ay ‘Mi = - a eee eRe DP PS a EDO LLL ADA PP LL A-ot- a aee ee ee ee we Pa Ti a ti tt oa a He a al . peeled eS ee ae a RR a al HARVARD UNIVERSITY REE iITAS LIBRARY OF THE Museum of Comparative Zoology ~ a] 7 7 p f, [ {J DIHOS ANG ihe 7) 40h i TULANE STUDIES Iny ZOOLOGY VO EP EINEES i 1966-1967 TULANE UNIVERSITY NEW ORLEANS TULANE STUDIES IN ZOOLOGY is devoted primarily to the zoology of the waters and adjacent land areas of the Gulf of Mexico and the Caribbean Sea. Each number is issued separately and contains an individual monographic study, or several minor studies. As volumes are completed, title pages and tables of contents are distributed to institu- tions 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. The editors of Tulane Studies in Zoology recommend conformance with the principles stated in chapters I and II (only) of the Style Manual for Biological Journals, 2nd ed., published in 1964 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. An abstract not exceeding three percent of the length of the original article must ac- company each manuscript submitted. This will be transmitted to Biological Abstracts and any other abstracting journal specified by the writer. The editors also recognize the policy adopted by the Federal Council for Science and Technology, and endorsed by the Conference of Biological Editors, that page charges for publication of scientific research results in scientific journals will be budgeted for and paid as a necessary part of research costs under Federal grants and contracts. Accord- ingly, writers crediting research grant support in their contributions will be requested to defray publication costs if allowable under the terms of their specific awards. Illustrations and tabular matter in excess of 20 percent of the total number of pages may be charged to the author, the levy applied being the excess above 10-point typesetting costs. Exchanges are invited from institutions publishing comparable series but subscriptions are available if no exchange agreement can be effected. Separate numbers or volumes can be purchased by individuals, but subscriptions are not accepted. Remittance should accompany orders from individuals. Authors may obtain separates of their articles at cost. Address all communications concerning manuscripts and editorial matters to the editor; Communications concerning exchanges, and orders for individual numbers to the Meade Natural History Library. When citing this series authors are requested to use the following abbreviations: Tzlane Stud. Zool. Price for this volume: $1.75. Harold A. Dundee, Editor Gerald E. Gunning, Associate Editor Department of Biology, Tulane University, New Orleans, Louisiana 70118, U.S.A. Harold A. Dundee, Director Meade Natural History Library, Robert A. Martin Tulane University, Assistant to the Editors New Orleans, Louisiana 70118, U.S.A. 4 NUMBER 1. POPULATION CHANGES IN RHESUS MONKEYS: CAYO SANTIAGO, CONTENTS OF VOLUME 13 1960-1964 __ "Gant a Roroed TWO NEW SPECIES OF THE GENUS CAMBARUS FROM ARKANSAS (DECAPODA, ASTACIDAE)_ nee Rolin Meee Rewme SPHOCIROLANA THERMYDRONIS, A NEW SPECIES OF CIROLANID ISOPOD CRUSTACEAN FROM CENTRAL COAHUILA, MEXICO_ Gerald A. Cole and W. L. iincklew SOCIAL ORGANIZATION OF THE SOUTH AMERICAN MONKEY, CAL- LICEBUS MOLOCH: A PRELIMINARY REPORT____. William A. Mason A COMPARATIVE BIOSYSTEMATIC STUDY OF FUNDULUS NOTATUS AND FUNDULUS OLIVACEUS (PISCES: CYPRINODONTIDAE) Jamie E. Thomerson® ANALYSIS OF A KEY ROLE IN A CAPUCHIN (CEBUS ALBIFRONS) GROUP Irwin S. Bernstein RESURRECTED NAMES FOR MEXICAN POPULATIONS OF BLACK- NECKED GARTER SNAKES, THAMNOPHIS CYRTOPSIS (KENNICOTT) Robert G. Webb SKELETAL AGE CHANGES IN THE CHIMPANZEE Ellis R. Kerley THE WESTERN ATLANTIC SWIMMING CRABS CALLINECTES ORNA- TUS, C. DANAE, AND A NEW, RELATED SPECIES (DECAPODA, POR- TUNIDAE) Austin B. Williams ETHEOSTOMA RUBRUM, A NEW PERCID FISH OF THE SUBGENUS NOTHONOTUS FROM BAYOU PIERRE, MISSISSIPPI : Edward C, Raney and inal D. ‘Suttkus A REVIEW OF THE COLUBRID SNAKE GENUS CHEMOPHORA COPE__. Kenneth L. Williams and Larry David Wilson STUDIES ON AMERICAN PARAGONIMIASIS. V. FURTHER OBSERVA- TIONS ON THE PRESENCE OF PARAGONIMUS IN FRESH-WATER CRABS FROM COSTA RICA, WITH NOTES ON SUSCEPTIBILITY TO CERCARIAE OF P. KELLICOTTL. Franklin Speandene. es- pera ante Alfr oa E. fewalles PERCINA AUROLINEA TA, ANEW PERCID FISH FROM THE ALABAMA RIVER SYSTEM AND A DISCUSSION OF ECOLOGY, DISTRIBUTION, AND HYBRIDIZATION OF DARTERS OF THE SUBGENUS HADROP- Royal D. Suttkus and John S. Ramsey PAGE 1 17 25 29 49 7a! 95 125 129 Printed in the U.S.A. at New Orleans, by HAuSER-AMERICAN TULANE STUDIES IN ZOOLOGY VOLUME 13 INDEX TO AUTHORS AND SCIENTIFIC NAMES (New taxonomic entities in boldface) Alosa alabamae, 140 Alouatta, 23, 24 Ambloplites rupestris ariommus, 140 Ammocrypta beani, 140 vivax, 140 Anguilla rostrata, 140 Antrolana lira, 20 Aotus, 24 trivirgatus, 23 Archillurbania nouvelli, 127 recondita, 127 Arizona, 110-112, 121 Artemia, 39 Ateles, 23 Bernstein, lrwin S., article, 49-54 Callicebus, 25-28 moloch, 23 ornatus, 24 Callinectes, 83-93 danae, 83-93 diacanthus, 84-86 ornatus, 83-93 sapidus, 98 similis, n. sp., 87-93 Cambarus causeyi, n. sp., 9, 11 diogenes diogenes, 14 ludovicianus, 14 hedgpethi, 14 setosus, 11 strawni, n. sp., 11, 14 zophonastes, 11 Campostoma anomalum, 139, 140 Cebus, 238, 24 albifrons, 49, 50 Cemophora, 103-124 coccinea, 103-124 coccinea, 112, 113 copei, 113-116 lineri, 116, 117 copei, 108, 113 doliata, 112, 113 coccinea, 112 doliata, 113 Cercopithecus aethiops, 49 Cirolandes texensis, 21 Cnemidophorus, 56 sexlineatus, 110 Cole, Gerald A., article, 17-22 Coluber coccineus, 103, 104, 112, 113 doliatus, 104 dumfrisiensis, 103, 113 elapsoides, 104 Conilera stygia, 17, 21 Coronella coccineus, 104 Cottus carolinae zopherus, 140 Diadophis, 108 Dorosoma cepedianum, 140 Hlaphe g. guttata, 108 quadrivitatta, 50 Elaps, 103 coccineus, 112, 118 Hricymba buecata, 140 EHtheostoma camurum, 95-98 histrio, 141 jordani, 139 moorei, 95-98 rubrum, n. sp., 95-102 rupestre, 140 stigmaeum, 140 tippecanoe, 96-102 zonale, 101 Eumeces, 108 Hutaenia aurata, 58 cyrtopsis, 56-69 faireyt, 61-69 proxima, 61-69 pulchrilatus, 63-69 sirtalis, 58-69 Fundulus notatus, 29-31, 33, 37-39, 41, 43, 44, 46 olivaceus, 29-31, 33, 35-39, 41, 43, 44, 46, 140 Gambusia marshi, 22 Graptemys, 108 Hadropterus, 129, 130 scierus, 141 Haldea, 108 Heterodon, 103 coccineus, 112, 113 Holbrookia maculata, 57 Hybopsis aestivalis, 140 amblops winchelli, 140 storeriana, 140 Hylobates lar, 49 Hypentelium etowanum, 139, 140 Ichthyomyzon gagei, 140 Justica, 139 Kerley, Ellis R., article, 71-82 Koford, Carl B., article, 1-7 Lagothrix, 23 Lampropeltis, 103, 110-112, 121 doliata doliata, 110 triangulum, 103, 105, 110, 121 elapsoides, 104, 110 syspila, 110 Lepomis auritus, 140 cyanellus, 140 macrochirus, 140 megalotis, 140 INDEX TO AUTHORS AND SCIENTIFIC NAMES—Continued Lichanura trivirgata, 56, 57 Loxothylacus texanus, 93 Lupa dicantha, 86 Macaca mulatta, 1, 49 nemestrina, 49 Maritrima. prolixum, 125-127 Mason, William A., article, 25-28 Micropterus coosae, 139, 140 punctulatus henshalli, 140 Minckley, W. L., article, 17-22 Moxostoma duquesnei, 140 Nocomis micropogon, 139, 140 Nothonotus camurum, 96-102 moorei, 96-102 rubrum, 95-102 tippecanoe, 96-102 Noturus hildebrandi, 101 munitus, 140 Notropis caeruleus, 140 callistius, 139, 140 euryzonus, 31 stilbius, 139, 140 trichroistius, 139, 140 wranoscopus, 140 venustus stigmaturus, 140 volucellus, 140 Nymphaea, 20 Oligodon, 103 Orconectes palmeri longimanus, 14 Pan, 71-82 Paragonimus, 125, 126 kellicotti, 125, 126 rudis, 126, 127 Percina, 129-145 aurolineata, n. sp., 129-145 caprodes carbonaria, 140 copelandi, 140 lenticula, 129-145 nigrofasciata, 129-145 nigrofasciata, 140 palmaris, 136, 139, 140 sciera, 101, 129-145 shumardi, 140 uranidea, 101 Phenacobius catostomus, 140 Philander opossum fuscogriseus, 127 Pimephales vigilax perspicuus, 140 Pituophis, 110-112, 121 Podostemum, 139 Pomatiopsis lapidaria, 125 Potamocarcinus magnus, 125, 126 Priatella phreatophila, 22 Procambarus, 12 blandingi acutus, 14 clarki, 126 simulans simulans, 14 Pseudemys, 108 Pseudothelphusa magna, 125 tristrani, 125 Ptychophallus montanus, 126 tristrani, 125, 126 tumimanus, 126 Ramsey, John S., article, 129-145 Raney, Edward C., article, 95-102 Reimer, Rollin Dewayne, article, 9-15 Rhinocheilus, 110-112, 121 Rhinostoma, 103 coccinea, 112, 113 coccineus, 112, 118 Saimiri, 23, 24, 27 Semotilus atromaculatus, 140 Simotes, 108 coccineus, 112 Smalley, Alfred E., article, 125-128 Sogandares-Bernal, Franklin, article, 125-128 Speocirolana bolivari, 18-21 pelaezi, 18-21 thermydronis, n. sp., 17-19, 20, 22 Stasiotes, 108, 104 coccineus, 112, 118 Stilosoma, 110-112, 121 Suttkus, Royal D., article, 95-102; article, 129-145 Tantilla c. coronata, 110 Terrapene c. bauri, 108 Thamnophis collaris, 55-69 cyclides, 55-69 cyrtopsis, 55, 56 collaris, 60-69 cyclides, 56-69 cyrtopsis, 56-69 ocellata, 69 postremus, 69 pulchrilatus, 61-69 sumichrasti, 68 dorsalis, 55-69 eques, 56-69 faireyi, 61-69 proxima, 61-69 pulchrilatus, 55-69 sirtalis, 55-69 ornata, 55-69 sumichrasti, 68 vicinus, 62-69 Thomerson, Jamie E., article, 29-47 Troglotrema acutus, 127 Tropidonotus sirtalis dorsalis, 56-69 ordinatus eques, 68 Uma, 110 Webb, Robert G., article, 55-70 Williams, Austin B., article, 83-93 Williams, Kenneth L., article, 103-124 Wilson, Larry David, article, 103-124 TULANE STUDIES . TON} ZOOLOGY MAK 28 1966 S- NA- Neon dow egal Volume 13, Number 1 March 17, 1966 POPULATION CHANGES IN RHESUS MONKEYS: CAYO SANTIAGO, 1960-1964 CARL B. KOFORD, Laboratory of Perinatal Physiology National Institute of Neurological Diseases and Blindness National Institutes of Health Public Health Service U. S. Department of Health, Education and Welfare San Juan, Puerto Rico p. 1 TWO NEW SPECIES OF THE GENUS CAMBARUS FROM ARKANSAS (DECAPODA, ASTACIDAE) ROLLIN DEWAYNE REIMER, Department of Zoology, University of Arkansas Fayetteville, Arkansas p. 9 SPEOCIROLANA THERMYDRONIS, A NEW SPECIES OF CIROLANID ISOPOD CRUSTACEAN FROM CENTRAL COAHUILA, MEXICO GERALD A. COLE and W. L. MINCKLEY, Department of Zoology, Arizona State University empe, Arizona p. 17 SOCIAL ORGANIZATION OF THE SOUTH AMERICAN MONKEY, CALLICEBUS MOLOCH: A PRELIMINARY REPORT WILLIAM A. MASON, Delta Regional Primate Research Center Covington, Louisiana p. 23 A COMPARATIVE BIOSYSTEMATIC STUDY OF FUNDULUS NOT ATUS AND FUNDULUS OLIVACEUS (PISCES: CYPRINODONTIDAE) JAMIE E. THOMERSON, Department of Biology, Tulane University New Orleans, Louisiana p. 29 TULANE UNIVERSITY NEW ORLEANS and adjacent land areas of the Gulf of Mexico and the Caribbean Sea. Each number is issued separately and contains an individual monographic study, or several minor studies. As volumes are completed, title pages and tables of contents are distributed to institu- tions receiving the entire series. TULANE STUDIES IN ZOOLOGY is devoted primarily to the zoology of the waters : 4 | 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. The editors of Tulane Studies in Zoology recommend conformance with the principles — stated in chapters I and II (only) of the Style Manual for Biological Journals, 2nd ed., — published in 1964 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. ‘Iwo copies, carbon or other suitable reproduction, must ~ accompany 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. An abstract not exceeding three percent of the length of the original article must ac- company each manuscript submitted. This will be transmitted to Biological Abstracts and any other abstracting journal specified by the writer. The editors also recognize the policy adopted by the Federal Council for Science and Technology, and endorsed by the Conference of Biological Editors, that page charges for publication of scientific research results in scientific journals will be budgeted for and paid as a necessary part of research costs under Federal grants and contracts. Accord- ingly, writers crediting research grant support in their contributions will be requested to defray publication costs if allowable under the terms of their specific awards. Illustrations and tabular matter in excess of 20 percent of the total number of pages may be charged to the author, the levy applied being the excess above 10-point typesetting costs. Exchanges are invited from institutions publishing comparable series but subscriptions are available if no exchange agreement can be effected. Separate numbers or volumes can be purchased by individuals, but subscriptions are not accepted. Remittance should accompany orders from individuals. Authors may obtain separates of their articles at cost. Address all communications concerning manuscripts and editorial matters to the editor; Communications concerning exchanges, and orders for individual numbers to the Meade Natural History Library. When citing this series authors are requested to use the following abbreviations: Tulane Stud, Zool. Price for this number: $1.50. Harold A. Dundee, Editor Gerald E. Gunning, Associate Editor Department of Biology, Tulane University, New Orleans, Louisiana 70118, U.S.A Harold A. Dundee, Director Meade Natural History Library, James R. Reed, Jr. Tulane University, Assistant to the Editors New Orleans, Louisiana 70118, U.S.A. the original to expedite editing and assure more rapid publication. Legends TULANE STUDIES IN ZOOLOGY Volume 13, Number 1 March 17, 1966 POPULATION CHANGES IN RHESUS MONKEYS: CAYO SANTIAGO, + 7O 1960-1964 LIBRARY CARL B. KOFORD, Laboratory of Perinatal Physiology a; 5 5F National Institute of Neurological Diseases and Blindness MAK & © \JOU National Institutes of Health Public Health Service HARVARD U.S. Department of Health, Education and Welfare NIVERSITY San Juan, Puerto Rico UNIVE! : ABSTRACT A population of introduced rhesus monkeys increased 16% annually dur- ing a five year period. Ultimately it consisted of six bands, five of which were formed by subdivision of one of the two original large bands. Divisions occurred during the fall mating season, apparently due to factors other than band size. Of the males at least three years of age about a third changed bands annually. This tendency to shift bore no constant relation to the size or the sex composition of the bands. Near- ly all births occurred January to June each year. Mortality rates were great- est in animals less than two years and over Six years of age, and in post pu- beral males. In the adult population the females outnumber males by two to one. Detailed long term studies of the numeri- cal and social relataions in animal populations are feasible only when the individuals are marked and their movements somewhat re- stricted. Such a population is the colony of rhesus monkeys (Macaca mulatta) on Cayo Santiago, a wooded 40-acre islet situated off the east coast of Puerto Rico. All the ani- mals are descendants of monkeys released there in 1938, and though most of their food has been provided by man, the animals are essentially wild. There were about 350 monkeys in 1940 when Carpenter (1942) studied their sexual behavior. Because of subsequent mortality and occasional heavy removals, there were only about 150 left in 1956 when Altmann (1962) commenced a 2-year study of monkey sociobiology. Large- ly because he tattooed the majority, I was able to account for every individual by the end of 1959, a year after I commenced work (Koford, 1963); at that time there were 277 monkeys. Since 1959 the population has grown about 16% per year, though this nat- ural increase has been partly offset by the removal of 84 animals, mostly young males. BAND COMPOSITION The current (mid-1964) population com- prises 482 monkeys. Four are solitary males and the rest live in six bands of 167, 142, 56, 53, 39, and 21 members. The largest band, A, has been an entity since early 1956 or before (Fig. 1). The other five bands, C, E, F, H, and I, were formed by the sub- division of a single band, B, in three stages which occurred about one year apart. Al- though two bands, A and C, are now larger than any which split, there have been no divisions in the past 4 years. Contrary to EDITORIAL COMMITTEE FOR THIS PAPER: Missouri Louisiana ton, Louisiana CLINTON H. CoNAwWay, Professor of Zoology, University of Missouri, Columbia, HANS KUMMER, Research Associate, Delta Regional Primate Center, Covington, EMIL W. MENZEL, JR., Research Associate, Delta Regional Primate Center, Coving- 2 A (C E | H IP vt o on rm © a N o on o a ° o o G on 0 o D ae 1 i \ \ t — { ' \ \ / y = { 1 \ \ Wea / © { 4 \ 7, 0 1 ] \ / 2 { ! \ / oe bef VB! 1 1 { \ I n fea 0 | | I ; 2 Eo bel | 1 = ' : I 1 100 MONKEYS © I 4 I | o PS ! j t | EVOLUTION OF BANDS Fgure 1. Changes in the number and size of bands during the pericd 1956 to 1964. Capital letters are designations of bands. Horizontal widths indicate numbers of ani- mals. Broken outline indicates early period of incomplete data. expectation, it was not always the largest band which split. At the time when band G divided, in 1960. it was a third smaller than band A or C. Notably at least two of the divisions occurred in fall, when mating and social tensions were highest. REPRODUCTION No monkeys have been added to the col- ony; numbers have increased only through births. The birth dates of nearly all born from 1960 to 1964 (454) are known within one day. These births occurred from late December to late July, within a span of 210 days. But in any one year a period of 130 days included practically all births (at least 94%). In spite of the increase in numbers born each year from 70 in 1960 to 119 in 1964, there has been no increase in the length of the birth season. In fact, the sea- son with the most births was the shortest, only 126 days (Fig. 2). Tulane Studies in Zoolog) Vol. 13 From year to year the time of the birth season has varied moderately. For the 5-year period, the range of the initial birth date has been 46 days (December 29 to Febru- ary 13), and of the median date 34 days (March 2 to April 5), while the final birth date, sometimes more than a month after the others, has ranged 65 days (latest, July 26). The interval from the last birth of one year to the first of the next year has varied from 165 to 224 days. Apparently some of the annual variations in birth season were caused by differences in weather and its in- fluence on the nutritional quality of plant foods, because delay in the start of heavy rains in spring has been accompanied by delay in the onset of mating in summer (Koford, 1965). In any one year, the distribution of births was also influenced by the survival of in- fants born the previous year, because multi- parous females that failed to reproduce, or which lost their infants before the mating season, tended to come into estrus and con- ceive earlier than others. In adults the dif- ference was probably caused by the fact that lactation delays the onset of reproductive cycles (Hartman, 1932). Nevertheless, mon- keys conceiving for the first time (normally at 314 years of age) give birth at about the 1964 119 BIRTHS JAN FEB Distribution of all (119) birth MAR APR MAY Figure 2. dates during 1964. Ordinate is number of births per one-third-month period. No. 1 same time as lactating adults. For the 5 year period, the peak month of births to non- lactating adults occurred in February, a month earlier than the peak for other fe- males (Fig. 3). Therefore, when reproduc- tion is poor and early infant mortality is high, the following birth season tends to be early. There are exceptions and a few non- lactating adult females breed late (Fig. 4). The overall birth pattern also differed among the bands (Fig. 4). In 1964, the spread of initial births among the six bands was 28 days, and of the median birth dates, 24 days. That birth season was unusually compact; in other years, the respective spreads have been as much as 67 and 33 days. Even between the two largest bands, in a single year (1960) the initial dates (for 21 and 25 births) differed by 32 days. Con- siderable variations are to be expected, of course, in view of individual differences in the virility of males, irregularities in the fre- quency, intensity, and duration of estrus in females, and the complex social interactions within the breeding population during the mating season (Conaway and Koford, 1964). Population growth is influenced by fer- tility, which varies from year to year and 50 40 % OF BIRTHS DEC, WAN - FEB Figure 3. season (shaded bars, N = N = 359), 1960-1964. Rhesus Monkeys 3 among bands. Over a period of 5 years the reproductive rate, or ratio of births to the number of mature females, has ranged from 78% (1960) to 86% (1964), with an ap- parent tendency to increase. For newly ma- ture females, 4 years old, the mean reproduc- tive rate has been the same as for older ones (81% of 89 vs. 82% of 460). On rare occasions (7 in 6 years) a 3-year-old female has given birth. So far, there has been no obvious relation between the size of bands and high reproductive rate; the band with the highest mean rate (93% ) was the third largest. MORTALITY Births tended to increase numbers about 25% annually, but deaths partly offset this increment. Considering mortality rate to be the number dying during the year as a per- centage of the number alive at the start of the year, the mean mortality rate for the years 1960-1963, excluding infants, was 6.7% (N=1235), with little year to year variation (5.9% of 320 in 1961, to 7.3% of 372 in 1963). Mortality tended to be high in the old and young; for yearlings the mean rate was 9.8% (N=235) and for animals 7 NOT LACTATING (N®95) — LACTATING (N= 395) MAR APR MAY JUN JUL Distribution of births to adults that were not lactating during the mating 95) and to other females, lactating or primiparous (open bars, 4 Tulane Studies in Zoology 1964 BIRTH S N DEC JAN FEB MAR APR MAY JUN JUL Figure 4. Distribution of birth dates by bands, 1964. Shaded squares indicate births to adults that were not lactating during the preceding mating season. Ordinate as in Fig. 2. N = 179192 104131 81 90 l2 MALES Mm aoe ° FEMALES o~ ——_ 8 > = a4 C:ECAeS-S “UY:R?S) Figure 7. Mean proportion of each male age class departing from bands each year, 1960-1963. N is pooled number of males in each class at the start of the year. MATING BIRTHS NO. MALES DEPARTING JF MA M J J MONTH Figure 8. Month of male departures (126) from bands, 1960-1963. Males becom- ing solitary are included. A S O N OD August to November. The restlessness of males during the mating season was also in- dicated by frequent fighting. The interchange of males among bands was complex, even in single years (Fig. 9). Size, sex ratio, or identity of bands did not affect interchanges; the frequency of depar- tures seemed to depend on social tensions among certain individuals, and in part on the readiness of other bands to accept the Movements of 49 males among Figure 9. bands, or between band and solitary status (S), during 1963. Only changes persisting at least one month are shown. Areas of cir- cles are proportional to the number of sexu- ally mature animals (at least 3 years old) in each band at midyear. One male re- mained solitary throughout the year. Tulane Studies in Zoology Vol. 13 aliens. Over the four-year period, the num- ber of males departing from each band con- stituted from 10% (band A, N=263) to 21% (band H, N=72) of the sexually ma- ture animals (both sexes, at least 3 years old), except for the smallest band in which the loss was 42% (band I, N=46). As to losses versus gains, these were about equal in the four largest bands (A, C, E, and F) but losses were about twice gains in the two smallest bands (H, I). Lastly, in terms of the sexually mature population of the island, the proportion leaving bands each year has varied from 10% to 15%, with no regular increase in spite of a 45% increase in numbers. Of the male departures from bands, 17% (N=126) were to solitary status, and of the 18 monkeys involved, all save two adults were 4 to 7 years old. Apparently the tend- ency of a young male to depart from his band is inhibited by the presence of his mother, for of the 2- and 3-year-old males having a mother in the same band only 13% (N=111) departed, whereas for orphans of the same age, 42% (N=24) departed. For older males, the presence of the mother had no apparent influence on the rate of de- parture. Females did not become solitary, even temporarily, and less than 3% of the females changed bands. Evidently the social forces causing females to shift bands are different from those affecting males, because most (10 to 14) changes occurred in the birth season, from January to May, rather than in the mating season. These movements in- volved eight females, at least 3 years old, none of which became a member of more than two bands. Three females changed bands about one month before giving birth, and five were accompanied by infants. SUMMARY Over a period of five years the number of free-ranging monkeys in a provisioned island population increased 16% annually, and in mid-1964 there were 482 animals in six bands. Five of these formed by subdivision of a single large band, whereas a second large band did not divide. Divisions oc- curred during the fall mating season and apparently depended on social factors other than band size. Nearly all births occurred from January to June, and in spite of a 70% increase in numbers born the length of the No. 1 birth season did not increase. Among years the initial birth dates varied 46 days and the median dates 34 days. These variations were presumably caused by annual differ- ences in weather and vegetation, and in the proportion of non-lactating parous females, which tended to breed early. Social factors also probably influenced the time of mating, for the spread of birth dates among bands in some single years was as great as for the entire population in different years. The ratio of births to the number of mature fe- males ranged from 78% to 86%, with an apparent tendency to increase. Excluding infants, mortality reduced num- bers at a mean rate of 6.7%. Animals less than 2 years and over 6 years of age suffered the highest mortality. Commencing about puberty, females survived better than males, so that in the adult population females out- number males more than two to one. Of the monkeys at least three years old, each year about a third of the males changed bands, principally in the mating season. This tendency to shift bore no constant relation to either the size nor the sex composition of the band, but among immatures orphans shifted more readily than other males. These data suggest that under natural con- ditions it is chiefly the adolescent and adult Rhesus Monkeys 7 males which tend to disperse and distribute genetic material among all bands in a region. Verification of the theoretical implications of these and other characteristics of the island colony will require similar informa- tion for wild populations of rhesus and other primates. LITERATURE CITED ALTMANN, S. A. 1962. A field study of the sociobiology of rhesus monkeys, Macaca mulatta. Ann. N. Y. Acad. Sci. 102: 338- 433. CARPENTER, C. R. 1942. Sexual behavior of free-ranging rhesus monkeys, (Macaca mulatta). J. Comp. Psychol. 33: 113-162. CoNAWAY, C. H. and C. B. Kororp 1964. Estrous cycles and mating behavior in a free-ranging band of rhesus monkeys. J. Mammalogy 45: 577-588. HARTMANN, C. G. 1932. Studies in the re- production of the monkey Macacus (Pith- ecus) rhesus, with special reference to menstruation and pregnancy. Contrib. Embryol. Carnegie Instit. 23: 1-162. Kororp, C. B. 1963. Group relations in an island colony of rhesus monkeys. In: Primate Social Behavior (Edit. C. H. Southwick). Van Nostrand; Princeton. pp. 136-152. seen ---------- 1965. Population dynam- ics of rhesus monkeys on Cayo Santiago, pp. 160-174. In: Primate Behavior; Field Studies of Monkeys and Apes (Edit. I. DeVore). Holt, Rinehart and Winston; New York. TWO NEW SPECIES OF THE GENUS CAMBARUS FROM ARKANSAS (DECAPODA, ASTACIDAE) ROLLIN DEWAYNE REIMER, Department of Zoology, University of Arkansas Fayetteville, Arkansas! ABSTRACT Two new species of the genus Cam- barus from Arkansas are described. Cambarus causeyi from northern Ar- kansas seems to have its closest affini- ties with members of the Asperimanus Group. Cambarus strawni from the southern slope of the Ouachita Moun- tains appears to be closely related to members of the Diogenes Section. The first specimens of two undescribed species were collected in 1963, during a survey of the crawfishes of Arkansas (Reimer, MS). I had hoped to obtain more individuals before publishing my findings, but since I have not been able to return to the areas it seems best to proceed with the description of these two new species. My appreciation is extended to Dr. Hor- ton H. Hobbs, Jr., United States National Museum, for the verification of these two species, and to Dr. Kirk Strawn and Dr. David Causey, both of the University of Arkansas, for their assistance while collect- ing in Arkansas. I am pleased to name these species in honor of Drs. Causey and Strawn. Cambarus Causey, new species Diagnosis—Body pigmented; eyes nor- mal. Rostrum short, excavate, lacking lateral spines; acumen indistinctly delimited at base, with dorsally projecting knob at tip: areola open but narrow, approximately 28 times longer than wide: lateral branchio- stegal spines absent; suborbital angle absent; antennal scale widest distal to midlength; ' Present address: Department of Biolo- gy, Tulane University, New Orleans, Lou- isiana 70118. chela slightly depressed; hooks on ischiopo- dites of third pereiopod of male only; first pleopod of Form I male terminating in two distinct processes bent at more than 90 de- gree angle to main shaft; annulus ventralis as figured (Fig. 2); prominent setiferous punctations over most of carapace and pere1opods. Holotypic male, Form I—Body subovate. Abdomen narrower than thorax (7 & 9 mm, respectively), shorter than carapace. Cephalic section of telson with one spine on each side. Width of cephalothorax less than depth in region of caudodorsal margin of cervical groove (8.5 & 9 mm, respectively ). Areola open but narrow; 28.0 times longer than wide. Cephalic section of cephalo- thorax 1.4 times longer than areola. Rostrum (Fig. 6) excavate dorsally; acu- men indistinct but with upturned corneous tip; marginal spines absent; tip extending cephalic to penultimate segment of peduncle of antennule. Postorbital ridges well devel- oped, with prominent groove extending al- most its full length. Subrostral ridges mod- erately developed. Suborbital angle, branchi- ostegal spine, and lateral spines absent. Sur- face of carapace densely punctate and mod- erately setiferous. Epistome (Fig. 7) equal in length and breadth, with indistinct cephalomedial pro- jection; cephalolateral edges elevated ven- trally. Eyes normal. Antennules of usual form. Antennae of usual form and only slightly longer than carapace. Antennal scale ( Fig. 4) reaching to distal end of penultimate segment of antennule; widest distal to mid- EDITORIAL COMMITTEE FOR THIS PAPER: JoE B. BLACK, Associate Professor of Biology, McNeese State College, Lake Charles, Louisiana JOSEPH F. FITZPATRICK, JR., Assistant Professor of Zoology, Mississippi State Uni- versity, State College, Mississippi HORTON H. Hosss, JR., Senior Scientist, United States National Museum, Washing- ton, D.C. 10 Tulane Studies in Zoology Se 6 Figures 1-8. Cambarus causeyi, new species: 1. lateral view of first pleopod of holo- typic male; 2. annulus ventralis; 3. mesial view of first pleopod of holotypic male; 4. an- tennal scale of holotypic male; 5. lateral view of carapace of holotypic male; 6. dorsal view of carapace of holotypic male; 7. epistome of holotypic male; 8. holotypic male. length (approximately 2.4 times longer than wide Vee Right chela (Fig. 8) slightly depressed and with palm only slightly inflated. Inner margin of palm with one row of six tuber- cles. Tubercles on lower surface of palm indistinct. All surfaces of chela bearing setiferous punctations, with setae of various lengths within same punctation. Fingers slightly curved ventrally from their bases, upper surface of chela of gaping along proximal two-thirds of their length. Inner margin of immovable finger with five distinct tubercles: proximal four noncorneous and placed on mesial border, first much smaller than distal three; tubercle at base of distal third of finger, corneous (like tip of fingers), placed laterally. Lower surface of immovable finger without tuber- cles but with single row of setiferous punc- tations. Opposable margin of movable finger with four tubercles; third from base largest. No. 1 Upper surface of movable finger with four tubercles near base. Carpus longer than wide (6.6 & 4.7 mm, respectively ) with well defined longitudinal furrow above; two prominent spines located distally (dorsal distal edge-mesial distal edge), remainder of surface lacking tuber- cles; all surface with setiferous punctations. Merus with two rows of tubercles along ven- tral margin, converging proximally, six in mesial row, three in lateral row, becoming larger distally; tubercles absent from dorsal surface; setiferous punctations on dorsal and ventral surfaces, diminishing laterally. Hooks present on ischiopodites of third pereiopods only. First pleopod (Figs. 1, 3) reaching base of third pereiopod and terminating in two distinct parts. Central projection corneous, blade-like, strongly curved caudomesiad at more than a right angle to main shaft. Mes- ial process noncorneous, bulbiform with small nipple-like apical projection, curving caudolaterally at about same angle as central projection. Allotypic female —Differs from holotypic male in the following respects: Rostrum longer, with lateral edges upturned dorsally to give deeper excavation. Inner margin of palm with five tubercles. Two rows of tu- bercles along ventral margins of merus, three in mesial row, seven in lateral row. Ce- phalic section of telson with three spines on each side. ; Annulus ventralis (Fig. 2) situated at caudal end of a deep, narrow V-shaped ster- num. Sternum partially hidden by numerous long setae. Annulus ventralis longer than wide with centrally located fossa; sinus ex- tending cephalically and caudally from fossa. Male, Form I1—Unknown. Measurements (in millimeters ).— Holo- Allo- type type Carapace Height 9.0 12.9 Width 8.5 1225 Length 20.4 Bailey Areola Width 0.3 0:2 Length 8.4 12.8 Rostrum Width 2D 326 Length 2.8 4.8 Chela Length inner (Right) margin of palm 5.7 Et Width of palm 6.0 10.4 Length outer margin of palm 15.5 23.5 Length of dactyl 9.1 14.4 New Species of Cambarus 1] Caudal sinus normal, i.e., lateral surfaces slope away from sinus region. Cephalic sinus with lateral surfaces folding mesially as if to close over top of sinus. Type locality and ecological notes—Type locality is four miles west of Sandgap, Pope Co., Arkansas, on State Highway 124; from a spring and natural pond. The spring is located on the western side of a gently slop- ing hill and drains into a small shallow pond about 200 feet away. The hillside and pas- ture were dotted with several large rocks, under which could almost always be found a burrow. Unconcealed openings to burrows were scattered over the area. No chimneys were observed. Most burrows had several openings lead- ing to a common tunnel which descended al- most vertically. None of the burrows was dug to the bottom. The holotype was taken from a burrow (top capped by a rock) on the hillside. The allotype and paratype, the latter an immature female, were taken from burrows near the natural pond. The pond was clear and contained a large number of frog and toad eggs but lacked crayfish. Disposition of types —The holotypic form I male and the allotypic female are deposited in the U. S. National Museum, numbers 116678 and 116679, respectively. A juvenile paratypic female also is deposited in the U. S. National Museum. Relationships —Cambarus causeyi seems to have its closest affinities with Cambarus setosus Faxon, 1889, and Cambarus zophon- astes Hobbs and Bedinger, 1964, members of the Asperimanus Group in the Ozark Region. Similarities include the narrow areola, setiferous punctations, single row of tubercles on inner margin of palm, and shape of the chela. Cambarus causeyi can be distinguished by normal coloration, norma! eyes, and absence of lateral spines on the rostrum and sides of carapace. In addition the mesial process of the first pleopod of the Form I male is bulbiform. The shape of the epistome and annulus ventralis is quite different from either C. setosws or C, z0- phonastes, Cambarus strawni, new species Diagnosis—Rostrum excavate, lacking marginal spines, with indistinct acumen; areola obliterated; spines absent along sur- face of carapace; suborbital angle absent; antennal scale widest at (or slightly distal WA to) midlength; chela strongly depressed; carapace and chela lacking conspicuous seti- ferous punctations; hook on ischiopodites of third pereiopods only, of male; first pleopod of Form I male terminating in three distinct processes bent at approximately a 90 degree angle to main shaft; annulus ventralis as figured (Fig. 17). Holotypic male, Form I—Body ovate. Abdomen narrower than thorax (8.7 & 13.4 mm, respectively ) and shorter than carapace. Cephalic section of telson without spines. Width of cephalothorax equal to depth in region of caudodorsal margin of cervical groove. Areola obliterated in middle; ce- phalic section of cephalothorax 1.5 times as long as areola. Rostrum depressed; upper surface exca- vate; margins converging from base; acumen only slightly delinated basally by small tu- bercle; marginal spines absent; tip extend- ing cephalically to base of penultimate seg- ment of peduncle of antennule. Subrostral and postorbital ridges moderately developed with postorbital ridges terminating ce- phalically without spine. Suborbital angle, branchiostegal spines, lateral spines absent. Punctations on dorsal surface of carapace few, increasing in number laterally. Epistome (Fig. 11) wider than long and terminating cephalically in small cephalo- median projection; cephalolateral margins slightly rounded. Eyes normal. Antennules of usual form. Antennae broken (see allotypic female). Antennal scale small, reaching slightly be- yond tip of rostrum; widest at midlength; approximately three times longer than wide (2.8 & 0.8 mm, respectively ). Right chela (Fig. 18) depressed, palm slightly inflated. Inner margin of palm with two rows of tubercles; inner row of eight much more prominent than outer of five. Tubercles on lower surface of palm indis- tinct. Palm moderately punctate. Fingers curved ventrally from bases, gaping along entire length; both slightly punctate. Inner margin of immovable finger with three tu- bercles; first from base being largest. Op- posable margin of movable finger with four tubercles; second from base largest, termi- nating arch originating at base of finger. Upper surface of movable finger with single row of seven indistinct tubercles. Carpus longer than wide (9.6 & 6.7 mm, respectively ) with well defined longitudinal 2 Tulane Studies in Zoology Vol. 13 furrow above; dorsal crest with row of four indistinct tubercles; mesial surface with se- ries of 13 irregularly placed spines, largest being near distal dorsal margin; punctations sparsely scattered over all surfaces. Merus with two rows of tubercles along ventral sur- face converging proximally (8 in lateral row—12 in mesial row); one row of 15 tubercles on dorsal margin becoming indis- tinct proximally. Punctations few on dorsal and ventral surfaces. First pleopod (Figs. 9, 13) reaching to base of third pereiopod, terminating in three distinct parts. Central projection corneous, blade-like; curving caudomesiad at slightly more than right angle to main shaft. Mesial process slightly more than right angle to main shaft of pleopod. Third distinct proc- ess lying between and mesial to mesial proc- ess and central projection; thin, slightly corneous, half as long as central projection and mesial process. According to its posi- tion, third process is probably homologous to the cephalic process in Procambarus (Hobbs, personal communication ). Allotypic female—Allotypic female dif- fers from the holotypic male in the follow- ing respects; tubercles on movable finger not as well defined, antennae approximately equal to length of carapace (one broken), slight variation in size and arrangement of tubercles on carpus and metus. Annulus ventralis (Fig. 17) situated at caudal end of deep V-shaped sternum and divided into right and left sides by promi- nent sinus leading caudally from cephalo- mesial fossa. Morphotypic male, Form 11—Morpho- Measurements (in millimeters ).— Mor- Holo- Allo- — pho- type type type Carapace Height 12.4 a7 9:2 Width 12.4 12.5 9.4 Length 28.6 28.7 Pile Areola Length 11.2 te? 8.4 Rostrum Width 4,3 4.3 3.0 Length 4.8 4.9 3.6 Chela Length inner (Right) margin of palm 6.5 5.5 3.4 Width of palm bul 4.5 2.8 Length outer margin of palm 19.5 16.3 9.8 Length of dacty] WAS 10.2 6.9 No. 1 New Species of Cambarus 15 16 Figures 9-18. Cambarus strawni, new species: 9. lateral view of first pleopod of holo- typic male; 10. mesial view of first pleopod of morphotypic male; 11. epistome of holotyp- ic male; 12. lateral view of first pleopod of morphotypic male; 13. mesial view of first pleopod of holotypic male; 14. antennal scale of holotypic male; 15. dorsal view of cara- pace of holotypic male; 16. lateral view of carapace of holotypic male; 17. annulus ven- tralis; 18. upper surface of chela of holotypic male. 14 Tulane Studies in Zoology type differs from holotype in following re- spects; two rows of tubercles on inner mar- gin of palm of right chela with ratio of 6:5 (5 in inner row), third and fourth tubercles on movable finger indistinct, slight variation in size and arrangement of tubercles on car- pus and merus, hooks on ischiopodites of third pereiopods greatly reduced in size, first pleopod (Figs. 10, 12) with only two terminal elements (distal half of right pleo- pod broken). V artations—Few variations observed. The more obvious are: second tubercle lacking on immovable finger of one specimen, third and fourth tubercles lacking on movable finger of two others. Tubercle number on dorsolateral crest of carpus varies from four to seven, and on one, tubercles not in defi- nite row. Mesial surface of carpus of one small female with 10 tubercles. Epistome shorter and with rounder cephalolateral bor- ders in some specimens. Annulus ventralis on small female (probably immature ) swol- len and not resembling that of allotype. First pleopods of both second form males lacking third terminal process. In one male Form II, mesial process twice as long as cen- tral projection. One paratypic female with spine on left side of cephalic section of telson. Type locality and ecological notes—The type locality is 2.7 miles north of Dierks, Howard County, Arkansas, on State High- way 4; a small marshy area in the Saline River drainage. All specimens were taken from burrows. The burrows were in a low area adjacent to a small permanent creek. During the wet period of the year this area is quite boggy and standing water is com- mon in places. During the dryer parts of the year the surface water is absent and the area loses its boggy nature. The soil is a sandy clay. The stream nearby is shallow, clear, fast running, and paved with a rocky bottom. Procambarus simulans simulans (Faxon), 1884, Procambarus blandingi acu- tus (Girard), 1852, and Orconectes palmeri longimanus (Faxon), 1898, were taken from the stream. The holotypic male, allotypic female, and one paratypic female were taken on June 22, 1963. Another collection at the type locality on February 17, 1965 yielded the morpho- typic male and two paratypes (male Form II, female ). Two female paratypes were also taken on Vol. 13 June 22, 1963 from the headwaters of the Cossotat River about four miles west of Umpire, Howard County, Arkansas, on State Highway 4. Color—tThe carapace is primarily olive tan with the abdomen being slightly lighter. Ventral surfaces are cream colored. Dactyl, suborbital ridge, lateral borders of the ros- trum, and articulating areas of the pereio- pods grade from a deep olive green to black. The colors are more vivid in the older or larger individuals. Burrows —Two types of burrow construc- tion have been observed. The burrows at the type locality consisted of a maze of in- terconnecting tunnels with the primary tun- nel proceeding almost vertically. The bur- rows west of Umpire, Arkansas lacked the maze and the primary tunnel went down in a spiral manner. Some openings at both locations were capped with chimneys. Disposition of types —tThe holotypic form I male, allotypic female, and morphotypic form II male are deposited in the United States National Museum, numbers 116675, 116676, 116677, respectively. The five para- types (four females and a form II male) have been retained in my personal collection. Relationships —With the exception of having three terminal projections on the first pleopod of the Form I male, Cambarus strawnt appears to be most closely related to members of the Diogenes Section. Cambarus diogenes diogenes Girard (1852), Cambarus diogenes ludovicianus Faxon (1884), and Cambarus hedgpethi Hobbs (1948) are members of the Diogenes Section which oc- cur in the same area. Cambarus strawni can be distinguished from these three species by the three terminal projections on the first pleopod of the Form I male, shape of the antennal scale, shape of the annulus ven- tralis of the female, tubercle arrangement on the fingers of the chelae, and in the case of Cambarus diogenes diogenes and Cambarus diogenes ludovicianus, by the absence of suborbital angles. LITERATURE CITED FAXON, WALTER 1884. Description of a new species of Cambarus to which is added a synonymical list of known species of Cam- barus and Astacus. Proc. Am. Acad. Arts and Sci., 20: 144. In GARMAN, S. 1889. Cave animals from southwestern Missouri. Bull. Mus. Comp. Zool. 17 (6): 237 No. 1 _a---------------------.---- 1898. Observations on the Astacidae in the United States National Museum and in the Museum of Compara- tive Zoology, with descriptions of new species. Proc. U. S. Nat. Mus. 20: 643-698. GIRARD, CHARLES 1852. A revision of the North American Astaci. Proc. Acad. Nat. Sci. Phila., 6: 88. Hosss, H. H., Jr. 1948. A new crayfish of the genus Cambarus from Texas with notes on the distribution of Cambarus New Species of Cambarus 5 fodiens (Cottle). 983224, Proc. U. S. Nat. Mus., . and M. S. BEDINGER 1964. A new troglobitic crayfish of the genus Cambarus (Decapoda, Astacidae) from Arkansas with a note on the range of Cambarus cryptodytes Hobbs. Proc. Biol. Soc. Wash., 77: 9. REIMER, ROLLIN [MS] The Crayfish of Ar- kansas. Unpublished M. S. dissertation, submitted 1963, University of Arkansas. March 17, 1966 SPEOCIROLANA THERMYDRONIS, A NEW SPECIES OF CIROLANID ISOPOD CRUSTACEAN FROM CENTRAL COAHUILA, MEXICO GERALD A. COLE and W.-L. MINCKLEY, Department of Zoology, Arizona State University Tempe, ABSTRACT Speocirolana thermydronis is describ- ed from a single specimen taken in thermal waters of an isolated bolson in northern Mexico. The type locality is high in the Sierra Madre Oriental, sug- gesting that the genus Speocirolana may be of pre-Tertiary origin. The species occurs in an area of high en- demism where special habitats afford- ed by springs permit an ancient fauna to persist. Extens ve spring-fed marshes, lakes and streams occur in the intermontane basin sur- rounding the town of Cuatro Cienegas in central Coahuila, México. This bolson, lo- cated in the structural axis of the Sierra Madre Oriental, has long been isolated, and evidence exists for a number of faunal vasions, resulting in marked endemism in aquatic, and to a lesser extent in terrestrial, organisms (Webb, ef al/., 1963; Hubbs and Miller, 1965). In April 1964, a field party from the University of Colorado Museum visited the area. Their collections included a single cirolanid isopod, which is here de- scribed as a new species. We are prompted to describe this form on the basis of a single specimen for a number of reasons. First, extensive field work in the basin since 1958, including three expeditions since 1964 with unsuccessful searching for additional isopod material, indicates the rarity (or difficulty in finding) more speci- mens of this species. Second, the area of original discovery has been severely modi- fied and may soon be destroyed. Third, de- scription of the isopod may stimulate addi- tional work in the largely-unexplored_ bol- Arizona son region of northern México. And, fourth, its description further emphasizes — the unique, endemic nature of the biota of that region. Speocirolana thermy dronis, n. sp. Figs. 1-21 The new species is assigned to Speociro- lana Bolivar (1950), a name originally pro- posed as a subgenus of Czrolana Leach, but elevated to generic rank by Bowman (1964). Generic characters of the new species agree with those given by Bolivar, with minor exceptions. The major character of the genus is the first three pairs of pereopods prehensile and pereopods 4-7 ambulatory. These features are otherwise unknown in North American troglobitic cirolanids, ex- cept in the poorly-known Conzlera stygia Packard (1900) from near Monterrey, Nuevo Leon, México. The last form, when rediscovered, may prove to be a species of Speoctrolana. Type Material and Etymology—vThe type material of Speocirolana thermydronis con- sists of a single female specimen, measuring 15 mm from the tip of the head to the end of the telson. It was collected on 12 April 1964, in the complex habitat associated with Pozos de la Becerra, a large warm spring with its source lying 9.8 miles (13.7 km) south-southwest of Cuatro Cae The specimen was collected by Mary L. Alle- sio of the University of Colorado, and was referred to us for study by Clarence J. McCoy, now of the Carnegie Museum. The specimen now is housed in the United EDITORIAL COMMITTEE FOR THIS PAPER: THOMAS E. BOWMAN, Associate Curator, Division of Marine Invertebrates, Smith- sonian Institute, Washington, D. C. MILTON A. MILLER, Professor of Zoology, University of California, Davis, California Ly 18 Tulane Studies in Zoology States National Museum, catalog number 113054. The name is derived from the Greek “thermydron,’ a warm spring, and alludes to the habitat of the animal in thermal waters of the Cuatro Cienegas basin. Diagnosis —Characters that separate Speo- cirolana thermydronis from S. pelaezi (Bolt- var) and from S. bolivari (Rioja) are as follows: clypeus of head with posterior ends rounded, shaped in gentle bow, and without deep notch anteriorly; labrum with mandibular margin almost straight; first an- tennal flagellum with 14 articles, with length of peduncle divided by length of flagellum 0.9; second antenna with 35 flagellar ar- ticles, extending posteriorly to seventh pe- reon segment when deflected; palp relative- ly short; lacinia of maxilliped paddle-shaped, bearing 12 plumose setae; pleopods with all endopods lacking terminal setae; inner sur- faces of endopods of uropods spineless; and, telson shaped as a broad shield, terminating in acute point. Description and Compartsons.—Our com- parisons of Speocirolana thermydronis with S. pelaezi and S. bolivari are based on illus- strations and data given by Bolivar (1950) and by Rioja (1953); some additional data were graciously provided by Thomas E. Bowman. All three forms are eyeless and unpigmented except for brown masticatory surfaces on the mandibles and darkened tips on the claws of the pereopods. The ratio of body length to greatest body width is 3.5 for thermydronis and near 2.7 for the other two species. The head of thermydronis 1s sub-pentagonal in shape, rounded anteriorly (Fig. 1); bolivari has a similar head, while that of pelaezi is broader and shorter. The inferior frontal lamina of the head of thermydronis is similar to that of pelaezt in projecting forward between the antennal bases to a rounded terminus (Fig. 3). The lamina of bolivari is long and pointed. The clypeus and the labrum of thermydronts are distinctive (Fig. 3). The former is bow- shaped, gently rounded to its smooth pos- terior terminations, and the latter has an almost-straight mandibular margin. The first pereonite lacks epimera in all three forms. Epimera also appears absent from the second and third pereonite of thermydronts (Fig. 1); they probably were lost in preservation or in preparation of the Vol. 13 specimen for study, are not visible from dorsal view, or are fused to the segments. Epimera on pereonites 4-7 are well devel- oped in all three species, with strong, acute, postero-lateral angles. Five segments are visible in the pleon of all three species anterior to the pleotelson. The first two segments are sub-equal in length and twice as long as the third seg- ment. The fourth and fifth visible segments are abruptly shorter and narrower than the first three. Pleon segments 1-3 have acute postero-lateral angles. The telson of thermydronis is slightly longer than broad and ends in an acute pos- terior tip (Figs. 1, 7). In pelaezi the telson is broadly rounded and that of bolivari 1s truncate. The first antenna is about the length of the peduncle of the second antenna in thermydronis (Figs. 1, 14) and in pelaezz; it appears shorter than that of bolivari. The first antenna has three peduncular joints in all three species, but in thermydronis it has 14 flagellar articles (Fig. 14) as opposed to about 20 articles in pelaezt and 22-28 in bolivari. The ratio of length of the antennal peduncle to length of the flagellum is about 0.9 in thermydronis. In pelaezt and bolivari this ratio is nearly 1.3. S. thermydronts re- sembles pelaezt in having the terminal setae of the first antenna relatively uniform in length; bolivart bears a number of short setae on the terminal flagellar article and a single elongate seta. The distal flagellar ar- ticles of all three species are invested with aesthetases. The second antenna of thermydronis is elongate, reaching back to the seventh pere- onite when deflected (Fig. 1). This condi- tion prevails also in bolzvarz, but the second antenna of pelaezi reaches only to the fifth peronite. There are 35 articles in the flagel- lum of thermydronis, 48-52 in bolivari, and 30 in pelaezi. The terminal articles of the flagellum in thermydronis (Fig. 10) and bolivari are elongate, whereas those of pelaezt are shortened. The antennal ped- uncle of thermydronis comprises five clearly delimited joints. In S. thermydronis the mandibles (Fig. 6) are asymmetrcial, with the right incisive process overlapping the left ventrally. The lacinia mobilis of the mandible is sub- triangular, bearing on its margin about 34 short, cone-shaped teeth. The second seg- Vol. 1 ment of the mandibular palp in thermy- dronis is invested with setae on its distal two-thirds (Fig. 2); in the other two species setae occur on the distal half to two-fifths of this segment. The first and second maxillae are similar in all three species. The outer lamina of the first maxilla bears 10 toothed distal spines (Fig. 5). The inner lamina has three sparsely-plumose spines and two setae (one New Cirolantd Isopod 19 long, one short). The second maxilla (Fig. 4) has three conspicuous endites. The outer two are provided with large, incurved spines on their distal margins. The inner endite has numerous setae of differing sizes, some of which are plumose. The maxilliped of thermydronis (Fig. 9) differs from that of bolivari and pelaezi. The four distal articles of the palp are armed on their inner surfaces with strong. setae. Figures 1-9. Speocircolana thermydronis, female holotype. 1. dorsal aspect; 2. mandib- ular palp; 3. inferior frontal lamina, labrum, clypeus; 4. second maxilla; 5. first maxilla; 6. mandible, lacina mobilis, pars molaris; 7. telson; 8. left uropod, ventral aspect; 9. maxilliped; 9a. lacinia of maxilliped. 20 Tulane Studies in Zoology There are stout spines on the distal outer corners of the second and third articles of the palp segments, but only fine, hair-like setae occur on the lateral margins of all but the terminal article (which is naked). The outer surface of the fourth joint of the palp in bolivari and pelaezi is setose. The lacinia (Fig. 9a) is ovoid, with 12 plumose setae and three coupling hooks. The pereopods (Figs. 11-13, 15, 16) are dimorphic in all three species. The first three pairs are prehensile and sub-cheliform, perhaps more so in ¢hermydronis than in the others. The last four pairs are ambula- tory, without obvious morphological spe- cialization. Pereopods 1-3 in thermydronts have the palmar margins of the propodus armed with two, three, and four stout spines, respectively; however, this character seems variable, in pelaezi at least, and may be of little worth. Pereopods 4-7 become progress- ively longer from front to back in all three forms. The first pleopod of thermydrontis is nar- rower than the succeeding ones (Fig. 17). The expodites of all the pleopods have distal, plumose setae (Figs. 17-21); these are sparsely developed on the fourth. The third exopod has an incomplete transverse suture, and the fourth and fifth exopods have complete sutures. All endopodites lack distal setae; thus Bowman (1964) may have erred in implying that setose first and_sec- ond pleopodal endopods and the lack of setation on the remainder characterizes the genus Speocirolana. In females of bolzvart, at least the endopods of the first pple lack setae (fide, Rioja, 1953: lam. 3, Fig. 31). Rioja’s illustration (lam. 3, Fig 30) of the first pleopod of a male of bolivari shows profuse setation on the endopod. Bolivar (1950: Fig. 9), on the other hand, shows setation on the endopod of the first pleopod of a female of pelaezz. The char- acter needs further study. The uropodal base of thermydronis has setae On its inner, proximal margin (Fig. 8). The distal part of the inner margin is pro- longed, and bears an apical spine that reaches to the middle of the endopod and almost to the end of the telson. In pelaezi this spine etxends past the end of the telson. In bolivari, it extends only about half the length of that structure. The inner surface of the endopodite of thermydronis bears a few scattered, short setae; in pelaezi there Vol. 13 are three stout spines on this surface and bolivari has five such spines. The uropodal exopods and endopods of all three forms have their apices armed with short terminal setae. Type Locality—Prior to December 1964, Pozos de la Becerra was one of the largest and most complex aquatic habitats in the Cuatro Cienegas basin. The laguna was elongate and irregular in shape, with depths ranging to more than 10 m at the largest spring inflows. In areas of inflow the bot- toms were of gravel and rubble. Other areas had bottoms of deep calcareous silt; most silt bottoms were covered by dense beds of waterlily (Nymphaea). The laguna originally measured about 25 m in width at its narrowest place, ranging to more than 150 m wide, and was perhaps two km long. Water was always extremely clear. Temper- atures taken in the sources on seven differ- ent occasions ranged from 29.4 to 32.2° C. Water levels did not change perceptibly in the period 1960-64, and one estimate of dis- charge was about 1.34 m*/second at the outlet channel. The laguna suffered some modification in 1961 through construction of a bathing facility. In 1964, however, the laguna was drastically modified by construction of a canal, and the water level fell 46 cm in about two days. In April 1965, the down- cutting of the uncontrolled canal outlet had apparently stabilized, with the laguna sur- face lowered more than a meter. This re- sulted in drainage of extensive marshes that were associated with the spring, and reduced the over-all surface area of water and marsh from perhaps 10 km? to less than 0.2 km*. Swimmers had muddied the laguna in April 1965, and the silty bottoms had been greatly disturbed. Many formerly gravel bottoms were silted and most of the Nymphaea beds were dried or uprooted. Only the inflows of the largest springs remained clear of silt. Origin of Speocirolana thermydronis.— The troglobitic cirolanid isopods of the Western Hemisphere, with the exception of Antrolana lira Bowman (1964), occur in an arc surrounding the Gulf of México (Rioja, 1953; Bowman, 1964). A. lira lives in the Appalachian Valley of Virginia and does not enter directly into the problem of origin for S. thermydronis. The distribution of the species of Speocirolana in Cuatro Cie- negas and in the Valles-Mante area of San Vol. 1 Luis Potosi and Tamaulipas, and the poorly- known Conilera stygia (also with the first three pairs of pleopods prehensile), from Monterrey, Nuevo Leon, form a compact triangle in the northeast of México. These, and Crrolanides texensis Benedict from the New Czirolanid lsopod 21 San Marcos area of Texas, all are in the area inundated by the sea that filled the mid- Cretaceous Mexican Geosyncline (Maldondo- Koerdell, 1964). S. bolivari, 8. pelaezt, and C. stygia all occur near the edge of the Gulf Coastal Plain, slightly inland from areas of Figures 10-21. Speocirolana thermydronis, female holotype. first pereopod; 12. second pereopod; 13. third pereopod; 14. first antenna; 15. fourth periopod; 16. seventh pereopod; 17. first pleopod; 18. second pleopod; 19. third pleopod; 20. fourth pleopod; 21. fifth pleopod. 10. second antenna; 11. i) i) mid-Tertiary inundations (West, 1964). The transgressions by the sea in Oligocene may have affected the last three species, but undoubtedly excluded the Sierra Madre Ori- ental Axis in which S. thermydronis occurs. The Cuatro Cienegas basin, in addition to a number of endemic vertebrates ( Hubbs and Miller, 1965), holds a unique molluscan fauna. This includes a number of genera and species, yet to be described, that show few relationships to other living forms of México, or elsewhere (Dwight W. Taylor, pers. comm.), and therefore indicate a great age for aquatic habitats of the area. Ciro- lanid isopods in freshwater cave habitats are generally thought to be derived from popu- lations of marine forms that are relicted by marine regressions (Bowman, 1964). The discovery of S. thermydronis high in the Sierra Madre Oriental may indicate a pre- Tertiary origin of the genus; as part of an ancient plateau fauna it has been able to persist in the special habitats afforded by the springs. It seems doubtful that distributions of epigean animals, such as fishes, will shed much light on the origins of S. thermydronts. However, it is worth noting that Miller and Minckley (1963) found the endemic platy- fish of the Cuatro Cienegas basin (NXzpho- phorus gordont Miller and Minckley) shar- ing many characters with X. vartatus xiphid- mm (Gordon) of the Rio Soto la Marina system. They suggested that an overland dispersal of the aquatic animals might have occurred, utilizing stream captures, from southeast, to the Cuatro Cienegas area. This is substantiated by the presence of Gam- busta marshi Minckley and Craddock in the Rio Salado system (the stream that now drains the Cuatro Cienegas basin), a species whose closest relatives also are in the Rio Soto. Ja Marina-Rio Panuco complex (Minckley, 1962). The presence of a new cirolanid in the Cuatro Cienegas basin, a specialized, caverni- colous catfish (Priatella phreatophila Car- ranza, 1954) at Muzquiz, Coahuila, north of Cuatro Cienegas, and the relative wealth of cavernicolous animals in the limestone Ed- ward’s Plateau Region of Texas, points out a need for further exploration in that area for troglobitic organisms. Until extensive surveys are made and additional collections are obtained little can be done in synthesis of the over-all fauna of the region. We Tulane Studies in Zoology Vols defer speculation on the intra-generic rela- tionships of S. thermydronzs until additional material can be obtained. Acknowledgments—We thank Clarence J. McCoy of the Carnegie Museum for bring- ing the specimen of S. thermydronts to our attention and for permission to describe it. Work in the Cuatro Cienegas basin has been greatly facilitated by the able assistance of José Lugo, Jr. of Cuatro Cienegas, and by a Grant, GB-2461, from the National Science Foundation to Minckley. This work also was supported by N.S.F. Grant GB-154 to Cole. LITERATURE CITED BOLIVAR Y PIELTAIN, C. 1950. Estudio de una Cirolana cavernicola nueva de la re- gion de Valles, San Luis Potosi, México. Ciencia, 10 (11-12) : 211-218. BowMaN, T. E. 1964. Antrolana lira, a new genus and species of troglobitic cirolanid isopod from Madison Cave, Virginia. nt. J. Speleology 1 (1-2) : 229-236, pls. 50-57. CARRANZA, J. 1954. Descripcion del primer bagre anoftalmo y depigmentodo encou- trado en aquas Méxicanas. Ciencia 14 (7-8) : 129-186, lam. 1. HusBgs, C. L., and R. R. MILLER. 1965. Studies of cyprinodont fishes, XXII. Vari- ation in Lucania parva, its establishment in Western United States, and descrip- tion of a new species from an isolated basin in Coahuila, México. Misc. Publ. Mus. Zool., Univ. Mich. 127: 1-111, pls. 1eSe MALDONADO-KOERDELL, M. 1964. Geohistory and paleogeography of Middle America. In, Handbook of Middle American Indi- ans, Univ. Texas Press, Austin. Pp. 3-32. MILLER, R. R., and W. L. MINCKLEY. 1963. Niphophorus gordoni, a new species of platyfish from Coahuila, México. Copeia 1963 (3): 5388-546. MINCKLEY, W. L. 1962. Two new fishes of the genus Gambusia (Poeciliidae) from northern México. Copeia 1962 (2): 391- 396. PacKkArp, A. A. 1900. A new eyeless isopod crustacean from México. Proc. Amer. As- soc. Adv. Sct. 49: 228. RrogA, E. 19538. Estudios carcinologicos. XXX. Observaciones sobre los cirolanidos cavernicolos de México (crustaceons, iso- podos). Ann. Inst. Biol., Mexico 41 (1) 141-170, lam. 1-6. WEBB, R. G., W. L. MINCKLEy, and J. E. CRADDOCK. 1963. Remarks on the Coahu- ilan box turtle, Terrapene coahuila (Tes- tudines, Emydidae). Southwestern Nat. 8 (2): 89-99. West, R. G. 1964. Surface configuration and associated geology of Middle Ameri- ca. In, Handbook of Middle American Indians, Univ. Texas Press, Austin. Pp. 33-83, March 17, 1966 SOCIAL ORGANIZATION OF THE SOUTH AMERICAN MONKEY, CALLICEBUS MOLOCH: A PRELIMINARY REPORT WILLIAM A. MASON, Delta Regional Primate Research Center Covington, Louisiana ABSTRACT Callicebus moloch ornatus, a small diurnal South American monkey, was the subject of an 11-month field study. Groups usually consist of an adult pair and one or more young, and each group occupies a definite and fixed territory. Adjacent groups often meet, usually in the early morning, in specific locations at the boundaries of their territories. As opposing pairs approach, each ani- mal draws closer to its mate. Male and female then sit with sides touching, fac- ing the opposing pair a few meters away. There follows an elaborate vocal exchange in which all animals partici- pate. Calling is accompanied by tail- lashing, arching, piloerection, and short rushes which may develop into a chase. One animal rarely overtakes another during a chase and even when this oc- curs, fighting is not severe. Mateships appear to be stable, although animals will occasionally copulate with members of adjacent groups, especially during the seasonal peak of sexual activity. Ar- rival of young had no obvious effect on the pair bond. In one group in which reliable identification of sexes was pos- sible, the male carried the infant at vir- tually all times, except when it was being nursed. I. INTRODUCTION Few New World primates have been studied intensively from a naturalistic point of view and little systematic information is available on the forms and yarieties of social organization in these monkeys. Of the more than 15* genera of platyrrhine monkeys, only one, the howler monkey (Alowatta) has been the subject of a major field investiga- tion (Carpenter, 1934). Groups of howler * Authorities differ somewhat in number of genera assigned to platyrrhine monkeys. EDITORIAL COMMITTEE FOR THIS PAPER: monkeys on Barro Colorado Island in Pana- ma averaged about 18 animals, with an upper limit of 35. Observations of red spider monkeys (Afeles), indicate that they are frequently found in small subgroups, but join with others in larger groupings of as many as 40 individuals (Carpenter, 1935). More limited observations on Cebus, Lago- thrix, and Saimiri suggest a similar tendency in these primates toward the formation of large groups (Bates, 1944; Fooden, 1963; Kuhlhorn, 1939). All New World monkeys, however, do not conform to this pattern. Moynihan (1964), for example, notes that Panamanian night monkeys (Aotws trivirgatus) are most often found in pairs or in small family units, and the same characteristic is suggested for marmosets (Hampton, 1964; Stellar, 1960). However, there have been no_ systematic field studies of any South American primate which displays a tendency toward a small, “family unit” type of social organization. Such a tendency was strongly evident in Callicebus moloch, the subject of the present report, and was a major consideration in the selection of this species for study in the field. Il. RANGE AND PHYSICAL APPEARANCE The range of Amazonian and Orinocoan species of Callicebus extends north from the upper Rio Paraguay basin in Mato Grosso, Brazil and Paraguay into Colombia, Vene- zuela, and extreme northwestern Brazil; and it extends west from Rio Tocantins in the state of Para, Brazil into parts of Bolivia, Peru, Ecuador, and to the base of the Andes in Colombia (Hershkovitz, 1963). Callicebus is a small monkey, comparable C. R. CARPENTER, Research Professor of Psychology and Anthropology, Pennsylvania State University, University Park, Pennsylvania MARTIN H. MoyYNIHAN, Director Smithsonian Institution Canal Zone Biological Area, Balboa, Canal Zone ie) Uo 24 Tulane Studies in Zoology the night monkey, but strongly diurnal in habit. A Colombian form, Callicebus moloch ornatus (Hersh- kovitz, 1963) is the subject of this report. This subspecies is distinguished by deep auburn fur covering the throat, chest and forearms, and by a striking white band across the forehead which contrasts sharply with the dark face and chestnut red crown. The appealing and colorful aspect of these ani- mals is enhanced in some individuals by white gloves and ear tufts. Males and fe- males are similar in size and coloration and it is only after close association with specific groups that reliable identification of sexes is sometimes possible. in size to Aotus, Ill. Srupy SITE The study site was in the Ilanos or plains area of eastern Colombia, near the town of San Martin. The region consists of natural grassland and additional thousands of acres of artificial pastures which have been cleared in recent years. In spite of this, a substan- tial number of forested areas remain as nar- row galleries along the rivers and in the form of groves surrounded by savannah. Many of these tracts contain stable popula- tions of Callicebus and other monkeys (Alow- atta, Aotus, Cebus, Saimtrt). Three groves, ranging in size from ap- proximately 3 to 17 acres, were selected for study. At one time all were part of the same large tract, but they were isolated from each other around 1950 when large sections of Vol. 13 the forest were cleared. The study areas were left as a water Conservation measure, and the monkey population had apparently been undisturbed for more than 14 years. The largest of the three forests, Socay, was studied intensively from May 1964 through March 1965. The only resident monkeys in this forest were Callicebus. The present re- port is a preliminary description of the so- cial organization and behavior of these monkeys. IV. POPULATION CHARACTERISTICS The first census of Socay Forest, com- pleted on August 2, 1964, indicated that there were 28 monkeys living in nine small groups of from two to four animals each. In the three- and four-animal groups it was clear in most cases that only two of the ani- mals were fully mature. In December 1964 the first birth was noted and five additional animals were born in the next three months. However, owing to the presumed death of one infant and the disappearance of two subadult animals, the total population had increased to only 31 by March 1965. No new groups were formed, but as the result of births, mean group size rose from 3.1 to 3.4. The results of the first census and the final census are presented in Table 1. V. GROUP ACTIVITIES AND INTERGROUP RELATIONS Throughout the study period each group occupied a definite and fixed area which TABLE I May-August 1964 Group First Count Composition 1 5/23/64 2A,J, (Iz) * 2 5/27/64 2A,J,I2 3 5/25/64 2A,I: 4 6/25/64 2A,5 5 7/ 3/64 2A,J 6 6/25/64 2A a 8/ 2/64 2A,137,1J5? 8 5/25/64 2A,I, 9 6/27/64 2A,I; mean/gerp = adult Population and Groun Composition, Socay Forest March 1965 Total Composition Total 3 2A,I 3 4 2A,2J 4 3 2A,J 3 3 2A,J.(1*) 3 3 2A,J,1 4 2 2A,I 3 4 2A,J,I 4 3 2A,J 3 3 Paleo fr 4 28 31 31 3.4 late infant, no longer carried, considered J in March census INGE J = juvenile or subadult I. * = died or presumed dead J = presumed to have left forest ? = age class uncertain Vol. 1 Figure 1. first census, August 2, 1964. A = Monkey Social Organization GRP 1 2A, J, (Ip) Location of groups in Socay Forest. eS) VN N SOCAY FOREST 9-64 Group composition at completion of adult, J = juvenile or subadult, I. = late infant, no longer carried. Age class of animals in group 7 was uncertain. contained its food and lodge trees. The ap- proximate location of groups is shown in Figure 1. The nine groups making up the population of Socay Forest constitute a com- munity in which each group has frequent contact with its immediate neighbors. Fur- thermore, even those groups which are sel- dom if ever in visual contact may communi- cate directly by means of the elaborate vocalizations which play such a prominent part in the social life of Callicebus. The most interesting social relations, however, are the face-to-face encounters between neighboring groups. Activity begins at the first hint of dawn. Usually, the initial signs of life are faint stirrings in the lodge tree, the sounds of urination and defecation, followed by one or two moaning vocalizations. Frequently, these lead into a longer series of calls in which both adult and immature animals may participate. The call is elaborate and the arrangement of its several components varies from one occasion to the next (Moy- nihan, in press). Calling often begins with a series of moaning u-ah notes which in- crease rapidly in tempo and intensity. At the end of each note there is a sharp chuck or squeal. This form of calling may con- tinue for a minute or more and is often terminated by a series of rather slow, em- phatic O-O-O-O sounds. All vocalizations may cease at that point or the entire se- quence may be repeated. In anywhere from a few minutes to half an hour after the first call, the group leaves the lodge tree and be- gins to feed on fruits and berries, which form the bulk of the Callicebus diet. Groups from neighboring areas frequent- ly meet as they move toward food trees. These are not chance encounters, nor do they seem to be merely the result of attrac- tion to a common food source. The animals actively converge toward each other and usually meet in the same areas at the bound- aries of their territories day after day. As 26 Tulane Studies in Zoology the opposing pairs approach, each animal draws closer to its mate. Characteristically, male and female sit with sides touching and face the opposing pair in an adjacent tree. One of the animals (male?) begins to make the same low-pitched moans that were heard as the day began. As he vocalizes he presses against his mate and often looks toward her or touches her lightly with his hand; she, together with the members of the opposing pair, may join with him in a sustained call. During vocalization both animals stand and face the opposing pair. Often they seem to swell to half again their normal size as the result of arching their backs, stiffening or bowing of the arms, and piloerection. The animals shake vigorously with the effort of calling and the impression of tension and agitation is enhanced by the rhythmic tail- lashing that often accompanies the call. The vocal interchange may continue with- out pause for five minutes or longer. At times only one monkey or one pair will call; often, however, all animals call in concert. The volume of sound is unexpected in view of the relatively small size of the animals; on a quiet day one can hear the sounds from a mile away. While they call, the opposing pairs often edge closer together; individual animals leap forward or up a few feet, then suddenly, with no apparent forewarning, one monkey dashes toward the opposing pair. It may withdraw at once whether there is a counter- rush or not, or the rush may develop into an extended chase in which the males are prob- ably most often involved. During a chase the customary pathways are abandoned and the monkeys race across the tops of the trees, make perilous crossings, descend to a few feet from the ground, and sometimes fall or leap to the forest floor where pursuit con- tinues until the fleeing animal scampers up a vine or slender tree. Rarely is an animal caught, and even when this happens the con- sequences are not severe. There is no ex- tended fight; the pursuer pushes and slaps at its victim, may bite him once or twice, there are a few squeals and it is over. The chase is not an invariable climax to the meeting between groups. At times one or more of the participants seem to lose inter- est and begin to feed, or simply sit quietly. Sometimes both members of a pair sit side by side with their tails twined. This pattern is more often seen, however, during rest Vol. 13 periods and is especially common in the evening when the animals settle down for the night. Another response which is sometimes as- sociated with the interchanges between groups is chest rubbing. Callicebus has a small patch of glandular tissue over the chest which is probably implicated in this re- sponse. The animal grasps a branch with its hands and draws or pushes itself slowly forward dragging the chest along the sur- face. Often it pauses at the end of a stroke and sniffs or mouths the area it has just rubbed. Frequently, the chest is rubbed with the hands alternately in slow, down- ward strokes. Inasmuch as chest rubbing seems to occur most frequently after an en- counter between groups, it might be as- sumed to constitute a form of territorial marking. The response does not always oc- cur in boundary areas, however, and no ani- mal has been observed to react to a limb rubbed by another. The entire episode from the first meeting of the groups until they resume feeding may occupy less than five minutes to as much as half an hour or more. After the early- morning face-to-face encounters have ended and the animals have returned to their re- spective areas, there sometimes occurs a re- markable vocal pattern. This is a short call, sounding from a distance very much like the gobbling of a turkey. Its most interest- ing feature, however, is not its sound quality, but the fact that it is given almost simul- taneously by several groups. Typically, one animal or a single group begins the call and within a fraction of a second it is taken up by other groups and passes rapidly through the forest as a kind of chain reaction. Each burst of calling ends abruptly in less than 10 seconds, but the call is usually repeated at frequent intervals for five minutes or more. Preceding and following each call there is a series of brief, high frequency, whistling notes. The early-morning encounters and vocal chain reactions are similar to the territorial disputes that have been so carefully described for birds. But factors other than the defense or maintenance of territory are definitely involved, and this is particularly evident when the female is sexually receptive. Under these circumstances instead of fleeing from the onrush of the opposing male, as she usually does, she may sit quietly as he ap- Vol. 1 proaches, or even approach him. Mutual genital inspection may follow or they may copulate without preliminaries—or at least attempt to do so before her mate counter- charges and interrupts the act. At the same time that the female is showing obvious interest in neighboring males she may be indifferent to the sexual advances of her own mate. During a long episode I ob- served, the male repeatedly solicited the fe- male. He reached toward and manipulated her genitals, sniffed or licked them, and fre- quently attempted to mount. In spite of his persistence, the female did not accept him. She fled twice, but each time was overtaken before she had joined the male of a neigh- boring group. On this and other occasions the male was observed to place himself be- tween his mate and the intruding male, or to restrain her as she attempted to move away from him. Although it is possible that a female may remain with a neighboring male indefinite- ly, it seems more likely that such associations are brief. In one instance a female stayed with the male of a neighboring group for several hours before returning to her mate. This was the longest liaison of this type | observed. More often, extra-pair associations last only a few minutes, as in the following episode: A male and female of neighboring pairs approached each other near the bound- aries of their territories, copulated, then separated, the male moving off while the female returned to her customary partner. Although her mate had twice rushed the pair during copulation, he showed no unusual re- action to the female upon her return. On other occasions, however, mild aggression has been observed. VI. THE PAIR BOND In spite of the sexual attraction of both male and female to opposite-sexed members of neighboring groups, there are many indi- cations that the bond between mates is strong and enduring. Although most ani- mals could not be identified individually, for those pairs in which it was possible, part- ners remained together from the first con- tacts until observations terminated as much as 10 months later. Ordinarily, the members of a pair follow each other closely and the animals are usually found within a few meters of each other during feeding, travel- ling, and resting periods. Generally speak- Monkey Social Organization 27 ing, the attraction 1s mutual. Either sex may follow the other and leadership changes fre- quently throughout the day. Often when one animal has finished feeding it will sit at the edge of the food tree and wait until Its mate approaches before moving on. Grooming, nuzzling, gentle grasping, and sitting for long periods of time with sides pressed together and tails twined, provide further evidence of the breadth of social ties. When the animals become separated and lose contact there are various signs of dis- tress, particularly whining vocalizations which, if ineffective, may lead to more elaborate calls. The arrival of infants has provided addi- tional information on the relationship be- tween mates. In a single pair which I was able to observe intensively and at close range it was clear that the male almost al- ways carried the infant except when it was being cleaned or nursed by the mother. At times the transfer seemed to be accomplished by the infant while the adults sat in contact. On many occasions, however, I saw the mother remove the infant from the male’s shoulders, lick its genitals (which stimulates urination), and place it on her ventral sur- face. When nursing was completed the in- fant climbed to the mother’s shoulders from which it moved to the male or was removed by him. In groups of three and four mon- keys, I suspect that immature animals also carry the infant occasionally, but the burden probably devolves chiefly upon the adult male. VII. FACTORS REGULATING GROUP SIZE Socay Forest is a relatively small area and most of the available space is utilized by the present occupants. One might ask how this spatial limitation will affect social or- ganization as the population becomes more numerous. One possibility is that group size will increase. This may have occurred in the two smaller forests which were included in the study. One of these contained five Cal- licebus monkeys and a single male Sa:muzrz; the other contained at least five and possibly six Callicebus monkeys. In both forests the animals were organized into a single group. Animals were sometimes scattered through- out the forest during feeding, but moved to- gether on some occasions and appeared to occupy the same lodge tree at night. It seems likely, however, that five or six ant- 28 Tulane Studies in Zoology mals is approaching the upper limit for Cal- licebus groups, even though the factors that determine this limit are imperfectly under- stood. Pressure by mature adult on subadult and young adult animals probably is im- portant. In at least three groups in Socay Forest there were definite indications that young adult or near-adult animals were the occasional targets of adult aggression. In one of these groups (Group I) a young male disappeared from the parental group several months before the end of the study period and was not seen again. This oc- curred well in advance of the birth of an infant into the group. In another case (Group 7) a young adult remained with its group, which included a sibling, until the birth of an infant, then disappeared. In a third group of four animals (Group 2) a young adult was the object of sporadic adult aggression extending over a period of sev- eral months, but was still closely associated with the group when the study ended. These observations suggest that there is no single factor, such as the arrival of infants, which plays a critical part in forcing young adults from the natal group, but rather that the process of separation occurs slowly, over a period of several months. There is no indication of what has be- come of the animals in Socay Forest which have left the natal group. It seems unlikely that they could have remained in the forest undetected. One possibility is that these animals have moved across the savannah to other, less crowded, forests. Although there is no direct evidence for this hypothesis, I have on several occasions seen Callicebus for brief periods on the forest floor and local residents claim that migration of individual animals across the savannah does occur. VIII. ACKNOWLEDGMENTS This work was initiated while the author held a National Institutes of Health Special Fellowship. Support was provided by Na- tional Institutes of Health Grants FRO0O164, GM11328 and TW00143. I am grateful to Dr. J. C. S. Paterson, Director of the ICMRT Vol 13 program in Cali, Colombia, and to Dr. J. H. Esslinger, Dr. A. D'Alessandro, and other members of the ICMRT staff in Cali for assistance and support. Dr. Carlos Lehman kindly suggested the study site and was help- ful in many other ways. I am indebted to Mr. and Mrs N. Ponomoreff, former prc- prietors of Hacienda Barbascal, for their hos- pitality and many kindnesses, and to mem- bers of the Botero family, present owners of Barbascal. I also wish to thank Dr. M. Moy- nihan for valuable discussions of Callicebus behavior during his visit to the study site, and for the opportunity to read his unpub- lished manuscript “Communication in Cal- licebus”’. IX. REFERENCES CITED Bates, M. 1944. Saimiri monkey as an ex- perimental host for the virus of yellow fever. Am. J. Tron. Med. 42: 83-9. CARPENTER, C. R. 1934. A field study of the behavicr and social relations of howling monkeys. Comp. Psychol. Monogr. 10: 1- 168. : 1935. Behavior of red spider monkeys in Panama. J. Mammal. 16: 171-80. FOoopEN, J. 1963. A revision of the woolly monkeys (genus Lagothrix). J. Mammal. 44 :213-47. HAMPTON, J. K., Jr. 1964. Laboratory re- quirements and observations of Oedipo- midas oedipus. Amer. J. Phys. Anthropol. 22: 239-43. HERSHKOVITZ, P. 1963. A systematic and zoogeographic account of the monkeys of the genus Callicebus (Cebidae) of the Amazon and Orinoco River basins. Mam- malia 27: 1-79. KUHLHORN, F. 1939. Beobachtungen uber das Verhalten von Kapuzineraffen in frei- er Wildbahn. Z.f.Tierpsychol. 3: 147-51. (cited in Hill, W.C.O. Comparative anat- omy and taxonomy of primates, vol. 5, New York: Interscience, 1962). MOYNIHAN, M. 1964. Some behavior pat- terns of platyrrhine monkeys I. The night monkey (Aotus trivirgatus). Smithso- nian Mise. Coll. 146: 1-84. . Communication in Calli- cebus. In press. STELLAR, E. 1960. The marmoset as a lab- cratory animal: maintenance, general ob- servations of behavior, and simple learn- ing. J. Comp. Physiol. Psychol. 53: 1-10. March 17, 1966 A COMPARATIVE BIOSYSTEMATIC STUDY OF FUNDULUS NOTATUS AND FUNDULUS OLIVACEUS (PISCES: CYPRINODONTIDAE ) JAMIE E. THOMERSON,! Department of Biology, Tulane University New Orleans, Louisiana ABSTRACT Western populations of Fundulus no- tatus (Guadalupe, Colorado, Brazos, San Jacinto, Trinity, Neches and Sa- bine rivers) tend to have more dorsal (average 9.9) and anal (average 12.8) rays. Eastern populations (Mississippi Valley, Lake Pontchartrain drainage, Pearl River and Tombigbee River) tend to have fewer dorsal and anal rays (average 9.2 and 12.1 respectively). Trends in caudal ray number are roughly parallel. Lateral-scale number is generally 34 or 35. Northern fish tend to have broader heads and shorter snouts than southern fish. Dorsolateral spots, if present, are irregular, or dif- fuse, or both. The westernmost population of Fun- dulus olivaceus (Navasota River, Tex- as) has fewer dorsal and anal rays (9.4 and 12.0 respectively). Popula- tions from the Trinity River east to the Mississippi Gulf drainage average 9.8 dorsal rays. Dorsal ray number de- creases clinally to the east to a low average of 8.9 for Choctawhatchee Bay drainage samples. Trinity and Neches river populations average 12.5 anal rays. Anal ray number decreases clin- ally to the east to a low average of 11.2 for Choctawhatchee Bay drainage sam- ples. Caudal ray number tends to be high from the Alabama River west, and low to the east. But average lateral- scale number is greatest (35.7) for the Chattahoochee sample. Northern fish tend to be more robust than southern fish. Dorsolateral spots are the best character to separate F’. notatus from FE. olivaceus. With rare exceptions, F’. olivaceus has discrete, regular, black dorsolateral spots. Reproductive isolation is primarily ecological and is reinforced by homo- 1 Present address: Faculty of Biological Sciences, Southern Illinois University, Ed- wardsville, Illinois. specific mating preference. The two species are interfertile and produce fertile and distinctive hybrids in the laboratory. Only two natural hybrids are known. The two species have broad- ly overlapping ranges but are seldom syntopic. Syntopic associations are un- stable and character displacement is not demonstrated. Ecological prefer- ences are not uniform throughout the ranges of Ff’. olivaceus and F. notatus, but F’. notatus seldom occurs in “black- water”’. INTRODUCTION Fundulus notatus (Rafinesque) and F. olivaceus (Storer) are two of the most widely distributed species of the North American cyprinodontid genus Fandulus (Fig. 1). Both sexes of these two species have a single black, more or less solid, lateral stripe originating at the tip of the snout and terminating at the distal end of the caudal peduncle or on the caudal fin. This char- acter separates them from all other presently recognized species of Fundulus. Both F, notatus (personal communication, Clark Hubbs) and F. oltvaceus (Miller, 1955) have been reported from brackish water near the Gulf of Mexico, but they ordinarily swim at the surface of the water near the margins of freshwater lakes, rivers, ponds and streams. Both species occur in many different types of environment and both are quite variable. Almost all char- acters that have been used to separate them have considerable overlap and are unreliable over large parts of their sympatric range. Most early workers considered them to be conspecific (Garman, 1895; Jordan and Evermann, 1896; Forbes and Richardson, 1920; Hubbs and Ortenburger, 1929). EDITORIAL COMMITTEE FOR THIS PAPER: CLARK Huss, Professor of Zoology, University of Texas, Austin, Texas RoBerT R. MILLER, Curator of Fishes, University of Michigan Museum of Zoology, Ann Arbor, Michigan RoyAL D. SutrKus, Professor of Biology, Tulane University, New Orleans, Louisiana 749) Uo =) Figure 1. Known ranges of Fundulus no- tatus and Fundulus olivaceus and distribu- tion of material examined. The dashed line outlines the range of fF’. notatus and the dotted line outlines the range of F’. oliva- ceus. Solid circles are F’. olivaceus locali- ties and open circles are F’. notatus locali- ties. Solid triangles are I’. olivaceus locali- ties and open triangles are F’. notatus lo- calities plotted from data given by Brown (1956). Kuhne (1939) used F. ”. notatus and F. n. olivaceus as subspecific names in his study of the fishes of Tennessee. Moore and Paden (1950) considered F. notatus and F, olivaceus as separate species and presented evidence that they can occur together without interbreeding in the IIli- nois River of Oklahoma and Arkansas. Jur- gens and Hubbs (1953) and Knapp (1953) recognized both species in Texas. Brown (1956) compared specimens from several areas and concluded that both species are valid. Braasch and Smith (1965) reached the same conclusion from their study of the two forms in the upper Mississippi Valley. Their paper includes a synonomy for both species. This study was undertaken in an attempt to answer the following questions: How do various systematic characters vary from population to population? Do populations in the area of sympatry exhibit character Tulane Studies in Zoology Vol. 13 displacement (Brown and Wilson, 1956) ? Are F, notatus and F. olivaceus capable of hybridizing? Do they hybridize in nature? If the two forms are in fact valid species, what isolating mechanisms keep them sepa- rate? The results of this study indicate that F, notatus and F, olivaceus, though complete- ly interfertile, are valid species separated in nature primarily by different ecological re- quirements and secondarily by behavioral reproductive isolation. RANGE Fundulus notatus The westernmost Gulf drainage inhabited by F. notatus is the Guadalupe River system in southern central Texas. Hubbs, Kuehne, and Ball (1953) considered Landa Park Lake in Comal County as closely defining the upstream limit of the species in the Guadalupe River. It is abundant in the mainstream of the Colorado River at Austin in Travis County and is known from tribu- taries from Colorado County (Clark Hubbs, personal communication) upstream to Llano County (TU uncatalogued). Fandulus notatus is distributed generally throughout the Brazos (upstream to Palo Pinto and Eastland counties), San Jacinto, Trinity, and Neches rivers (Clark Hubbs, personal com- munication). It also occurs in isolated bayous and creeks draining into the Gulf between the Brazos and Sabine rivers. I have no rec- ords of F. notatus from any of the rivers between the Sabine River and the various eastern tributaries of the Mississippi River. This area is hereafter referred to as “western Louisiana drainage”. The southernmost record of F. notatus in the Mississippi Valley is approximately 20 miles west of New Orleans, Louisiana (TU 387). The species ranges north to south- eastern Wisconsin, southern Michigan, northeastern Iowa, and western and central Ohio; west to eastern Oklahoma and Kansas (Brown, 1956); and east to Jackson County, Tennessee (UMMZ 125113). A few specimens have been taken from the Lake Pontchartrain tributaries: Amite (TU 35697), Natalbany (UMMZ 182345), and Tangipahoa (UMMZ 182344) rivers. Fundulus notatus has been collected from the Pearl River from the vicinity of Jackson, Mississippi (UMMZ. 170728) and down- stream. The species has not been recorded from Vol. 1 any of the small drainages between the Pear] River and the Tombigbee River (hereafter called Mississippi Gulf drainage), but is known from the Tombigbee drainage of the Alabama system (Brown, 1956). I have no record of F. notatus from the Alabama River. Fundulus olivaceus The range of F. olivaceus overlaps the range of F, notatus in large part but 1s displaced to the east (Fig. 1). The westernmost popu- iation known is trom the Navasota River, ‘texas (see Material Examined). an eastern tributary of the Brazos River. No records are available from the rest of the Brazos system. Ine species is abundant along the Gulf Coast eastward to the Choctawhatchee Bay drainage of western Florida and Ala- bama. In the Apalachicola (Chattahoochee ) drainage F. olivacews occurs in a limited area south of the fall line in Alabama and Georgia. This distribution 1s strikingly sim1- lar to that given by Suttkus (1955) for Notropis euryzonus. Fowler (1945) re- corded F. notatus (—F. olivaceus?) from the Okefenokee Swamp. Since I have been unable to verity this record or obtain addi- tional records of F. olivaceus east of the Apalachicola drainage, | doubt that the spe- cies occurs east or tnat drainage. In the Mississippi Valley, F. olivacews ranges north to southern Missouri and Illinois, west to eastern Oklahoma, and east to western Ken- tucky (Brown, 1956) and to Anderson County, Tennessee (Cornell University 19148 cited by Brown, 1956). GEOGRAPHIC VARIATION Methods Dorsal and anal rays were counted at their bases. All rays were counted, rather than considering the last 2 rays as a single ray as is often done. This is the same method used by Brown (1956) and I have incorporated Brown’s counts into my data. Caudal ray count is total branched rays plus 2. Since both species show early ontogenetic increase in the number of branched caudal rays this character was not recorded for individuals smaller than 30 mm standard length. Lateral- scale counts were made as described by Brown (1956). Head width was measured at the rear of the orbits, and body depth was taken as the vertical distance from the origin of the anal fin to the dorsal surface of the body. Caudal peduncle depth and Fundulus Biosystematics 51 y length of depressed dorsal fin were taken as described by Hubbs and Lagler (1947). Color pattern was recorded for specimens over 30 mm standard length. No attempt was made to quantify such patterns in more than a general manner except to estimate the number of dorsolateral spots for FP, olivaceus. This estimate was obtained by counting the number of spots on the lett side of the dorsum, above the lateral stripe, below the dorsal midline and within the standard length. Dorsum pattern was _ re- corded as uniform, intermediate, or cross- natched. Lateral stripe shape was recorded as smooth, rough, or with few, several or many vertical extensions. Tone of the band and extensions was recorded as dark, inter- mediate, or light. Dorsum spotting in F. notatus was recorded as none, few, many, or blotched. Spots in the dorsal, caudal and anal fins were recorded as many, normal, few, very few, and none. Predorsal line was recorded as absent, light, dashed, partial, or complete. These categories were not par- ticularly satisfactory because of the high variability in coloration from sample to sample. However, their use helped guard against forming false impressions of the dis- tribution of color patterns in a given sample. Methods of statistical comparison follow those given by Cazier and Bacon (1949) and Mayr, Linsley and Usinger (1953). Meristic characters were compared with the use of a standard 2 x n Chi square test in addition to graphical comparison following Hubbs and Hubbs (1953). Fundulus notatus Number of dorsal rays (Table 1, Fig. 2) and number of anal rays (Table 2, Fig. 2) serve to separate F, notatus into two groups: a western group which includes populations from the Guadalupe, Colorado, Brazos, San Jacinto, Trinity, Neches, and Sabine rivers; and an eastern group composed of popula- tions from the Mississippi and Great Lakes drainages, Lake Pontchartrain drainage, Pearl River, and Tombigbee River. The differ- ences between these two groups do not war- rant the recognition of subspecies. Samples from populations making up the western group have 10 as the modal dorsal ray number and an observed range of vari- ation of 8 to 12. The modal dorsal ray num- ber is 9 in all samples from the eastern group and the observed range of variation 32 Tulane Studies in Zoology Vol 13 TABLE 1 Number of Dorsal Rays in Fundulus notatus : : Number of Dorsal Rays ; Populations ae ess 9 10 ial Ly N M Sey 283 Guadalupe R., Tex. 8 39 20 67 OFZ .64 .16 Colorado R., Tex. 2 44 6 71 9.8 .59 14 Brazos R., Tex. 25 68 7 100 9.8 .54 11 San Jacinto R., Tex. 11 54 34 1 100 10.2 .66 13 Trinity R., Tex. 155 32 3 50 9.8 O71 16 Neches R., Tex. 40) 60 100 9.6 .49 09 Sabine R., Tex.-La. 2 38 54 6 100 9.6 63 i183 Total Western 2; 158 352 76 1 588 929 66 .06 Mississippi R. 8 116 AT ilzal OF? 45 07 L. Pontchartrain Dr. 2 33 5 40 rl 41 13 Pear! R., La.-Miss. 3 68 29 100 9.3 .50 .10 Tombigbee R., Ala. 3 24 9 36 9.2 57 .19 6 On2 50 .05 Total Eastern 16 345 is from 8 to 10. A line drawn between 9 and 10 dorsal rays separates 73% of 588 western specimens from 74% of 345 eastern specimens; average separation 73.5%; co- efficient of divergence 0.60. The two groups are similar in observed range of variation (11 to 14) in anal ray number but have different modes. All sam- ples from western populations except those from the Brazos River (mode 12) are modal at 13. All samples from eastern populations are modal at 12. A line drawn between 12 and 13 anal rays separates 63% of 588 western specimens from 85% of 337 eastern specimens; average separation 74%; coef- ficient of divergence 0.60. Caudal ray number (Table 3) varies from 11 to 17. With the exception of the San Jacinto sample (mean 13.3, mode 13), the western group samples have a high average number of caudal rays (13.8 to 14.2). The Mississippi River sample (mean 14.2, mode 14) resembles the western group for this character. The rest of the eastern group samples are modal at 13. The Lake Pont- chartrain drainage sample of 16 specimens (mean 13.4) is significantly different from the Mississippi River sample (Chi square 14.2, 4 degrees of freedom, P less than 0.01), but not from the Pearl River sample (mean 12.9) or the Tombigbee River sample (mean 13.4). The pattern of caudal ray number distribution roughly parallels the patterns seen in dorsal and anal ray number with the exception of the Mississippi River sam- ple which has high mean caudal ray number and low mean dorsal and anal ray numbers. Observed range of variation in lateral TABLE 2 Populations ital lz 13 Guadalupe R., Tex. 6 39 Colorado R., Tex. 26 41 Brazos R., Tex. 2 53 37 San Jacinto R., Tex. 30 62 Trinity R., Tex. 1 16 29 Neches R., Tex. 44 51 Sabine R., Tex.-La. 2 35 57 Total Western 5 210 316 Mississippi R. 7 124 16 L. Pontchartrain Dr. 3 34 2 Pearl R., La.-Miss. 3 68 29 Tombigbee Re Alar 2 23 10 Total Eastern BO 249 57 Number of Anal Rays in Fundulus notatus Number of Anal Rays 14 N M S.D. 25.E 20 67 Sta bs 14 4 71 12:7 153 12 8 100) 0 t2-b G7 13 8 100 12.8 58 12 4 50 12.7 64 16 5 100) 912.6. SE5se ea 6 100 =: 12.7 62, ete 51 588 12.8 65 205 167 11220" od .08 39 1250) ees G ete 100 12:3 48) eG 1 36 DE) G4) 1 352) ai PAL | Vol. | ap ee — J 4 : Rie ts ; apt 2 =O Ge i Ouahin Ss 5 10 <¢ t wie Santee onan ro) 100, © PPHer 20 2 SS as Se i2 nt) io LJ om — z PENN) Ww Oo oc WwW a 100 DORSAL RAYS 100 40] 3 LJ 20 iS PER CENT 100 ANAL RAYS Distribution of dorsal and anal Figure 2. ray numbers in Fundulus notatus. Each bar represents 100% of the sample and is aligned on the break between 9 and _ 10 (dorsal rays) and 12 and 13 (anal rays), the most common numbers. Numbers above each bar are total sample size. The samples are arranged in order from west to east. These data are tabulated in tables 1 and 2. Fundulus Biosystematics 33, scale number is 32 to 37 (Table 4): Braasch and Smith (1965, Fig. 2), however, reported a variation of 30 to 38 in their samples from the upper Mississippi Valley. Their method of counting gave counts | scale greater than the method I used. Most samples are modal at 34 or 35, but the Colorado River sample is modal at 36 and the Mississippi River sample is modal at 33. There is no broad pattern of geographic variation, such as those seen in ray number for the dorsal, anal and caudal fins. Specimens from the upper Mississippi Val- ley and Great Lakes drainage tend to have broader heads, shorter snouts, and deeper caudal peduncles in comparison to specimens from other populations, but these characters are highly variable in all samples and do not seem to warrant the recognition of northern and southern groups. The Mississippi Valley and Great Lakes drainage populations seem to be similar throughout on the basis of dor- sal, anal, and caudal ray numbers as well as lateral scale number. In general, F. notatus may be divided into an eastern group characterized by a high percentage of low dorsal, anal and caudal ray numbers and a western group character- ized by a high percentage of high aumbers of dorsal, caudal and anal rays. No such pat- tern is seen in lateral scale number. The Mississippi River populations are character- ized by low dorsal and anal ray numbers, high caudal ray number and low lateral scale lateral scale number. Fundulus olivaceus Differences in dorsal ray number (Table 5, Fig. 3) may be used to divide F. olivaceus into three population groups. The most common dorsal ray number in samples from TABLE 3 Number of Caudal Rays in Fundulus notatus Number of Caudal Rays Populations ile: 12 13 14 Guadalupe R., Tex. 2 5 IL Colorado R., Tex. 2 6 30 Brazos R., Tex. 4 5 Oath San Jacinto R., Tex. 1 13 25 21 Trinity R., Tex. 2 4 14 Neches R., Tex. 2 9 13 Sabine R., Tex.-La. 2 9 21 Mississippi R. 2 6 Pall L. Pontchartrain Dr. 10 6 Pearl R., La.-Miss. 1 13 28 7 oO i Tombigbee R., Ala. 2 9 1 15 16 17 N MM SD * 29H 5 1 30 14.0 .88 32 i 3 5p di4ets eomnned i Geers! pil, etds y eOOmmees 8 68 otomeleO.0 25 14, 1 35.0 Ide? wieGiwen eo 6 30 13.8 86 34 10 Ag i AISige ai s0n= 204 20°. a 50 7 142° 285 2a 16. 13:4 1 s50n 225 ) Bit. Pitao en es 2 20 13.4 .89 .40 34 Tulane Studies in Zoology Volhi3 TABLE 4 Number of Lateral Scales in Fundulus notatus Number of Lateral Scales Populations 32 33 34 36 olf N M SsDe ye 2sebe Guadalupe R., Tex. 2 24 11 1 38 34.3 oT Sil) Colorado R., Tex. 3 a 17 23 2 52 35.3 .95 .26 Brazos R., Tex. 2 26 14 ff il 50 34.6 61 .20 San Jacinto R., Tex. 1 13 Zi 1 36 34.6 61 .20 Trinity R., Tex. 1 6 19 8 34 34.0 .70 .24 Neches R., Tex. 1 6 20 2 39 34.1 a6 PAD Sabine R., Tex.-La. 5 11 21 6 2 45 34.8 199 .30 Mississippi R. 6 24 9 3 1 53 33.0 394 .26 L. Pontchartrain Dr. 4 14 18 34.8 .44 All Pearl R., La.-Miss. 18 32 1 51 33.0 47 20 Tombigbee R., Ala. 9 3} 3 20 33.8 44 .18 the Navasota River population is 9. Sam- ples from the Trinity, Neches, Sabine, west- ern Louisiana drainage, Mississippi, Lake Pontchartrain drainage, Pearl, and Missis- sippt Gulf drainage are modal at 10 dorsal rays. There is clinal intergradation in this character from the Mississippi Gulf drain- age population (mode 10) into the popula- tions at the eastern limits of the range, where the mode is 9. Observed range of variation is from 8 to 11. A line drawn between 9 and 10 dorsal rays separates 64% of 100 Navasota River specimens from 77% of 970 Trinity to Mis- sissippi Gulf drainage specimens; average separation 70.5%; coefficient of divergence 0.37. A line drawn between 9 and 10 sepa- rates 77% of 970 Trinity to Mississippi Gulf specimens from 93% of 110 Chocta- whatchee Bay specimens; average separation 85%; coefficient of divergence 0.89. The mean dorsal ray number of the Apalachicola population (9.2) is significantly higher than that of the Choctawhatchee Bay populations (8.9); Chi square (2 x 3; 2 degrees) of free- dom) 11.2, P less than 0.01. A sample of 100 specimens from the southern Mississippi Valley is significantly different from a sample of 70 specimens from the northern Mississippi Valley popu- lations; Chi square (2 x 4, 3 degrees of free- dom) 11.7, P less than 0.01. A sample of 100 specimens from middle Mississippi trib- utaries (White River, Tennessee River) is intermediate between the northern and southern Mississippi Valley samples. These data indicate a clinal intergradation from low average dorsal ray number (9.6) in the southern Mississippi Valley to a higher aver- age number (9.9) near the northern limits of the range in the upper Mississippi Valley. However Braasch and Smith (1965, Fig. 2) TABLE 5 Number of Dorsal Rays in Fundulus olivaceus Number of Dorsal Rays Populations 8 9 Navasota R., Tex. 2; 62 Trinity R., Tex. 13 Neches R., Tex. 1 20 Sabine R., Tex.-La. 22, Western La. Dr. 20 S. Mississippi R. 1 40, M. Mississippi R. 22 N. Mississippi R. 11 L. Pontchartrain Dr. 19 Pearl R., La.-Miss. 1 30 Miss. Gulf Dr. 25 Total Trin. R.-Miss. G. 3 222 6 Alabama R. 4 132 1 Pensacola Bay Dr. 63 Choctawhatchee Bay Dr. 18 84 Apalachicola R. 5 69 11 N M S.D. 258. 1 100 9.4 54 sata 19 100 10.1 .56 Sika 9 100 929 56 sill 8 100 SM 53 mile 7 100 ae) 50 sul 6 100 9.6 61 wli2 5 100 9.8 .45 .09 3 70 a) 44 olla 4 100 9.8 56 silat 5 100 etl .56 aati 2 100 9.8 47 .09 68 970 9.8 .53 .03 2 241 9.4 .55 07 100 9.4 49 -10 110 8.9 .48 .10 100 9.2 .02 10 Vol. 1 give a mean dorsal ray number between 9.6 and 9.7 for upper Mississippi Valley F. olivaceus. This suggests that Mississippi Valley populations are reasonably homo- a ad a NAVA M.GULF PENS CHOC APAL. nC) io LJ 98 8B PER CENT DORSAL RAYS 100 i4L] 3] 2a iZ io {il 10 100 100 100100 '09 PER CENT loo] ANAL RAYS Figure 3. Distribution of dorsal and anai ray numbers in Fundulus olivaceus. Drawn as in fig. 2. These data are tabulated in tables 5 and 6. Fundulus Biosystematics 35 genous for this character. Further study is needed to clarify this point. Samples from eastern tributaries to the Mississippi were compared with those from the western tributaries. No significant dif- ferences were found. The total observed range in variation in anal ray number is from 10 to 14 (Table 6, Fig. 3). Number of anal rays shows a pat- tern of variation similar to that seen in dor- sal ray number, but the samples from the Mississippi Valley do not differ significantly. The sample from the Navasota River is modal at 12 as is the sample from the Trin- ity River, but a line drawn between 12 and 13 anal rays separates 86% of 100 Navasota specimens from 48% of 100 Trinity speci- mens; average separation 67%; coefficient of divergence 0.48. Average number of anal rays drops from a high of 12.5 in the Trinity and Neches samples to a low of 11.2 in the Choctaw- hatchee Bay sample. This trend does not continue to the east. Although the average number (11.6) of the Apalachicola popu- lation is low, it is higher than the Choctaw- hatchee Bay average. The difference be- tween the Choctawhatchee Bay and Apala- chicola populations are probably real; Chi square (2 x 4, 3d. f.) 13.0, P less than 0.01. Two different rates of clinal intergrada- tion are involved: (1) a change of average number of anal rays from 12.5 in the Trinity and Neches rivers into an average number of 12.1 for samples from the Mississippi Gulf drainage. This change in average num- ber is the result of an eastward increase in percent of specimens with 12 anal rays and Table 6 Populations 10 11 12 Navasota R., Tex. 9 et Trinity R., Tex. 3 49 Neches R., Tex. 2 52 Sabine R., Tex.-La. 4 50 Western La. Dr. 5 63 Mississippi R. 32 180 L. Pontchartrain Dr. 8 {fal Pearl R., La.-Miss. 6 64 Miss. Gulf Dr. 9 70 Alabama R., Ala. 33 67 Pensacola Bay Dr. 44 55 Choctawhatchee Bay Dr. 2 76 22, Apalachicola R. 0m Number of Anal Rays in Fundulus olivaceus "Number of Anal Rays 13 14 N M S.D. 2S.E. 14 100 12.0 48 .10 47 1 100 12.5 56 silat 45 1 100 12.5 96 Aaa 46 100 12.4 .o9 a2 32 100 12.3 .05 51 57 1 270 12.1 08 .O7 21 100 12.1 ol .10 29 1 100 12.2 .O7 ela 21 100 12.1 09d sei 18 118 fae 65 12 1 100 LIES 2 .10 100 11.2 45 .09 2 100 11.5 54 11 36 Tulane Studies in Zoology Vol. 13 TABLE 7 Number of Caudal Rays in F undulus olivaceus Kmper of Caneel Rane Populations LAP? 13 14 15) eG a7. N M S:De 253 Navasota R., ‘Tex. 10 31 17 2 60 14.2 US .20 Trinity R., Tex. 4 6 22 28 1 61 14.3 94 24 Neches R., Tex. 1 1 10 BO 1 1 61 1520 .65 SiG Sabine R., Tex.-La. 3 18 18 1 40 AA yp s25 Western La. Dr. il 9 ifal 29 2 52 14.4 .89 a2 Mississippi R. 11 35 86 9 3 144 14.3 .69 sabal L. Pontchartrain Dr. 1 113% Bye 38 4 82 14,3 .19 sll 7 Pearl R., La.-Miss. 4 15 24 DAL 4 2 76 142) 108 PAR Miss. Gulf Dr. 8 22, 28 1 59 14.4 Alte 19 Alabama R., Ala. 2 12 14 28 AGA IRON PAs Pensacola Bay Dr. 5 26 2 6 39 ee, 87 .28 Choctawhatchee Bay Dr. 5 32 ila 2 50 alas) BA All Apalachicola R. 1 1 16 13 ial 42 13.8 .96 .30 a decrease in percent of specimens with 13 anal rays. The percentage of fish with 11 anal rays is reasonably constant in all these samples. (2) A reduction of average num- ber of anal rays from 12.1 in the samples from the Mississippi Gulf drainage to an average number of 11.2 in the samples from the Choctawhatchee Bay drainage. This drop is the result of an eastward trend toward an increase in percent of specimens having 11 anal rays and a concurrent decrease in per- cent of specimens with 13 or 12 anal rays. A line drawn between 11 and 12 anal rays separates 919% of 100 specimens from the Mississippi Gulf drainage from 78% of 100 specimens from Choctawhatchee Bay drain- age; average separation 84.59%; coefficient of divergence 0.92. Observed range of variation in caudal ray number is 11 to 17 (Table 7). The Nava- sota River F. olivaceus sample is modal at 14, samples from the Trinity River east to the Alabama River are modal at 15, and samples from drainages east of the Alabama River are modal at 13. Though the modal numbers of the Trinity and Navasota sam- ples are different, the means, 14.3 and 14.2 respectively, are not significantly different. A line drawn between 13 and 14 separates 93% of 28 specimens from the Alabama River from 79% of 39 specimens from the Pensacola Bay sample; average separation 86%; coefficient of divergence 0.64. Lateral scale number ranges from 32 to 37 (Table 8) and 34 or 35 lateral scales are most common. Distribution of lateral scale number varies considerably from population to population. The Navasota, Trinity, Neches, and Sabine River samples are simi- lar (mean: 34.4, 34.8, 34.4, 34.8 respective- ly). With the exception of the Neches sam- ple, bimodal at 34 and 35, these samples are Table 8& Number of Lateral Scales in Fundulus olivaceus Number of Lateral Seales Populations 32 33 34 35 Navasota R., Tex. 1 6 19 WP Trinity R., Tex. 2 15 27 Neches R., Tex. 2 15 ifs) Sabine R., Tex.-La. 5 11 PAI Western La. Dr. 14 19 8 Mississippi R. 6 OH LON, 41 L. Pontchartrain Dr. 2 15) 26 28 Pearl R., La.-Miss. 5 32 29 Miss. Gulf Dr. 1 ef sil 11 Alabama R., Ala. 12 21 2 Pensacola Bay Dr. 2 15 23 Choctawhatchee Bay Dr. 4 37 Apalachicola R. 2 17 36 37 N M S.D. 2S.E. 2 50 34.4 sD ood 8 52 34.8 75 oral 32 34.4 56 .20 6 2 45 34.8 98 29 2 45 33.9 92 .28 3 1 194 34.0 .78 let 3 74 34.2 91 oil 7 73 34.5 .78 nalts d 64 34.0 .78 .20 35 33.7 O38 18 9 49 34.8 ol 29 11 52 35.1 .55 15 30 3 52 35.7 -66 18 Vol. 1 modal at 35. The western Louisiana drain- age sample is modal at 34 as is the Missis- sippi River sample. The Lake Pontchartrain drainage sample is almost bimodal at 34 and 35 (26 and 28 specimens respectively ). These 3 samples have significantly lower means (33.9, 34.0 and 34.2 respectively ) than the 4 western samples. There is a clinal decrease in average number of lateral scales from the Pearl River (34.5) east to the Alabama River (33.7) and a clinal increase in average lateral scale number from the Alabama River to a high of 35.7 for the Apalachicola River sample. Brown (1956) contrasted Mississippi Val- ley specimens with specimens from the Gulf Coast. The former were said to have deeper and wider bodies and deeper caudal pe- duncles than the latter. He did not examine specimens from the lower Mississippi Val- ley. Southern Mississippi Valley fish resem- ble Gulf Coast drainage specimens and do not have the deep, wide body and deep cau- dal peduncle characteristic of northern speci- mens. The following generalizations may be made. There is some correlation in variation in fin ray number in the unpaired fins. With the exception of the Navasota River popu- lation, western populations have a_ higher percentage of high fin ray numbers than eastern populations. The average dorsal ray number is nearly uniform from the Trinity River population east to the Mississippi Gulf drainage and drops off clinally to the east. Intergradation from high average anal ray number in the west to low average anal ray number in the east involves a low rate of clinal intergradation from the Trinity popu- lation to the Mississippi Gulf drainage popu- lation followed by a rapid rate of clinal intergradation eastward. Western popula- tions (Navasota included) have high aver- age caudal ray number and populations east of the Alabama River have low average cau- dal ray number. The Alabama River sample has the lowest average lateral scale number and the Apalachicola River sample has the highest, a reverse trend to those shown by fin ray numbers. ECOLOGY The published information available on the ecology and distribution of F. olivaceus and F. notatus is somewhat confusing, in part because each statement has been based Fundulus Biosystematics 3 on a study of either or both species in a limited area. In a study of the fishes of Tennessee, Kuhne (1939) stated, “Two sub- species occur in Tennessee, the northern (and more upland) I. 2. notatus and the southern lowland F. ». olivacews (Putnam). The latter has a flatter head and the body is marked by strong blackish spots.” Moore and Paden (1950) noted that the two spe- cies sometimes occurred together in the Illi- nois River. Knapp (1953), in regard to F. olivaceus in Texas, stated: “Where its range overlaps with F. notatus the two are usually ecologically separated, F. olivaceus being typically a quiet water form. Near the coastal plain this species inhabits swifter water.” He also stated, “In Texas F. notatus is to be expected in headwaters and fast streams. Braasch and Smith (1965) stated that F. notatus is more likely to be found in still water and F. olivacews in fast water in the upper Mississippi Valley, and that the two seldom occur together in that area. My examination of the hundreds of col- lections of the 2 species catalogued at Tulane University, University of Michigan Museum of Zoology and the Texas Natural History Collection at the University of Texas shows that in fact the two species are seldom col- lected at the same locality at the same time. When collections lumping specimens from several different habitats are excluded, it is clear that F. notatus and F. olivaceus are rarely syntopic (Rivas, 1964: 43). Braasch and Smith (1965) reported mixed collections from the same general area taken 64 years apart, but my observations cf two areas where syntopy has occurred indicate that syntopic associations at a particular lo- cality are unstable and transient. One area is in the mainstream of the Amite River in Louisiana and the other is in a small tribu- tary of the Navasota River in Texas. The first area Hs visited 4 times and the second 3 times (Table 9). Where numbers are given, all ee seen were collected and the sample is probably an accurate reflection of the Fandulus population at that time. Where catalogue numbers are not given, specimens were retained alive and mixed with specimens from other localities for use in another study. The differences in ecological preferences that are responsible for this separation of the two species are not clear at present. Fundulus olivaceus is abundant in the 38 Tulane Studies in Zoolog Vol. 13 TABLE 9 ; Variation in relative abundance of Fundulus notatus and Fundulus olivaceus in 2 areas of occasional syntopy messes and Date a F. “notatus F’. olivaceus Water Level Amite Rive er 28 Nov. 1963 abundant 0 low 14 May 1964 — about equal numbers — high 3 July 1964 2 0 ee, high, rising 25 Oct. 1964 19 (TU 35697) 22 (TU 35691) low Navasota River 30 March 1963 7 5 low 19 June 1964 abundant 0 low 19 Nov. 1964 1 (TU 37114) G (TU: 37116) high, dropping “blackwater” streams along the Gulf Coast sistent difference from the macrohabitat oc- and F. notatus is rare in this type of stream, but neither species seems to show any pref- erence for a particular type of habitat over a large area. Knapp (1953) has character- ized F. notatus as a headwater form, but in the Guadalupe and Colorado rivers it is ex- cluded from the headwaters. In the Pearl River drainage F. notatus is generally ex- cluded from the small tributaries and is most abundant in the main channel of the river, but all the records I have from the Tombig- bee River drainage are from small tribu- taries. Clark Hubbs (1957) pointed out that F. olivaceus in Texas is an eastern ele- ment of the Texas fauna whose western lim- its correspond primarily with the western limits of the Mixed Pine Oak Region. Ex- ceptions to this statement are the records from the Navasota River in Brazos and Grimes counties (see Material Examined ). These records are slightly west of the Mixed Pine Oak Region but from a drainage with previously recognized eastein faunal affin- ities (Knapp, 1953). Where only one of the two species is present, it is able to occupy suitable habitats along the margins of lakes, rivers, ponds, and streams. Either species may inhabit large or small bodies of water and in the southern part of their ranges either F. notatus or F. olivaceus may be found at the edges of still or rapidly flowing water. Ordinarily neither species is found in riffle areas, although either may Occur in quieter areas at the margins of riffles. In the area of sympatry the suitable habitats are divided between the two species and contact between the two is rare. With the exception of the general exclusion of F. notatus from “blackwater” areas, it appears that the macrohabitat occu- pied by F. notatus does not exhibit any con- cupied by F. olivaceus. In general, F. notatus is abundant in Texas and in the western and northern Mississippi Valley, and relatively rare in the southern Mississippi Valley and east of the Mississip- pi. F. olivaceus may be less abundant than F, notatus in Texas, but is by far the more common in the south, and is about as abun- dant as F. notatus in the middle Mississippi Valley. The abundance of F. olivaceus in the southern tier of states from Louisiana to western Florida is to some extent cor- related with the occurrence of “blackwater” streams and swamp lakes in that area, but F, olivaceus is also abundant in other types of habitat. EXPERIMENTAL HYBRIDIZATION In reference to F. notatus and F. olivacens, Bailey, Winn, and Smith (1954) suggested: “It is possible that a more thorough study may prove them to be the genetic variants of a single species.” Many individuals of both species have an intermediate appear- ance suggestive of either hybrid origin or introgression. The spotted individuals of F. notatus and the individuals of F. olivaceus which have very few, or very small spots could be interpreted as hybrids if opposing evidence were not available. The most fruitful approach to this type of problem seemed to be the production of known hybrids for comparison with sus- pected natural hybrids. These experiments were intended to help determine the nature of the isolating mechanisms (if any) exist- ing between the two forms in question. At the beginning of this study I believed F. notatus and F, olivaceus to be conspecific, but the results of these experiments render this position untenable. Vol) 1 Matertals and Methods In February, March and April, 1964, 3 sets of hybridization experiments were per- formed: (1) a female F. olivaceus, from the Bogue Falaya River at the Delta Regional Primate Center near Covington, Louisiana, was crossed with a male F. notatus from the Colorado River at Austin, Texas; (2) a fe- male F. notatus, from Hildebrandt Bayou near Beaumont, Texas, was crossed with a male F. olivaceus collected with the F. oli- vaceus female used in experiment one; and (3) a female F. olivaceus from a tributary to the Cahaba River in Alabama was crossed with a male F. notatus from Hildebrandt Bayou. The choice of fishes used in these experiments was governed by the necessity of using individuals unquestionably belong- ing to One species or the other. Each cross was carried out in a 5 gallon Metaframe brand aquarium filled with aged New Orleans tap water. Some of the water used in changing or adding water came from a temporary rainwater pond near Lake Pontchartrain. This was necessary because New Orleans tap water was lethal to these fishes during late January and Hee 1964. A nylon yarn mop weighted with ¢ lead sinker was placed in a bottom aide corner of the aquarium to provide a spawn- ing site. Both green and white mops were used with good results. Filtration and aeration were provided by a standard small inside box filter in each aquarium. These filters were charged with glass wool, char- coal and fine gravel. The fish were usually fed twice a day and at least once a day with Tetra-Min brand flake dry food, frozen raw beef liver, or frozen adult Artemia. Live mosquito larvae and cladocerans were fed when available. Each aquarium was illumi- nated with a 15 watt incandescent bulb in addition to daylight from an east window. A timer turned the lights on at 0530 and off at 2030 to simulate a 15 hour day. The fishes spawned readily, generally in the late afternoon. Each morning the mops were removed from the aquarium, squeezed partially dry and inspected visually while running the strands of the mop through the fingers. Eggs were removed with the fingers and placed in clean aged aquarium water in a plastic container (11 cm square and 2 cm deep). Acriflavine dye was added to the incubation water to serve as a fungi- Fundulus Biosystematics 39 cide. Dead eggs were discarded and newly hatched fry (F-1) were transferred to small rearing tanks and fed twice daily on live newly hatched Artemia nauplii. Dead Ar- temia and detritus were removed from the rearing tanks periodically and the water was partially replaced with aged aquarium water. When the fry began to show signs of crowding (differential and decreased growth rates) they were transferred to 10 or 20 gallon Metaframe brand aquaria and had scraped frozen beef liver, frozen adult Ar- temia and Tetra-Min added to their diet. The fry were distributed at the rate of about one fish per gallon of water. The F-1l’s were transferred to outdoor concrete pools. The fry from experiment one were transferred on 28 May and the fry from experiments two and three on 22 July 1964. They were from 4 to 5 months old and averaged about 30 mm total length when transferred. The offspring from experiment one were removed from the pools on 20 September 1964, 21 were preserved on 21 September, and 5 pairs were retained for further experiments. The fish from experi- ment two were removed and preserved on 7 December 1964, and the fish from experi- ment three were transferred to a laboratory aquarium on 28 September 1964 and _pre- served 13 December 1964. While in the pools, fish were fed daily with Tetra-Min or frozen adult Artemia to supplement natural foods present in the pools. Results Experiment one resulted in 34 F-1 off- spring; 2 died before reaching maturity, | was lost in transfer, 2 have been kept alive and 29 preserved for study. A comparison of the F-1 hybrids from experiment one with samples from the parent populations is given in Table 10. These hybrids are in- termediate and distinctive in general appear- ance (Fig. 4). F-1 hybrids from experiment one were able to produce F-2 hybrids and backcrosses to both parent species with no sign of reduced sex drive or fertility. These experiments will be discussed further in a later paper. Though the hybrids are intermediate in general appearance, analysis of individual characters (Table 10) shows that they are intermediate only in pattern of the dorsum and in number and shade of dorsolateral 40 Tulane Studies in Zoology Vol. 13 F Figure 4. A. male Fundulus olivaceus (SL 54.9 mm) and B. a female F. olivaceus (SL 17.8 mm), from tne Bogue Falaya River. C. a male, and D. a female F-1 hybrid from ex- periment one (SL 54.6 mm and 45.0 mm respectively). E. a male, and F. a female Fun- dulus notatus from the Colorado River (SL 49.4 mm and 42.8 mm respectively). Photo- graphs by Forrest Jack Hurley. Vol. 1 spots. Other characters resemble one parent species, both parent species, or neither. The parent population samples and the F-1 hy- brids do not show significant differences in dorsal ray number, caudal ray number, body depth, caudal peduncle depth of both males and females, or male dorsal fin length (se Fig. 5). The F-1 hybrids resemble Colo- rado River F. notatus in anal ray number and lateral scale number, but resemble Bogue uw ° DORSAL FIN LENGTH AS % OF S.L. n fe} 35 40 45 so 55 60 STANDARD LENGTH IN MILLIMETERS Figure 5. Comparison of dorsal fin length in Fundulus notatus (open circles), Fundulus olivaceus (solid circles), and F-1 hybrid (triangles) males. See Table 10 for comparison of females. Falaya River F. olivaceus in female dorsal fin length, female anal fin spotting, and number of lateral band extensions in males. Hybrid males have more heavily spotted anal fins than males from either parent population. The 29 F-1 hybrids are sig- nificantly broader-headed than samples from either parent population. This character seems to be intrinsic in the hybrids and not an artifact caused by laboratory conditions. Specimens of both parent species raised under similar conditions have not shown a similar effect. Of 16 characters studied, the F-1 hybrids from experiment one are inter- mediate in 3, extreme in 2, similar to both parent populations in 6, similar to F. notatus in 2 and similar to F. olivaceus in 3. DISCUSSION Fundulus olivaceus generally has discrete, regular black spots on the dorsolateral sur- face of the body. Fundulus notatus may be Fundulus Biosystematics 4] immaculate or may have dusky spots or blotches. Though other characters may be of use in separating these species in a given area, this 1s the only one of the many pro- posed by Hubbs (in Moore and Paden, 1950: 88) and by Brown (1956: 131) which will serve to separate the two species everywhere. Head shape, snout length, body depth, caudal peduncle depth, shape of male dorsal and anal fins, shape and number of spots in the unpaired fins, color of the dor- sum, predorsal stripe, and male lateral band extensions may be different for the two spe- cies at a given locality, but these characters are not conservative and their use as key characters confuses rather than clarifies. The spots characteristic of F. olivaceus are generally more numerous in males than in females and large males tend to have more spots than small males. The spots vary from less than 0.2 mm to over 1.0 mm in diameter. They are usually round but may be dash-shaped or slightly irregular. The most striking development of these spots was seen on large males from the Pensacola Bay drainage samples. An attempt to use numbers of spots to define subpopulations of F. olivaceus failed because variation in most samples is too great. This spotting characteristic, though the most reliable for use in separating F. oli- vaceus and F. notatus, breaks down in rare instances. A few collections from the south- ern states, from Texas to Florida, have in- cluded individuals of F. olivacews with few dorsolateral spots. Braasch and Smith (1965) reported similar collections from the upper Mississippi Valley. Males in these collec- tions have as few as 15 spots. Females usu- ally have between 5 and 15 spots and some females have 4, 3, 2, 1, or none. These rare immaculate females may be morphologically indistinguishable from F. notatus females, and are identified as extreme variants of F. olivaceus rather than F. notatus because: (1) they sometimes occur outside the range (as presently known) of F. notatus, (2) they are not associated with immaculate males, and (3) they are associated with F. olivaceus which have a low number of dorso- lateral spots. One of these females taken alone would certainly be identified as F. notatus and records of F. notatus based on single females should be critically examined. F. notatus may have an immaculate dor- sum or may have dusky spots or blotches on Volk. 13 Tulane Studies in Zoology yaep AuBvul 04 Mof - 6 Aysnup ‘ Moy - & euou - 5 “Vy 3I] 5 pue Pp Oo} BIPIULI9}UT yaep AuBUl - P OJ BIPIULIO4UI - P AOA MO} 0} BUOU - f ‘syodg urns.10q dUVIPIULII4UT Aysnp ‘u.toyiun OU VIPIULI94Ul poyoyey-Ssso.o uie}}eg wns10g SNAIDAYO * GT 03 G GT 01 0T Sse] 10 G suotsueyxy edits SNAIDAYO * Moy Moj auoUu 4 sjodg uly [euy LOYPION Moy Auvul auoUu 10 Moy ATOA 9 sjodg uly [euy SHEIDQYO A 8a" 8h GGG ra Ie 69° 0°GG 61 TL 6ST G06 vI 6 ‘IS/1d squoieg 40g ‘G ‘BIJ ‘saydwies useaMjoqd sUdLeTJIP JUBIIFIUSIS ON OTST syuered YO 6S" Gai Vor VI 6G" 6F° LOT 61 06" Le Oe Si 6 IS/dd0 syuoled YO 8a" 8g" Cue oT 61" 0&8" 6° OT OT OF" 08 Geek 9T 2 IS/ddO syuered YO 0g vel G81 6G 66" 66" T'61 6G 09° ve 1 vst 0G % S®'IS/da LOYPION ZI Sh" mals Os Ga" 89° [Sei 6G Oe 6S" esl 6& Yo S®'TS/ MH SN}D}OU * Ie EG) GPS VL 6S" L8° 0°Sé 66 96° G6" 6s 6g So[BIG [B19}e'T syuored 430 JEU 6L° 6 Vl 68 6G E9% ima! 6G Vo 68° aaa! sits) skey [epneD S1N}D}OU * OT Se [EGlE OOT 61" 6F° Lol 6G ol’ eo" Lot TL sAey [BUY spud el 40d ce 9¢° 8°6 OOT 6T" 6 6°6 6G vl 6g 8°6 TL Saat eee e[quiesey ce Se ‘ds = N ‘aSe dS a N ‘aSs AS x N Layered 81000110 “Hf Behera ee oe spiiqaH T-a SN}D}0U * gee 1010 spuqay Savin [-A paunay jipalany YPN Snadvarljo snjNpUNn PUD snzD}oU snjnpUNny {0 842018 Ol alavy jojuawng {0 wosrupdUoy) Vol. 1 Fundulus Biesystematics 43 Figure 6. the only Fundulus in the collection. both Fundulus notatus and F.. olivaceus. Photographs by Forrest Jack Hurley. the dorsum. Females are often more heavily spotted than males. Usually individuals with spots or blotches also have strong develop- ment of the predorsal line, which may be broken up into dashes or spots which com- plement the rest of the dorsal pattern. In contrast, F. olivaceus seldom has strong de- velopment of the predorsal stripe. As has been pointed out by Clark Hubbs (1963) and others, closely related species generally are easier to cross than more dis- tantly related ones. The ease with which these two species hybridize in the laboratory supports the idea that their close morpho- logical similarity reflects close phylogenetic relationship. As the hybrids are intermediate in several characters and resemble one parent or the other in other characters, it is clear that the hybrids are not of gynogenetic origin nor are they “false hybrids” in which the paternal chromatin has no effect. The sex ratio of the hybrids (experiment one ), 10 males to 21 females, is not significantly different from a theoretical 50:50 sex ratio of 15 males and 15 females. Hubbs and Drewry (1959) have discussed the pitfalls encountered in interpreting results of hy- brid experiments. That F. notatus and F. olivaceus are inter- fertile in the laboratory is not necessarily evidence that they are conspecific, but is evidence that isolating mechanisms (if any) which separate them in nature are ecological or behavioral rather than cytological. My examination of some 12,000 speci- Two natural hybrids, both males. UMMZ 161253 (SL 39.5 mm, above) was AM OL BiAlPAD) (SL 31.2 mm, below)) was taken with mens of F. notatus and F. olivaceus has un- covered two natural hybrids, both males (TU 37120 and UMMZ, 161253), Fig. 6): The extreme rarity of F. notatus x F. oli- vaceus hybrids in nature is evidence of the effectiveness of isolating mechanisms. Pau- city of natural hybrids may be due to hybrid non-survival but several lines of evidence point against this. Moore and Paden (1950) collected breeding pairs of F. notatus and F. olivaceus from an area of syntopy and found no heterospecific pairs. This observa- tion indicates that when the ecological bar- riers which ordinarily separate the two spe- cies break down, isolation is maintained by the preference of the fish for mates of their own species. Heterospecific matings are easily obtained in the laboratory but seem to occur rarely in nature. The robust nature of the laboratory hybrids suggests that nat- ural hybrids are probably able to compete with either or both parent species. The paucity of natural F-1 hybrids sug- gests that introgressive hybridization is not at present an important contributor to vart- ation in either species. Thus F. oltvaceus with small spots, or few spots, or both, are probably not of hybrid origin. The same is true of heavily spotted or blotched F. no- tatus. The two species are similar in range of variation in dorsal. anal and caudal ray num- bers and in lateral scale number. Although there is considerable intraspecific variation in average number of caudal rays, popula- 44 Tulane Studies in Zoology tions of F. notatus are not significantly dif- ferent from populations of F. olivaceus from the same area, with the exception of the populations from the Pearl River (com- pare Tables 3 and 6). Populations of the two species have significantly different dis- tributions of lateral scale numbers in the Trinity, Mississippi, and Pearl rivers. Lateral scale number does not show a coordinated pattern of geographic variation in the 2 species. It is thus probable that local inter- specific differences in this character are for- tuitous and of little importance. With the exception of the Navasota River population of F. olivaceus, western popula- tions of F. notatus and F. olivaceus tend to have high dorsal ray number, and this char- acter does not serve to separate them. The trend to high dorsal ray number extends to the Mississippi Gulf drainage in F. olzvaceus. High dorsal ray number “western group” F. olivaceus and low dorsal ray number “east- ern group’ F. notatus show statistically sig- nificant differences in the Mississippi River (see also Braasch and Smith, 1965), Lake Pontchartrain drainage, and Pearl River. In the Alabama system populations of the two forms are not distinguishable on this char- acter because FP. olivaceus dorsal ray num- bers are beginning to drop clinally toward the east. With the exception of the Navasota River F. olivaceus both species tend to have high anal ray numbers in the west and low anal ray numbers in the east. Although there seems to be a slight difference in pattern of variation between the two species, this char- acter serves only to separate the Navasota River F. olivaceus population from the Bra- zos River F. notatus population. Fundulus notatus and F, olivaceus are two closely related species which are sympatric and occasionally syntopic over a large area. Both species have completely allopatric populations, F. notatus in the west and F. olivaceus in the east (Fig. 1). In this kind of distribution the phenomenon of character displacement is often seen. Brown and Wil- son (1956) defined character displacement as follows: “Character displacement is the situation in which, when two species of ani- mals overlap geographically, the differences between them are accentuated in the zone of sympatry and weakened or lost entirely in the parts of their ranges outside this zone.” The probable causes were said to Vol. 13 be reinforcement of reproductive isolating mechanisms and ecological displacement. Character displacement would thus be the result of a long term process, and would be most pronounced where it was the result of relatively stable contact of populations of two closely related species over a period en- compassing a large number of generations. My observations on areas of occasional syn- topy do not support the hypothesis that F. olivaceus and F. notatus populations main- tain stable contact over the period of time necessary for character displacement to oc- cur. I suspect that individuals or populations of both species may migrate considerable distances. Individuals collected syntopically were perhaps allotopic the day before and vice versa. Braasch and Smith (1965) were unable to show statistically significant differences between allopatric and sympatric samples from the upper Mississippi Valley, but they felt that apparent parallel divergence of two independent characters (dorsal ray number and lateral scale number) indicated some interspecific interaction between sympatric populations. My data do not support this conclusion because the patterns of geo- graphic variation in either species seem to be TS of the presence or absence of the other species. I have been unable to see evidence of character displacement, either between syntopic populations and allotopic populations in the same drainage or between populations of one species inside and outside the range of the other. The pos- sibility of character displacement in species recognition ability is presently being studied but results to date are inconclusive. MATERIAL EXAMINED Complete locality data are given for only those collections which represent range extensions or other significant distributional data. Other collections are listed by drainage, state, county and museum number. Complete data on most collections are given by Thom- erson (1965). The following museum abbreviations are used: University of Alabama Ichthyological Col- lection, UAIC; Florida State University, FSU; Private collection of William Smith-Vanez, WSV; Texas Nat- ural History Collection, University of Texas, TNHC; Tulane University, TU; University of Michigan, Mu- seum of Zoology, UMMZ. In addition to standard abbreviations for states and compass directions, the following are used: Co(s).—County (ies), Par(s).— Parish (es), Cr.—Creek, Dr.—Drainage, Hwy.—High- way, mi.—mile(s), R.—River, and trib.—tributary. In addition to the specimens listed here all specimens of the two species catalogued TNHC and TU (to Vol. 1 May 31, 1965), most of the specimens catalogued by UMMZ, and a few UAIC specimens were examined for verification of identification. Some 12,000 speci- mens of the two species were examined and counts were made on 933 F. notatus and 1621 F. Olivaceus from the material listed below. Distribution of these collections is shown in Fig. 1. Fundulus notatus Gusdalupe Dr.—Texas: Medina Co.; TNHC 1865 (12)—Bexar Co.; TNHC 5336 (43)—Gonzales Co.; TNHC 301 (5)—Caldwell & Guadalupe Cos.; TNHC 86 (13), TNHC 177 (1), TNHC 235 (3), TNHC 330 (3)—Hays Co.; TNHC 3037 (13). Colorado Dr.—Texas: Bastrop Co.; TNHC 3733 (11), TNHC 3795 (7), TNHC 5266 (16)—Travis Cou TNHG 225° (4); DNHE 526 (2), INHE 540 (22), TNHC 1078 (1), TNHC 1098 (2), TNHC 1895 (5), TNHC 2552 (2)—Llano Co.; Honey Cr. on Ray Smith Ranch midway between Llano and Marble Falls, TU uncatalogued (19). Brazos Dr.—Texas: Grimes Co.; TU 37115 (1)— Bell’ Co:; TNHC 1021 (2), TNHC 3773 (12), TU 35996 (41)—McLennan Co.; TNHC 3467 (1), TNHC 4165 (4), TU 36002 (17)—Bosque & Hill Cos.; TNHC 4246 (1)—Bosque Co.; TU 2989 (9)— Erath Co.; TNHC 1992 (39)—Eastland Co.; TNHC 993 (20). San Jacinto Dr.—Texas: Montgomery Co.; TNHC 1118 (42), TNHC 1199 (18)—San Jacinto Co.; TNHC 2377 (36), TNHC 2843 (4). Trinity Dr.—Texas: Polk Co.; TNHC 5445 (3)— Madison & Walker Cos.; TU 4891 (13)—Ellis Co.; UMMZ 92385 (6), UMMZ 92388 (2)—Dallas Co.; UMMZ 120152 (4)—Tarrant Co.; UMMZ 97503 (7), UMMZ 170048 (1), UMMZ 170060 (5). Neches Dr.—Texas: Anderson Co.; TU 3479 (6) —Cherokee Co.; TU 17686 (46), TU 17788 (11)—- Rusk & Shelby Cos.; TU 3854 (6)— Shelby Co.; TWNs315 (20). Sabine Dr.—Louisiana: Sabine Par.; TU 33312 (2), TU 33798 (1), TU 33938 (34), TU 34005 (6), TU B078. (21), TU 35461 (11), TU) 35567 (2)—De Soto Par.; TU 35544 (11)—Texas: Panola Co.; TNHC 6063 (10)—Harrison & Panola Cos.; TNHC 3190 (6). Mississippi and Great Lakes Dr.—Louisiana: St. Charles Par.; Ponds in Bonnet Carre Spillway S of Hwy. 61, 20 mi. W New Orleans, TU 387 (5)—La Salle Par.; UMMZ 181940 (12)—Oklahoma: No- wata Co.; TU 15532 (39)—Tennessee: Lake Co.; UMMZ 103367 (2)— Davidson Co.; UMMZ 88106 (6), UMMZ 104498 (3), UMMZ 177568 (7)— Marshall Co.; UMMZ 121387 (13)—Jackson Co.; UMMZ 125113 (10)—Michigan: Washtenaw Co.; (Huron R.) UMMZ 72255 (19). Lake Pontchartrain Dr.—Louisiana: East Baton Rouge Par.; Amite R. at Hwy. 190, 7 mi. E of in- tersection Hwy. 190 & Hwy. 61, TU 35697 (19)— Tangipahoa Par.; Tangipahoa R. at Hwy. 35, 1.5 mi. E Amite, UMMZ 182344 (1)—Tangipahoa Par. (?); Trib. to Natalbany R., UMMZ 182345 (20). Pearl Dr.—Louisiana: Washington Par.; TU 37118 (54)—Mississippi: Marion Co.; TU 37119 (36)— Lawrence Co.; TU 26408 (29)—Hinds & Rankin Cos.; UMMZ 170728 (3). Tombighee Dr.—Alabama: Sumter Co.; TU 7497 (3), UMMZ 163741 (12)—Mississippi: Kemper Co.; UMMZ 157774 (3)—Noxubee Co.; TU 3755 (11), UMMZ 113885 (6). Fundulus Biosystematics 45 Fundulus olivaceus Navasota Dr.—Texas: Grimes Co.; Trib. to Nava- sota R. 5.1 mi. NE Navasota at Hwy. 190, TU 37116 (7)—Brazos Co.; Wickson Cr. UMMZ 129747 (3), Peachtree Cr. borrow pits 12 mi. S College Station, UMMZ 129764 (1), UMMZ 129951 (16), and UMMZ 138235 (5); Peach Cr. at Hwy. 6, UMMZ 129807 (80), Sand Cr. trib. to Navasota R. E of Kurten, UMMZ 129916 (3), Navasota R. and borrow pits 4 mi. E Kurten, UMMZ 129935 (7). Trinity Dr.—Texas: San Jacinto Co.; TNHC 1169 (9)—Polk Co.; TNHC 5446 (29), TNHC 2415 (40), TNHC 6039 (20)—Anderson Co.; TU 3792 (2). Neches Dr.—Texas: Hardin Co.; TU 21615 (57), TU 21761 (28)—Tyler Co.; TU 21686 (23). Sabine Dr.—Texas: Newton Co.; TU 14045 (3)— Panola Co.; TU 3496 (10)—Louwisiana: Vernon Par.; TU 777 (6) TU 5136 (6)—Sabine Par.; TU 793 (7), IU $075 (21), TU 33764 (18), TU 33924 (13)8 GU: 33941 (2). TU 3717. (He Western Louisiana Dr.—Louisiana: Allen Par.; TU 1327 (16), TU 14458 (4)—Rapides Par.; TU 1344 (11), EU 2059 (5); TU 2065 (13), TU 3459 (23), TW 4982 (19), FU 5769 (Ch) LU 5830 Ce nw 5871 (4). Southern Mississippi Dr.—Louisiana: East Feliciana Par.; TU 5231 (19) TU 30884 (43)—Mississip pi: Lincoln Co.; TU 28889 (7)—Holmes Co.; TU 3695 (10). Middle Mississippi Dr., (Arkansas and Tennes- see rivers) —Arkansas: Garland Co; TU 14249 (12) —Yell Co.; TU 15487 (73)—Pope Co.; TU 13770 (13), TU 26925 (3)—Mississippi: Tishomingo Co.; TU 4197 (4)—Tennessee: Benton Co.; TU 14715 (5). Northern Mississippi Dr.—lllinois: Massac Co.; TU 3530 (25)—Johnson Co.; UMMZ 175870 (7). Lake Pontchartrain Dr.—Louisiana: East Baton Rouge Par.; TU 35691 (22)—East Feliciana Par.; TU 4656 (39)—St. Tammany Par.; TU 967 (33), TU 9811 (14), TU 14512 (18). Pearl Dr.—Louisiana: St. Tammany Par.; TU 25823 (25)—Mississippi: Pearl River Co.; TU 14107 (20), TU 23549 (40)—Marion Co.; TU 28415 (44) —Copiah Co.; TU 28872 (22). Mississippi Gulf Dr.—Mississippi: Hancock Co.; TU 4771" (20), TU 7653 (6), TU 27078) (12) — Harrison Co.; TU 27094 (3)—Jackson Co.; TU 28087 (7)—Stone Co.; TU 16348 (21), TU 28095 (1)—Lamar Co.; TU 1622 (5)—Perry Co.; TU 1231 (5)—Covington Co.; TU 28582 (9), TU 28593 (1) —Lauderdale Co.; TU 7484 (3). Alabama (including Tombigbhee) Dr-—Alabama: Washington Co.; TU 33910 (2), UMMZ 163597 (84)—Clarke Co.; TU 32522 (11)—Monroe Co.; TU 2644 (15)—Crenshaw Co.; TU 2586 (8)—Wil- cox Co.; TU 3065 (9), TU 3431 (14) Marengo Co.; TU 24544 (1)—Perry Co.; TU 25969 (19)—Bibb Co.; TU 19423 (6), TU 24688 (17)—Talapoosa Co.; TU 12174 (7)—Tuscaloosa Co; TU 27548 (4), TU 30192 (3), TU 30212 (6)—Blount Co.; UMMZ 168616 (18)—Marion Co.; TU 30235 (21)—Missis- sippi: Lee Co.; TU 2443 (10). Pensacola Bay Dr.—Florida: Santa Rosa Co.; TU 20941 (7)—Okaloosa Co.; TU 23691 (30)—Walton Cor LU 24762 (67); Choctawhatchee Bay Dr.—Florida: Holmes Co.; TU 2277 (37)—Alabama: Dale Co; TU 2348 (28) —Henry Co.; TU 3902 (60). Apalachacola (Chatahoochee) Dr.—Alabama: Rus- 46 Tulane Studies in Zoology FSU 6591 (15), FSU 6656 6423-1 (2), WSV 6438- FSU 6643 (11), Early sell Co.; UAIC 1097 (1), (77), TUL 10717 (3), WSV 1 (3), WSV 6445-1 (44)—Lee Co.; TU 12099 (3), WSV 6414-3 (1)—Georgia: Coss WAIC 1134 (1): Fundulus notatus x F. olivaceus Mississippi Dr.—Louisiana: Cr. trib. to Little River at Chatham, (1). Pearl Dr.—Mississippi: Copiah Co.; 37120) (1). Jackson Par.; Castor UMMZ 161253 Pearl River 3 mi. E Georgetown, TU CONCLUSIONS 1. Fundulus notatus and F. olivaceus are highly interfertile in the laboratory and pro- duce fertile and vigorous hybrids. Their ranges broadly overlap in nature and they are occasionally syntopic. The rare occur- rence of natural hybrids (about 1 in 6000) indicates that strong isolating mechanisms separate the two species. 2. The exact nature of the isolating mechanisms is not known. The rare occur- rence of syntopy indicates that 1solating mechanisms are primarily ecological. How- ever, no general statement of ecological parameters seems to have more than local validity and either species seems to occupy all favorable habitats when the other 1s absent. 3. Observations of syntopic populations indicate that strong behavioral isolation ts operative. No evidence is available to sug- gest intergradation or introgressive hybridi- zation. 4. Morphological character displacement is not demonstrated and is probably rare or absent as a result of the unstable nature of the syntopic associations. 5. A single character, the presence or ab- sence of distinct, regular, dark black dorso- lateral spots, separates specimens of F. olt- vaceus from specimens of F, notatus except in rare instances. Other characters may be used to separate local populations, but do not have range-wide utility. There is broad intraspecific overlap in all characters studied and the recognition of subspecies is not war- ranted, ACKNOWLEDGMENTS This paper is in large part a revision of a dissertation (Thomerson, 1965) submitted to the Graduate School of Tulane University in 1965 in partial fulfillment of the Ph.D. degree. Drs. Royal D. Suttkus (Chairman), Andrew A. Arata, Gerald E. Gunning, and Merle Mizell were the committee for the Vol. 13 dissertation. Drs. Reeve M. Bailey, Clark Hubbs, and Robert R. Miller offered much helpful discussion of the problem, and ex- tended every courtesy during visits to their institutions. The study would not have been possible without free access to the Tulane University collection under the care of Dr. Royal D. Suttkus. The following persons made loans of specimens from their institutions: Dr. Clark Hubbs, Texas Natural History Col- lection, University of Texas; Dr. Robert R. Miller, University of Michigan Museum of Zoology; Dr. H. T. Boschung, University of Alabama Ichthyological Collections; Dr. Ralph Yerger, Florida State University. William Smith-Vanez loaned specimens from his private collection. The following persons made particular effort to help obtain fishes for the study: Lawrence and Dorothy Blackburn, Beau- mont, Texas; James D. Satterfield, Georgia State University; Michael Dahlberg, John S. Ramsey, and Francis L. Rose, Tulane Uni- versity; Bernard Halverson, Mobile, Ala- bama; George Maier, Chicago, Illinois; Dr. Jerry McClure, Miami University, Oxford, Ohio; James E. McShane, Houston, Texas; and Dr. Richard L. Stone, New Orleans, Louisiana. Dr. James E. Bohlke and other members of the staff of the Academy of Natural Sciences of Philadelphia searched the cata- logued material of that collection for the specimens from the Okefenokee Swamp cited by Fowler (1945). Lawrence Bayless and John Ramsey helped maintain the ex- perimental animals during periods when | was off campus. Gilbert Pogany and Clyde D. Barbour provided major photographic assistance in preparation of the figures. My wife, Kathleen A. Thomerson, helped proof- read the manuscript and did all typing. The study was carried out during the ten- ure of a National Aeronautics and Space Ageaaien Fellowship administered by the Graduate School, Tulane University. Laboratory space and the major portion of the equipment needed were made available by the Department of Biology, Tulane Uni- versity. Items of field equipment, and travel assistance were provided under auspices of several research grants administered by Dr. Royal Suttkus (National Science Foundation G-9026; Department of Health, Education, and Welfare WP-00082-01-05 and 3-T1- Vol. 1 Fundulus Birosystematics 47 ES-27-01S1, 03S1). A research support Zool. (John A. Moore, ed., Washington grant from the Society of the Sigma Xi helped defray research expenses. REFERENCES CITED BAILEY, REEVE M., H. E. WINN and C. L. SMITH 1954. Fishes from the Escambia River, Alabama and Florida, with ecologic and taxonomic notes. Proc. Acad. Nat. Sct. Phila. 101: 109-164. BRAASCH, MARVIN E., and P. W. SMITH 1965. Relationships of the topminnows Fundulus notatus and Fundulus olivaceus in the Upper Mississippi River Valley. Copeia (1): 46-53. BROWN, JERRAM L. 1956. Identification and geographic variation of the cyprinodont fishes Fundulus olivaceus (Storer)) and Fundulus notatus (Rafinesque). Tulane Stud. Zool. 7(38) : 119-134. BROWN, WILLIAM L., and E. O. WILSON 1956. Character displacement. Syst. Zool. 3(2): 49-63. CazigER, M. A., and A. BAcon 1949. Intro- duction to quantitative systematics. Bull. Amer. Mus. Nat. Hist. 93: 347-388. FORBES, STEPHEN A. and R. E. RICHARDSON 1920. The fishes of Illinois. Nat. Hist. Surv. Ill. (2nd ed.) 3: i-exxi, 1-357. FOWLER, HENRY W. 1945. A study of the fishes of the southern Piedmont and Coastal Plain. Monogr. Acad. Nat. Sci. Phila. 7: i-vi, 1-408, figs 1-313. GARMAN, SAMUEL 1895. The cyprinodonts. Mem. Mus. Comp. Zool. 19: 1-179, pls. 1- WA, HUuBBS, CARL L., and CLARK HUBBS: 1953. An improved graphical analysis and com- parison of series of samples. Syst. Zool. 2: 49-56, 92. s . and K. F. LAGLER 1947 (2nd printing 1949). Fishes of the Great Lakes region. Bull. Cranbrook Inst. Sci. 26: 1-186, pls. 1-26, figs. 1-251. : -. and A. IT. ORTENBURGER 1929. Fishes collected in Oklahoma and Arkansas in 1927. Publ. Univ. Okla. Biol. Surv. 1: 45-112. Hugs, CLARK 1957. Distributional patterns of Texas fresh-water fishes. Southwest- ern Naturalist 2 (2-3) : 89-104. Bed wa ave : 1963. The use of hybridi- zation in the determination of phyloge- netic relationships. Proc. XVI Int. Cong. D. C.) 4: 103-105. and G. E. DREWRY 1959. Survival of F hybrids between Cyprino- dont fishes, with a discussion of the cor- relation between hybridization and phylo- genetic relationship. Bull. Inst. Mar. Sci. U. Texas 6: 81-91. , R. A. KUEHNE, and J. C. BALL 1953. The fishes of the upper Gua- dalupe River, Texas. Tew. 216-244, JORDAN, DAvip STARR and B. W. EVERMANN 1896. Fishes of North and Middle Ameri- Case. 2 DULL UES. Nats Mus Ai PG 1: i-ix, 1-1240. JURGENS, KENNETH C. and CLARK HuBrRs 1958. A checklist of Texas fresh-water fishes. Texas Game and Fish 11: 12-15. KNAPP, FRANK T. 1953. Fishes found in the fresh-waters of Texas. Ragland Studio and Litho Printing Co., Brunswick, Ga., Jour. Sc. 5: i-vii, 1-166. KUHNE, EUGENE R. 1939. A Guide to the Fishes of Tennessee and the Mid-South. Tenn. Dept. Cons., Nashville, Tenn., 1- 124. E. LINSLEY and R. L. USINGER 1953. Methods and Principles of Systematic Zoology. McGraw-Hill Book Co., New York, i-ix, 1-328, figs. 1-45. MILLER, ROBERT R. 1955. An annotated list of the American cyprinodontid fishes of the genus Fundulus, with the description of Fundulus persimilis from Yucatan. Occ. Pap. Mus. Zool. Univ. Mich. No. 568: 1-25, 1 pl. Moore, GEORGE A. and J. M. PADEN 1950. The fishes of the Illinois River in Okla- homa and Arkansas. Amer. Midl. Net. 44: 76-95. RivAs, Luis R. 1964. A reinterpretation of the concepts “sympatric” and “allopatric” MAYER, ERNST, G. with proposal of the additional terms “syntopic” and “allotopic”. Syst. Zool. 18(1): 42-43. SuTTkuS, RoyaL D. 1955. Notropis eury- zonus, a new cyprinid fish from the Chat- tahoochee River system of Georgia and Alabama. Tulane Stud. Zool. 3(5): 85- 100. THOMERSON, JAMIE E. 1965. Experimental hybridization, geographical variation and distribution of the closely related top- minnows, Fundulus notatus and Fundu- lus olivaceus. Ph.D. dissertation, Tulane University, unpublished. March 17, 1966 oe aad »> = oe as a7 > €4 a sph is ie > “Tn! a a Sa nin a 7 —_ Cs i ==. TULANE STUDIES IN ZOOLOGY VOLUMES 1 to 6, 1953-58 For contents and prices of these volumes see previous issues. VOLUME 7, 1959 1 An illustrated key to the crawfishes of Louisiana with a summary of their distri- bution within the State (Decapoda, Asta- cidae), by George Henry Penn, pp. 8-20 (April 28, 1959). Comparison of the chromatophorotropins of two crayfishes with special reference to electrophoretic behavior, by Milton Finger- man, pp. 21-30 (April 23, 1959) .........$0.60 A review of the seabasses of the genus Centropristes (Serranidae), by Rudolph J. Miller, pp. 33-68 (July 9, 1959) Digenetic trematodes of marine fishes from the Gulf of Panama and Bimini, British West Indies, by Franklin Sogandares-Ber- nal, pp. 69-117 (August 24, 1959) ........ Parasites of the commercial shrimps, Pen- aeus aztecus Ives, P. duorarum Burkenroad, and P. eetiferus (Linnaeus), by Dwayne ieee use, pp. 123-144 (October 19, eoeeeesece The larva of the oak toad, Bufo quercicus Holbrook, by BH. Peter Volpe and James L. Dobie, pp. 145-152 (October 19, 1959) .... Complete volume, including title page, 1.00 -60 table of contents and index (unbound) ....§2.65 VOLUME 8, 1960-61 Number 1 Studies on the backswimmers of Costa Rica (Hemiptera ; Notonectidae), by John L. De Abate, pp. 1-28 (April 29, 1960) ........ ; Three Ascocotyle complex trematodes (Het- erophyidae) encysted in fishes from Louisi- ana, including the description of a new genus, by Franklin Sogandares-Bernal and 1900) F. Bridgman, pp. 81-89 (October 28, Age and growth of the spot, Letostomus wanthurus Lacépéde, by Bangalore I. Sun- dararaj, pp. 41-62 (October 28, 1960) The breeding habits of the mole salaman- der, Ambystoma talpoideum (Holbrook), in southeastern Louisiana, Ad C. Robert Shoop, RP. 65-82 (December 2, 1960) alinity relations of some fishes Poe me . Ren- Aransas River, Texas, by William fro, pp. 83-91 (December 2, 1960) Ecology of the rice rat, Oryzomys palustris Harlan), on Breton Island, Gulf of Mexico, with a critique of the social stress theory by Norman C. Negus, Edwin Gould, an at K. Chipman, pp. 93-128 (May 10, eoereeeerseserseseseseseeeeseeseod A quantitative study of the movement of Paramecium caudatum and P. multimicro- nucleatum D. F. Sears, and Lila BHilve- back, pp. 127-189 (May 81, 1961 Nine digenetic trematodes from t e Altantic Coast of Panama, by Franklin Sogandares- Bernal and Lucy McAlister Sogandares, pp. Pose (MAY. S150 1961)i