0.35 to 0.15 ml 02/hr. Active metabolism of some fish (Salvelin- us sp.) may be four times the resting rate (Fry and Hart, 1948a). The activity patterns and distributional success of fish were most often reflected by their metaboUc rate. For example, percid darters have several times greater metabolic rates than most sluggish fish (Vernberg, 1954). Caillouet (1967a) concluded that death of channel catfish following hyperactivity was caused, at least in part, by lactic aci- dosis. However, Black (1955) and Caillouet (1967b) suggested that increased blood lact- ic acid in exercised black bullheads was in- sufficient to cause this condition. Black (1955) attributed this to a low Bohr Effect, i.e., increase in blood lactic acid had little effect on the oxygen combining capacity of the blood. Buddenbrock (1938) suggested lactic acid concentration might lead to tissue hypoxia. Thus fish with relatively low meta- boUc requirements, enjoying the advantage of a low positive Bohr Effect, should be favored in situations where the dissolved oxygen-carbon dioxide ratio imposes hypox- ial limits on less favored species. Early studies by Birge (1907) showed that waters in which dissolved oxygen was abundant produced no gas other than carbon dioxide in appreciable amounts. McCay (1925) and Powers (1937) implicated carbon dioxide in fish poisonings. This gas is im- portant from the standpoint of its relation to gas exchange at the gills — since it cannot be excreted osmotically against a gradient — and in its relationship to the oxygen combin- Nos. 3-4 Black Bullhead 105 ing capacity of the blood and the premature unloading of oxyhemoglobin (Matthew, 1921). W.G. Moore (1942) measured the oxygen regimens of 13 species of fish by suspending them in live boxes at various depths. Black bullheads were the least sus- ceptible to hypoxia at the highest carbon di- oxide tensions encountered. Even the brown bullhead was unable to endure oxygen ten- sions of 2.8 to 3.7 ppm for 24 hours in waters of high carbon dioxide tension. Powers, et al. (1938) showed that fishes could absorb oxygen to low levels in wide ranges of carbon dioxide. The species Micropterus dolomieui Lacepede, Am- bloplites rupestris (Rafinesque) and Perca flavescevs (Mitchill) were able to extract oxygen down to 0.30 and 0.40 ml/1 in carbon dioxide tensions of 0.15 to 21.00 mm Hg. Comparable data for ictalurid fishes have not been presented. Natural mortality for a given body of water, and species complex, during a specific period of time, appears to be one of the most difficult of all fish population param- eters to estimate with any degree of statisti- cal accuracy (Ricker, 1944, 1958; Regier, 1962). Two basic analysis procedures have been proposed and subsequently elaborated to provide mortality estimates. The first pro- vides estimates for populations that obey certain well-defined restrictions: the popu- lation must remain unexploited throughout the study period; it must be affected by neither recruitment nor emigration; it must be assumed homogeneous with respect to species and age. The second method is de- signed to provide estimates for exploited populations and becomes rather involved by comparison. It is not surprising, then, that few attempts have been made to estimate differential mortality in freshwater situ- ations. It is usually difficult to observe a population in extensive enough detail to assure than the necessary assumptions have been met. As a result, most of the mortality that has been reported has been of the chance observation type; however, even this type of information can prove important to the management of a fishery (Bailey, 1955). Bennett (1948) sampled Gale Lake, Illinois, previous to and after a known winter-kill during which approximately 80 per cent of the population was decimated. Analysis of the persistent population point- ed to a conclusion now basic to fisheries management; population mortalities are seldom, if ever, complete, and the remaining portion can radically alter the fishery con- dition. Gale Lake bullheads, along with carp and buffalofish, survived the mass-mortality, and subsequently proliferated to such an extent that the lake no longer provided sport fishing. Powers (1937) commented on the factors involved in the sudden mortality of fishes, relating the aperiodic influence of bacterial oxygen depletion-carbon dioxide elevation curve to unexplained fish kills, and conjectured "derangement" resulting from continuous adjustment to different carbon dioxide tensions as an important mortality factor. MATERIALS AND METHODS Study Area Characteristics Sample populations utilized during this study were contained within the adjoining central Kentucky counties of Madison, Fayette and Clark in the Kentucky River drainage. The area is underlain by Ordovici- an and Silurian limestone (Hall and Palm- quist, 1960). Elevation ranges from 500 feet on valley floors to as much as 1,000 feet on ridge -tops on shale and limestone hills; major drainages range in elevation from 800 to 900 feet. Graybrown podzoHc soils underlain by soft limestone or highly calcareous shale are conspicuous on steeper ridges (Austin, 1965). The continuity of underlying strata lends some degree of similarity to major streams and artificial impoundments throughout the area, but differences in topography and land-use patterns affect stream habitats and the faunal complexes they contain. Conse- quently, several streams, were surveyed before the study areas listed below could be chosen. Since it was desirable to compare certain data from different habitat types, the bullhead populations of Wilgreen Lake and several smaller impoundments were also utilized. 106 Tulane Studies in Zoology and Botany Vol. 20 Streams: Streams throughout the study area were surveyed to ascertain the character of their bullhead populations. Two of these, Boone Creek in Clark and Fayette counties, and Strodes Creek in Clark and Bourbon coun- ties, were considered to have suitable bull- head populations. The Boone Creek drainage consists of first through fifth order streams from its headwater just north of U.S. Route 60 to its confluence with the Kentucky River, 0.8 km east of Clay's Ferry, Ken- tucky. Land adjoining lower order streams (first through third) was being utilized pri- marily as permanent pasture, with only a small portion under crop tillage. Third order streams contained numerous silt bottom pools that proved excellent habitats for iso- lated populations of black bullheads. Two third order tributaries, the East and West forks of Jones Creek, which converge approximately 10 km from the intersection of Cleveland and Sulfur Well roads, were sampled through the study period. Strodes Creek, located north of Win- chester, Kentucky (38° 02' latitude: 84° 12' longitude), is a tributary of Stones Creek which, in turn, flows into the South Fork of the Licking River in Bourbon County. The entire southern drainage joins the main fork of the Licking River at Falmouth, Kentucky. The streams in this drainage, especially Strodes Creek and its tributaries, arc maxi- mally influenced by agricultural use. Mean- dering stream beds are extensively silted, pools are abundant and stream flow is mini- mal during most of the year. All tributaries sampled contained large numbers of black bullheads. Johnson Creek, the most promi- nent tributary of Strodes Creek, was sampled during the spring and summer of 1971. Ponds: Farm ponds, and other small earthen im- poundments, were surveyed to determine the feasibility of their use for this study. Seventeen ponds were considered; of these, the five listed below were found to contain black bullhead populations and were select- ed for further study. Ponds \, H, and III were located on the Central Kentucky Wild- life Management Area situated 4.2 km east of Kingston. Kentucky (37° 38' latitude; 84° 14' longitude). Pond IV was located on Fair- view Farm approximately 2.4 km south of the intersection of state routes 1156 and 25 near Whitehall, Kentucky. Pond V was in Fayette County, Kentucky, 0.8 km south of state route 421 at 37° 54' latitude; 84° 20' longitude. Pond I: This pond had a surface area of approximately 0.65 acre (.26 ha), with a maximum spring depth of 2 m. The entire surface area was of open water; no marginal or emergent vegetation occurred. The pond accepted run-off from a denuded 3.0 acre barnlot. Pond II: Formed within a natural valley approximately 0.4 km wide, this pond had a surface area of 0.5 acre (.20 ha) and a maxi- mum depth of 2.6 m. It received a spring-fed stream which flowed throughout the study period. Emergent vegetation (primarily Sagittaria sp. and Scirpus sp.) and mats of filamentous algae (Spirogyra sp.) covered approximately 25 to 30 per cent of the sur- face through spring and summer. The bottom deposits consisted of pulpy peat and muck that reached depths of 0.9 m. Pond III: This pond was small and shal- low, with a surface area estimated at 0.125 acre and maximum spring depth of 1.0 to 1.2 m. It was located within a natural de- pression that contained some water through- out the year, but received the flow from adjoining pastures only during heavy rains. Emergent vegetation (predominantly Sagit- taria sp.) was prominent over the entire surface. Pond IV: With a surface area of 0.50 (.20 ha) acre and a maximum depth of 1.5 m, this pond also drained pastured areas. Aquatic vegetation was limited to insignifi- cant accumulations of pondweed (Pota- tnogetoii sp.): however, during spring and early summer it was extensively covered with algal mats (Spirogyra, Oedogonium and Zygnerna). Bottom deposits were of ooze to a depth of 1 .0 m. Pond V: This was a pond of recent origin (four years) located at the base of a natural valley. The open water area covered approxi- Nos. 3-4 Black Bullhead 107 mately 0.25 acre (.10 ha) acre, and the maxi- mum spring depth was 1.5 m. Lakes: Wiigreen Lake, formerly known as Taylor Fork Lake, is located in Madison County, Kentucky, with spillway and access point located approximately 4 km from the inter- section of Barnes Mill and Curtis roads (37° 54' latitude: 84° 20' longitude). The lake has been inadequately surveyed surveyed; conse- quently most morphometric data has not been detailed. The shore Une extends an esti- mated 42 km, containing 175 surface acres of water (71 ha). The early spring maximum depth was between 24 and 26 m. Shoal areas (water less than 1.2 m deep) comprised an estimated 15.0 to 20.0 per cent of the total area: aquatic vegetation, other than fila- mentous algae, was conspicuously lacking from shoal waters. Wiigreen Lake was sampled during the period July 1970 to June 1971. Capture Methods Several methods were employed to cap- ture specimens. Most pond sampling was during winter months (Nov. 1970 to Feb. 1971) using wire traps. Larvae and young-of- the-year were sampled by means of a modi- fied meter net, and by hand netting along shore -lines. Stream populations were sam- pled with a 15 foot (4.57 m) common-sense seine of 1/4 inch (6 mm) linear mesh. Sam- ples were taken from Wiigreen Lake during the fall of 1970, and from February through June 1971. using: 5-cm inner mesh trammel nets established across major channels; wing nets, 3.8- and 2.5-cm mesh funnels and 1.3-cm wings, fished in waters less than 3 m deep: bag seines trawled in affluents and shallow coves: and 230-volt electric shockers operated along shore. Population Dynamics Food Habits. Food habits were assessed for all populations studied, i.e., from ponds, lakes and streams, and for all age groups within the populations. The terminology of Hubbs (1943) for establishing age classes was followed. According to this analysis, tran- sition from lower to higher age class occurs on either January 1 or July 1 (Hile, 1948). Since the black bullheads in this study were found to form annuli during the month of June, July 1 was used as the transformation date. To determine feeding selectivity, benthic and planktonic analyses were carried- out during all seasons of the year, and the results scaled to reflect annual availability. The weighted percentage of availability was then compared to percentage estimates of appear- ance of the same items in the stomach con- tents of representative samples to form an electivity ratio for each item. The electivity ratio used by Priegel (1970) is represented by the equation: r;-P; E =■ r^ + Pi Where rj is the relative percentage of any food item in the stomach, and Pi is the rela- tive percentage of the same item available to the fish. Values of (E) range from -1 to +1, and a value of 0 was expected for a food item when no selective process was operative. In- vertebrate identification followed Eddy and Hodson (1961) and Pennak (1953). Age ayid Growth. Representative samples from all study areas were investigated to determine age and growth characteristics. Age determinations were made by modifica- tions of procedures suggested by Lewis (1949) and Sneed (1951). Lewis aged black bullheads by viewing the annular markings on vertebral centra, and Sneed was able to age channel catfish by the number of an- nular depositions on pectoral and dorsal spines. In the latter case, these sections were made near the base of spines, a drop of xylol added, and the preparation viewed under indirect light. In this study a more facile modification of Sneed's procedure was suc- cessfully employed. Instead of making a thin section, the pectoral spine was simply sever- ed near its base with a high-speed rotary cutter (Dremel Model 42), the cut portion polished if necessary by the use of a high- speed jeweler's stone, and the entire spine mounted in styrofoam for observation under 108 Tulane Studies in Zoology and Botany Vol. 20 indirect light. Age estimates, when compared to preparations by Sneed's method (1951), were identical. Total length (tip of snout to tip of tail) and a standard length (tip of snout to the flection point at the caudal peduncle) were determined on a standard mm-measuring board. All weights were determined on a tri- beam balance permitting a precision of ±0.05 g. The coefficient of condition (K) was determined for representatives of pond, lake and stream populations. The condition factor was that proposed by Hile (1936), represented by the equation: _ lOOW where, K is the coefficient of condition, W is the weight expressed in grams and L is the standard length expressed in centimeters. Since the coefficient of condition was de- signed as a comparative value, it has little significance as an absolute number. Weight-length relationships for each habitat were determined according to the ex- pression Log W = a + n Log L, where W is the w/eight in grams and L is the length in millimeters. The resultant regression equa- tions were analyzed as to the homogeneity of their slopes. In addition, correlation co- efficients were employed to establish the relationships between weight and length, age and weight, and age and length for each population. Population Estimates. Population esti- mates were made for pond populations to relate degree of crowding and intraspecific competition to population success. These estimates were made by employing the mark-recovery procedure detailed by Fredin (1950) which modifies the Peterson method by considering each sample as a separate esti- mate. By this method, usually referred to as a Schnabel estimate (Ricker, 1975), a more accurate estimate is assured by the process of minimized squares according to the ex- pression (where marking is continued into the recapture period): S n2 (m + u) S(nm) (1) where, N is the total population estimate, m is the number of fish marked during the first sampling period, u is the number of un- marked fish in the second sample, n is the number of marked fish in the pond, and S represents summation. Ricker (1958) noted that (1) was most effi- cient when the value n/N was 0.5. Where this assumption could not be made, and in cases where subsequent estimates were made over long periods of time (one month or more), the standard Peterse"!! mark-recovery index was utilized (valid only if M is constant). According to this procedure, the total popu- lation (P) is equal to a ratio of the number of marked and unmarked fish in a pond according to the expression: P = M (U + R)/R where, M is the number of fish marked dur- ing the first sampling period, U is the number of unmarked recoveries and R is the number of marked recoveries. Natural Mortality. The bullhead popula- tion in Pond I was followed throughout the study period to set limits on the extent of natural mortality, and to determine the period or periods when fish are most vulner- able to the factors of natural mortality. The procedure used to calculate natural mortali- ty followed those of Regier (1962) for ex- ploited populations, and will not be detailed here. Although this pond was not exploited through fishing effort, sampling in the pond was considered exploitation, and the mortal- ity was analyzed accordingly. Reproductive Potential Gonadal Development. Fish were re- moved from Pond I and II during the period August 1970 through July 1971 to obtain an index for male and female gonad develop- ment after procedures suggested by Lari- more (1957). Since Dennison (1970) estab- lished gonad development curves for male and female black bullheads in a lake popula- tion, greater emphasis was placed on pond Nos. 3-4 Black Bullhead 109 development during this study. However, estimates were made throughout the spring and summer of 1971 for lake and stream populations, and the results were superim- posed on pond fish curves. The gonadal development index is simply a ratio of gonad weight to body weight. Samples from ponds were analyzed at bi-weekly intervals from August 1970 through February 1971, and weekly thereafter. From these values the high and low points of the reproductive cycle were determined. In addition, gross morphology of the gonads was described for each stage in the development sequence. Fecundity. The several methods used for analysis of egg production in fish (Davis and Paulik, 1965; Boyar and Clifford, 1967; Behmer, 1969; Larimore, 1957) can be grouped under three headings: dry weight measure, volumetric measure, and egg dis- placement. Each of these methods was analyzed in contrast to total egg counts to determine the most suitable method for bull- head fecundity determinations (Table I). Subsequently, all determinations were made by the egg-weight and displacement methods of Burrows (1951b); the dry -weight method was excluded. Since all methods tested over- estimated the actual number of eggs (minus any adhering germinal tissue), all determina- tions were corrected by factors cor- responding to the magnitude of over-esti- mation. Displacement determinations were multiplied by a factor of 0.90, volumetric methods by 0.79, and egg-count determina- tions by a factor of 0.98 (Table I) to account for adhering germinal tissues. All weights obtained during this procedure were determined by means of a Mettler 400 ana- lytical balance with ±0.0005 g precision. Induced Spawning. Eight pairs of black bullheads, designated as possible spawners, were brought into the laboratory and segre- gated into breeding pairs. The fish ranged in weight from 130 to 145 g, and in standard length from 168 to 220 mm. They were maintained on a diet of adult crayfish re- inforced with liver. The male and female of each pair were separated from each other in 50-gallon (189 1) aquaria by imposing 1.25 cm mesh weld-wire partitions. The photo- period was varied according to the environ- mental photoperiod, and temperature was increased at the rate of 1°C per day to a maximum of 7.1°Q,. In accordance with a procedure detailed by Sneed and Clemens (1960), each pair was treated with dried carp pituitary which had been previously prepared by desiccation in four 12 hour changes of acetone, then stored in sealed vials. Each fish received 4.0 mg dried pituitary per 24 hour period until signs of pre-spawning behavior appeared, or for a total of 10 intramuscular injections. Pitui- tary glands used in the study were obtained from carp seined or trapped in local streams; these were not separated as to sex or stage of reproductive development. Pituitaries used were not over two hours old when dried. Physiological Ecology Oxygen Consumption. Oxygen consump- tion rates were determined for age groups post -larva through age group I (to formation of second annulus) at three acclimation temperatures, 7°, 12° and 23'C. Acclimation was accomplished for each group by subject- ing ten or more fish to progressively decreas- ing temperatures in a Sherer-Gillett Environ- mental Chamber. Once the desired tempera- ture was reached, fish were held at this temperature for an additional 7 to 10 days. Notably, however, post-larval specimens (approximately 2 days post-hatching) could not be accUmated to 7°C. Consequently, determinations for this group were made only at 12°and23'<:. Oxygen consumption rates (standard rates) were determined by immersing fish held in weld-wire cages consistent with their total length, into 5.54 liter glass aquaria. The aquaria were sealed with a plexiglass cover and silicon sealant, and filled to capacity through a single 7.5 mm hole drilled to ac- cept the BOD probe from a YSI Model 54 Oxygen Analyzer. Subsequent to filling, fish were allowed to accustom themselves until locomotory activities minimized before initial oxygen readings were made. Prior to each experimental run, demineralized water in the chambers was recharged with oxygen to a level of 8.0 to 9.0 ppm. (Oxygen con- 110 Tulane Studies in Zoology and Botany Vol. 20 Table 1. Preliminary analysis of egg count methods. Column A= total counts; Column B= estimation by displacement methods; Column C=estimation by volumetric methods; Column D= individual egg weight m_e_thod. SL= standard length (mm); D= deviation from total count; X= mean of deviations. SL LOCATION A TOTAL COUNT B ESTIMATE D C ESTIMATE D D ESTIMATE D 169 POND 1891 1923.0 +32 2312.5 +493.5 1900 +9 176 POND 1927 1985.0 +59 2404.0 +477.0 1943 +16 176 POND 1914 1922.0 +8 2322.0 +408.0 1923 +9 210 POND 2880 3845.0 +965 3520.0 +640.0 2992 +112 156 POND 1600 1748.0 +148 2104.0 +405.0 1623 +23 195 POND 2880 3312.0 +432 3415.0 +535.0 2903 +23 177 PONT) 3015 3150.0 +135 3271.0 +156.0 2975 +40 X = 2301.0 X = 235 X = 473.4 X = 33 % ,dev = 10.21 % dev = 20.54 1 dev = 1.75 Correction 100 - . 1021 100 - .2057 100 - .0175 Factor (.90) (.79) (.98) sumption rates were read in mg 02/g/hr for purposes of comparison.) Consumption rates at the three acclimation temperatures were expressed as functions of stage of develop- ment, and the corresponding QiQ values between each pair of acclimations were established according to the equation of Prosser and Brown (1962). In addition, oxy- gen consumption per fish was expressed as a function of weight at 23° and 12°C. For age groups with weight ranges greater than two standard errors of mean weight, consump- tion rates were converted to the rate of an individual of mean weight according to the equation: M m (Wi, X M)/W2 Where, M,-n is the metaboUc rate of a mean weight fish, Wm is the mean weight, M is the metabolic rate of any fish in the sample, W is the weight of any fish in the sample, and a is the slope of the weight versus consumption rate curve for any size group (Smit, et al., 1971). In addition to routine rates, oxygen con- sumption was measured as a function of de- creasing oxygen tension and increasing car- bon dioxide tension in order to establish threshold values. Water used for these pro- cedures was first "stripped" of oxygen down to 4.0 ppm (2.8 ml/1 at 20°C) by subjecting it to a jet of nitrogen gas. Fish were then allowed to spontaneously reduce oxygen levels to hear-asphyxiation levels. Rates were continuously monitored by a procedure ana- logus to that described above. Nos. 3-4 Black Bullhead 111 Temperature. Lower incipient lethal tem- peratures, critical thermal maxima and tem- perature preference ranges were determined at the three acclimation temperatures listed above. Lower lethal temperatures were estabhshed by subjecting groups of ten or more fish (post-larva through age II) to progressively high temperatures from a mini- mum corresponding to the freezing point of water. The point in time at which 50 per cent mortality (LT50) occurred was re- corded, and the temperature at which no further death occurred with time was con- sidered the lower lethal temperature. While under observations the fish were disturbed only for periodic aeration of the water. Critical thermal maxima (incorporated into the data analysis as upper lethal temper- atures) were determined by a procedure similar to that described by Hutchinson (1961). Tri-necked distillation flasks were fitted into hemispherical heating mantles which could be heated at any desired rate by means of a variable voltage transformer. Animals to be tested were placed in suffici- ent demineralized water to allow free move- ment. Alternate ports of the flask were pro- vided with an air hose to aerate and circulate the water, and the probe from a YSI Model 4S2 Tele-Thermometer. Temperatures were recorded at the beginning of each run (to) and at the CTMax (tf), which was the point at which an animal failed to maintain an up- right position. All animals used for CTMax determinations were acclimated at 7°, 12°, and 23°C for periods of 7 to 10 days. Sub- sequent to determination, the previously unfed animals were allowed to recover in water held at their prior acclimation temper- ature. Temperature preference ranges were determined for age groups post-larva through age I, since design of the apparatus did not facilitate the handling of older fish. Animals previously acclimated to the temperatures designated above were placed in a 10-foot galvanized gradient chamber. The chamber, located in a walk-in refrigerator of 5° ambi- ent temperature, was so constructed that temperature could be graduated from 3° to 27'XI through 12 previously marked inter- vals. Ten to fifteen animals of comparable age and size were placed in the chanber at the point corresponding to their prior accli- mation temperature. At hourly intervals, for a total of nine hours, the animals were ob- served and their positions in the gradient recorded. This was easily accomplished with- out causing disturbance, since the fish could be observed from outside the refrigerator. In order to eliminate light as a factor in range selection, determinations were made under full illuminations on fish previously accli- mated to constant light. Correlative Studies. In conjunction with laboratory and field studies, water chemistry analysis was accomplished for each habitat sampled following Hach Chemical Company procedures. Table 2 contains a complete list- ing of determinations, and the range of each year specified periods of time. Statistical analyses followed the pro- cedures of Sokal and Rohlf (1969), and Downie and Heath (1961). RESULTS During the period July 1970 through July 1971, a total of 623 adult black bullheads (age groups I through VI) was taken from the study area. Fifty-six per cent of these (349 specimens) were obtained from ponds, 36% (224 fish) from streams, and the re- maining 8% (50 fish) from Wilgreen Lake. In addition to adults, 300 post-larvae and juveniles were taken from the three habitat types. Food items consumed by 292 adult black bullheads, the availability of each item in the habitat expressed as a weighted volume per cent, and the corresponding electivity ratio of each, are presented in Table 3. Although a great variety of food items was taken by adults, there was evidence to suggest feeding selectivity. Table 3 shows that chironomid larval and pupal stages made up a predomi- nant part of the adult diet; however, this item had no greater selection value (0.1 to 0.5) than some other less common foods. Food items taken most frequently dif- fered from one habitat to another. Pond populations took chironomids and additionally relied heavily on physid and 112 Tulane Studies in Zoology and Botany Vol. 20 0) ■C) CO ID W +^ (1) o ft •H X c 0) CO 0) CO t>D x: c M o e tj c 0) nS x: +-> CO U hD cSS ID M c O -H (D ID o o m ■d o cti -P ID 2 -s<: -P c o (D OO c c . O -H 0) o U T) bDH tt) r-i r-i C 0) c tci CO to ID u X ID CO >5 tt) M ^ M M '-3 d +-> -x; n! M M H 3 0) to O Pi c xi 0) o +J H Pi o Cm C^ • O On'^ r-{ S CO Ph CO C- — >> 3 rH i-^ C nJ O C --H ni -d rH O r-i f^ ^H e -P (D to Pi Jh e 0) Pi ID x: x: -p o c U .H > M • x: c ^ -P rt rH ttj f-i < s: o (M o o o Q -P ID CO XI o OO OLOOOOOOfNloO CTl X- hOvO •LOLOOL/IOCTlCTlo ^ O tNl(VJ0OrH(VlhOrsl(N|rsi(Sl CT^ o •<3- vD 00 O 00 O •K * -K * Or-IOvOI^vOOt-~-« « « II ...... t^Oi— li— li— li— lr-IC3 O O LT) LO I— lOC^t-O^fNltNl,— I* II ...... * * * * i-ir^OOOOOO* * * o o o en I o LO o o o o (NI o LO OO o" (Nl o I^ o t— 1 r~- (VI o o o "* 'S- ^ * * * *o 1 1 r ■K ■K 1 1 ^ o o OO OO 00 r^ LO LO C o o 1—1 rH :— 1 o o r-0 "a- "* O) CTl LO Ol OO Ol cri ro o o 00 I ^ 1 o LO o \o 1 VO 1 o ^0 o o o o 1 o M- LO 'C >o r-- LO LO (Nl o 'a- ■^ vO r-~ i~~- r- LO OO vO o vO UO VO vO LO LO vO vO 00 LO LO LO ^o O o hO en »£) t-^ 00 1— 1 ■—1 o (Nl 00 eg t^j 1—1 1 ■K * 1— 1 r-i 00 f— 1 1 t— 1 Ol CTi 00 1 * * ■K (Nl o rsj o \0 •^ vO o OO vO 00 LO o 00 rsi o OO CTi o^ ■^ LO LO O INI -^ O t^ vO (Nl ^ ,-1 O ■^ o (Nl (Nl tn ^ 00 •^ o o 1— 1 .—1 1 OO 1 1—1 1 rH vO (Nl (Nl (Nl ■^ •^ •=!■ 00 o 00 00 vD 00 00 OO 00 oo.,_i,_i • ••.•.•».....(— 5 fv-) CTlO^vOi— IO(NlO'3-(OtOhOOOO_j _i ■-I 1-1 ^ "^ •-' O ■^^ ^ E' <=> •-' o « r-- r^ 1—1 1— ii— I r--v^ "« «,_( i^ i-ir--r^ i-(OC3 (D»(-it^«s^t~^ .-It-- i-itN.r-~t---.xi>-iaj(u>,C "f^ «.r^ j->EiUX)^CS« ^ ••x:i-i """I'j'U-CiG'ecSax: •H -tuu-H '-(uiuxD^-'OiuSaCtj D.^3«D.rt33Dga)uo5J alj S CJ3 g Nos. 3-4 Black Bullhead 113 Pi e QJ O HO C (U bO r^ O X U z; -P •H •H P< :g p£) ■o O iO vD o vO (NJ O o LO r~- LO ^0 LO vD vD >r3 vO 00 1- o (NJ o i-( o O o (Nl (NJ (NJ o o o o >* rt •D- 00 ■* •rr 00 't '^ (» 00 00 oo oo 00 00 CO CO (Nl o O O LO o o o o o o o <=^ <='. <==. lo'lOI-^-'lOLOIOLOLOLOLOI-'^'-"^ r-H rt O C= CD LO LO (— ) O • • • "* IlHLor^i-((TioooOLnLO(Ni oooi— ( 1 ..(111 .,..•1 tOOO'^(JlLOOLO(NltO'3- "^tOLO a^ 00 I-- o o J^ CD r-~ o o 1—1 o t~^ f^ "— I ""^ O - t~^ t-- r-- I- . « « r-( i-(r~-t~~t~^ - X^ U dj ^ >~ S.(fl9933uoa) TOdjw 00 oo 00 o (Tl o O ^ LO t-~. o >o LO r-~ CO r-i "a- rH o LO CTi LO LO CTl to (NJ 1— ( rH rH t-~ LO -* (Ti LO LO to Ol LO O cn OO o "^r (Nl •T3 C •T3 O c < 5! 1- X •a, «i 2 S 114 Tulane Studies in Zoology and Botany Vol. 20 Table 3« Food consumption of 292 adult black bullheads expressed as (a) percentage of total complement, (B) represents percentage of each item available and (C) is the electivity value of each available item. Food Organism Ponds Wilgreen Lake A ^ B 1 Streams A B C A B C Annelida Oligochaeta 6.1 1.5 0.6 8.8 3.5 0.4 2.9 1.9 0.2 Pelecypoda Spaeriidae 3.2 12.4 -0.6 O.k 7.4 -0.9 13.1 15.3 -0.1 Gastropoda Planorbidae 2.1 0.8 0.5 2.8 5.5 -0.3 2.9 1.4 0.4 ftiysidae 5.5 0.8 0.7 16.0 5.5 -0.6 5.8 6.4 -0.1 Rotifera 0.0 3.2 -1.0 0.0 7.8 -1.0 0.0 » • • • Gladocera 0.0 9.3 -1.0 0.0 12.5 -1.0 0.0 ♦ • • • Copepoda 0.0 6.5 -1.0 0.0 3.5 -1.0 0,0 * • • • Ostracoda 0.0 1.3 -1.0 0.0 7.8 -1.0 0.0 ♦ > a • Amphipoda l.k 2.5 -0.3 4.9 1.7 0.4 2.9 5.3 -0.3 Isopoda 0.0 1.1 -1.0 0.0 0.3 -1.0 0.7 0.5 0.2 Decapoda 1.5 1.5 0.0 0.0 0.6 -1.0 5.1 7.8 -0.2 Ephemeroptera 1.5 0.3 0.6 1.1 1.1 0.0 3.6 2.9 0.1 Plecoptera 0.0 0.0 0.0 0,0 1.5 1.1 0.2 Odonata Zygoptera 3.0 2.1 0.2 4.2 1.1 0.6 0.7 2.0 -0.5 Anisoptera 6.7 1.5 0.6 2.1 1.9 0.1 3.6 3.0 0.1 Trichoptera 0.0 0.0 0.0 0.0 3.6 2.9 0.1 Diptera Ghironomidae '+7.2 il5.2 0.0 55.5 19.3 0.5 39.0 11.9 0.5 Tabanidae 2.3 0.6 0.6 0.0 0.5 -1.0 0.7 0.1 0.7 Tipulidae 0.0 0.0 • • • 0.0 0.0 3.6 0.7 0.7 Simulidae 0.0 0.0 • • • 0.0 0.0 . . . 8.0 2.9 0.5 Unidentified 2.0 0.8 0.5 4.9 1.4 0.6 0.0 0.0 Hemiptera. Gorixidae 0.0 0.8 -1.0 0.0 0.9 -1.0 2.2 1.4 0.2 Amphibia Ranidae (Larva) 1.5 1.8 -1.0 0.0 0.0 0.0 « Pisces Lepomis macrochirus 3.8 5.6 -0.2 0.0 * • • • 3.6 3.6 0.0 Micropterus salmoides 2.3 0.5 0.7 0.0 ♦ • • ■ 0.0 ♦ 0.0 Pimephales promelas 13.0 7.8 0.3 0.0 * • • • 2.2 1.0 0.4 P. notatus 0.0 0.0 0.0 * • • • 2.9 4.6 -0.3 Gyprinus carpio 0.0 0.0 0.0 0.0 • • • 0.7 0.2 0.7 Notropis sp. 0.0 0.0 . . • 0.7 ♦ • • • 8.0 7.4 -0.4 * Density Not Determined Nos. 3-4 Black Bullhead 115 planorbid snails, sphaeriid clams, dragonflies and the young of fish species (especially the minnow Pimephales promelas). Adults from Wilgreen Lake consumed appreciable num- bers of the amphipod Hyalella, larval damsel- flies, physid snails and oligochaetes. Adults taken from streams consumed a wider range of food items than did populations in either lake or ponds, but still relied heavily (39%) on chironomids. Although of lesser import- ance, tipulid and simulid larvae, decapod crustaceans, mayflies, sphaeriid clams and physids were found more frequently than other food items. Fingernail clams (Sphaeri- idae) formed a notable portion of the food consumed in ponds and streams, but appear to have been selected against in both situa- tions. A predominant percentage of stom- achs investigated contained some fila- mentous algae (181 of 282: 64%) from March through May, but this item was not considered in Table I because its true signifi- cance as a food item was not determined. Adult feeding changed markedly with season. According to our observations, from a peak during late spring, feeding diminished with the progression of summer, increased again in early fall, and appeared to cease entirely during winter months. The stomachs of 31 fish captured from ponds during 15 December 1970 to 1 February 1971 con- tained no food. Wire traps used to capture these specimens were usually left unraised for 12-hour periods; however, peptic di- gestive rates during winter are sufficiently low (Nordlie, 1966; Baur, 1968) that some food should have remained if the fish were actively feeding. Measurable quantities of food did not appear in the stomachs until mid-March, corresponding to temperature increases above 10°C. Larval and juvenile feeding appeared even more selective than that of the adults. Table 4 lists the food items consumed by 68 larval bullheads from Wilgreen Lake and Pond L Since larval food analysis was not accomp- hshed for stream populations, these data are not comparative. They did indicate, how- ever, that food was selected far more closely according to its position in the habitat than its relative size. TABLE 4. FOOD ITEMS OF 68 LARVAL BULLHEADS: COLUMN A= EACH ITEM AS PERCENT OF TOTAL FOOD COMPLEMENT: COLUMN B= THE AVAILABILITY OF EACH ITEM AS HIHCENT OF TOTAL: AND COLUMN C= THE ELECTTVITY RATIO, Food Organism ABC Rotifera 1.8 12.5 -0.8 Cladocera 4.8 14.3 -0.5 Copepoda Copepodids 6.0 3.6 0.3 Nauplil 2.2 23. 1 -0.8 Ostiacoda 48. 0 9,0 0.7 Amphipoda 34. 0 8.0 0.6 Diptera Chironomldae 1,4 22.0 -O.9 Annelida Oligochaeta 0.1 4.1 -1.0 Nematoda 0.1 3.0 -0.9 Miscellaneous 1.2 ♦ ,,.. * Density not determined In both habitats sampled, ostracods, amphipods and adult copepods made up the bulk of the diet. Pond larvae relied more heavily on copepods than their lake counter- parts, apparently because of the differential abundance of ostracods and amphipods between the two areas. Nauplii, the most abundant planktonic component, and chiro- nomid larvae, the most prevalent benthic food, were strongly selected against. In addition, selection against other benthic ele- ments seemed to support the observation that most larval feeding was pelagic. Larval feeding was observed on several occasions during July and August 1970 and June and July 1971. Aggregations consisted of curious admixtures of different aged indi- viduals. Approximately 25 to 30 % of the pods observed during June 1971 consisted of yolk-sac stages (determined from capture percentages) which according to the time required to absorb the yolk, were considered two to three days younger than the others. Forming aggregations did not begin feeding immediately, but continued to wander until a visual orientation pattern was achieved. Pods observed in Wilgreen Lake were always oriented on an axis facing the shoreline. Once disturbed, the larvae broke up into less compacted masses, submerged and swam toward open water, where they regrouped on the same imaginary axis. Feeding appeared to be restricted to a limited area 116 Tulane Studies in Zoology and Botany Vol. 20 about the axis of orientation, and continued at the surface during daylight hours. Larvae were always found in the few areas where aquatic vegetation and floating algal mats were most prominent. Most observed feeding was at the water surface. However, the inclu- sion of annelids, free-living nematodes and chironomid larvae in the stomach contents suggested that some bottom feeding also occurred. The mean ponderal indices for adult fish sampled from each habitat during April, May and June 1971 were: ponds, (54 fish) K = 2.55, standard error 0.38; Wilgreen Lake, (23 fish) K = 3.00, standard error 0.51; streams, (32 fish) K = 2.50, standard error 0.49. The K factors were comparable for bullheads from the three habitats; a two-way analysis of variance indicated that no signifi- cant difference occurred at the 1.0 per cent level of confidence (F = 132.50/-2.31). Weight-length relationships from more extensive sampHng (372 specimens) through- LENGTHImm) Figure 1. Length-weight relationships of 372 adult black bullheads from pond 2, Wilgreen Lake and Boone and Johnson creeks. Curve extrapolated to the point where weight approaches zero. Solid circles=ponds; solid stars=lake; open circles= streams. 211' O) ISi- X O lij 5 •r.42 E E 151 ~ X O z UJ -SI AGE(years) Figure 2. Mean weight and length at the time of annulus formation of 372 adult black bullheads (age groups I-IV) for ponds, lakes and streams. Each curve accompanied by a coefficient of corre- lation (r) computed using all measurements about the line indicated. Dashed lines=length versus age; solid Iines=weight versus age; solid circles=ponds; open stars=lake; solid stars=streams. out the study period were established. The relationship for 167 pond specimens (age groups I through VI) fit the equation: Log W = -4.8498 + 3.1135 Log L; the relationship for 45 Wilgreen Lake bullheads of compa- rable age was Log W = —4.1812 Log L; and the same relationship for 100 stream speci- mens was Log W = i-2.4453 + 1.9968 Log L. An analysis of homogeneity of slope for the three regression curves indicated significant differences in the weight-length relationship at the 1.0 per cent level (Fg = 29.6948/F2,l03 = 4.82). The correlation between growth in length and weight was particularly high for ponds (r = 0.94) and Wilgreen Lake (r = 0.97), but not for streams (r = 0.68). The length-weight curve illustrated in Fig. 1 represents pond, lake and stream popu- lations and is interpolated to express the wieght at a point where length approaches zero. The curves in Fig. 2 illustrate mean Nos. 3-4 Black Bullhead 117 weight and mean length (mm) as functions of age at the time of each annulus forma- tion. Each curve is accompanied by a cor- responding coefficient of correlation com- puted for all measurements within the range. In computing these data, July 1 was taken as the date of annulus formation (year class I), although experimental evidence indicated that most black bullheads (27 to 31 or 87 per cent) formed the annulus before the third week in June. Correlation between growth in length, growth in weight and age in years, as indi- cated in Fig. 2, are quite variable among the three habitats. The most rapid growth rates were apparent in the Wilgreen Lake and stream populations, which also had the most favorable length-weight relationships and mean ponderal indices for all age groups studied. Growth in ponds was rapid to the time of second annulus formation, but showed a definite lag between the second and third year. In general, past the age of lii- M U- 41- (D w IN- (0 M. «D a 3 2 o < M- M- Wilareen lake Pondf Figure 3. Age composition of pond, lake and stream populations detcrmmed trom capture percentages. two years, pond bullheads had much lower growth rates than their counterparts in the other two habitats. No bullheads older than six years were taken during the study, and it is apparent from Fig. 2 that fish approaching the age of six already had begun to show unfavorable growth rates. Fig. 3 illustrates the population composi- tion for the habitats sampled based on catch data. These data show definite population trends, i.e., survival through age three (the age at which most bullheads reached repro- ductive maturity) in ponds and streams was nearly linear, followed by an inversion of the trend after first spawnings. The polymodal nature of the Wilgreen Lake population may not represent the true population trend, but may reflect sampling bias associated with several different sampling methods, e.g., electro-fishing and trammel netting were more selective for older fish. The populations in Ponds I and IV were subjected to mark-recovery estimations of population levels. Pond I, in conjunction with natural mortality estimates, was ana- lyzed three times during the period October 1, 1970 to July 10, 1971. Pond IV was esti- mated only once, prior to the 1971 spawn- ing season. Pond I represented the special situation in which no fishing mortality (p) occurred dur- ing the period of investigation, but samples of the population were removed periodical- ly. With certain carefully chosen assump- tions, this situation was made to fit the Model II-B analysis of Regier (1962). The model was derived from the fact that mortal- ity rates are either zero, infinite or directly proportional to the number of individuals in the population. Since the natural mortality (q) was assumed finite and constant, and the fishing mortality (sampling mortality in this case) was measurable over short periods of time, (q) was derivable through step-wise interation. An estimate of the natural mor- tality was obtained by partitioning popula- tion extremes about the calculated popu- lation line, and drawing from that line the tangent to a horizontal axis. Tnen, q = -2.303 c tan O, where c was a correction factor between the calculated slope b' and 118 Tulane Studies in Zoology and Botany Vol. 20 the true slope be determined from linear measure of the unit cycle on the semi-loga- rithmic plot (Fig. 4). The natural mortality coefficient, cal- culated according to the graphic method illustrated in Fig. 4, and the relationship q = -2.303 c tan O was 0.89. Extrapolated along the line to + 1, the population curve extends for a period of one year. Thus the coefficient 0.89 represented the probability that any fish in the population would die at any designated instant in time throughout the year. Observations carried out at Pond I indicated that black bullheads were more vulnerable during spring and early summer than during other seasons of the year. Popu- lation estimates determined on October 1, 1970 (680 individuals; 413 fish per ha) and April 15, 1971 (561 individuals; 341 fish per ha) suggested that the population remained relatively stable during winter months. Dur- ing the week of 13 May 1971 a mass mortali- ty occurred in this pond, which claimed an estimated 200 fish, or 35 per cent of the original population. A similar and coincident die-off occurred in Pond V. This mortality occurred after the calculations discussed above were performed. Reproductive Development The reproductive development curve for pond bullheads was examined during the 1970-71 season. Fig. 5 depicts the gonad development sequence, where points are plotted as gonad weight-body weight ratios. Although the sequence of development was not followed throughout the season for lake and stream populations, spring and summer ratios for these populations are superim- posed on pond curves in Fig. 5. Five stages were apparent in the development of pond bullheads. Table 5 presents delimitation of each stage, accompanied by its significance and gross morphological appearance. 900 800 ♦- 700 09 E 600 m 500 400- •2 300- a a. 200- N F M A Time (months) M Figure 4. Graphic Ulustration for determination of the natural mortality coefficient (Q) by the method of stepwise iteration about the true population line. Nos. 3-4 Black Bullhead 119 The gonads of both males and females were at their lowest point during the post- spawning period; however, this period was apparently relatively short. Males taken dur- ing late August and early September had gonad/body weight ratios that averaged 0.7 per cent, as opposed to 1.2% for mid-August males. Female ratios varied more than those of males, but showed a corresponding in- crease (5.8 to 6.2%). The weight ratios of both sexes decreased sharply between Sep- tember 15 and October 15, and remained low (0.3 to 0.4 for males and approximately 5.2 for females) throughout the winter. Gonad development in the spring was not an instantaneously initiated response. Although the reflex, once triggered, appears to have been irreversible, there was considerable fluctuation in the mean response with time, seemingly related to fluctuating temper- atures. Mature females reached their peak gonad development (12.1%) on 23 June, and re- mained relatively high until sampling was terminated on 10 July 1971. The 1970 indices remained high (10.0 to 11.5%) until mid-July. Mature males appeared to reach a reproductive peak later than females. An o o 12- 11- 10- 9- 8- 7- 6- 5- 4 3- 2- Wilgreen Males Boone Creek Males'^ -25 -20 ■30 -15 ni 9 t 3 o o -10 -5 N J M M Figure 5. Gonad development curves for black bullheads obtained from ponds for the 1970-71 cycle. Imposed on the curves are corresponding records for Wilgreen Lake and Boone Creek obtained during spring and summer 1971. See Table 2 for temperature measurements. 120 Tulane Studies in Zoology and Botany Vol. 20 n a o -p H s, o Pi a o p< a o u u TJ nj (U X! 3 O o -p C a o H > ^ o 0) x: o nJ 0) o o a (6 o o c Q) P< 0) e5 o CM bD CO o 0) o c cd o •H tH •H c bo •H 0 o o 0) o c 0) p< -p O 0) g tiD O -P 4) cn > cd -^ o U ft w 0) (0 CQ 0 >> o o u c 0) w w cd 0 u CO •H -P 0 w t:3 C C •H (d I (d O w td > O CQ •H — -P 0 a< H S 0 A bD o •- (d 0 .H Cd H bD o C 6 o •^^ rH M Cd ft o o ft -p cd 0 -p • •rH T) e 0 •H N 0 u^ -p • • cd 0 O -H -P U -H o ^ x: -P 6 » •H CntH >> S o >> Cd rH 0 .- 0) u 0 e o ^ 0 ^ M •• rH -P M C X c O 0 •H ,a bD ^ - fH « >1 I x; o u I I cd cd ^ > -p — C rH 0 r-\ ft (d (0 bD bD 0 0 o cd 0 ft ft cd bD bD C C O .H H U ^ 0 o x: 0 H-> > c •H 0 bD o to rH Cd M 0 C U cd Cd M -p u cd rH H rH 3 cd O S O •r-i C rH I e Ch -H x: -p bD C cd — 0 0 U M o •H C C O 0 w cd 0 o c •H c o d 6 H O ft ^ -P Ch cd bD-H a u cd 0 cd ■H U o ■H W -P rH C 0 e H 0 cd Ti U > 73 3 o CO CO bD c •H - — s ft 0 CM O bD 0 ^ > CO 0 Q g s . O M to bD hJ to O C 0 3 •H X -rt 0 O 0 CO M 0 Oi fH x: tH H-> o CO pq o 3- 0 CO 0 CO Tj to cd O 0 (0 U rH rH ft g 0 u X 0 0 Ch o CO +j o 6 tJ cd C •H cd C cd C 0 M •H ft (d o > e rH o e 0 > • x) o >^ CO rH 0 bDrH 0 bD 3 ^ :^ ^ (i-i XI 0 bQ •H U Cd rH 3 O O O U O £ O u Ch >jrH rH 0 x: >-, O ■H 0 U CO 3 H tH H O 0 O cd x: 0 lo -p ft • fto e 0 cd lA ^ C CM ^ o • rH boo C O •- - rC P 0 to t3 O -rl XI o o S -P -H U 0 cd 3 rH T) 3 O 0 CO -P cd •H > x; 2 xi 0 x: CO to cd •H 0 ■s^ •H C ft-H C ^-^ bo O t^ C -H -^ -H -P C -rl S -d cd C ft O CO O 0) ^ CO p o I X ]) CO 0 1 bDrH to >5 1 •^^ rH 3 0 rH -H rH rH Cd to !h CO ,^ 5Pt:) r-\ 0 H Ch C X> 3 O U P &• rH 0 •H 0 to H S rH d 0 to Cd C 0 fl tS > rH Tj M > 0 x: o x: -p g 0 0 >irH — [/] ■ a. •rl 0 ■^ Cd • Ti -H c X bD e +^ •H '5 •H C O -H o •H H-l T) M cd fn 0 tH 0 a o e •p c 0 bD H^ 0 cd -H •H 0 ft • •* (d bO rH tJ Xl M 0 bD ^'^ ^ Cm C O 2, r-l 0 •H C 0 •H X) M C •H -d o o CO > U ^ M 0 cd -rl c +^ O cd c 0 0 0 0 -P bD CM sa c =§ x: to c g 0 •H •H -J s -p s <: M n bq 1 >1 to C cd •H H bD to 0 0 J«i bD cd E CO x) 0 o a; Cm 0' 0 a^ •H -H C 0 rH C C 3 11 bD 0 0 O cd — C Xi " XI'--. 0 ft •S e •rl T-H ft •H cd M--^ cd rH 0 C ft t:) 0 0 0 ft cd bD+^ d (d -rl O -P J^ •H C CM tJ CO X « O^ — ^^ >1 u^ ^1 - — 0 -p > C 0 0 0 ft 0 CO « ^ I , s ^ 0 " > o e z • 0 tH -P O ft-P u u o <^H X) 0 to 3 0 0 O 0 0 > -p o c O tH Nos. 3-4 Black Bullhead 121 TABLE 6. FECUNDITY OF BLACK BULLHEADS FROM PONDS, WILGREEN LAKE AND BOONE AND JOHNSON CREEKS. STANDARD ERROR IN MM: Sr- = STANDARD ERROR OF ESTIMATED MEAN. PONDS SL = 190(178-208) WILGREEN LAKE SL = 198(186-220) STREAMS SL = 182(164-200) N Egg No. S^ N Egg No. S_ N Egg No. S- 23 2552 163 7 3372 184 17 2645 237 1(2^^2) = 3.3373/2.2829 t^ ^. = .3212/2.1785 t = 2.4265/2.3101 average peak of 02.1% was recorded on 1 July 1971. Although no male samples were taken after 5 July 1971, results from the 1970 season suggested that males retain this condition for approximately 30 days. Immature males and females had gonad development curves that mirrored those of adults throughout a good portion of the cycle. However, after reaching a plateau well below the levels found necessary for success- ful spawning, material previously mobilized in the gonads was rapidly reabsorbed. At the time of spawning immature gonads again re- flected the refractory condition (Table 5). The fecundity of 47 mature females of comparable size -ranges from pond, lake and stream populations was determined by a combination of displacement and egg-count methods. Table 6 presents the mean fecun- dity of comparable size-range females, stand- ard errors of mean egg number estimate and Student "t" comparisons between each sample. Pond females had lower mean fecun- dity than females in either of the other two habitats. Analysis by Student "t" compari- sons indicated significant difference between the mean fecundity value for Wilgreen Lake and the other two populations; however, the limited number of mature spawners analyzed from the former population could have masked the true relationship. Mean fecun- dity in ponds and streams was not signifi- cantly different (P > .05). Attempts to determine egg and hatching mortality in the laboratory were unsuccess- ful. Subsequent to treatment with dried carp pituitary, females exhibited a sequence of behavioral changes that suggested repro- ductive readiness. These females went through stages of appeasement when con- fronted by a male, and two of the five re- maining after the fifth injection constructed nests. However, males similarly treated with dried pituitary remained refractory even though long day photoperiods and tempera- ture regulation were imposed. On 4 July three of these pairs were termi- nated and their gonad/body weight ratios determined. One female (SL 207 mm, 245 g) had a ratio of 13.1 per cent, exceeding the upper limit for females taken from the wild. The remaining females (168 mm SL; 102. 1 g and 172 mm SL; 134.2 g) had ratios of only 1.2 per cent and 3.24 per cent, respectively. Males ranging in standard length from 180 to 185 mm, and in weight from 132 to 139 g, had ratios of 1.5, 1.9 and 2.0, mirroring the re productively mature condition seen in the wild (Fig. 5). Physiological Ecology Table 7 considers routine oxygen con- sumption rates for the first four free swim- ming stages of the life cycle of black bull- heads acclimated at 7°, 12° and 23'<:. Since postlarvae and juveniles were separated by 122 Tulane Studies in Zoology and Botany Vol. 20 TABLE 7. ROUTINE OXYGEN CONSUMPriON RATES FOR STAGES POSTURVA THROUGH AGE II AT 7°, 12° AND 23 C WITH THE CORRESPONDING Q.^ VALUES BETWEEN ACCLIMATION TEMIERATURES. Postlarva and Juveniles Yearlings Age Group II (/ 0 (B) (c) oxygen oxygen oxygen consumed consumed consumed Temp °C (ml O^/hr) Temp °C (ml O^/hr) Temp °C (ml O^/hr) (1) 23 0.28 23 0.17 23 0.09 (a) %0 1.78 ^0 1.18 ^0 0.12 (2) 12 0.15 12 ^0 0.1i^ 1.01 12 ^0 0.11 (b) 0.95 (3) 7 * 7 0.13 7 0.11 (c) Significance tests (Student "t") by column and row: Items indicated were significant at ^ level of confidence. Column B 1-3 Row 1 A-C B-C Row 2 B-C Row 3 B-G Not determined at this temperature only a few days from the transformation date, their consumption rates were grouped. The values listed in Table 7 are accompanied by Qio values computed between readings at the different acclimations. The lower portion of this table contains Student "t" comparisons for individual listings according to column and row. These data appear to represent a uniform graded series of oxygen consumption for the life stages examined. Table 7 shows that significant differences occurred only be- tween the extremes of observation, i.e., between postlarvae and Age II at 23°, year- lings and Age II at 23°C, and between year- lings at 12° and 23°C. The mean consump- tion rate for Age II at 23'XI! was lower than at the other acclimation temperatures; how- ever, the means are not widely separated over the range tested, and there was no sig- nificant difference between any two readings in the series. The QlQ values computed between each pair of acclimations ranged from a high of 1.78 between larval means at 1 2° and 23°C, to a low of 0.12 between the means of Age II at these same acclimation temperatures. Oxygen consumption increases in a nearly linear fashion when plotted on a double logarithmic grid (Fig. 6), where the lines are fitted by a least squares method. Such a plot, if representing a true linear relation- ship, presumably should show the magnitude of temperature effect on oxygen consump- tion by changes in the slopes of the lines for different temperatures. Fig. 6 reveals that there was less than an ideal lineation be- tween the log ot oxygen consumption (mg) and the log of weight in grams. However, since the tendency for consumption to de- crease with respect to the log of weight after the first year was constant for the two tem- peratures, the slopes of the two regression lines were analyzed as to their homogeneity. There was no significant difference apparent between the slopes (Fs = 1.0018/F(1, 38) = 7.35). Oxygen consumption rates were also Nos. 3-4 Black Bullhead 123 measured tor fish ranging from 20.15 to 37 mm as functions of decreasing oxygen con- centrations. It was found that these bull- heads could extract maintenance oxygen down to 0.01 ml/hr in oxygen tensions of 0.2 to 0.5 ppm. Below these levels, asphyxi- ation was imminent, and the runs were term- inated. Carbon dioxide tensions of 10.5 to 21.0 ppm were routinely determined by titration with O.OlN NaOH at the points of minimum maintenance. Mean critical thermal maxima (CTMax) for the four age groups, postlarva through age II, are graphically illustrated in Fig. 7. Here also, physiological tolerance was re- lated to stage of development. Postlarvae, acclimated to 12° and 23°C, had critical thermal maxima that increased 0A4X!. per degree increase in acclimation temperature, in contrast to 0.09, 0.14 and 0.23 for juve- niles, yearlings and age II, respectively, over the same range. There were significant dif- ferences in the effects of acclimation tem- perature on different age groups. All four groups had significantly different means at 23° and 12X1, (P = 0.05); however, only the age II group had a significant difference in the mean of 7° acclimation when compared to the two higher means (P < 0.05), al though the spread may not have been ade- quate to demonstrate difference. Temperature-preference ranges were determined for the first three age groups (above). Selected frequencies, along with 5.0- M E E 3 M C e e • M >t H o 1.0- 0.5- 12" 1^ 0.1 I I I I III 1 — MM Mil 1.0 10.0 Weight (g) 1 — r TTTT 50.0 Figure 6. The influence of weight on routine oxygen consumption at 12° and 23'C. Solid circles= 23°log Y=-0.189 + 0.075 log X; open stars=121og Y=-0.042 + 0.1124 log X. 124 Tulane Studies in Zoology and Botany Vol. 20 TABLE 8. ANALYSIS OF TEMPERATURE PREFERENCE OBSERVATIONS FOR POSTLAHVAL, JUVENILE AND YEARLING BLACK BULLHEADS AT 7°, 12° AND 23°C. M= MEDIAN FREQUENCY SELECTED, N= NUMBER OF ANIMALS OBSERVED, R = DURATION OF EACH OBSERVATION IN HOURS, x^/x^ .05 FOLLOWING EACH COLUMN COMPARES THE SELECTED FREQUENCIES OF ALL GROUPS AT THE SAME ACCLIMATED TEMPERATURE, AND x^/x^ .05 FOLLOWING EACH ROW GOMIARES THE SELECTED FREQUENCIES OF INDIVIDUAL GROUPS AT THE THREE ACCLIMATION TEMPERATURES* . POSTLARVAE WERE NOT ACCLIMATED AT 7°C. 7^ 12^ 23 N R M Postlarva N R M Postlarvae x^/x^ .0 5 N R M Postlarvae 10 9 25.1-27.0 10 9 23.1-25.0 17.89/3.84 Juveniles Juveniles Juveniles 10 9 17.1-19.0 10 9 21.1-23.0 10 9 19.1-21.0 7.92/4.82 Yearlings Yearlings Yearlings 9 9 19.1-21.0 10 9 19.1-21.0 10 9 19.1-21.0 26.02/4.82 x^/x^ .05 28.67/3.84 30.68/4.82 35.55/4.82 In order to compare the selection frequencies, the assumption was made that groups representing the same population with respect to temperature preference should have the same median frequency. After making this assumption, the frequencies were treated with the nonparametric "Median Test" of Downie and Heath (I96I). methods of comparison, are presented in Table 8. Generally, temperature preference inflected downward with progressive stage of development. In addition, acclimation to lower temperatures shifted preference tem- peratures upward for the first two life stages. Preference temperatures ranged from 23.5° to 27.0°C for postlarvae; 18.0° to 21.0°C for juveniles, and remained relatively unchanged with acclimation, 19.1° to 21. Ot, for year- lings. These data were gathered over nine- hour periods in continuous light. Postlarvae, brought into the laboratory from 23°C envi- ronmental temperature, showed a character- istic ascendence toward the temperature shown in pig. 8; however, after 10 hours in the chamber these fish almo-st instantane- ously shifted to a lower temperature (17.5° to 19.0°C). Two more sets of larvae were acquired to determined the character of this shift, and on each separate run the same phenomenon was observed. These data, along with critical thermal maxima and lower lethal temperature points (determined as the point at which no further death occurred with time) were represented as incorporations into the over-all thermal picture. Fig. 8 displays the thermal polygons of postlarvae, juveniles and Age II. Since lower lethal temperatures were not deter- mined for yearling bullheads, only their preference was imposed on the graph. Fig. 8 shows that each stage had a distinctly dif- ferent temperature polygon. Postlarvae had the highest upper mean limits, but this in- crease was compensated by the upward in- flection of lower limits. Larvae measured at 23° had a lower lethal temperature of 9°C, and at 12°C acclimation, a lower limit of yt. Juveniles acclimated at the same tem- peratures had lower lethal limits at 2.5° and 4.0°C, and Age II individuals had limits of Nos. 3-4 Black Bullhead 125 0.5° and 1.0° at 12° and 23'X:, respectively. The two latter groups, when acclimated at 7°C. had lower lethal temperatures corre- sponding to the freezing point of water. Pre- liminary analysis suggested that the true lower limit was probably more closely re- lated to the supercooling capacity of the water. Samples of blood serum, washed with heparin, centrifuged and suspended with equal volumes of water in an ethylene glycol bath suggested that serum had a freezing point of 0.8°C lower than the freezing point of water. However, this investigation was too limited to merit consideration here. DISCUSSION Population Dynamics There has been disagreement in the few past studies that have sought to determine black bullhead feeding habits, although the differences may be more apparent than real since the individual autliors used different e 36- iS 34 Acclimation Temperature °C Figure 7. Mean CTMax values for stages post- larva through age II at 7°, 12° and 23°C. Post larvae were not acclimated at 7°C; consequently, only the means obtained at the other acclimation tempera- tures are included. Solid circles=post larvae; open stars=juveniles; solid stars=yearlings; solid tri- angles=age group II. terminology and were not principally in- terested in electivity and the relative abun- dance of food. Several authors (Trautman, 1957; Clay, 1962; Smith, 1949) assessed the feeding habits of this species as truly omniv- orous. Forney (1955) and Viosca (1931) suggested that feeding probably was selec- tive, since great numbers of chironomids were consumed. Rose and Moen (1951), on the other hand, contended that this tenden- cy to rely on chironomids simply reflected the preponderance of this item in the food supply. Kutkuhn (1955) listed a variety of food items taken by adults; however, no single item was assumed to predominate in the food consumed. None of these investi- gations employed selective feeding analysis. Our study clearly demonstrated that certain foods were selected for, and others against, and that although chironomid larvae and pupae did make up a sizeable percentage of the total complement (39.0 to 55.0%) in lake, pond and stream habitats, their selec- tivity ratio was no greater than less abundant foods (Table 3). A notable percentage of specimens cap- tured prior to the spawning season 1971 had eaten measurable quantities of filamentous algae and plant debris. Forney (1955) indi- cated that there was a necessity for this species to feed on plant material in order to reach reproductive readiness, and Raney and Webster (1940) noted a sequence of feeding in early spring which included several plant foods. We could not substantiate the former contention during this study. One of the ponds most heavily sampled during the sur- vey contained no plant material except members of the Cyanophyta, yet the bull- heads in this pond reached reproductive readiness, as expected, during the 1969-70 and 1970-71 seasons. Feeding intensity, but not pattern, changed markedly throughout the year. Stream populations took a greater variety of food items than did individuals in either of the other two habitats. Possibly this re- sulted from greater availability rather than enhanced selection. Larval feeding appeared to be even more selective than that of adults. Larvae and 126 Tulane Studies in Zoology and Botany Vol. 20 juveniles relied heavily on the aquatic amphi- pod Hyalella, ostracods and adult copepods, but selected strongly against the nauplius larva of the latter and the truly benthic com- ponents relied on most heavily by adults. This was essentially the same food consump- tion described by Ewers (1931) and Forney (1955). However, the former authors con- cluded that peaks of feeding activity occur- red just before dawn. Early morning samples taken during this survey indicated that rela- tively less feeding occurs during early morn- ing than during mid-day. With respect to food selection, no other studies have indi- cated preferences as functions of available items. We concluded that items were select- ed for, not according to their availability, but according to the position and abundance they assume in the habitat. Observation of larval feeding suggested that it occurred along a visually or otherwise determined orientation axis. Since it is well documented that young bullheads have poor visual acuity, and correspondingly advanced chemical perception (Bowen, 1931), orienta- tion is believed to result from a combination of sensory mechanisms. However, orienta- ting larvae appeared to establish an axis of the Y-axis type described by Ferguson, Landreth and Turnipseed (1965). Disturbed larvae tended always to reaggregate toward this imagined line which formed perpendic- ular to the shoreline. Feeding also appeared to be restricted to a rather limited locus about this same axis of orientation, with the bulk of feeding occurring during daylight hours. Ponderal indices measured during April, May and June 1971 for specimens from pond, lake and stream populations were highly comparable from one habitat to another (P = 0.05). Since these ratios, re- ported throughout the range, are quite vari- able from one state to the next, there seems little value in making such comparisons here (see Carlander, 1968). ComparabiHty from one habitat to another is, however, impor- tant. Length-weight regressions for the same populations were found to be heterogene- ously distributed, as indicated by analysis of their slopes (Fg - 29.69/F(2, 103 = 4.82). In addition, length-weight correlations for the three populations were quite variable, the higl: ist values obtaining in Wilgreen Lake (r = 0.68). It was difficult to attach signifi- cance to this finding since variance between the separate populations could lead to lower values. Investigators in other parts of the black bullhead's range have reported both high and low length-weight relationships (Dennison, 1970; Lewis, 1949; Carlander, 1968). Our findings did serve to point out that members of the single species tend to show deviate population trends when en- countered in different habitats. The size in length and weight at the time of annulus formations were compared. Generally, increase in length was more close- ly related to age than increase in weight (Fig. 2). The size at the time of each annulus formation showed greater mean differences in ponds where there was a growth lag be- tween the second and third year. This find- ing is in line with the conclusion drawn by Houser and Collins (1962), who found that growth rates in Oklahoma bullhead popula- tions were lowest in ponds. It also appears to augment the observations of Moen (1959) that black bullhead growth is retarded ac- cording to the degree of Lntraspecific com- petition, which is doubtlessly more influenti- al in ponds than in streams. Three of the ponds utilized during this study. Ponds I, IV and V, were overcrowded with bullheads. A fourth pond, Pond II, was overpopulated by the bluegUl sunfish. However, the effect of overcrowding by this second species was not reflected in bullhead condition. Population estimates in Pond I indicated the conditional and genetic effects of crowding. On October 1, the population of this pond was estimated at 681 fish, or 1,021 fish per surface acre of water (431/ha). Hastings and Cross (1962) and Fredir (1950) reported the presence of similar sizect populations in turbid ponds, and commented on the relationship of crowding to weight gain. The age-weight cor- relation determined for ponds during this study (heavily weighted by crowded pond specimens) was quite low (r = 0.42). A great number of abnormalities were observed in the Pond I population, apparent- Nos. 3-4 Black Bullhead 127 ly resulting from overcrowding. Several fish w^ere captured which lacked one or more fins, and one individual captured during July 1971 had no paired fins at all. In addition to abnormal fish, most of the breeding popula- tion in this pond suffered from bacterial tin rot during spring and early summer. Micro- scopic examination revealed the presence of a proliferative streptococcus bacterium. Apparently, then, from samplings of three different populations within a compa- rable area, the greatest population success is probably achieved in ponds, as indicated pre- viously by Houser and Collins (1962). How- ever, this does not imply that greatest popu- lation survival also occurs in ponds. From observations during the period September 1, 1970 to July 10, 1971 a natural mortality Acclimation Temp "C Figure 8. Relation between mean upper and lower lethal temperature limits expressed as functions of acclimation temperature for stages postlarva, juvenile and age II. Included are the preferred temperatures of postlarvae, juveniles and yearlings similarly expressed as functions of prior acclimation. Solid stars= postlarvae; open stars=juveniles; solid circles=yearlings; solid triangles^age group II. 128 Tulane Studies in Zoology and Botany Vol. 20 coefficient was computed for the population of Pond I. The result suggested that each adult fish in the population had only an 11 per cent chance for survival throughout the year. Apparently, there are no natural mor- tality estimates available with which to compare this one. Taken as an absolute value, however, it does indicate that mortali- ty was high for the period of one year. How- ever, further estimates of population sizes suggested that mortality was neither infinite, nor evenly distributed throughout the year. A population estimate made during the week of 15 April 1971 disclosed the presence of 562 bullheads, 83% of the original estimate that had survived through the winter. Thus, heavy mortality did not occur during winter months. This seems to augment the con- tention of Forbes and Richardson (1908) that black bullheads enter a relatively quies- cent state during periods of prolonged cold. Nordlie (1966) further noted that peptic digestive rates could be directly related to temperature regimens, increasing or decreas- ing with temperature over relatively short periods of time. Fish were captured through- out the winter, and on one occasion the ice was broken in order to remove traps con- taining bullheads. These fish had no food in their stomachs, but were active enough to enter traps. Thus, although retaining some active impulses, winter bullheads appeared to survive reasonably well without food by the conformity of certain physiological mechanisms to reduced temperatures. During the second week in May 1971 a naturally occurring mass-mortality was observed in Ponds I and V. This occurred during a period of heavy rainfall and un- seasonally cool weather, and the water tem- perature in Pond I was reduced 7.0°C. Al- though the overall water chemistry picture did not change appreciably, the dissolved CO2 concentration increased from 4.5 to 14.0 ppm over a two-day period (Table 2). An estimated 200 fish died in Pond I, 29% of the original population estimate. The bulk of the mortalities were older fish, and mature individuals approaching the breeding condition. Powers (1937) commented on similar mortalities and implicated CO2 cnanges as the most instrumental factor in mortality. This contention, however, could not be substantiated during our investi- gation. Black bullheads older than six years were not taken during this study. This seemed to indicate that the longevity for members of the species was relatively short, and in Fig. 2 one sees that bullheads approaching the age of six had already begun to show un- favorable incremental growth. Rose and Moen (1951) followed a single black bull- head year class through 10 consecutive years in Lost Island Lake, Iowa. Few individuals survived past the age of seven, and those that did remain showed unfavorable growth rates. Reproductive Potential As noted in Table 5, material began mobilizing in the gonads of pond bullheads shortly after the spawning season; however, most of this extragonadal tissue had been reabsorbed before temperature increases in spring initiated the developmental stage. From the quiescent stage brought on by cold it could be surmised that this same sequence should prevail in all populations. However, stream bullheads spawned at least one month previous to the other populations, with peak gonad/body weight ratios attain- ing during early and mid-May (Fig. 5). Two apparent explanations might account for this difference in spawning time and develop- mental sequence: (1) although constant rise in temperature appeared to initiate the re- flex in ponds (Fig. 5), and in lakes (Den- nison. 1970). a separate and distinct set of conditions might prevail to initiate the spawning response in streams; (2) as Aron- son (1955) disclosed for the cichlid Tilapia heudiloti, which spawns throughout the year but has spawning peaks corresponding to the vernal and autumnal equinox, each individu- al may have a temperature threshold distinct from others in the populations, producing noticeable differences when different habi- tats are considered. Pond and lake spawners also react differ- ently to temperature. Dennison (1970) indi- cated that once initiated the response in Clear Lake, Iowa, was nearly linear and un- Nos. 3-4 Black Bullhead 129 interrupted. In both situations, once initi- ated, the developmental sequence appeared to be irreversible. In Wilgreen Lake, where temperature fluctuations were not marked over short periods of time, the gonad/body weight curves showed interrrupted increases. In contrast, corresponding ratios for pond spawners fluctuated directly with tempera- ture change (see Table 2). This probably was a parametric manifestation of the alternate speeding-up and slowing-down of physiologi- cal rate functions in response to rather wide- ly fluctuating temperatures (Table 2). Re- gardless of the influence, pond spawning lagged behind that of Wilgreen Lake, with pond males and females reaching the most favorable gonad/body weight ratios (2.1 and 12.1%) on July 3 and June 23, respectively. The above comparison shows that the habitat imposed its own limit on breeding time and the sequence of reproductive devel- opment. Thus, it is not difficult to reckon if one considers that different bodies of water mediate environmental modifiers differently and prompt the establishment of genetic stocks that adjust the breeding sequence to the most favorable period of the year. Fecundity determined during this study also showed considerable variability when all three habitats were considered. Greatest mean fecundity per individual attained in WUgreen Lake (mean = 3,372), with ponds exhibiting the lowest value (mean = 2,552), and streams intermediate (mean = 2,645) (Table 6). Mean values between the lake and other populations were significantly differ- ent at the 5% level of confidence. Few fecundity estimates have been pub- lished for black bullhead populations and the reports that have appeared have repre- sented lake populations. Thus, the data obtained during this study probably are more significant when the different habitats are compared. Pond fecundity apparently is reduced coincident with the limits of intra- specific competition, similar to the situation that accrues with growth (Moen, 1959). This assumption is difficult to make for streams which had equally low mean fecundity. Data for Johnson and Boone creeks were com- bined to arrive at the value listed in Table 6; however, from individual estimates made in Boone Creek during the 1969-70 season (7 specimens) we assumed that all overall fecundity was greater for this stream (mean = 3,845) than for the other. Johnson Creek presented a situation similar to that observed in ponds. The water was turbid during the period of sampling, and on a single night of seining over 500 bullheads were hauled from a .20 km section. Therefore, overpopulation here seems to have had an effect on mean fecundity similar to that recognized in turbid ponds. Attempts to induce spawning in black bullheads proved unsuccessful. There has been considerable controversy over method- ology when employing pituitaries to induce spawning. Early investigators (Hasler, Meyer, and Field, 1939; Pickford and Atz, 1957) insisted that only fresh preparations should be used. Indian biologists, however, have achieved considerable success with dried pituitary preparations administered through intramuscular injection (Alikuhni, Sukum- aran and Parameswaran, 1962; Alikhuni and Sukumaran, 1962), The Indians indicated that great numbers of the silver carp (Hypo- thalmichthys molitrix) had to be obtained to acquire a relatively few fish that could be artificially stimulated.^ Others (Sneed and Clemens, 1960) preferred the intraperitoneal route of injection, but concluded that greater dosage rates are required. Wallace (1967) used the methods detailed by Sneed and Clemens (loc. cit.) and was successful in inducing black bullheads to spawn; however, his results were based on the observation of a single breeding pair re- tained in a 400 gallon (1514 1) gallon aquari- um from the end of one spawning period to the beginning of the next. Perhaps in con- trast to the refractory nature of spawners held in aquaria during this study, that a pair of fish kept over an extended period may retain an endogenously controlled repro- ductive rhythm, such as that described for certain small mammals and birds (Pengelley and Asmundson, 1971). If so, such fish could conceivably spawn naturally. The method used during the present study followed that of Sneed and Clemens 130 Tulane Studies in Zoology and Botany Vol. 20 (loc. cit.) and Wallace (loc. cit.). This method, which suggests the administration of 4 mg/lb (1.82 mg/kg) per day of dried pituitary, was devised for channel catfish, the authors suggesting that breeders should be obtained as close to the spawning season as possible. Other authors (Pickford and Atz, 1957) suggested different stages of repro- ductive readiness during which riverine fishes should be obtained in order to obtain the best results. Although our experimental fish- es oozed gametes when strong pressure was applied to the abdomen, they did not re- spond to pituitary treatment. We have no explanation for the failure of black bull- heads to spawn in aquaria. When our experi- ments were terminated on July 5, the catfish exhibited gonad/body weight ratios similar to those observed in the wild. This would seem to indicate that the attainment of re- productive readiness was not the chief limit- ing factor. Physiological Ecology Several investigators have commented on the black bullhead's ability to persist in situ- ations that are limiting to most other fishes (Black, 1955, 1959; Lewis, 1949,; Traut- man, 1957; Bennett, 1948), because of com- bined behavioral and physiological adapta- tions which enhance the ability of rate func- tions to remain relatively constant over a wide spectrum of environmental gradients. However, information is conspicuously meager in the area of black bullhead physio- logical ecology. Temperature thresholds and oxygen consumption rates have been deter- mined in the brown (Brett, 1944) and the yellow bullheads (Morris, 1965), but only incidental treatments have been reported for the black bullhead. As indicated by Brett (1971), temperature is probably the most important controlling factor in distribution and life styles of most fish species. For this reason, tolerance points relative to different acclimation temperatures must be estab- lished, and extrapolated from them to obtain the thermal survival picture. How- ever, Brett (1959) and Hart (1952) ques- tioned the imposition of temperature toler- ance points that fail to reflect environmental survival, i.e., the physiological and ecological limits are often widely separated. Further- more, fishes rarely live in conditions close to their survival thresholds. In general, temperature tolerance poly- gons occur as a graded series when more than one stage in the life cycle is considered (Fig. 8). Extremely high upper limits (39^^ for postlarvae acclimated at 23°C), and lower limits restricted to a narrow range about the freezing point of water for older animals, assure this species a position in most habi- tats. We observed as well that the critical thermal maxima curves for four different life stages were also graded, each stage exhibiting a relatively flattened curve with respect to closely related species (Brett, 1956). The oldest group tested (age II), and two inter- mediate stages, exhibited relatively good compensatory ability with respect to tem- perature change; whereas postlarvae at the two acclimations 12° and 23^ showed less ability to compensate for temperature change through alteration in rate functions. Since the postlarval stage could not be ac- climated to 7"^, we suggest that black bull- heads are not hatched with the ability to compensate for wide change in temperature. However, since they are born during a period of the year when environmental tempera- tures are highest, the tolerance to high tem- peratures (Fig, 7) probably is most impor- tant. As indicated by the curve for juveniles (Fig. 7), after attaining a few days of age (10 to 15 days) the biochemical structure is suf- ficiently enhanced to permit compensations over a wider temperature range. Brett (1956) constructed thermal poly- gons similar to the ones we depict in Fig. 8 for the brown bullhead. The black bullhead, as determined by us, appears to have a greater tolerance range at comparable accli- mation temperatures. Hart (loc. cit.) and Brett (1959) con- cluded that temperature tolerance is not of direct ecological significance, since fishes rarely confront such temperatures. However, our work indicates that black bullheads can or do live close to their thermal limits during at least a portion of summer and early fall. The temperature of Ponds III and IV reached 33°C even in the deepest parts on Nos. 3-4 Black Bullhead 131 several days during August and September, Even if temperature tolerance fails to reflect a true ecological condition, it does indicate the ability of a species to alter rate functions in connection with temperature changes, which suggests that such species function ettcctivcly over very broad thermal zones. The temperature preferences of three age groups ot black bullheads were investigated (Fig. 1). Different stages in the life cycle were found to have significantly different preference ranges over the acclimations indi- cated above (P < .05, Table 8). In addition, a phenomenon was observed in larvae which suggested the influence of temperature on life style at this stage. Postlarval bullheads held in the selection chamber for nine-hour periods behaved much like older fish. How- ever, after ten hours in the chamber, they suddenly shifted from one preferred temper- ature to a lower one (from 23.5° to 17.5- 19'Xi; at 23°C acclimation). After subsequent experimental runs demonstrated the same response, we theorized that this was an envi- ronmental artifact, i.e., that larvae apparent- ly had free running rhythms of activity. After spending daylight hours near the water surface they sought lower strata of water where seclusion was more nearly assured. We suspect that the response observed in the chamber was the same as that which oc- curred in the wild. Larvae apparently sought water in which to remain secluded by seek- uig water temperatures associated with that stratum. The reciprocal explanation could also be advanced according to the time of day. Three levels of oxygen consumption have been described for fish: standard, routine, and active '(Fry, 1968). Routine rates, which may be defined as the oxygen consumed by a fish whose only movements are spontane- ous (Beamish and Mookerjii, 1964). Such readings are most frequently recorded, and their values fall between the extremes of active and standard metabolism. The rates of oxygen consumption at different acclima- tion temperatures are still another, and perhaps better, indicator of change in rate- function with time. Morris (1965) stated that brown bull- heads, when moved to 27" from 12°C accli- mation, compensated by lowering metabolic levels from 0.35 to 0.15 ml 02/g/hr. The routine rates measured during the present study indicated that black bullheads do not display such compensatory shifts over a 16°C change in temperature. However, since their rates were significantly low even at the high- est acclimation temperature (Table 7), and since all rates were measured at 20'XI!, such radical compensatory shifts were not antici- pated. We believe that the black bullhead is able to survive in extremes of temperature by having a relatively flattened metabolic rate curve, which hardly doubles for each 18°F increase in temperature as indicated by McCay (1925). Beamish and Mookerjii (1964) and Basu (1959) suggested that oxygen consumption increased linearly with weight when plotted on a double logarithmic grid. This appeared to hold for our first two groups (Table 5) but not for age II individuals, whose abso- lute consumption fell below the best fit line (Fig. 6). However, since we assumed that the effect was constant from one acclimation temperature to another, the only factor operating to account for changes in con- sumption is temperature. From an analysis of the homogeneity of the slopes plotted for 12° and 23°C acclimation, we determined that there is no significant change in routine consumption for an eleven degree change in temperature (Fs = I.OOI8/F.99 (1, 38) - 7.35). According to Prosser and Brown (1962) chemical reaction rates usually are more than doubled for each 10°C increase in tem- perature. Thus, Qio values of less than 2 would indicate a relatively insensitivity to temperature. The QiQ values determined for black bullheads (Table 7) ranged from 1.78 for postlarvae to a low of 0.12 for age group II. The most important observation drawn from metabolic-rate determinations was the fact that black bullhead rates decreased to a maintenance level of 0.01 ml 02/g/hr at 0.2 to 0.5 ppm dissolved oxygen before showing signs of asphyxial stress. However, this should not be considered unique. Larimore 132 Tulane Studies in Zoology and Botany Vol. 20 (1957) reported similar low levels for the warmouth, Chaenohryttus gulosus, and Powers (1938) found that other fishes, such as the smallmouth bass, Micropterus dolomi- eui, could extract oxygen to 0.3 - 0.4 ml/1. We doubt, however, that this fish could have established maintenance levels as low as that determined for the black bullhead. The black bullhead is known to survive in habitats that present far less than the ideal conditions of temperature and dissolved oxygen. Presumably this is possible through a combination of ecological and physiologi- cal adaptations. The ability to subsist on a variety of foods, or on a very limited food supply, and the ability to become quiescent during periods of prolonged temperature de- crease are significant adaptations to hfe in some of the habitats occupied. In addition, comparatively favorable high and low lethal temperature ranges, assure the species a position in most situations that might prove limiting to less favored species. Low meta- bolic rate-temperature curves imply at least one rate function that changes minimally with temperature, and the ability to extract maintenance oxygen in the presence of low ambient levels favor survival in most extreme situations. SUMMARY AND CONCLUSIONS Although the black bullhead is widely dis- tributed in the United States it has not been studied extensively. Our study was designed to facilitate a better understanding of the species' life history and physiological re- sponses in pond populations, as compared to other known habitats. Conclusions from data derived during April 1970 through July 1971 are summarized as follows: 1. Adult black bullheads subsist on a variety of foods, but exhibit feeding selectivity in ponds, lakes and streams. Chironomid larvae, compris- ing a predominant portion of the adult diet, are not more strongly selected for than other less available foods. 2. Larval feeding occurs in a rather re- stricted locus about an axis of visual orientation. Foods taken most fre- quently by young are the aquatic amphipod Hyalella, ostracods and adult copepods. Larval feeding was found to be highly selective. 3. Mean ponderal indices for pond, lake and stream populations do not differ significantly (P < .05). However, log-transformations of length-weight relationships have slopes for which the hypothesis B = 0 does not hold at the 99% level of confidence. Most favorable growth rates occurred in Wilgreen Lake, and least favorable in ponds. The age-weight correlation for pond populations was quite low (r = 0.42). 4. Populations attained high levels in ponds [1,021 fish per surface acre (= 413/ha) in one 0.65 (= .26 ha) acre = impoundment] . The effect of crowd- ing was significantly reflected in population condition, number of genetic abnormalities and lowered mean fecundity. Also apparently cor- related with high population levels was a high mortality coefficient (0.89 for Pond I of this study). 5. Mortality is appreciably higher dur- ing some portions of the year than others. Greatest mortality occurred in Pond I during spring and early summer. Population estimates indi- cate that mortality is relatively low during the winter. 6. Gonadal development and attain- ment of spawning condition in ponds lag behind that of lakes and streams. Most favorable gonad/body weight ratios occur in June for mature females (12.1 per cent) and July for mature males (0.21 per cent). In con- trast, stream spawners reach develop- ment peaks during mid-May, and Wil- green Lake spawners attain the most favorable index during mid-June. 7. Fecundity for females ranging from 164 to 220 mm SL is lowest in ponds (mean - 2,552), highest in Wilgreen Lake (mean = 3,372) and intermedi- ate for combined samples from creeks (mean = 2,645). Nos. 3-4 Black Bullhead 133 8. Mature bullheads brought into the laboratory during the month of May could not be induced to spawn even after repeated treatment with dried carp pituitary gland. The gonad/body weight ratios of these fish on the termination date, 5 July, were similar to levels observed in wild pop- ulations. 9. Temperature tolerances were suffici- ently high and low that few habitats could be considered restrictive to bullhead survival on the basis of tem- perature alone. The CTMax of age groups postlarva through Age II ranged from 36.5° to 38.2° at 23'C acclimation. CTMax curves over a temperature range of 16°C were rela- tively flattened, suggesting little rate-function change with tempera- ture. Lower lethal temperature grad- uated upward for each descending stage of the life cycle. Preference ranges for the first three life stages were significantly different from one another at the 5% level of confi- dence. 10. Metabolic rate-temperatures curves, QlO values, and absolute consump- tion at different temperatures suggest that metabolic rates change very little over a 16'C acclimation range. Black bullheads of 20.15 to 37.00 g reached the maintenance oxygen level at 0.1 ml 02/g/hr in 0.2 to 0.5 ppm ambient oxygen. LITERATURE CITED ALBAUGH, D.W. 1969. Sources of growth varia- tion among individual black bullhead Ictalurus melas and channel catfish Ictalurus punctatus. Trans. Amer. Fish. Soc, 98: 35-44. ALIKHUNI, K.H. and K.K. SUKUMARAN. 1962. Preliminary observations on chinese carp in India. Proc. Indian Acad. Sci., 60: 171-188. ALIKHUNI, K.H., K.K. SUKUMARAN and S. PARAMESWARAN. 1962. Induced spawning of the Chinese carps Ctenopharyngdon idellus (C & V) and Hypothalmichthys molitrix (C & V) in poads at Cuttack, India. Current Science, 32:103^106. ARONSON, L.R. 1955. Factors influencing the spawning frequency in the female cichlid Til- apia macrocephala. Amer. Mus. Novitates 1484:1-26. AUSTIN, M.E. 1965. Land resources regions and major land resource areas of the United States. Soil Const. Ser. Handbook no. 296. 82 pp. BAILEY, R.M. 1955. Differential mortality from high temperature in a mixed population of fishes in southeastern Michigan. Ecology, 36: 525-28. BARNICKOL, P.G. and W.C, STARRETT. 1951. Commercial and sport fishes of the Mississippi River. 111. Nat. Hist. Surv. Bull., 25(5): 267-350. BASU, S.P. 1959. Active respiration offish in rela- tion to ambient concentrations of dissolved oxygen and carbon dioxide. J. Fish. Res. Bd. Can., 16: 175-212. BAUR, R.J. 1968. Digestion rates of Clear Lake black bullheads. 30th Ann. Midwest Conf. (mimeo). 4 pp. BEAMISH, F.W.H. and P.S. MOOKERJII. 1964. Respiration of fishes with special emphasis on standard oxygen consumption. I. Influence of weight and temperature on respiration of gold- fish Carassius auratus L. Can. J. ZooL, 42: 166-175. BEHMER, R.J. 1969. A method of estimating fecundity; with data on river carpsuckers, Carpiodes carpio. Trans. Amer. Fish. Soc, 98: 523-24. BENNETT, G.W. 1948. Winterkill of fishes in Illi- nois Lake. 111. Nat. Hist. Sur. Bio. Note no. 19: 9 pp. _-^___ 1952. Pond Management in Illinois, in Pond management symposium, J. Wild!. Mgr. 16: 249-253. BIRGE, E.A. 1907. The respiration of an inland lake. Trans. Amer. Fish. Soc, 36: 223-241. BLACK, E.C. 1953. Upper lethal temperatures of some British Columbia freshwater fishes. J. Fish. Res. Bd. Can., 10: 196-210. ________ 1955. Blood levels of hemoglobin and lactic acid in some freshwater fishes following exercise. J. Fish. Res. Bd. Can. 12: 917-929. BOWEN, E.S. 1931. The role of the sense organs in aggregations oi Amieurus melas. Ecol. Monogr., 1: 1-35. BOYAR, H.C. and R.A. CLIFFORD. 1967. An automatic device for counting dry fish eggs. Trans. Amer. Fish. Soc, 96: 361-63. BRETT, J.R. 1944. Lethal temperatures of fresh- water fish and temperature tolerance of young salmon. J. Fish. Res. Bd. Can., 9: 265-323. __^_____ 1956. Thermal requirements of fish. Quart. Rev. Biol., 31: 75-87. • .. 1959. Principles of thermal require- ments of fish — three decades of study — 1940 1970. U.S. Public Health Second Seminar in Biol. Problems in Water Pollution. 1971. Energetic responses of salmon to temperature. A study of some thermal relations in the physiology and freshwater ecology of 134 Tulane Studies in Zoology and Botany Vol. 20 sockeye salmon (Oncorhynchus nerka). Amer. Zool., 11: 99-113. BRODY, S. 1945. Bioenergetics and growth. Chap. 13-14, in Metabolism in relation to body tem- perature, Reinhold Publishing Co., New York. 1023 pp. BUCK, H. and F. CROSS. 1952. Early limnological and fish population conditions of Canton Reservoir, Oklahoma, and Fishery management recommendations. Res. Found., Oklahoma A & M College, Spec. Publ. BUCKMANN, A. 1929. Methodikfischeribio- logischcr untersuchungen meeresfischem. in Abderhalden Emil (ed). Handbuch der Biol. Arbeitesmeth Abt. IX. Methoden der Erfors- chung derleistungen des tierischem Organsims. Tier. 6., Helft 1., Lief 307. BUDDENBROCK, W. von. 1938. What physiologi- cal problems are of interest to the marine biolo- gist in his studies of the most significant prob- lems of fish? Conceil. Permintern. Exploration Mer. 101: 1-7. BULLOCK, T.H. 1955. Temperature adaptation in poikilothermic animals. Biol. Rev., 30: 311-342. BURGER, J.W. 1939. Some experiments on the relation of the external evirnonment to the spermatogenetic cycle oi Fundiihis heteroclitiis. Biol. Bull., Woods Hole. Marine Lab. Rept., 77: 96-103. BURROWS, R.E. 1951a. A method for the enu- meration of salmon and trout eggs by displace- ment. Prog. Fish. Cult., 13: 25-30. 1951b. An evaluation of methods of egg enumeration. Prog. Fish. Cult., 13: 79-85. BUSER, J. and M. BLANC. 1949. Action de la lumiere sur L'osteogenese reparatrice che les poissons teleosteens. Vie. Ct. Milieu. , 1 : 252-54. Caillouet, C.W. JR. 1967a. Hyperactivity, blood lactic acid and mortality in channel catfish. Iowa St. Univ. Agr. Home Ec. Exp. Sta. Res. Bull., 551: 898-915. 1967b. Blood lactic acid concentration in black bullheads Ictalurus melas after trans- portation and forced exercise. Iowa St. J. Sci., 42: 1-7. CARLANDER, K.D. 1968. Handbook of fresh- water fishery biology. Vol. I. Iowa State Univ. Press., Ames. 751 pp. and C. SPRUGEL. 1948. Bullhead management: shallow lake investigations, no. 42. Progr. Reg. la. St. Coop. Wildl. Res. Unit., 112-117. CLARK, C.F. 1960. Lake St. Marys and its man- agement. Ohio Dept. Nat. Resources Bull. no. 551. 60 pp. CLAY, W.M. 1962. A field manual of Kentucky fishes. Ky. Dept. Fish & Wildl. Resources Publ. 147 pp. CLEMENS, H.P. and K.E. SNEED. 1957. The spawning behavior of the channel catfish /cfa/u- riis punctatus. U.S. Fish & Wildl. Ser. Spec. Sci. Rept. Fish. no. 219. 11 pp. COHEN, H. 1946. Effects of sex hormones on the development of the platyfish Platypoecilus maculatus. Zoologica N.Y. 31: 121-28. CORTHELL, R.A. 1961. Coos-Coquille District Annual Report. Oregon St. Game Comm., Fish. Div., 1961: 273-87. COWLES, R.B. and CM. BOGERT. 1944. A pre- liminary study of the thermal requirements of desert reptiles. Bull. Amer. Mus. Nat. Hist,, 83: 261-96. CREASER, C.W. 1926. The structure and growth of the scales of fishes in relation to the interpre- tation of their life history, with special refer- ence to the sunfish, Eupomotis gibbosus. Univ. Mich. Mus. Zool. Misc. Publ. 17: 1-38. DARNELL, R.M. and J. MEIEROTTO. 1965. Diurnal periodicity in the black bullhead, Ictalurus melas. Trans. Amer. Fish. Soc, 94: 1-9. DAVIS, A.S. and G.E. PAULIK. 1965. The design, operation and testing of the photoelectric fish egg counter. Prog. Fish. Cult., 27: 185-92. DENDY, J.S. 1945. Depth distribution of fish in relation to environmental factors. Norris Reser- voir Rept., Reelfoot Lake Biol. Sta., 9: 111-135. 1946. Water temperature and spring fishing, Norris Reservoir Tennessee. Tenn. Acad. Sci., 21: 89-93. DENNISON, S.G. 1970. The reproductive potenti- al of the black bullhead, Ictalurus melas (Rafin- esque), in Clear Lake, Iowa. Unpublished Masters Thesis. Iowa St. Univ. 62 pp. DOWNIE, N.M. and R.W. HEATH. 1961. Basic Statisticsl Methods. Harper and Row Publishers, New York. 325 pp. EDDY, S. and A.C. HODSON. 1961. Taxonomic keys to the common animals of the north cen- tral states. Burgess Publ. Co., Minneapolis. 161 pp. EWERS, L.A. 1931. Summary report of the Crusta- cea used as food by the fishes of the western end of Lake Erie. Trans. Amer. Fish. Soc, 63: 379-90. FERGUSON, D.E., H.F. LANDRETH and M.R. TURNIPSEED. 1965. Astronomical orientation of the southern cricket frog Acris gryllus. Copeia 1965: 58-66. FOGEL, N.E. 1964. Summation of four years of creel census July 1959 through June 1963 on Cahe tailwaters. South Dakota Dept. Game, Fish and Parks. 18 pp. FORBES, S.A. and R.A. RICHARDSON. 1909. The fishes of Illinois. 111. Nat. Hist. Surv., State Lab. Hist. 357 pp. FORNEY, J.L. 1955. Life history of the black bullhead, Ameiurus melas, in Clear Lake, Iowa, la. St. J. Sci., 30: 145-162. FOWLER, H.W. 1917. Notes on the breeding habits of local catfishes. Copeia no. 42: 32-36. Nos. 3-4 Black Bullhead 135 FREDIN. R.A. 1950. Fish population estimates in small ponds using the marking and recovery technique. la. St. J. Sci., 24: 363-384. FRY, F.E.J. 1968. Responses of vertebrate poikilo- therms to temperature, pp. 375-409 in A.H. Rose (Ed.) Thermobiology. Academic Press, New York. and J.S. HART. 1948a. Cruising speed of goldfish in relation to water temperature. J . Fish. Res. Bd. Can.. 7: 169-75. 1948b. The relation of temperature to oxygen consumption m the goldfish. Biol. Bull., 94J 66-67. GRAYBILL, J.R. and H.F. HORTON. 1969. Limited fertilization of steelhead trout eggs with cryo-preserved sperm. J. Fish. Res. Bd. Can., 26: 1400-1404. HALL, F.R. and VV.N. PALMQUIST JR. 1960. Hydrobiological investigation. Dept. Int., U.S. Geological Sur., HA-19, Sheet 1. HAR, W.S. 1975. Comparative physiology: hor- mones and reproduction in fishes. Ann. Rev. Physiol. 27: 51-70. HART, J.S. 1952. Lethal temperatures offish from different latitudes. Publ. Ontario Fish. Lab., 72: 1-79. HASTINGS, C.E. and F.B. CROSS. 1962. Farm ponds in Douglas County, Kansas. St. Biol. Surv. Kansas, Misc. Publ., no. 29, pp 1-21. HATHAWAY, E.S. 1927. The relation of tempera- ture to the quantity of food consumed by fishes. Ecology, 8: 428-34. HILE, R. 1936. Age and growth of the cisco, Leucichthys artedi (Le Sueur), in the lakes of the northeastern highlands, Wisconsin. Bull. U.S. Bur. Fish., 48: 211-317. 1948. Standardization of methods of expressing lengths and weights of fishes. Trans. Amer. Fish. Ser. 75: 157-164. HOUSER, A. and C. COLLINS. 1962. Growth of black bullhead catfish in Oklahoma. Okla. Fish. Res. Lab. Rept. no. 79. 18 pp. and B. GRINSTEAD. 1961. The effect of black bullhead catfish and blucgill removals on the fish population of a small lake. Proc. S.E. Assoc. Game & Fish Comm., 1 0 pp. HUBBS, C.L. 1943. Terminology of early stages of fish. Copeia 1943: 260. HUTCHINSON, V. 1961. Critical thermal maxima in salamanders. Phys. Zool. 34: 93-125. JENKINS, R.M. 1959. Standing crops of fish in Oklahoma ponds. Proc. Okla. Acad. Sci., 37: 164-173. JONES, J.W. and G.M. KING. 1952. The spawning of the male salmon parr (Salmo salar, Linn, Juv.). Proc. Zool. Soc. London, 122: 615-619. KUTKUHN,J.H. 1955. Food and feeding habits in a dredged Iowa lake. Proc. la. Acad. Sci., 62: 576-88. LARIMORE, W.R. 1957. Ecological life history of the warmouth (Centrarchidae) 111. Nat. Hist. Surv. Bull., 27: 83 pp. LEWIS, W.M. 1950. Fisheries investigations of two artificial lakes in southeastern Iowa II. Fish populations. la. St. J. Sci., 24: 287-324. LILLY, N.R. 1969. Hormones and reproductive behavior in fish. pp. 73-115 in "Fish Physiol- ogy", Vol. Ill, ed. W.S. Hoar and D.J. Randall. Academic Press. LOWE, C.H. and W.G. HEATH. 1969. Behavior and physological responses to temperature in the desert pupfish Cyprinodon maculariiis. Physiol. Zool. 1969: 53-9. LYNCH, T.N., P.A. BUSCEMI and B.C. LEMONS. 1951. Limnological and fishery condition of Two Buttes Reservoir, Colorado. Colo. St. Game& Fish Publ. (1953). 93 pp. MATTHEW. A.P. 1921. Physiological chemistry. Wm. Wood & Co., New York. 1 1 54 pp. McCAY, CM. 1925. The role of oxygen in water pollution. Calif. Fish & Game, 11: 157-61. MEDLEN, A.B. 1951. Preliminary observations on the effects of temperature and light upon repro- duction in Gambusia affinis. Copeia 1951: 148-56. MOEN, T.E. 1959. Notes on the growth of bull- heads, la. St. Consr. Comm. Quart. Biol. Rept., 11: 29-31. MOORE, W.G. 1942. Field studies on the oxygen requirements of certain freshwater fishes. Ecology, 23: 319-29. MORRIS, R.W. 1965. Thermal acclimation and metabolism of the yellow bullhead Ictalnrus mitalis (Le Sueur). Physiol. Zool., 38: 219-227. NOBLE, G.K. and K.F. KUMPF. 1936. The sexual behavior and secondary sexual characteristics of gonadectomized fish. Anat. Record, 67 (supp I) 115 pp. NORDLIE, E.G. 1966. Thermal acclimation and peptic digestive capacity in the black bullhead Ictalnrus melas (Rafincsque) Amer. Midi. Nat. 75: 416-24. NORRIS, K.S. 1963. The function of temperature in the ecology of the period fish Girella nigri- cans (Ayres). Ecol. Monogr., 33: 23-62. PENGELLEY, E.T. and S.J. AMUNDSON. 1971. Annual biological clocks. Sci. Amer., 224: 72-79. PENNAK, R.W. 1953. Freshwater Invertebrates of the United States. Ronald Press Co., New York 769 pp. PICKFORD, G.E. and J.W. ATZ. 1957. The physi- ology of the pituitary gland of fishes. N.Y. Zool. Soc. Publ., 613 pp. POWERS, E.B. 1937. Factors involved in the sud- den mortality of fishes. Trans. Amer. Fish. Soc, 67: 271-281. , H.H. ROSTORFER, L.M. SHIPE and T.H. ROSTORFER. 1938. Relation of respira- tion of fishes to environment XII. Carbon di- oxide as a factor in various physiological respi- ratory responses in certain freshwater fishes. J. Tenn. Acad. Sci. 13:23 25- 136 Tulane Studies in Zoology and Botany Vol. 20 PRECHT, H., J. CHRISTOPHERSEN and H. HENZEL. 1955. Temperatur und Leben. Springer- Verlag, Berlin. 514 pp. PROSSER, C.L. and F. BROWN. 1962. Compara- tive animal physiology. W.B. Saunders Co., Philadelphia. 688 pp. RANEY, E.G. and D.A. WEBSTER. 1940. Food and growth of the young of the common bull- head (Ameiurus nebulosus). Trans. Amer. Fish. Soc, 79: 205-209. REGIER, H.A. 1962. On estimating mortality co- efficients in exploited fish populations, given two censuses. Trans. Amer. Fish. Soc, 91: 283-94. RICKER, W.E. 1944. Further notes on fishing mortality and effort. Copeia 1944: 23-44. 1958. Handbook of computations for biological statistics of fish populations. Fish. Res. Bd. Can. Bull., 119: 300 pp. 1975. Computation and interpretation of biological statistics offish populations. Fish. Res. Bd. Can. Bull. 191: 382 pp. ROBINSON, O.H. and A.P. Rinfret. 1957. Matura- tion of the infantile testis in rainbow trout (Salmo gairdnerii) produced by salmon pitui- tary gonadotropins administered in cholesterol pellets. Endocrinol, 60: 559-6 2. ROSE, E.T. and T.E. MOEN. 1951. Results of in- creased fish harvest in Lost Island Lake, Iowa. Trans. Amer. Fish. Soc, 80: 50-5 5. SCHMIDT, P.J., B.S. MITCHELL, M. SMITH and H. TSUYUKI. 1965. Pituitary hormones of the pacific salmon I. Response of gonads of im- mature trout (Salmo gairdnerii) to extracts of pituitary glands from adult pacific salmon (Oncorhynchus). Gen. & Comp. Endocrinol., 5: 197-206. SHIELDS, J.T. 1957a. Report of fishery investi- gations during the second year of impoundment of Gavins Point Reservoir, South Dakota. 1956. S.D. Dept. Game, Fish and Parks. 346 pp. 1957b. Report of fishery investigations during the third year of impoundment of Gavins Point Reservoir, South Dakota. 1957. SX). Dept. Game, Fish and Parks. 1-60. SIMCO, B.A. and F.B. CROSS. 1966. Factors affecting growth and production of channel catfish Ictalurus pnnctatus. Univ. Kansas Mus. Nat. Hist. Publ., 17: 191-256. SMITH, H., J.M. AMELINK-KOUSTAAL, J. VIJVERBERG and J.C. von VAUPELKLEIN. 1971. Oxygen consumption and efficiency of swimming goldfish. Comp. Biochem. Physiol. 39: 1-28. SMITH, L.L. JR. 1949. Cooperative fishery survey of the Upper Mississippi River. Trans. Amer. Fish. Soc, 76: 279-82. SNEED, K.E. 1951. A method for calculating the growth of channel catfish Ictalurus lacustris pnnctatus. Trans. Amer. Fish. Soc, 80: 174-83. and H.P. CLEMENS. 1960. Use of fish pituitaries to induce spawning in channel cat- fish. U.S. Bur. Fish. Spec Sci. Rpt. Fish. no. 329. 12 pp. SOKAL, R.R. and F.J. ROHLF. 1969. Biometry. W.W. Freeman Co., San Francisco. 776 pp. SWINGLE, H.A. 1950. Relationships and dynamics of balanced and unbalanced fish populations. Alabama Polytec Inst. Agr. Exp. Sta. Bull. no. 274. 74 pp. TA VOLGA, M.C. 1949. Differential effects of estradiol, estradiol benzoate and pregninolone on Platypoecilus maculatus. Zoologica N.Y., 34: 215-237. TAVOLGA, W.N. 1955. Effects of gonadectomy on prespawning behavior in males of the gobiid fish, Bathygobius seperator. Physiol. Zool. 28: 218-233. TRAUTMAN, M.B. 195 7. The fishes of Ohio. Ohio State Univ. Press. 683 pp. UNDERHILL, J.C. 195 2. The effect of tempera- tures on food consumption in fishes with special reference to the brown bullhead Amei- urus nebulosus (Le Sueur). Unpublished Masters Thesis, University of Minnesota. VAN OESTEN, J. 1929. Life history of the lake herring (Leucichythys artedi, Le Sueur) of Lake Huron as revealed by its cales with a critique of the scale method. U.S. Bur. Fish. Bull. no. 44: 265-428. VERNBERG, F.J. 1954. Oxygen consumption by tissues of teleosts of different size. Biol. Bull., 106: 360-70. VIOSCA, P. JR. 1931. The bullhead Ameiurus melas catalus as a dominant in small ponds. Copeia 1931: 17-19. WALLACE, C. 1967. Observation on the repro- ductive behavior of the black bullhead Ictalurus melas. Copeia 1967: 852. WALLEN, I.E. 1955. Some limnological considera- tions in the productivity of Oklahoma farm ponds. J.Wildl. Mgt. 19: 450-62. August 25, 1978 NOTES ON THE BREEDING BEHAVIOR, EMBRYOLOGY AND LARVAL DEVELOPMENT OF CYPRINODON VARIEGATUS LACEPEDE IN AQUARIA^ MAURICE F. METTEE, JR. Geological Survey of ALibama P. O. Drawer O, University, Alabama 35486 and EUGENE C. BECKHAM, III Tulane University, Museum of Natural History Belle Chasse, Louisiana 70037 ABSTRACT Several adult CyprinoJoti variegatus were placed in aquaria, maintained at 23°C, where re- productive behavior was observed. After each spawning the eggs were collected and observed to record the developmental stages. They were 1.4 - 1.6 mm in diameter, spherical, and covered with semi-adhesive threads. Embryological development proceeded rapidly at first but slowed in the later stages. Hatching was first observed between 120-125 hours after fertilization. Newly hatched prolarvae were 3.0 mm in total length, had 24 myomeres, black eyes and an apparently fully developed mouth. The heart was visible and the heart rate averaged 124 beats per minute. A tin told was present that contained seven or eight caudal but no dorsal or anal fin primordia. One or two tin ray primordia were present in each pectoral tin. The vertical tms were developed and the fin told had disappeared on individuals over 8.2 mm standard length. Pelvic fin buds began to develop last (at about 8.1 mm SL) and all rays were present at SL 9.0 mm. All individuals possessed most adult characteristics at 12-13 mm SL. INTRODUCTION Various aspects of the early life history of Cyprirtodon variegatus have been reported by several investigators; however, none have given a complete description of the sequence of events beginning with the spawning be- havior and extending to the development of juvenile individuals. Newman (1907) pre- sented the first data on the spawning activi- ties of Cyprirtodon variegatus in aquaria. His observations were later supplemented by Raney, et al. (1953) based on observations in south Florida lagoons. Kuntz (1916) gave the most complete account of the embryol- ogy of the Cyprirtodon variegatus but em- phasized information that would aid in the ready identification of the eggs and larvae rather than details of its embryology and development. Hildebrand (1917), Hilde- brand and Schroeder (1928), Nichols and Breder (1927) and Simpson and Gunter (1956) reported on various aspects concern- ing the life history of C. variegatus along the Atlantic and Gulf coasts. Most of these data were summarized by Breder and Rosen (1966). The two most recent comprehensive studies on estuarine larval fishes are those by Lippson and Moran (1974) and Scotten, et al. in 1973. The data presented on C. varie- gatus in those reports, however, is essentially a summary of earlier studies. The purpose of this report is to give a complete account of the reproductive se- quence of Cyprinodon variegatus including spawning behavior, embryology and larval development. Where possible, our observa- tions and results are compared with those of earlier studies. MATERIALS AND METHODS Specimens oi Cyprinodon variegatus were collected from a tide pool to Mobile Bay, Mobile County, Alabama on 7 April 1974. They were placed in a styrofoam container and transported to the laboratory at the Uni- versity of Alabama and several males and females were placed in each of two 40-L aquaria filled with freshwater and a clean sandy substrate. Water temperature was EDITORIAL COMMITTEE FOR THIS PAPER: DR. JAMES N. ATZ, Associate Curator of Ichthyology, American Museum of Natural History, New York, New York 10024 DR. C. M. BREDER, 6275 Manasota Key Road, Englewood, Florida 33533 137 138 Tulane Studies in Zoology and Botany Vol. 20 maintained at 23°C ±1°. Fishes were fed brine shrimp (Artemia) nauplii each day. The room that housed the aquaria was main- tained on a 14-hour light/10-hour dark photoperiod using banks of fluorescent lights controlled by an automatic timer. This lighting was augmented with daylight. After each observed spawning, the eggs were removed from the aquarium and placed in small culture dishes. Eggs obtained from unobserved spawns were sorted into separate dishes based on their stage of embryonic development. In the initial stages of the study, observations on eggs were made con- tinuously in order to record all embryo- logical changes, however, after the develop- ment of several groups had been thoroughly B D Figure 1. Developing eggs of CyprUiodon varie- gatiis. A. One-celled stage; B. Two-celled stage; C. Four-celled stage; D. Eight- celled stage; E. Early blastuia; F. Late blastula. monitored and a schedule of the principal embryonic stages had been devised, eggs were checked only periodically. A written and photographic record of the develop- mental process was kept throughout the study. Terminology employed in the descrip- tions of eggs, embryological stages and larvae is similar to that used by Hubbs (1943), Rugh (1962), and Blaxter (1969). Eggs were photographed with a 35 mm single lens re- flex camera through a dissection microscope using tungsten light. On about the third day after hatching, the prolarvae were fed freshly hatched brine shrimp (Artemia) nauplii. This food supply was maintained throughout the study. After each feeding period, uneaten nauplii were removed from the culture dishes to avoid problems of fungus growth associated with water spoilage and disease. During the period after hatching, one or more individuals were periodically preserved in a 5% formalin solution for observation and measurement. The schedule for preserva- tion of specimens was irregular and depend- ent on the amount of morphological change that occurred with increased age. As the larvae grew older and assumed adult char- acteristics, fewer specimens were required for preservation. During the study 89 larvae were preserved which ranged in age from one to 160 days. Observations on changes in body shape, fin development and coloration were made on each specimen. The following morphological measurements were made with a vernier stage micrometer attached to a dissecting microscope. STANDARD LENGTH (SL) - length from the tip of the snout to the posterior end of the notochord or the hypural plate. HEAD LENGTH (HL) - length from the tip of the snout to the rear edge of the gill cover. MAXIMUM DEPTH (MAXD) - greatest body depth. SNOUT-TO-DORSAL-FIN-ORIGIN LENGTH (SNTDO) - length from the tip of the snout to the anterior base of the dorsal fin. Nos. 3-4 Notes on Cyprinodon 139 SNOUT-TO-VENT LENGTH (SNTVT) - length from the tip of the snout to the anterior edge of the vent. The snout-to-vent length as a percentage of standard length (SNTVT/SL x 100) was also calculated to determine relative proportional growth. Morphometric data on the larvae were punched on computer cards and ana- lyzed on an IBM 260 computer to obtain mean data for each age group. Product moment correlation values (r) were also computed and plotted for various pairs of these data. BEHAVIOR Within a period of 7 to 10 days after the fishes were placed in the aquaria, one of the males assumed a dominant role and estab- lished a territory, that occupied approxi- mately one -half of the total area of the tank and extended from the top to the bottom. Any individual that approached the territory was quickly confronted by the male and ulti- mately sent fleeing to the opposite end of the aquarium. Eventually, one of the females was allowed to enter the territory un- molested. Once together, however, the male and female did not spawn immediately. The male seemed unattentive to the female and con- centrated principally on maintaining his ter- ritorial perimeter. Meanwhile the female swam about the area in a seemingly non- chalant manner. Periodically, she proceeded to the bottom of the aquarium and scooped up a mouthful of sand. She held it in her mouth for an instant as if to test it, then spit it out and continued swimming about the k J^<^- ■> '>?' •■•.'■J \l M <#• ^:-.-! ^ ..-> N r. Q c :4 K .r^^. •■•y T^ ■• ■■' ■■.■ ■•';■ •'•';• ^■^ M. Figure 2. Developing eggs of Cyprinodon varie- gatus. G. Early neurula; H. Mid neurula; I. Late neurula; J. Six-somite stage; K. Ten-somite stage; L. Free-tail prolarva. Figure 3. Larvae of Cyprinodon variegatus Newly hatched prolarva (2.7 mm SL); N. 3-day old larva (3.5 mm SL); O. 8-day old larva (4.7 mm SL); P. 12-day old larva (5.6 mm SL). 140 Tulane Studies in Zoology and Botany Vol. 20 territory. This behavior was repeated several times. After spitting out one of these mouth- fuls, the female shoved her snout into the sand at the same spot and threw her head sideways, thereby raising a small cloud of sand particles up off the bottom. Once he observed this behavior, the male's attentions changed immediately. He no longer seemed interested in defending his territory; instead he positioned himself by the side of the female and began to swim in unison with her. During this time, his body and fin coloration intensified tremendously. The dorsal part of his body and head turned almost black. On each shoulder there was an iridescent blue patch that measured approxi- mately 3 to 5 mm wide and 8 to 12 mm long. The entire ventral surface turned a bright, brassy orange. The dorsal and caudal ■.■i :r-- Q -rpr-r^ Tr> ^ ^ >1 Figure 4. Larvae of Cyprinodon variegatus. Q. 16-day old larva (6.5 mm SL); R. 26-day old larva (7.5 mm SL); S. 40-day old larva (8.1 mm SL); T. 98-day old larva (14.4 mm SL). fins were moderate to bright yellow and each had a distinct black margin. The anal and pelvic fins were yellowish orange. The pectoral fins remained essentially clear. The color of the female changed very little during this time. The dorsal part of her body became slightly darker and her venter as- sumed a pale yellow while her pectoral, caudal and dorsal fins remained essentially clear. The black spot in her dorsal fin inten- sified only slightly. After a brief period of swimming side by side, the pair settled to the bottom where the female had previously nosed the sub- strate. Suddenly, both fishes began to tremble violently. Simultaneously, they waved their caudal fins rapidly from side to side, creating a small cloud of sand particles in the water. At this time, the eggs and sperm must have been released and covered with sand, but the process occurred so rapid- ly that it was not actually observed. After the spawning act was completed, the male and female swam away from the area and any eggs that remained uncovered were quickly eaten by the other fishes that had congretaged around them. Two activities that have been observed in the aquaria have not otherwise been re- ported in the literature. They appear, how- ever, to be important aspects of the spawn- ing ritual, because of the unfailing frequency with which they occurred. The first activity concerns the apparent control of the spawn- ing time and place by the female. During each pairing, the male indicated no intent to spawn with the female until after she had "signaled" him by running her snout into the sand. Secondly, Newman (1907) indi- cated that the male almost forced the female to spawn by trapping her either in a corner of the aquarium or at the bottom. After he had captured her, he then "held the female just forward of her caudal fin, using chiefly his very strong dorsal fm." Neither the present authors nor Raney et al. (1953) ever observed the male to hold the female, even during the height of the spawning act. Repeated observations of the substrate testing by the female followed by egg burial led to the conclusion that these two activi- Nos. 3-4 Notes on Cyprinodon 141 ties were closely related and that they may represent an ecological adaptation to the heavy predation and/or other environmental conditions in the natural habitat o( Cyprino- don variegatus. The same result was probab- ly obtained by the nest construction follow- ed by spawning that has been described by Nichols and Breder (1927) and Raney et al. (1953). Both of those reports described how a nest consisting of a clean shallow de- pression was constructed by a fish, usually a male, during which time he vigorously wiggled his venter against the substrate. The spawning act that followed invariably dis- turbed bottom sediments and debris which, upon settling, probably covered the eggs. The end result of this activity was that the eggs were out of sight and predation on them was probably reduced. The spawning of Cyprinodon variegatus was observed on four different occasions during this study: 10:15 a.m. (21 April), 11:45 a.m. (1 May), 8:00 p.m. (15 April) and 9:45 p.m. (22 April). Raney et al. (1953) found that spawning activity of C. variegatus diminished in the late afternoon and ceased after dark. Nocturnal spawning of the fishes in the present study probably was due to the lengthy light period main- tained by fluorescent light banks that masked the normal effects of dusk and darkness. Spawning did not occur as a single act but consisted of a period of minutes during which a male and female spawned as many as six or seven times. Similar observations were made by Raney et al. (1953) in Florida. E E c ■D c (T3 CO 20t 19 17-1 15 13 11- 9- 7- 5- 3 N-89 r=.989 B ^A A .« A §^ AC A H BA AA CDA A^ To" 40 A A A A A A 60" "bo" 100 120 140 160 Age (Days) Figure 5. Growth of young Cyprinodon variegatus expressed as standard length. A= one observation, B= two observations, etc. 142 Tulane Studies in Zoology and Botany Vol. 20 The number of eggs deposited during any one pairing could not be determined because of the rapidity with which the act occurred and their being covered. Moreover, it was not uncommon for the eggs of one spawning to be placed either close to or on top of the eggs from a previous one deposited during the same spawning period. Consequently, when the eggs were collected, they could not be assigned to one spawn or the other. Best estimates indicate that between 8 to 10 eggs were deposited per spawn. EMBRYOLOGY Two hundred and twenty-four eggs of Cyprinodon variegatus were collected and observed during the study. The newly laid eggs averaged 1.1 to 1.3 mm in diameter, were spherical, yellowish in color and demersal. Each egg contained a yolk that averaged 0.9 mm in diameter and each yolk contained one large, round oil droplet (0.2 mm average diameter) and numerous smaller ones of various sizes randomly distributed throughout it. The surface of each egg was covered with numerous tiny adhesive threads like those noted by Kuntz (1916). The threads caused eggs to stick to each other and any objects that they touched. The following is a composite description of the various embryological stages that were observed during the study. The letter assign- ed to each stage corresponds to the drawings in Figures 1 and 2. A. One-celled stage (45 minutes after ferti- lization) Each egg has swelled slightly due to the infusion of water so that the average Q. QJ Q E 13 E •X 71 A- 3- 2- 1-A N-- 89 r=.977 A A AA AA A AA A AA AAA BAA AA T^ A A B. A AA A AB AA A A A BCABA Ag CC A A AAA A A A —I — 17 I 19 T — I- 5 T -T- 7 9 — I— 11 —I — 13 15 Standard Length (mm) Figure 6. Proportions of young Cyprinodon variegatus expressed as maximum depth vs standard length. Legend as in Fig. 5. Nos. 3-4 Notes on Cyprinodon 143 diameter is 1.5 mm. A distinct perivitel- line space is visible around each yolk. A prominent crescent-shaped structure, the blastodisk, is present on one side of the yolk. B. Two-celled stage (75 minutes) As a result of the first mitotic division, two blastomeres of equal size are formed. The chorion has become taut so that it is difficult to press downward on the egg without it being deflected to one side. The area immediately below the blastomeres now contains only a few oil droplets. C. Four-celled state (2 hours) The second cleavage occurs at right angles to the first and the resulting four blastomeres are equal in size and approx- imately one-quarter the dimensions of the previous two. Approximately one- fourth of the yolk below the blastomere is essentially clear of oil droplets. D. Eight-celled stage {IVi to 3 hours) The third cleavage occurs at right angles to the second and forms two rows with four blastomeres in each row. The cells appear to be slightly unequal in size. Approximately one-third of the yolk is free of oil droplets. E. Early blastula (4 hours) The early blastula stage is formed as the result of continuous asynchronous cell division at several planes. With each succeeding division, cell size has de- creased. F. Late blastula (6 to 7 hours) Cell division has continued and a prominent berry -like structure called the morula is present on the side of the yolk. 6n E .^4 DO C 3- 2- 1- N=89 r= .989 AAA A AA AA ^8A ACBE A B A CAA A AAA A V^* AC A ABA A A A AA A 3 — r- 5 -r- 7 T- 9 — r- 11 A A — r- 13 Standard Length (mm) A A A 15 T I 17 — r- 19 Figure 7. Proportions of young Cyprinodon variegatus expressed as head length vs standard length. Legend as in Fig. 5. 144 Tulane Studies in Zoology and Botany Vol. 20 G. Early neurula (10 hours) At the beginning of this stage, the morula becomes flattened on top and its base starts to spread out evenly over the surface of the yolk. Soon, the leading edge of this layer of cells (the germ ring) will thicken and slightly constrict the yolk. H. Mid neurula (15 hours) The blastoderm has migrated over approximately one-half of the yolk. The constriction of the yolk by the germ ring is more apparent. 1. Late neurula (22 to 23 hours) The uncovered portion of the yolk is seen as a prominent bulge. The embry- onic shield, which indicates the future longitudinal axis of the embryo, is visible as a thickened area on the blastoderm. The neural keel is present on the embry- onic shield. J. Six-somite stage (36 hours) Six pairs of somites have developed at about the middle of the embryo. Its an- terior end may be recognized by the optic placodes on either side of the head. There is a distinct dome on top of the head in the region of the midbrain. The large oil droplet is now positioned against the ventral side of the yolk oppo- site the developing embryo. K. Ten-somite stage (48 to 50 hours) The optic placodes are more circular around their edges and for the first time a faint outline of a pupO is present. The area over the midbrain has increased in size and there is also a slight bulge over the area of the hind brain. The first melanophores have developed on the dorsolateral portions of the yolk and each adjacent side of the body. The yolk has decreased slightly in size. Immedi- ately anterior to it is a small cavity that contains the structure that will become the heart. The heart is not beating at this time. The notochord is visible as a rod that extends from the rear edge of the brain posteriorly to the end of the body. L. Free-tail prolarva (90 hours) The ventral edge of the posterior end of the body has completely separated from the yolk, and this forms the basis for the name of this stage. The tail has oriented to one side of the body because of the occasional wriggling of the pro- ■ larva inside the egg. The eyes appear to % be fully formed and the iris of each is turning black. A small pectoral fin bud is present on either side of the body im- mediately posterior to the head and dorsal to the yolk. Melanophores are present on the snout and the dorsal and lateral sides of the body. Three or four concentrations of melanophores on the back form dorsal saddles. The heart is pumping at 104 to 106 beats per minute. The blood is light pink in color and the circulatory pathway throughout the head and body is visible. Twenty to 22 somites are present in the posterior three-fourths of the body, many of which occur as myomeres of muscle. A continuous fin fold is present; it arises on the dorsal midline above the yolk, ex- tends posteriorly around the tail and anteriorly on the ventral midline to the vent. The first indications of two or three caudal fin rays are present in the fin fold. THE HATCHING PROCESS The earliest time at which hatching oc- curred was between 120-125 hours after fertilization, the temperature having been 23°C ±1°. The majority of the prolarvae had hatched by 150 hours but a few required as long as 160 to 170 hours. Prior to hatching, the movements of each prolarva became more frequent and violent inside the egg. During one of these periods, the chorion ripped and this treed the posteri- or end of the fish. After a brief rest, the prolarva vigorously shook its head from side to side and eventually managed to free itself from the egg. For the first few hours, the prolarvae re- mained motionless, lying on their sides on the bottom. Periodically, one or two indi- viduals darted ahead for a short distance along the bottom or swam upward in the Nos. 3-4 Notes on Cyprinodon 145 water column in a short burst of speed. As quickly as they started, however, they stopped and drifted back to the bottom. DESCRIPTION OF THE PROLARVA The newly hatched prolarva of Cyprin- odon variegcitus was tear-drop in shape ex- cept for the large round yolk on its ventral side (Figure 3M). Total length averages 3.0 mm and standard length 2.7 mm (Table 1). The eyes were black with a gray iris. Kuntz (1916) noted that the head of Cyprinodon variegatus was not deflected; however, it was deflected on all ot the newly hatched pro- larvae that were observed in the present study. It is not certain if the references to the non-deflected head by Lippson and Moran (1974) and Scotton et al. (1973) were based on Kuntz (1916) or on original observation. A large bump occurred on top of the head over the enlarging brain. A small, incompletely formed mouth was visible in a terminal position on the snout. The rear edges of the opercles were free and moved as the prolarva breathed. The body of the prolarva was translucent pale yellow and the dorsal and lateral por- tions of it and the head were covered with randomly distributed melanophores. Seven or eight irregular saddles of melanophore concentrations appeared on the back. The total number of myomeres in the body varied from 23 to 26. Postanal myomeres ranged from 16 to 19 but averaged 17 or 18. Inside the yolk and located along its an- terio-ventral wall was a small but distinct oil 11 - 9\ E E o I c 03 to o Q I o I O c CO 7- ^ 5- 3- N-89 r =.993 A B BAAA AAA A A 8 A B AA AA BAAA A AA A BAA AB DA A B „AAAA^A BACA AAA A A A -T -T — r- 5 -T— 19 -r- 7 -r- 9 — ?— 11 I 13 — r- 15 — f— 17 Standard Length (mm) Figure 8. Proportions of young Cyprinodon variegatus expressed as snout-to-dorsal-fin-origin length vs standard length. Legend as in Fig. 5. 146 Tulane Studies in Zoology and Botany Vol. 20 droplet. The heart was easily seen inside the pericardial cavity which was located immedi- ately anterior to the yolk in the area of the isthmus. The heart beat averaged 124-125 beats per minute. The blood was light pink, the erythrocytes were distinguishable and their pathway throughout most of the body was easily observed. A continuous finfold was present on the dorsal midline, slightly behind the head. It extended around the posterior end of the body and continued anteriorly on the ven- tral midline to the vent. Six or seven caudal fin ray primordia were present but no dorsal or anal primordia could be seen. A small pectoral fm bud that contained no rays was present on either side of the body immedi- ately posterior to the opercle. No pelvic fin buds were observed on the newly hatched prolarvae. LARVAL DEVELOPMENT Growth of Cyprinodon variegatus larvae was rapid. Age vs standard length is pre- sented in Table 1 and Fig. 5; various mor- phological measurements in relation to standard length are presented in Table 1 and Figs. 6-9. Figures 3 and 4 depict significant morphological stages during larval develop- ment. YOLK: Shortly after hatching, the yolk began to decrease noticeably in size. On day 3 (Figure 3,N), only a small portion was left and the oil droplet was not discernible. By day 8 (Figure 3,0), the yolk had completely disappeared. FINS: Dorsal fin ray primordia were first apparent on 9- and 10-day old larvae. Twelve -day old larvae (Figure 3,P) possessed 13i E 11-1 E DO Q4 c > I o 3 o c CO 7- 5- 3- N-89 r^.994 BAAA BA A a^a' A ABBA A AA AA D A AAA — r- A ABA CAC C ADAA — r— 5 T- 7 9 —I — 11 —I — 13 15 Standard Length (m m) A A — I — 17 AA — I — 19 Figure 9. Proportions of young Cyprinodon variegatus expressed as snout-to-vent length vs standard length. Legend as in Fig. 5. Nos. 3-4 Notes on Cyprinodon 147 c/: CO < O U < > O Q O o O ix: fac C II rH O Q rt ^; Xl (D +-I > be C O 0) -P c > o 3 o c I— I OS H o o w T3 O C o o o o X X 0) o o w Cd CO O l-H p ^ < 0) ITS c a CO x: +-> C x: +-> fciC c 0) CO > H . IZ; rH CO CO x: c c3 bJ3 C C •- CS 01 x: +j rH -P W a C > •H T3 JQ bn •H E o E ■H -H C X > •H rt -H 'VH S "D 0) 5-1 X o <- X +-> C a; 0 t— I X} •J CO \ o H O > rH CO Q X > CO g CO :z; CO CO 0) bC ooccoo'^t— ioootr>'^a)ioot^mcsicDcot>-t^occccco'^'^ r-HOOOt^OOOOrHTfiOt*OXOOCV)Ttr-l r-li— lOOr— Ir— Ip-lrHr— IrHi— IpHCvJr— (CMCMWCCICSlCO'^U^U^'^O i-lrHrHr-IMCSIOJCMCOCOCO'^'^'^'^'^tOiniOCDTJC^r-IOqN rHcO'^iniOOO^t^COiOXCVI"<^0005OON000005Jcoco^'^'^^inin(£)t£)c^t^t^t^ooxcT5a)rHT}<;ch-oo rH i-H fM rH N C») COC0t^CSl'^'^NrHt*rriO0^inr-IT}<(£)'<1 0) cc t: c re 0) > o in c o TJ re a 0 u +-> c