IN F O R M A T IO N TO USERS This dissertation was produced fro m a m icrofilm copy o f the original docum ent. W hile the most advanced technological means to photograph and reproduce this docum ent have been used, the q u ality is heavily dependent upon the q u ality of th e original subm itted. Th e fo llo w in g explanation o f techniques is provided to help you markings or patterns which m ay appear on this reproduction. understand 1. T h e sign or "targ et" fo r pages apparently lacking from the docum ent photographed is "Missing Page{s|". If it was possible to obtain the missing page(s) or section, they are spliced into the film along w ith adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you com plete continuity. 2. When an image on the film is o b literated w ith a large round black m ark, it is an indication th at the photographer suspected th a t the copy may have m oved during exposure and thus cause a blurred image. Y ou w ill fin d a good image of th e page in the adjacent fram e. 3. When a map, drawing or chart, etc., was part of the m aterial being p h o to g ra p h e d th e photographer fo llo w ed a definite m ethod in "sectioning" the m aterial. It is custom ary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from le ft to right in equal sections w ith a small overlap. If necessary, sectioning is continued again — beginning below the firs t ro w and continuing on un til com plete. 4. T h e m ajority of users indicate th at the textual content is o f greatest value, however, a som ewhat higher q u ality reproduction could be m ade from "photographs" if essential to the understanding o f the dissertation. Silver prints o f "photographs" may be ordered at additional charge by w riting the O rder D epartm ent, giving th e catalog number, title, autho r and specific pages you wish reproduced. University M icrofilm s 300 North Z eeb R oad Ann Arbor, Michigan 48106 A Xerox Education C om pany I I 72 30,009 - MCGREGOR, Don L., 1936THE OSTRACODA OF GULL LAKE, MICHIGAN: SELECTED ASPECTS OF THEIR ECOLOGY, Michigan State University, Ph.D. , 1972 Limnology U n iv e rs ity M icro film s, A XEROX C om pany , A n n A rb o r, M ich ig an THE OSTRACODA OF GULL LAKE, MICHIGAN: SELECTED ASPECTS OF THEIR ECOLOGY By DON L. MCGREGOR A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1972 PLEASE NOTE: Some pages may indistinct Filmed as University Microfilms, have print. received. A Xerox Education Company ABSTRACT THE OSTRACODA OF GULL LAKE, MICHIGAN: SELECTED ASPECTS OF THEIR ECOLOGY By Don L. McGregor This study presents both qualitative and quantitative description of selected aspects of the ecology of the free-living ostracods In­ habiting littoral and sublittoral areas of Gull Lake, Michigan. Most data reported were collected over a period of one year. Twenty-four species representing 10 genera of the superfamllles Cypridacea, Darwlnulacea and Cytheracea were identified from samples totalling nearly 33,000 ostracods. Twelve species are reported from Michigan, and one from North America, for the first time. species are non-swimmers and seven are free-swlmmers; Seventeen of the latter, all but one were restricted to the littoral region (0-9 m ) . Almost 98 percent of the ostracods collected from sediment cores were found in the top five cm sediment interval. Most species of ostracods from Gull Lake reproduce sexually, all are dioecious and all but two are oviparous. Cypridopsis vid u a . Darwlnula stevensonl and I). pagliolii are known to be parthenogenetic and the last two species carry their young through embryologlcal development to the first, and occasionally second, instar. The maximum number of uterine eggs and/or young, recorded for different species from a sample of 7,533 females, ranged from three (Darwlnula pagliolii) to 69 (Candone crogmanlana). The maximum number of eggs and young Don L . McGregor reported for each species Is believed to be a more realistic estimate of the actual reproductive performance of most females than were the reported means. A majority of the free-swimming species reproduce during the summer months* but some have more than one generation per year. glaucus was found only from May to September. Cyprlnotus Many non-swlmmlng species, excepting those of Darwlnula and two species of Candona, re ­ produce during the Fall, Winter and Spring. had an annual reproductive cycle; Most species of Candona Candona ohloensls had an annual reproductive cycle In the littoral areas and a biannual cycle In the sublittoral. The reproductive cycles of free- and non-swlmmlng species exhibited little overlap. Ostracod diversity, In number of species and Individuals was greatest In the littoral areas of Gull Lake. Candona was the dominant genus, In number of species, and undoubtedly accounted for most of the standing crop biomass of non-swimming species. Free- swlmming species appeared to account for most of the standing crop biomass in the littoral areas during the summer months, whereas, in the late Fall and Winter non-swlmmlng species presumably were dominant in biomass and number of species. Statements regarding ostracod biomass, based upon visual observation, are supported by information on species size, composition, and density. For those species of Candona whose life histories were studied In sufficient detail, the proportion of males always was greatest during the initial phase of the reproductive cycle* Most evidence indicated that males have a shorter life span than females. Greater mortality rates of males resulting from predation by fish w a s strongly indicated by analyses of stomach and gut contents of the yellow perch, Perea Don L . McGregor flavescens. Percentages of males of two species of Candona found In the stomach contents of perch were much higher than those ever en­ countered in sediment core or dredge samples. Ostracods representing 13 species and six genera, were found in three species of fish, P_. f laves cens (yellow perch) , Lepomis macro chi rus (blueglll) , and Osmerus mordax (smelt) from Gull Lake. Detailed Information is presented on the life history, reproductive performance and cycle(s), temporal and spatial distribution, population density, size and age structure, sex and juvenile-adult ratio, pa r a ­ sites, and predators of most species of the ostracod fauna of Gull Lake. Literature reviewed on the above topics focuses attention on our limited knowledge of the biology of freshwater Ostracoda. ACKNOWLEDGMENTS Acknowledgment Is made to the Department of Zoology and the W. K. Kellogg Biological Station of Michigan State University for providing laboratory facilities, equipment and financial assistance throughout most of this investigation. Most of this research was supported by a National Institutes of Health predoctoral fellowship, 1-F1-GM-22, 970-01 and 02, from the Institute of General Medical Sciences to which I owe my sincere thanks. Appreciation is expressed to members of my Doctoral committee, D r s . T. Wayne Porter (Chairman), Marvin M. Hensley, Robert C. Ball, and Jane E. Smith for their council, encouragement and constructive criticism. I thank Dr. T. W. Porter, in particular, for generous financial assistance. I thank Drs. J. Whitfield Gibbons, Donald C. McNaught, Messrs. Richard G. Burbidge, Peter H. Rich, John L. Hesse, Sigurd Nelson, Robert L. Llppson, and Arthur Wlest for assistance in the field sampling. Temperature data on Gull Lake were provided by Dr. D. C. McNaught, and Messrs. R. G. Burbidge and J. L. Hesse. Dr. J. W. Gibbons and Mr. R. L. Llppson, respectively, collected Darwlnula stevenson1 from Mississippi and Florida. 1 am grateful to the late Professor Edward Ferguson, Jr., Lincoln University of Missouri for verifying my identifications of certain free-swimming ostracods from Gull Lake. Additional pro­ fessional and technical assistance was provided by Drs. Robert V. ii Keeling and David G. Darby, University of Michigan; Dr. Vernon W. Proctor, Texas Tech University; Dr. Iraja D. Pinto, Unlversldade do Rio Grande Do Sul, Porto Alegre, Brazil; Messrs. Eugene Howell, Sr., Eugene Howell, Jr., and Alfred Pshea, formerly of the H. K. Kellogg Biological Station. A prototype of the Glamacher frozen core sampler was kindly loaned to me by Drs. Matthew Hohn and Laverne Curry of Central Michigan University. I owe a debt of gratitude to Dr. Richard G. Wiegert, University of Georgia, for providing laboratory facilities and financial assistance during the last phase of this work. Sincere thanks go to Dr. Wiegert for many hours of enjoyable, rewarding discussion and for his unselfish assistance and kindness during my stay at the University of Georgia. I thank the Department of Zoology and Biological Sclencas Division of the University of Georgia for many services. The assistance of Mrs. Virginia Mullen also is acknowledged. Special thanks are due the students and faculty of the W. K. Kellogg Biological Station of Michigan State University. Drs. Robert G. Wetzel and George H. Lauff contributed to many aspects of this work that cannot be summarized here. 1 acknowledge Dr. R. G. Wetzel and Mr. Harold L. Allen for their kindness in translating several manu­ scripts . Mr. H. L. Allen and M t b . Dolores Johnson assisted, respectively, in the preparation of certain figures and tables and Mr. Arthur Wiest provided many services for which I am truly grateful. Thanks are extended to many of my colleagues for their stimulating ideas and helpful suggestions during the course of the investigation; am especially grateful to Drs. J. Whitfield Gibbons, David R. Osborne, Ronald J. Pfohl, and Donald W. Tinkle. Hi Mrs. Bernadette Henderson also I deserves my warmest personal thanks for years of considerate and faith­ ful assistance. Finally, I thank my wife, Geraldine, for generous assistance and for her many contributions to this study. Through her efforts I have been afforded the opportunity to work toward professional goals that are not punctuated simply by Ri.D. iv TABLE OF CONTENTS INTRODUCTION . ................................................... HISTORICAL REVIEW..................................................... Page 1 3 General........................................ 3 Genus Darwinula...................................... ......... 7 METHODS AND MATERIALS................................................ 14 Collecting M e t h o d s ............................................ 14 Separation Techniques. 15 .................. Dissection and Preservation.................................. 16 STUDY A R E A ............................................................ 19 RESULTS................................................................ 22 General........................................................ 22 Gull Lake Ostracoda......................... 23 Depth Distribution in Sediment.............. 26 Life History St u d i e s ......................................... 27 Darwinula stevensoni....... 27 Darwlnula pagliolii................ 54 Candona ohioensls................................... 60 Candona rawsoni....... 70 Candona acu t a . 77 .................. Candona inoplnata. . ................... 82 Candona elllptica................................... 89 Candona porter!. 96 ..... Candona aff. caudata............................... v 100 Page Other Non-Swimming Species.......................... 105 Free-Swimming Species...................... 110 Cypridopsie vidua............................... 110 Physocypria pustulosa.......................... 113 Cypria turner!.................................. 117 Other Free-Swimming Species............... 121 Fish Stomach Analyses......................................... 125 DISCUSSION............................................................. 129 Distribution................................................... 129 Substrate....................................................... 134 Reproduction, Development, Life History...................... 136 Dispersal....................................................... 143 Density......................................................... 144 Trophic Relations....................................... 148 Predators.............................. *....................... 151 Parasitism . . 154 SUMMARY AND CONCLUSIONS................................... 156 REFERENCES CITED................ 166 vi LIST OF TABLES Table 1. 2. 3. Page Ontogenetic development and relative size range of lnstars of Darwlnula stevensonl. Modified from Scheerer-Ostermeyer (1940, p. 366) in the style of Kesling (1951a, pp. 94-95).. 13 Percentage of adult and juvenile Darwlnula stevensonl from samples collected at depths of <3 and 3 meters in Gull Lake. 29 Percentage of adult and juvenile Darwlnula stevensonl from samples collected at depths of 6, 9 and 12 meters in Gull Lake........ .............................................. . 31 4. Estimates of the mean number of adult and juvenile Darwlnula stevensonl per square meter at depths of 3, 6 and 9 meters along Transect AB during the period from 1 June-1 August 1966 (A * adult; J = juvenile)............................... 35 5. Age structure and reproductive potential of adult Darwlnula stevensonl In Groups 1, 2, and 3. Samples collected along Transect AB at depths 3, 6 and 9 meters during the sumner of 1966. Group I includes only those adults having at least one first instar young with movable appendages; Group 2 Includes adults with young which have no movable appendages; Group 3 includes adults with no young or eggs free in the carapace. Percentages with totals more or less than 100 42 percent result from rounding to nearest tenth.............. 6. Mean number of first and second reproductive Beason adults per square meter calculated from data presented in Table 5. Density estimates for second reproductive season adults are given on second line; first line gives estimates for first reproductive season adults.................................. 46 7. Geographic and depth variation In the number of parasitized adult Darwlnula stevensonl collected along four transects on 4 November, 1965.............................................. 48 8. Seasonal, geographic, and depth variation in the number of parasitized adult Darwlnula stevensonl collected along Tran­ sect AB, Station GI, and other localities in Gull Lake..... 50 9. Parasites of Darwlnula stevensonl in Groups 1, 2 and 3, collected at depths of 3 and 6 meters along Transect AB, during the summer of 1966. Explanation of Group designa­ tions given In Table 5 ....................................... vii 53 Table 10. 11. 12. 13. Page Percentage of adults and average number of young in gravid females of Darvinula pagliolii collected at Station GI in Gull Lake. Depth is <3 meters.............................. 56 Variation in the number of parasitized adult Darwinula pagliolii collected at Station GI in Gull Lake............. 58 Percentage of adult and juvenile Candona ohioensis from Gull Lake samples collected at depths o f <3, 3, 6, 9, 12, 15 and 18 meters.............................................. 62 Percentage of males and females in adult Candona ohioensis and average number of eggs present in the uterus of adult females. Sample number and location corresponding to dates are presented in Table 12 .................................... 65 14. Estimates of the mean number of adult and juvenile Candona ohioensis per square meter on 5 August and 4 November 1965. August estimates from replicated samples along Transect AB; November estimates represent means of samples from four tran­ sects (A = adult; J « juvenile)............ .................. 66 15. Estimates of the mean number of adult and juvenile Candona ohioensis per square meter at depths of 3-18 meters along Transect AB from 1 July-1 August 1966 (A “ adult; J = juvenile)............................... 68 16. Percentage of adult and juvenile Candona ravsoni in samples 72 collected from Gull Lake at depths of <3 to 15 meters... 17. Sex ratios of adult Candona ravsoni and average number of eggs present in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 16............... ................................ . 18. Percentage of adult and juvenile Candona acuta in samples collected from depths of <3 to 15 meters in Gull Lak e ... 76 78 19. Sex ratio of adult Candona acuta and average number of eggs present in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 18. 80 20. Percentage of adult and juvenile Candona inopinata in samples collected from depths of <3 to 18 meters in Gull Lake .............................. 84 Sex ratio of adult Candona inopinata and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 2 0 ......................................... 87 21. viii Table 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Percentage of adult and juvenile Candona elllptica in samples collected from Gull Lake.............................. 91 Sex ratio of adult Candona elllptica and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 2 2 ........ 94 Percentage of adult and juvenile Candona porter1 in sam­ ples collected from GullLake................................. 97 Sex ratio of adult Candona porter! and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are pre­ sented in Table 2 4 ............................................ 99 Percentage of adult and juvenile Candona aff. caudata in samples collected from Gull Lake......... 101 Sex ratio of adult Candona aff, caudata and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are pre­ sented in Table 2 6 ................................. 104 Adult-juvenile ratio and reproductive state of adult Cypridopsis vidua in composite monthly samples from Gull Lake. Samples include individuals collected at depths of from <1 to 9 meters on different dates and from different localities.......................... , . 22. Page Percentage of adult and juvenile Kiysocypria pustulosa from Gull Lake. Samples represent composite monthly collections from different localities and depths........... 115 Sex ratio and reproductive cycle of adult Physocypria pustulosa collected from Gull Lake.......................... 116 Percentage of adult and juvenile Cypria turner! in com­ posite monthly samples collected from different localities and depths in Gull Lake...................................... 119 Sex ratio and reproductive state of Cypria turner! lected from Gull I k e 120 col­ Percentage of adult and juvenile Cyclocypris ampla in composite monthly samples from Gull Lake. Only 3 indi­ viduals from depths greater than 9 meters.................. 122 Sex ratio and juvenile-adult percentages of Cyprlnotus glaucus collected from Gull Lake............................ 124 ix 112 Table 35. 36. Page Sumnary of the number, sex, and percentage of male ostra­ cods removed from 15 yellow perch (Perea flavescens) col­ lected In Gull Lake on 9 January 1966............. .......... 128 Estimates of the number of ostracods per square meter In Gull Lake. Juveniles identified only to genus were Included in calculating densities. (See text for addi­ tional explanation)............................................ 146 x LIST OF FIGURES Figure 1. 2. 3. 4. Page Map of Gull Lake showing sample localities and general lake morphometry. Depth contour Interval is 9 meters 20 Annual reproductive cycle and reproductive potential of Darwinula stevensonl In Gull Lake. The total number of specimens examined each month, from left to right, are: Jan. (23), Feb. (79), Mar. (175), Apr. (30), May (83), June (217), July (475 and 20), Aug. (181), Aug. (324), Sept. (152), Oct. (252), Nov. (477), Dec. (99)........... 39 Differential rates of egg deposition and development of young along Transect AB depth gradient during the summer of 1966. Number of animals represented by histograms at each depth, on consecutive dates, may be found in Tables 2 and 3 ................... 39 Highly generalized reproductive cycles of numerically dominant species in Gull Lake.............................. xi 141 INTRODUCTION The Ostracoda are among the most widespread and common groups of Crustacea In aquatic environments and have one of the best documented geological histories of any Invertebrate group. Ironically, the ecology and most other aspects of the biology of freshwater ostracods are scarcely known compared to a wealth of Information available on other entomostracans. In recent years, increased attention has been focused on the tro­ phic structure, population dynamics and energetics of many groups of benthlc organisms in diverse types of aquatic ecosystems. Yet, when one peruses the literature relating to studies of this nature on the Ostracoda, the paucity of information is indeed striking. The small size of most freshwater ostracods, difficulty in making specific determi­ nations, and the general disregard of this group by benthic ecologists may account for the limited number of ecological studies on the fresh­ water Ostracoda. The objectives of the present w ork were 1) to investigate the life history and population dynamics of two or more species of Cull Lake Ostracoda and 2) to complete a taxonomic survey of the ostracod fauna of this lake. This study presents both qualitative and quantitative description of selected aspects of the ecology of free-living ostracods inhabiting the littoral and sublittoral areas of Gull Lake* Though limited in scope and based almost entirely upon field studies, this investigation represents a preliminary attempt to establish a basic 1 framework for identifying and formulating more detailed, experimental approaches to the study of ostracod populations. HISTORICAL REVIEW General Literature on fossil and Recant Ostracoda Is extremely voluminous but widely scattered in many journals throughout the world. No attempt is made to review the major publications on freshwater ostracods or even those on a given genus. Much of the Information of concern to ecologists Is published In taxonomic reports or In the paleontological literature. Freshwater ostracods often are reported In studies principally concerned with marine or bracklsh-water faunas or In the diverse array of ecological, limnologlcal, and oceanographical literature. Some papers cited In this study are related only indirectly to the problems Investigated. How­ ever, these publications have been Invaluable In directing me to other Important literature and in most every Instance have provided greater Insight to the more general biological and paleontological features of the Ostracoda from which future questions will be generated. Many problems exist In identifying and classifying both living and fossil freshwater ostracods. Zoologists have devised hlerarchial classi­ fications based largely upon appendage and reproductive structure mo r ­ phology. Detailed studies of the carapace have been limited. Paleon­ tologists, on the other hand, generally are obliged to scrutinize the carapace of ostracods in considerable detail, In the absence of preserved soft parts. Consequently, rather divergent hierarchic classifications are used in the study of fossil and living Ostracoda. 3 Much of the work 4 on living ostracods, however, has been compiled by paleontologists. In 1942, Hoff described the ostracod fauna of Illinois. He presented a review of pertinent European literature and a rather com­ prehensive discussion of earlier studies of freshwater ostracods to that date in North America. Later, the classic study by Kesling (1951a) contributed additional information in this area and provided the most thorough description of the anatomy and biology of a freshwater ostracod that has been made in the United States. Because much of the classic and other important literature has been reviewed by the above authors the primary Intent here is to cite those studies that have been most useful in this Investigation. Addi­ tional studies are reviewed, where appropriate, in subsequent sections of this thesis. During the taxonomic phase of this study I relied heavily on the works of the following authors: Aim (1915), Brady and Norman (1889), Delorme (1967a, 1967b, 1967c), Dobbin (1941), Ferguson (1957 to 1968), Furtos (1933, 1935, 1936a, 1936b), Hoff (1942), Klie (1938), MUller (1900, 1912), Pinto and Sanguinettl (1956), Pinto and Kotzlan (1961), Sars (1922-1928), Sharpe (1918), Swain (1962), Tressler (1959), and Turner (1895). Studies by Hartmann (1964, 1965a, 1965b) and Kile (1935a, 1933, 1939) also were useful. The following general references were used extensively in many aspects of this study and are recommended to be­ ginning students, non-specialists, and others desiring detailed or general information on many aspects of the biology and paleontology of living and fossil Ostracoda: Howe (1962), Moore (1961), and Van Morkhoven (1962, 1963). A taxonomic study usually requires much more than simply a thorough understanding of the anatomical features used for 5 Identification. In fact* many taxonomic works assume some acquaintance with the principal biological features of a given group of organisms and cannot be used or readily interpreted without previous reference to such related literature. The studies on carapace terminology (Kesling 1951 b ) , carapace structure (Keeling 1951a; Sylvester-Bradley 1941), muscle scars (Van Morlchoven 1963; cuticle development Smith 1965; Benson 1967), (Harding 1965), anatomy, morphology, histology, (Kesling 1951a, 1957, 1965, Smith 1965; Rome 1947a; Cannon 1926, 1931, 1940), and larval development (Schrelber 1922; Scheerer-Ostenoeyer 1940; Kesling 1951a, 1953; Fox 1964a) have been indespensable aids. Paleontological literature has been consulted in many Instances for taxonomic and paleoecologlc Information. Although the fossil records of most Gull Lake Ostracoda are not considered In detail at this time, the value of the following works is not to be minimized (e.g., Trlebel 1941; Swain 1947; Winkler 1960, 1962; Dlebel 1961; Gutentag and Benson 1962; Benson and MacDonald 1963; Staplln 1963a, 1963b; Grundel 1964; Gutentag 1964; Frey 1964; Luttlg 1965; Schmidt and Sellman 1966; Benson 1967; Klassen et al. 1967; and Delorme 1968). Ecological studies on freshwater ostracods are limited In number and most present few data that are directly comparable wit h those data presented here. The Investigations of Hoff (1942, 1943a, 1943b), Ferguson (1944, 1958b), Moore relevant to the present work. (1939), and Kenk (1949) perhaps are most Other studies (i.e., Ai m 1915; Pinto and Purper 1965; Neale 1965; Latour et al. 1966; Sywula 1966; de Vos 1954) have provided data of a more general nature as pointed out In subsequent discussion. Numerous authors, e.g., Benson (1959, 1966), Elofson (1941), Hagerman (1966, 1967), Hullngs (1967), lies (1953), Kornicker (1959, 6 1961, 1963, 1965), Kornicker and Wise (1960), McHardy and Bary (1965), McKenzie (1965), Purl (1966), Puri et al. (1965), Keys (1963), Theisen (1966), and Tressler (1948, 1957b) have reviewed work, or have investi­ gated various aspects of the ecology of benthonlc and planktonlc Ostracoda. Most of these studies are concerned w i t h marine and brackish- water ostracods but provide Information pertinent to several general biological features of the Ostracoda (refer to discussion). The symposium edited by Purl (1965) represents a particularly valuable reference on the ecology and paleoecology of ostracods. Several investigators have reported ostracods from Michigan during the past 30 years. Moore (1939) Investigated the ostracod fauna of Douglas Lake and found 11 species. Later, Tressler (1947) published a checklist of Michigan Ostracoda which included 14 species be­ longing to 10 genera. Hinkler Kenk (1949), Cole (1953), Ferguson (1957), and (1960), also reviewed the distribution of numerous living and fossil ostracods. In 1959 Tressler published ostracods of the United States and listed 18 Michigan. Tressler's its preparation; a taxonomic key for the species from the state of (1959) key was published several years following thus, the checklists of Ferguson (1957, 1958c) re ­ presented the most complete faunal lists for Michigan ostracods through the year 1959. In 1968, Ferguson listed United States dis­ tribution records for 91 new species and 31 previously described species for the period from 1957 to 1966. No additional species were recorded for Michigan. Both Tressler (1959) and Ferguson (1957, 1958c) omitted certain species from their respective distribution records that were recorded in these publications. The following list represents the ostracod fauna of Michigan recorded by Tressler and Ferguson through 1959. 7 Darwlnula stevensonl (Brady and Robertson) Eucyprls crasaa (O.F. Muller) Eucvprls fuscata (Jurlne) Candona crogmaniana Turner Candona axilla Furtos Candona alnoaonl Sharpe (- _C, reflexa Sharpe) Candona Indigene Hoff Candona decora Furtos Candona truncate Furtos Cypria t u m e r l Hoff Cypria obeaa Sharpe Cypria opthalmlca (Jurlne) Cypria lacustrla Sara Cypria subglobosa Sowerby Physocyprla puatuloaa (Sharpe) Cvclocyprla serene (Koch) Cyprldopsla obese Brady and Robertson Cyprldopsis vidua (0. F. MUller) Cyprleercue tuberculatus (Sharpe) Cyprlcercua tlncta Furtos Cyprlcercus cheboyganensla Ferguson Cyprtnotus Incongruens (Ramdohr) Cyprols marglnata (Strauss) Llmnocythare sancti-patrlcl (Brady and Robertson) Recent Investigations by McGregor Wetzel (1968), McGregor and Kesling (1967, 1969), McGregor and (1969a, 1969b) and Wetzel and McGregor (1968) report the only published records of Gull Lake Ostracoda. These studies, excepting McGregor (1969), involved research topics other than those discussed here, and are not considered in further detail. This thesis presents the first and only complete record of the ostracods of Gull Lake. „ Genus Darwlnula A significant portion of the time and effort devoted to ful­ filling the ecological objective of this thesis involved a study of the ostracod Darwlnula stevensonl spread distribution, (Brady and Robertson)^. The w i d e ­ unique morphological characteristics, geologic somewhat revised version of the study on IK stevensonl has been published (McGregor 1969). This study was, however, an Integral part of the work reported here and Is Included for the purpose of complete­ ness . 8 history, parthenogenetic reproduction, retention of young through the first instar, and high incidence of parasitism in this species afforded the opportunity to investigate eeveral ecological aspects of the type species of the genus Darwinula. of potential interest to both zool­ ogists and paleontologists. Much of the historical information on this species is scattered throughout the world literature. I believe such information it most useful In this section of the thesis in contrast to that presented for other species in subsequent pages. Darwlnula stevensonl and other species of this genus are unique among the freshwater Ostracoda because they are the only freshwater representatives in the United States which retain the young through embryologlcal development to the first instar. Members of this genus are characterized by weakly calcified, unornamented valves, with muscle scars arranged in a rosette pattern differing from that found in all other extant or fossil Ostracoda. The furca is rudimentary, and the morphologically similar second and third pairs of thoracic appendages are dissimilar to the first pair. Additional morphologic and taxonomic characteristics of the genua Darwinula may be found In Swain (1961), Howe (1962), Van Morkhoven Brady (1963), and Hartmann (1965b). (1870) was the first to report on Darwinula stevensoni (Brady and Robertson). He figured the carapace and mentioned that the species would be named and described later in collaboration with David Robertson. Folycheles stevensonl was described 1870) later the same year, but was changed (Brady and Robertson, (Brady and Robertson, 1872) to Darwlnella stevensonl due to preoccupation of the former name. Descriptions based upon dried specimens in the 1872 publication were later emended (Brady and Robertson, 1874). Darwlnella. however, also was preoccupied and the name Darwinula was erected in 1885 (Brady and 9 Robertson). In 1889, Brady and Norman presented another description and designated Darwlnula stavensonl as tha genotype. Subsequent descriptions of D. stevansonl or Its synonyma have been given by various workers Including Vavra Aim (1915), Sars Swain and Gilby (1922-28), Turner (1909), Muller (1895), Furtos (1965), and Staplln (1963). (1912), (1933), Hoff (1942), Finto and Sangulnetti (1958) presented a redascrlptlon of the genotype of Darvinula in order to clarify some of the nomenclatural problems In the literature. Darwlnula stevensonl is one of the most widely distributed species of freshwater ostracods. Some of the distribution records for D. stevensoni are summarized below: United States Ohio Illinois Michigan Massachusetts Georgia Tennessee Kentucky Mississippi Florida South Carolina Virginia Texas Mexico Nicaragua Great Britain Austria Germany France Sweden Norway Switzerland The Netherlands Italy Poland Yugoslavia Turkey U.S.S.R. Furtos (1933) Hoff (1942) Tressler (1947), Moore (1939), McGregor and Wetzel (1968), McGregor (1969) Furtos (1935) Turner (1895) Cole (1966) Cole (1966) McGregor (1969) McGregor (this thesis) McGregor (this thesis) Elliott et al. (1966), Nichols and Ellison (1967) Cole (1966) Furtos (1936b) Swain and Gilby (1965), Hartmann (1959, from Swain and Gilby, 1965) Brady (1870), Brady and Robertson (1872), Brady and Norman (1889) Loffler (1963) Kile (1938) Sars (1922-28) Ai m (1915), Sars (1922-28) Sars (1922-28) Sars (1922-28) de Vos (1954) Fox (1965, 1966) Sywula (1965) Kile (1941), Petkovski (1960, 1961), Stan­ kov!^ (I960) Hartmann (1964) Bronsteln (1947), Hartmann (1964) 10 Iran Sumatra Java Africa Hartmann (1964) Klie (1933) Kile (1933) Hartmann (1964), Klie (1935a, 1939) The type locality of Darwlnula stevenaoni Is the East Anglian Fen District In England (Brady and Norman, 1889). Brady (1870) first reported D, stevenaoni from the Meuse and Scheldt rivers of England and Holland, respectively. Anglian Fen District Additional distribution records for the East (Brady and Robertson, 1870, 1872) were summarized by Brady and Norman (1889) . Include: Other areas inhabited by _D. stevenaoni Lakes Ohrld, Prespa, Dojran (Yugoslavia); Magglore, Mergozzo, Varese, Comabbio, Monate (Italy); IJssel (The Netherlands); Van (Norway); Pereslavskoye (U.S.S.R.); Ngebel, Bedall (Java); Toba (Sumatra); Lothing and many others (Great Britain); Nicaragua, Managua (Nicaragua); Erie, Douglas, Gull (United States). Several river habitats Include the Nene, Cam, Ouse, Deben (Great Britain), Warta (Poland) and Rappahannock (United States). The seasonal and spatial distribution of I), stevenaoni has received limited study. Aim (1915, p. 225) stated that this species should be found throughout the entire year on the basis of its 'vivipa­ rous* reproduction. Pinto and Purper (1965) illustrated the seasonal occurrence of Darwlnula in a recent study, but did not specify the species. Moore (1939) collected _D. stevensoni at depths of from 4-9 meters in Douglas Lake, Michigan. Klie (1933) reported specimens from depths of approxi­ mately 4 meters and Cole (1966) found them at depths of about 0.3 meter. 1 have collected adults and late instar juveniles at depths ranging from a few millimeters, near Saline, Michigan, to 12 meters in Gull Lake. 11 The geologic range of the genus Darwlnula extends from the Upper Carboniferous (Pennsylvanian) to Recent according to Swain (1961) and Van Morkhoven (1963). Swain (1961) also reported a questionable record from the Ordovician Period. et al. (1874), Wagner (1957), Several authors, notably, Brady and Staplin (1963b) have recorded I). stevenaoni from Quaternary deposits. Other extant species of Darwlnula have been reported from Romania (Danielopol and Vespremeanu, 1961), Africa Brazil (Pinto and Kotzian, (Klie, 1935a, 1939; Menzel, 1916), Java (Menzel, 1923; Klie, 1933), Patagonia (Delachaux, 1964), (Daday, 1902), Uruguay (Kile, 1935b), Holland 1928), and Rennell Island, Pacific Ocean (Harding, 1962). Sohn (1965), Pinto and Purper (1965), and Sandberg (1965) reported Darwlnula, respectively, from Lake Tiberias, Israel, Brazil, and Mexico, but did not mention the species encountered. have been described, Jurassic for example, from the Miocene Fossil species (Mehes, 1908), and (Bate, 1967). Darwlnula Stevenson! Is presumably a parthenogenetic species. Brady and Robertson (1870) briefly described and figured the copulatory apparatus of a male specimen, and later restated the description (18/4). Brady and Norman (1889) Incorporated the above description and figure in their monograph on the marine and freshwater Ostracoda of the North Atlantic and North-western Europe. Male characteristics other than the copulatory apparatus were not described in these papers nor have any subsequent Investigators reported finding male D. stevensoni. I have Isolated and reared numerous late instar individuals through at least one reproductive cycle and on all occasions only females were produced. In addition, no males have been encountered in more than 10,000 specimens collected from Gull Lake and other 12 localities in the present study. and Robertson The validity of the record by Brady (1870) must be considered highly suspect until adequate evidence is provided to prove the existence of males in this species. In 1953 Kesling noted some of the causes for variation in form within an instar, such as diet, effects of temperature, sexual dimorphism and individual variation. He also stated (1953, p. 101) that: The number of instars in the ontogeny of an ostracod appears to be constant for a genus. Insofar as is known, it is constant for certain families, but re­ latively few species have been studied and there may be exceptions. Later, Sandberg (1964) reviewed several systems of instar designation used by different authors and some of the problems encountered in their use when working with living and fossil species. In the present dis­ cussion the first instar is number one and later instars are numbered consecutively. The ontogenetic development of Darwlnula stevensoni has been discussed by Scheerer-Ostermeyer (1940). Some of the apparent dif­ ferences In the ontogenetic development of _D. stevensoni from that reported in other freshwater ostracods may be due, as Kesling noted (1951a, p. 94), to different interpretations of appendages. Because the principle concern here is the total number and relative size of the instar stages in this species, the sequence of ontogenetic development and relative size described by Scheerer-Ostermeyer (1940, p. 366) is presented in Table 1. pp. 94-95) I have followed the style of Kesling (1951a, in portraying the developmental process of D. stevensoni described by Scheerer-Ostermeyer necessary (1940). Some modification was owing to a different Interpretation of appendages. 13 Table 1. Ontogenetic development end relative size range of lnatars of Darwlnula stevensoni. Modified from Scheerer-Ostermeyer <1940, p. 366) In the style of Kesling (1951a, pp. 94 -95). Length Height ______ Instar______________________________________ (microns )_____(microns) 1 2 3 4 5 6 7 8 9 A1 A1 A1 A1 A1 A1 A1 A1 A1 An An An An An An An An An A1 An ~ Md * () “ (R)(L)“ Md Md Md Md Md Md Md Md Md Mx Mx til1 ) M x (LI) M x (LI) Mx LI (L2) L2 Mx LI L2 Mx Ll Mx LI L2 Mx Ll L2 (L3) L3 L3 L3 Antennuie Antenna Mandible Anlagen, Incomplete Right valve Left valve (Rp) Rp (Fc) (Fc) (Fc) (Fc) (Fc) (Fc) (Fc) (Fc) Mx LI L2 L3 Fc Rp 160-175 223(R)-217(L) 250 (R) 290(R)-270(L) 350(R)-320(L) 440(R)- 4 3 1 (L) 518(R)- 5 0 8 (L) 700 « 90-100 1 2 6 (R) 1 3 3 (R) 130 (R) 168 (R) 2 4 5 (R) 2 5 0 (R) Maxilla First thoracic leg Second thoracic leg Third thoracic leg Furca Reproductive organs METHODS AND MATERIALS Collecting Methods An Ekman dredge, free-fall corer with removable plastic Innerliner, modified Glemacher frozen core sampler (Gleason and Ohlmacher, 1965), and a modified aquatic sweep net were used for sample collection. The modified aquatic net was used in areas suitable for wading or for manipulation from a boat. Ekman samples generally were collected from depths of three meters or greater. Transect and individual sample localities are shown in Figure 1. Sediment cores obtained using the free-fall corer were removed from a plastic inner-llner, A.3 cm in diameter. This corer was allowed to free-fall from approximately 3 meters above the sediment-water inter­ face. The liner was marked for separation of the top 5 cm of sediment from the underlying 10 cm sediment Interval. Retrieval of cores was dependent upon a clay plug to hold the sediment in the liner. After removing the liner, water was siphoned off to a point about A cm above the substrete-water Interface and the core was allowed to slide slowly out the coring end. Sediment below the 15 cm Interval was discarded and that between the 5 cm and the 15 cm interval was collected in a labelled container. The upper 5 cm of sediment, and the water above, were collected in a second container. Both samples were immediately frozen in dry ice and were later thawed and sieved in the laboratory. The modified Glemacher frozen core sampler employs a mixture of dry ice and acetone in the freezing compartment for in situ freezing of 1A IS the sediment. The dry ice and acetone, separated prior to penetration of the sediment, are released by a lever and spring tripping mechanism when a messenger is lowered. Complete freezing of the 3 cm diameter sediment core occurs within three minutes following messenger release. The frozen cores, ranging in length from 18-23 cm, were sectioned at 5 cm intervals and sieved in the laboratory. All Ekman and net samples were sieved in the field using the modified aquatic net. This net accommodated a second, finer mesh, detach­ able nylon net for retention of ostracods. Samples were examined in the laboratory immediately following collecting or stored temporarily at 4°C. No preservatives were added as separation and examination of living animals proved to be more satisfactory. Additional methods of collecting ostracods included stomach and gut analyses of several species of fish from Gull Lake. This method occasionally yielded rather large numbers of identifiable ostracods. Crayfish also were collected and examined for conmensal ostracods of the family Entocytherldae. this thesis. The latter, however, are not considered in Numerous samples were collected by hand or in light traps for taxonomic purposes. Separation Techniques Samples were hand sorted and all non-swimming ostracods visible to the unaided eye were separated from the sediment by employing the water surface tension technique. to air; The sediment is drained and exposed the non^wettable ostracod valves are held by the water surface tension when the sediment is again Immersed. sample was first hand sorted and then On several occasions, a sorted again using other techniques, including treatment with weak acids and bases, differential heating, 16 saturated sugar and salt solutions, wet sieving through graded sieve series, behavioral responses to light quality and quantity, elutriatlon, and microscopic examination of sediment. The surface tension technique works extremely well for quantita­ tive separation of smooth shelled, non-swimming ostracods, such as Darwlnula and Candona. However, ostracods which have highly ornamented and sculptured valves usually cannot be separated adequately, at least quantitatively, from the sediment. free-swimming ostracods. This is also true for most of the The latter often are caught in the water sur­ face film, but usually manage to free themselves through rapid movements of their antennules. No attempt was made to quantify the numbers of free—swimming and highly ornamented burrowing ostracods in most samples. However, small numbers of these animals were removed for investigation of reproductive cycles, sex ratios and other life history characteristics. Collecting methods and organlsm-sedlment separation techniques also have been described by other authors (e.g., Kesling 1956; Tressler 1959; Van Morkhoven 1962; Fox 1964b; Delorme 1967b). Dissection and Preservation Most of the above sorting techniques induced gravid females to release their young or eggs and were, therefore, unsatisfactory. Screen­ ing or hand sorting ostracods held in the water surface film facilitated removal of females, with eggs and young intact, and provided many oppor­ tunities for observing various behavioral traits of different species. The ostracods were sorted beneath a dissecting microscope to remove the remaining debris and other organisms. Adult and juvenile Darwlnula were isolated from other species and placed in another container. 17 The former were separated Into Blze categories, counted, anesthetized with Tricalne Hethanesulfonate (MS 222) and dissected under highest power of a dissecting microscope. All other species were treated In the manner above but were killed with a 1:1, by volume, solution of 95% ethyl alcohol and water. Individuals In each of these species were separated Into juvenile and adult categories, counted and preserved. The preservative contained 70:20:10 parts, respectively, of 95% ethyl alcohol, distilled water, and glycerol. The number of young or eggs within the carapace, parasites, developmental stage of the young, and several other observations were recorded for each Individual of Darwlnula stevenaoni and D. pattllolli Finto and Kotzlan. Later, Individuals of all other species were cleared in glycerine. The sex and reproductive state of each specimen and the number of eggs within the uterus of females were recorded. Counts were made In trans­ mitted and reflected light for females with few eggs; however, several thousand females required dissection before accurate counts could be made. Several hundred whole mounts and needle dissections suitable for permanent mounts of Gull Lake Ostracods were prepared for taxonomic study. Approximately 1,000 ostracods were dissected in temporary glycerine mounts, but most were used for purposes not discussed hare. Permanent mounts were made in balsam, diaphane, and CMC-S non-resinous stain mountant. The latter, distributed through the General Biological Supply Incorporated, Chicago, Illinois, is an excellent medium for needle dissections of ostracods (refer to Darby, 1965). Minuten nadaln Insect pins mounted on long, slender wooden rods were used In 18 most needle dissections. Many details concerning techniques and procedures for dissection, preservation, staining, serial sectioning and photography of freshwater ostracods have been expertly described by Smith (1965). and useful techniques are discussed by Hoff Kesling (1956), Fox (1964b), and Darby Other interesting (1962), Ferguson (1965). (1946) STUDY AREA Gull Lake (Fig. 1) Is located In the southwestern corner of the Southern Peninsula of Michigan, in Kalamazoo and Barry counties, between latitudes 42*20*-42*30* N and longitudes 85*20'-85*30* W. The lake basin, of glacial ice block origin, was formed during retreat of the Wisconsin glacier which had covered Barry and Kalamazoo counties (Martin, 1957). The lake drainage basin is small and most water comes from springs and a few small streams. Gull Lake drains into the Kalamazoo River which, in turn, empties into Lake Michigan. The lake covers an area of 821*5 hectares and has a maximum depth of 33.5 meters Bacon, 1952). (Taube and Major physiographic features, morainic highlands and dissected glacial outwash plains in the area surrounding Gull Lake, reflect the impact of the recession of the Wisconsin glacier (Deutsch et a l . , 1960). The regional soils are the well-drained, Grey-Brown Podzolic or Brunizem Great Soil Groups described by Whiteside et al. (1963). Preliminary studies on nutrient cycling and primary production rates in Gull Lake have been initiated onlj recently and no detailed chemical or physical measurements have been published. Yet, the lake may be classified generally as oligotrophlc, possibly mesotrophlc, on the basis of its physiographic and chemical-physical properties, when compared with other lakes nearby. Gull Lake is well oxygenated at all depths throughout the year 19 20 GULL LAKE co. KALAMAZOO CO. INDEX M AP Figure 1. SCALE M C T C M Map of Gull Lake shoving sample localities and general lake morphometry. Depth contour interval la 9 meters. 21 except In the deepest portions of the hypolymnion during the latter phase of summer and fall stratification. Alkalinity, measured as mg/I CaCO^, generally ranges between 140-160, and the pH, between 7.5-8.5- The lake stratifies thermally by late May or early June and begins the fall turnover sometime in November. By mid-January the lake usually freezes over and remains ice-covered through most of March and, occasionally, early April. Secchl disc readings reach a maximum of about 10 meters and calcium carbonate deposits cover most of the littoral areas. Steep slopes on all sides of the lake reflect its glacial origin and a majority of the surface area Is more than nine meters in depth. RESULTS General Most data reported below represent studies undertaken during the period from August, 1965 to August, 1966. dates are noted In the text. Data collected on other All samples were collected at depths ranging from 0-18 meters and the majority from 0-12 meters. A total of 262 samples collected on 38 dates were analyzed. The former total Includes 142 Ekman, 84 aquatic net, 25 free-fall core, 7 frozen core and 4 fish samples. The few samples representing other collecting periods are not Included In the above total. About 150 additional samples from Gull Lake and other areas were used In studies not reported here. Most of the sampling was accomplished without assistance and often was dependent on equipment available for use, wind velocity over Gull Lake, or character of ice cover. Perhaps the most serious problem was the limitation on sampling at scheduled Intervals due to the generally unpredictable weather In the study area. A total of 32,726 free-llvlng ostracods were collected from Gull Lake. Excluding ostracods collected from fish stomachs, 23,325 adults and juveniles were Identified to species; of the genus Candona 7,765 early instar Individuals could not be Identified to species, at least with certainty, and are used only for density estimates discussed later. Most of the 1,636 ostracods retrieved from fish stomachs were Identified to species. About 50,000 ostracods collected before and 22 23 subsequent to the present work have been preserved for later study or used In Investigations cited above. Aliquot samples were examined from large samples of certain freeswlmning species. Therefore, adult-juvenile percentages were cal­ culated from a total of 21,685 Individuals. Of 11,835 adults In the latter, the sex of 11,060 Individuals was determined. The number of uterine eggs or young carried In the carapace were recorded for 7,533 females. Parasites were investigated In 4,192 Individuals of the genus Darwlnula. Gull Lake Ostracoda Twenty-four species of ostracods, excluding the commensal Entocytheridae, were Identified from Gull Lake. depths reported above. All were collected from Seventeen species are non-swimmers, restricted to burrowing or crawling over the substrate and rooted aquatic plants. Seven species are free-swlmaers, but remain near the substrate and only rarely are found In plankton samples. The latter also are capable burrowers. The following species were found In Gull Lake: Darwlnula stevensoni (Brady and Robertson) Darwlnula pagliolii Pinto and Kotzlan Candona ohioensis Furtos Candona rawsonl Tressler Candona acuta Hoff Candona Inopinata Furtos Candona e3liptlca Furtos Candona porter! sp. nov. Candona a f f . caudata Kaufmann 2U Candona scopuloaa Furtos Candona crogmaniana Turner Candona a f f . guburbana Hoff Candona decora Furtoa Candona distincta Furtos Candona sp. Cypridopsia vidua (0. F. Muller) Fhysocyprla puatuloaa (Sharpe) Cyprla turner1 Hoff Cyclocyprla amp la Furtos Cyclocyprla cruciate Furtos Cyprlnotua glaucua Furtos Potamocyprls amaragdlna (V^vra) Llmnocythere verrucosa Hoff Cytherlssa lacustrls (Sars) Only one of the above species, Candona porterl ap. n o v . , la undescribed. This species will be properly described, along with emen­ dations of certain other species, in a future report. Darwlnula pagliolii is reported for the first time in North America and re­ presents one of the two species of the family Darwinulldae known on this continent. Living representatives of Candona acuta, C.. aff. caudate, C. scopuloaa, elliptica, aff. suburbans. £. distincta, Candona sp., Cyclocyprla cruciata. Cyprlnotus glaucua. Potamocyprls amaragdlna. Llmnocythere verrucosa and Cytherlssa lacustrls are reported from Michigan for the first time. Candona ohioensis. C. rawsoni, C^. inoplnata and Cyclocyprla ample ware recorded previously in reports by this writer and others. 25 Species of Che genera Darwlnula and Candona were the only species quantitatively sampled In this study. and Cytherlssa are non-swimmers; Darwlnula. Candona. Llwnocythere however, the latter two genera are characterized by sculptured, variously ornamented valves unsulted for the ostracod separation technique described earlier. Species belong­ ing to the remaining genera are free-swlmmers. Candona Is among the most dominant genera of ostracods In number of species and individuals in most freshwater habitats. Because of the large number of species (over 200 ) presently assigned to this genus, however, identification of Individual species Is extremely difficult. Sexual dimorphism Is pronounced In most species of Candona: the valves and certain appendages on opposite sides of the body often differ in one individual more than they differ from the same structures in closely related species. cluding the adult instar. This genus has nine Instar stages In­ Thus, considering the valves alone, one Is confronted with 36 different characteristic sizes and shapes for a single species. There are numerous taxonomlcally important character­ istics of the carapace. In addition to a large number of Internal structures, used for identification. The point here is that much information of potential value in a life history or population study (i.e., age and size structure, growth rate, mortality, sex ratio and other characteristics) is lost when one neglects or cannot identify immature Individuals of a given species. I regret that such is true in this study for the earliest Instars of the genus Candona. Perhaps one of the benefits of this study has been recognition of certain genera and species of ostracods which have life history characteristics suited for more detailed pop­ ulation studies than accomplished here. 26 Depth Distribution In Sediment Seven sediment cores vere collected in situ, employing the modified Glemecher frozen core sampler* at a depth of 5 meters along Transect MN on 15 October* 1965. One of the 71 ostracods recovered was below the top 5 cm. On 19 January, 1966, five free-fall cores ranging In length from 34-42 cm were collected at depths of 9-15 meters along Transect IJ. Each core was sectioned at 5 cm intervals; the top 5 cm of these cores. 25 ostracods were found in Later, on 5 June, 1966, 20 free-fall sediment cores more than 20 cm in length were collected at depths of 3, 6 , 9, 12, and 15 meters along Transect AB. from only one core. Ostracods were absent Sixty— five adult and juvenile ostracods, represent­ ing 10 species and 6 genera, were found in the upper 5 cm of sediment; 3 were below the top 5 cm. The latter, however, were in two cores which had been labelled for possible contamination of the 5-15 cm sediment interval. The species collected from sediment cores were Darwlnula stevensoni. Candona ohioansis. elliptica. inopinata, C^. crogmaniana. £. aff. caudata. porter!. Cypridopsis v i d u a . Cypria turner!. Physocypria pustuloaa. and Cyclocypris ampla. Of 164 ostracods removed from the free-fall and frozen core samples only 4 or 2.4 percent were found below the top 5 cm of sediment. These data Indicate that many species of Gull Lake Ostracoda normally burrow in sediment at depths less than 5 cm and may be sampled adequately with the Ekman dredge or other bottom dredges suitable for use in loosely consolidated sediments. Although no data are available on the vertical distribution of other Gull Lake species, I do not believe it unreasonable to assume 27 their burrowing habits are similar to the above species. Life History Studies Darwinula stevensoni A total of 6,274 adult and juvenile Darwlnula stevensoni from Gull Lake and nearby localities we r e studied. The number of speci­ mens examined for parasites or young is summarized below: Total No. No. ReproParasites Examined Adults Juveniles ductlve Adults Juveniles ____________________________________________________ Potential____________________ Gull Lake 5993 2906 3087 2590 2169 1879 Other Areas 281 68 213 53 53 84 This species was present in eighty— seven Ekman, 15 core, and 22 net samples from Gull Lake, and 7 net and hand samples from other areas. Only one adult of 12 juvenile and 10 adult I), stevensoni present in cores obtained using the free-fall corer was found at depths below 5 cm. This specimen, however, was in one of two cores which previously had been labelled for possible contamination. Three cores at 3 meters, four cores at 6 meters and one at 9 meters contained D. stevensonl. Thirteen adults and twenty-seven juveniles were present in the frozen core samples; one juvenile was found below the 5 cm Interval. The secondary nylon monofilament screen cloth used with the modified aquatic net had a mesh opening of 202 microns. Consequently, only those instars with length and height dimensions greater than 202 microns were retained. Good agreement with expected instar development end size was observed in all juvenile samples. eighth and adult instars were present; Generally, only the seventh, earlier Instar Individuals always comprised less than one percent of the sample. Measurements of a limited number of eighth and ninth Instar animals were somewhat 28 greater than those reported by Scheerer-Ostermeyer (1940), but were in general agreement. Further discussion of juvenile-adult per­ centages will be in reference to the seventh and eighth juvenile instars and the ninth or adult instar. Darwinula stevensoni was abundant in the littoral areas of Gull Lake and reached a maximum density at a depth of about 6 meters through­ out the year. The population density decreased markedly between 6 and 9 meters and only a few individuals were found at depths greater than 9 meters. Although an unequal number of samples were collected from different depths the majority of specimens were from depths of 3 and 6 meters. Variation in number of individuals sampled at a given depth and date, or at different times during one month, probably represented a reasonable degree of sample bias. Much of the variability, however, reflected the geographic variation in population density and the number of late instar juveniles that had reached the size necessary for retention by the collecting process discussed above. The percentages of adult and juvenile ostracods in replicated and non-replicated samples from Gull Lake at depths of <3, 3, 6 , 9, and 12 meters are presented in Tables 2 and 3. At <3, 3, and 6 meters the per­ centage of adults was maximal during the sumner months and was followed by a sharp decline in the early fall. The percentage of adults con­ tinued to decrease through the fall and winter seasons and generally reached a minimum by early or late spring depending upon the depth from which samples were collected. More than 90 per cent of the total number of ostracods collected from all depths, were adults. in August of 1965-66, The adult component of most samples collected at depths of <3, 3, and 6 meters during late March, 1966 ranged between about 29 Table 2. Percentage of adult and juvenile Darwlnula stevensoni from samples collected at depths of < 3 and 3 meters In Gull Lake, Sample No. Date X Transect N Adult X Juvenile < 3 meters 1965 7- 9 7-19 8- 2 8- 2 8- 3 8- 3 9-14 9-14 9-14 9-22 9-27 10- 4 10-13 10-18 10-18 11- 1 12- 2 1966 1- 4 1-11 2-11 2-21 2-24 3-21 3-26 3-26 5-31 6-15 7- 2 33 36 39 40 44 45 61 62 60,64 65 66 70,71 72,73 78 79,80 81 102 109 111 117 124 129 134,135 138,139 140,141 172 186 207 KL MN KL GI GI GI GI GI GI GI GI GI A£ AB AB AB GI GI AB 42 4 8 49 2 43 13 8 99 86 22 164 64 105 98 75 46 100.0 100 . 0 100 .0 98.0 50.0 95.4 100 . 0 37.5 77.8 62.8 54.5 57. 3 60. 9 60. 0 55.1 49. 3 58. 7 37.2 45.5 42.7 39.1 40.0 44.9 50.7 41. 3 18 24 1 6 1 3 67 36 59 9 35 50.0 58.3 00 , 0 33. 3 00.0 33.3 25.4 30. 6 74.6 66 . 7 97.1 50.0 41.7 100.0 67. 7 100-0 67. 7 74.6 69.4 25.4 33. 3 2.9 100.0 97.4 100.0 35.0 47.3 31. 8 62. 3 0.0 2.6 0.0 65.0 52. 7 68.2 37. 7 0.0 0.0 0.0 2.0 50.0 4.6 0.0 62.5 22,2 3 meters 1965 88811111111- 3 3 5 4 4 4 4 41 46 50,53 82 92 97 87 MN KL AB AB EF GH CD 61 38 17 137 184 365 167 30 Table 2. (cont'd.) Sample Date_________ No. 1966 2-24 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 8- 1 131,132 142 156 164 174,180 187,188 197,198 212,213 223,224 Transect N AB AB AB AB AB AB AB AB AB 241 159 50 36 89 121 2 152 29 % Adult 23.2 24.5 18.0 27.8 29.2 24.0 100.0 88.8 89.7 % Juvenile 76.8 75.5 82.0 72.2 70.8 76.0 0.0 11.2 10.3 31 Table 3. Percentage of adult and Juvenile Darwinula stevensoni from samples collected at depths of 6 , 9, and 12 meters in Gull Lake. Sample % % Date__________ H o ._______ Transect________N_____ Adult________ Juvenile 6 meters 1963 *7-27 82 8- 3 8- 5 10-13 114 114 114 128 1966 2-21 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 8- 1 8 38 A7 51-54 76 83 93 98 105 125 148,149 157 165 175,181 189,190 199,200 214,215 225,226 1 100.0 0.0 AB EF GH EF 33 35 56 28 123 63 98 511 100.0 97.1 87.5 46.4 55.3 49.2 52.0 12.9 0.0 2.9 12.5 53.6 44.7 50.8 48.0 87.1 IJ IJ AB AB AB AB AB AB AB 82 248 54 44 79 189 338 177 185 24.4 28.2 38.9 56.8 74.7 47.1 47.6 76.8 76.8 75.6 71.8 61.1 43.2 25.3 52.9 52.4 23.2 23.2 KL AB 9 meters 1965 8- 3 8- 5 114 114 114 12- 8 1966 2- 5 3-30 5-19 6- 1 6-15 7-15 8- 1 48 55,56 84 94 89 106 KL AB AB EF CD EF 114 144 166 182 192 216,217 227,228 IJ AB AB AB AB AB AB 61 7 96.7 100.0 3.3 0.0 6 100.0 0.0 24 7 93 87.5 85.7 24.7 12.5 14.3 75.3 4 75.0 25.0 1 100.0 0.0 7 9 3 100.0 100.0 66.7 0.0 0.0 33.3 10 100.0 0 .0 14 100.0 0.0 32 Table 3. Date (cont'd.) Sample No. % Transect N % Adult_______ Juvenile 12 meters 1965 11- 4 12- 8 100 107 GH EF 1 1 100.0 100.0 1966 2-21 126 IJ 1 0.0 *1964 0.0 0.0 100.0 33 25-30 percent. By late May and early June the percentage of adults in the population again began to Increase (Tables 2 and 3). The seasonal change in the adult-juvenile ratio more clearly, of <3 meters. may be expressed for example, using data from samples collected at depths The combined numbers of adult D. stevensoni collected each month were divided by the total monthly sample to obtain the following adult precentages. percent of 102 collected; 61 percent percent In August of 1965, adults comprised 96 In September, 70 percent (N - 262); January, 1966, 55 percent (N = 106); May, 75 percent (N = 42); Octobejr, (N = 42^; March, 27 (N = 59), and July, 97 percent (N = 135). Many of the overwintering juveniles had molted to the ninth or adult instar by mid-May, 1966, and contributed to the high percentage of adults reported at depths of <3 meters. Development and m o l t ­ ing proceeded less rapidly at greater depths owing, in part, to the lower temperatures. Consequently, the adult component at 3 meters (Table 2) was low until the eighth instar juveniles completed the final molt in early June. At 6 meters the percentage of adults re­ mained somewhat higher during the winter months as a smaller fraction of the overwintering juveniles had attained the size of those juveniles at shallower depths. The high adult percentages at 6 meters during May and early June (Table 3) probably represented an overest imation in light of the reproductive potential data discussed later. The small number of specimens collected from depths of 9 and 12 meters made comparison of adult-juvenile percentages difficult (Table 3). This portion of the total population also is considered later in the dis­ cussion of reproductive potential. Estimates of ostracod density per square meter presented in Table 4 34 are useful In predicting the contribution of young by adults and assist in understanding the life cycle of Darwinula stevensonl in Gull Lake. Only data from replicated samples collected along Transect AB during the summer of 1966 are considered (Refer to Tables 2 and 3 and fig. 1). The percentages of adults and juveniles collected on 15 June1 July and 15 July-1 August at 6 meters depth, were almost identical (Table 3). Yet, when one considers the density estimates on these dates, changes in the size structure of the population become more evident. The number of adults and juveniles per square meter was almost double, on 1 July, the corresponding estimates for 15 June (Table 4). The small number of replicated samples may account for some of the variation in density* due, however, I believe the major difference in density is to the number of newly molted overwintering juveniles that were contributing to the adult and late Inster classed by 1 July. This explanation also is tenable for estimates on 1 July and 1 August at 6 meters. The abrupt decrease in juvenile density Is believed to result from the rapid contribution of Juveniles to the adult class by mid-July and early August * The slight decrease in adult density from 1 July to 15 July may reflect a replacement of second reproductive season adults by first reproductive season adults during the latter part of the summer. Study of the reproductive cycle, discussed below, indicated that at least two adult age groups were represented in the summer collect­ ions. The adult age structure exhibited marked changes in 15 July and 1 August samples and was comprised primarily of first reproductive season adults. The Juvenile density, on 1 June, at 6 meters, was much lower than expected and is believed to reflect considerable 35 Table A. Estimates of the mean number of adult and juvenile DarwInula Stevenson1 per square meter at depths of 3, 6 and 9 meters along Transect AB during the period from 1 June - 1 August 1966 (A “ adult; J - juvenile). 1 June A J A 15 June J 1 July A 562 1363 627 1990 1277 A32 1926 216A 43 22 J 3 meters A3 6 meters 3A8A 3831 A 15 July J 2921 29A3 367 687 1 August A J 562 65 3073 931 303 --- 9 meters 195 216 36 s ampling b las. The variable nature of the substrate and the small number of ostracods at depths of 3 and 9 meters, respectively, renders density estimates less reliable in terms of absolute numbers. Yet, high juvenile densities in June and an abrupt reversal of adult-juvenile densities later in the summer correspond to the general trend noted at 6 meters. The decrease in density at 9 meters compared with that at 6 meters is obvious. The reproductive potential and cycles of freshwater ostracods have received limited study. Observations of the number of eggs or young in one or a few specimens of _D. stevensoni out the literature. However, are scattered through­ 1 know of no study on the reproductive preformance of I>. stevensoni over a single reproductive cycle. Some difficulty was encountered in studying the reproductive potential of Darwinula stevensoni from field samples. ulation reproductive cycle is Initiated, When the pop­ successive addition and development of eggs within the carapace and ultimate release of first instar young proceeds at varying rates. The difficulty arises in distinguishing adults exhibiting the maximum production of young from those which are adding or losing young. The thin, translucent carapace of _D. stevensoni premits direct observation of the developing young during most of the reproductive season. However, eggs and young become so closely packed prior to release of the first individuals that accurate counts cannot be made without dissection of each animal. In addition, parasites attached to the hypodermis in the area in which the young are carried are easily confused with the developing eggs. Consequently, each adult ostracod was anesthetized and dissected alive, in order to obtain accurate 37 counts of the number of eggs or young and parasites. The annual reproductive cycle of _D. stevensoni and number of eggs and young in the carapace of each dissected female are presented in Fig. 2. These data represent combined monthly counts of 2587 individuals collected from Gull Lake. One specimen from 1964, two animals which died before young were counted, and several young released during observation were not included in Fig. 2. In Gull Lake the reproductive period of J). stevensoni begins in May and is effectively completed by October. Only 8 adults examined in late fall and early winter were found with young or eggs. The number of developing young per individual increased from a maximum of 3 in May to 15 in August. The sharp decrease in the number of young per individual from August to September reflects widely spaced sampling periods. Information derived from study of the reproductive cycle and repro­ ductive potential assists in understanding the juvenile-adult percent­ ages and density data presented earlier. On the other hand, the generalized reproductive potential and reproductive cycle presented in Fig. 2 do not reveal certain variables which are important in estimat­ ing population growth rates and the changing size and age structure in various components of the population. The differential rates at which eggs were released into the cara­ pace of adults at depths of 3, 6 and 9 meters during the sunnier of 1966, are shown in Fig. 3. These histograms show the retarded development and addition of young at Increasing depths. At a depth of 3 meters the number of eggs and young per individual progressed from a maximum of 4, on 1 June, to 13 on 15 July. By 15 July, a large number of the 3 and 6 meter adults had begun to release their young whereas those adults 16 ONE AOULT 12 JULY 1965 10 I 8 I NUMBER OF YOUNG/IND. t4 t JAN. FEft MAR APR. MAY JUL. I i 1 AUG. 1966 Figure 2. AUG. SEP OCT. NOV. DEC. 1965 Annual reproductive cycle and reproductive potential of Darwlnula atevensonl in Gull Lake. The total number of epeclmena examined each month, from left to right, are: Jan. (23), Feb. (79), Mar. (175), Apr. (30), May (83), June (217), July (475 and 20), Aug. (181), Aug. (324), Sept. (152), Oct. (252), Nov. (477), Dec. (99). 39 NUMBER OF YOUNG/IND. 14 • ONE ADULT 12 K) i 8 6 k 4 I I 2 O 1. 6 e I JUN. 6 6 15 JU N . 6 6 Figure 3. 9 1 » I 1 . 3 6 9 3 6 9 D E P T H . stevensoni. was present in the carapace of _D. pagliolii. The incidence of infection was much greater in the latter; 86.8 percent (N = 144) of the ostracods had one or more rotifers attached to the hypodermis in the posterior region of the carapace. Table 11 shows the percentages of infected adults in samples from Station GI. These data do not indicate if the rotifer caused a reduction in the average number of young produced by _D. pagliolii. All of the gravid females (Table 10) had released some young or were just beginning the reproductive cycle. Thus, data on the number of rotifers in the carapace of individual ostracods with maximum numbers of young and eggs are lacking. Observations of J). pagliolii in 1967 indicated that fewer young were produced by females with four or more rotifers in the carapace. gated. This problem still is being investi­ If the rotifer definitely is shown to reduce the number of offspring produced by infested ostracods, D. pagliolii certainly will 58 Table 11. Variation in the number of parasitized adult Darwlnula pagliolii collected at Station GI in Gull Lake. X No. No. Percent Parasites/ Date______ Adults______ Infected______ Infected Ind._____ 1965 9-14 9-22 10- 4 10-13 10-18 11- 1 12- 2 1966 5-31 1 12 12 14 20 2 79 100.0 91.7 66.7 100.0 95.0 100.0 91.1 4 75.0 . Range No. Parasites 4.0 2.1 2.3 2.1 3.6 3.5 3.2 4 0-6 0-5 1-4 0-6 3-4 0-7 2.0 0-3 59 have the highest incidence of parasitism (in the sense used here, see discussion) of any ostracod known. No rotifers were found in any Gull Lake ostracods except J). stevensoni and _D. pagliolii. Both of these species are slow, creeping benthic forms which do not have a modified third pair of thoracic legs for keeping debris and other organisms out of the carapace. The furcae also are rudimentary. The absence of appendages suited for forcibly removing the rotifers and the space available in the posterior region of the carapace where the young ostracods develop probably explains why the ostracod-rotifer relationship is found only in Darwinula. 60 Candona ohioensis Candona ohloensis was collected from 96 Ekman, 5 free-fall core, 8 aquatic net, and 3 fish samples. The number of specimens examined Is sumnarlzed below. Total No. Gull Lake Fish Analysts 2833 166 No. Adults 361 166 No. Juveniles 2472 No. Male 121 No. Female No. Females Examined for eggs 240 13 153 226 -- it* ohloensis was the largest ostracod species examined from Gull Lake, reaching a maximum length of slightly more than 2 inn. This species was collected at depths ranging from <1-18 meters during the study period, but was found at Tabular data for C^. depths of 32 meters on other dates. ohloensis and other Candona species are arranged first by depth and then by date of collection. This was done to show differences in the reproductive cycle(s) of species populations living in the littoral and sublittoral areas of the lake. ohloensis. for example, had two generations per year in the sub­ littoral and one generation per year in the littoral region of Gull Lake. Scanning of data inthe tables below is sidering the data for a given date made easier by first con­ at <3, 3, 6 and 9 meters and then those data for 12 meters or greater. In early August, 1965, 100 percent of 1065 specimens collected at depths of <3, 3, 6 and 9 meters were middle to late instar juveniles. By mid-September most <3 meter individuals had molted to the adult instar, but no females (N * 22) were gravid (Table 12). On 13 October 61 non-gravid females were found at 6 meters, whereas, by 18 October, 100 percent of the females from <3 meters were gravid and averaged 11.A eggs per Individual (Table 13). Samples were collected along four transects on 4 November at depths of 3, 6 , 9, 12 and 15 meters. In littoral samples all C. ohloensis were mature and most females were gravid. individual from 6 meters had 35 eggs; 4 females 24.5 eggs per individual. (0-9 m) One from 9 meters averaged Density estimates on 5 August and 11 November at depths of 3, 6 and 9 meters indicate high mortality of late instar juveniles and early maturing adults (Table 14). August estimates represent replicated samples along Transect AB and November estimates, the means of 4 transects. Reductions in the mean number of Individuals per square meter on the latter date were marked. The reproductive cycle of £. ohioensis in the littoral region apparently ended by early December as no adult or juvenile individuals were present in December samples. Additional evidence supporting this conclusion was the absence of £. ohioensis in all littoral samples through March of 1966. Early instars, abundant in 3-9 meter samples on 15 May 1966, were presumed to be offspring of the November 1965 adults. All ohloensis collected from 0-6 meter depths through the summer of 1966 were immature. Recently molted adult males were collected at a depth of 9 meters on 15 June 1966. One non-gravid and 2 gravid females were collected on 15 July and 1 August, latter depth. respectively, from the Most juveniles, at depths less than 9 meters, had molted to the seventh or eighth instar by 1 August 1966; presumably, these individuals would have entered their reproductive cycle by October or November as was found in 1965. In contrast to the annual fall generation of adults in the 62 Table 12. Date 1965 8 - 2 8- 3 8- 5 9-14 10-18 1966 7- 2 1965 8- 3 8- 5 11- 4 1966 5-19 6- 1 6-15 7- 1 7-15 8- 1 1965 8- 2 8- 3 8- 5 10-13 11- 4 1966 5-19 6- 1 6-15 7- 1 7-15 8- 1 Percentage of adult and juvenile Candona ohioenaIs from Gull Lake samples collected at depths of <3, 3, 6 , 9, 12, IS, and 18 meters. Transect or Location Sample No. N Percent Adult Perceni Juveni! 39 40 44 45 49 61 62 80 207 <3 meters 39 KL MN KL AB 61 62 80 AB 14 20 19 110 6 17 12 19 2 0.0 0.0 0.0 0.0 0.0 94.1 100.0 100.0 0.0 100.0 100.0 100.0 100.0 100.0 5.9 0.0 0.0 100.0 41 46 50 53 82 87 92 97 164 174, 187, 197 212, 223, 213 224 3 meters MN KL AB AB AB CD EF GH AB AB AB AB AB AB 113 97 153 80 4 2 1 1 50 69 58 1 90 16 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 181 190 200 215 226 6 meters 38 MN KL AB AB 76 AB CD EF GH AB AB AB AB AB AB 76 75 24 192 17 4 4 11 4 3 17 60 67 79 43 45 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 38 42 47 51 54 76 83 88 93 98 165 175, 189, 199, 214, 225, 180 188 63 Table 12. (cont'd.) Date 1965 8- 3 8- 5 11- 4 1966 5-19 6- 1 6-15 7- 1 7-15 8- 1 1965 11- 4 12- 8 1966 2-21 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 1965 10-13 11- 4 1966 3-30 5-19 6- 1 6-15 7- 1 7-15 1966 6- 1 * Core sample Transect or LocatIon Sample No. 48 52 55 84 94 166 176, 191, 201 , 216, 227, 85 90 95 100 107 126 145 150 159 167 169 177, 193, 203, 218, 75 86 91 96 101 146 151 168 178, 195, 205, 220 , N Percent Adult Percent Juveni' 182 192 202 217 228 9 meters KL AB AB AB EF AB AB AB AB AB AB 30 2 37 5 6 6 49 6 19 13 27 0.0 0.0 0.0 100.0 100.0 0.0 0.0 16.7 10.5 15.4 7.4 100.0 100.0 100.0 0.0 0.0 100.0 100.0 83.3 89.5 84.6 93.6 183 194 204 219 12 meters AB CD EF GH EF AB AB IJ AB AB IJ* AB AB AB AB 14 10 4 14 54 11 27 10 60 31 7 142 50 67 96 85.7 80.0 50.0 7.1 7.4 0.0 0.0 0.0 0.0 0.0 0.0 27.4 32.0 38.8 66.7 14.3 20.0 50.0 92.9 92.6 100.0 100.0 100.0 100.0 100.0 100.0 72.6 68.0 61.2 33.3 184 196 206 221 15 meters 75 AB CD EF GH AB IJ AB AB AB AB AB 6 2 2 17 21 3 11 2 63 59 65 51 0.0 100.0 100.0 94.1 38.5 0.0 9.9 0.0 4.8 10.2 23.1 54.9 100.0 0.0 0.0 5.9 61.5 100.0 90.1 100.0 95.2 89.8 76.9 45.1 18 meters AB 59 6.8 93.2 179-185 64 littoral region, data from 12 and 15 meter samples strongly Indicate a biannual turnover rate in the sublittoral population of £. ohioensis. No samples were collected at depths of 12 and 15 meters before November, 1965. Thus, data indicating the period in which the late fall-early winter reproductive cycle began are lacking. By early November, however, many juveniles had molted to the sixth-eighth in­ star and 66.7 percent (N * 84) of the individuals collected were mature (Table 12). Gravid and non-gravid females were present in samples from depths of 12 and 15 meters and the number of uterine eggs per gravid female averaged, respectively, 15.4 and 8.9 (Table 13). The rather low densities at depths of 12 and 15 meters on 4 November 1965 (Table 14) possibly reflect high juvenile and adult mortality, but also could mark the early stages of the reproductive cycle. Of 54 individuals collected on 8 December 93 percent were iranature; three of the four gravid females had from 27-34 eggs in the uterus (Table 13). No Ekman samples were taken in January, but analysis of fish stomachs of the yellow perch, Perea flavescens (Mitchill), yielded over 100 adult ohioensis. in gill nets at a depth of 15 meters. The perch were caught The condition of ostracods in the stomachs of most fish suggested that many had been eaten shortly before the fish were captured. The latter results are discussed in greater detail in other sections of this thesis. From late February, 1966 through 19 May, almost 100 percent of the specimens in 12 and 15 meter samples were early instar juveniles. A single adult male found on 30 March had just molted to the adult instar, indicated by the soft, flexible, incompletely calcified cara­ pace. By 1 June, 27.4 percent (N ■ 142), 4.8 percent 6.8 percent (N * 62) and (N = 59) of the ostracods, respectively, were mature in 65 Table 13. Date Percentage of males and femalaa In adult Candona ohloensis and average number of eggs present In the uterus of adult females. Sample number and location corresponding to dates are presented In Table 12. Number Adults 9-14-65 10-18-65 X Male Number Gravid Female Females Exam. Avg. No. X for X eggs per Female______ eggs_____ Gravid_____ female 28 19 21.4 10.5 78.6 89.5 8 0.0 100.0 < 3 meters 22 17 Range No. eggs 0.0 100.0 0.0 11.4 4-20 8 87.5 7.7 0-12 3 17 0.0 70.6 0.0 13.3 0-35 100.00 ----0.0 100.0 24.5 ----0.0 4.5 20-28 4-5 3 meters 11-4-65 6 meters 10-13-65 11-4-65 4 22 33.3 15.8 66.7 84.2 9 meters 11-4-65 6-15-66 7-1-66 7-15-66 8- 1-66 11 1 2 2 2 9.1 100.0 100.0 50.0 0.0 90.0 0.0 0.0 50.0 100.0 4 1 2 11-4-65 12-8-65 6- 1-66 6-15-66 7-1-66 7-15-66 23 4 39 16 26 64 13.0 0.0 56.4 31.3 61.5 35.9 87.0 100.0 43.6 68.7 38.5 64.1 12 meters 19 4 16 11 10 41 57.9 100.0 37.5 81.8 60.0 63.4 15.4 22.8 5.8 10.6 26.2 10.7 0-26 1-34 0-14 0-20 0-33 0-31 11-4-65 3-30-66 6- 1-66 6-15-66 7-1-66 7-15-66 33 1 3 6 15 28 6.1 100.0 33.0 50.0 80.0 53.6 93.9 0.0 67.0 50.0 20.0 46.4 15 meters 27 — 2 3 3 13 85.2 --50.0 33.3 33.3 61.5 8.9 --1.0 17.0 5.0 6.4 0-20 0-2 0-17 0-5 0-13 75.0 18 meters 3 33.3 16.0 0-16 6- 1-66 4 25.0 66 Table 14. Estimates of the mean number of adult and juvenile Candona ohioensis per square meter on 5 August and 4 November 1965. August estimates from replicated samples along Transect AB; November estimates represent means of samples from four transects (A ■ adult; J = juvenile). Depth (meters)________ August A --- J 5064 A --- J 4523 November 87 3 238 6 A 119 9 J 844 A 249 J 206 A 357 J 97 12 15 67 12, 15 and 18 meter samples. Greater than 50 percent of the animals collected on 15 July were mature (Table 12). Combined estimates of the mean number of adults and juveniles per square meter generally were of the same order of magnitude as juvenile densities at lesser depths (Table 15). Ostracod densities at 9 meters generally were less than at other depths. The largest number of eggs recorded for a single female was 35. Many females lived beyond periods of egg laying as evidenced by one or a few eggs in the uteri of females collected late in the reproductive cycle. The size and location of uterine eggs and condition of the ovaries clearly indicated whether or not individual females had already released the majority of their eggs. The maximum number of eggs recorded for C. ohioensis does not necessarily represent the maximum number of eggs that individuals may lay during a single reproductive cycle. However, these data may indicate differences in the relative importance of certain ostracod species in the benthic food web of Gull Lake. Females outnumbered males in most large samples of adults 13). (Table On November 4, 1965, more than 75 percent of the adults from all depths were females. During the sunnier of 1966, males were more abun­ dant than females in some samples. Of 55 eighth instar juveniles exam­ ined from 12-15 meter samples on 1 J u n e - 15 June 1965, 30 or 54.5 percent were males. This indicated that the sex of inmature ostracods did not differ significantly from a 1:1 ratio during the early phase of the sunnier reproductive cycle. However, males of most Candona species mature earlier and have a shorter life span than females. Of the total number of adult £. ohioensis examined, 33.5 percent were males. Large numbers of males in the population at the onset of the reproductive 68 Table 15. Depth (meters) Estimates of the mean number of adult and juvenile Candona ohloensis per square meter at depths of 3-18 meters along Transect AB from 1 July - 1 August 1966 (A ■ adult; J » Juvenile). June 1 A J June 15 A J July 1 A J 22 July 15 A J -- 1948 3 --- 1493 -- 1255 -- 6 ---1298 --- 1450 --- 1710 9 ... 1060 22 108 43 368 12 844 4393 346 736 563 887 1385 692 15 65 1298 325 1082 606 498 18 87 1190 130 1234 -- August 1 _A______ J --- 931 43 238 346 974 43 541 69 cycle may have some selective advantage In insuring fertilization of females soon after the latter mature. Feeding habits, involving size discrimination, by predators of ostracods may account for the rapid reduction of males in certain species populations in Gull Lake. Data relating to this assumption are discussed later. £. ohioensis also were collected from Lawrence Lake, Barry County, Michigan and from Pretty Lake, Indiana. 70 Candona rawsonl Candona rawsonl was perhaps the most easily recognized species of Candona in Gull Lake. The female has a prominent posterior flange on the left valve in the adult instar. Juveniles lack this flange, but have other distinctive carapace features which permit rapid identification. Late instar males were not easily identified. Candona scopulosa is very similar in shape and in external appearance to £. rawsonl. Males and late lnstar juveniles often had to be dissected to verify a tentative identification based on carapace morphology. Males and females are little more than 1 mm in length and exhibit considerable variation in size. The male is larger than the female. Juveniles or adults were collected in every month of the study period but not at all depths on a given date. 64 Ekman, 40 aquatic net and 2 fish samples* _C. rawsonl was found in The number of individuals examined is recorded below. Total No. Gull Lake Fish Analysis * 1887 28 No. Adults No. Juveniles 557* 1 1330 27 No. Male 122 1 No. No. Females exam. Female____ for eggs 423 -- 326 -- Sex of 12 adults not recorded C. rawsonl had a single reproductive cycle in both the littoral and sublittoral regions of Gull Lake. did not span the same time interval. The reproductive period, however, There appeared to be two distinct or at least effective breeding populations in the area of Transect AB. Sampling was insufficient to indicate whether this was true in other areas of Gull Lake. Large numbers of Individuals were found on the sandy, cobblestone substrates in shallow water near shore. The other 71 zone of high ostracod density was at depths of 12 and 15 meters. though Al­ rawsoni exhibited a highly clumped distribution in the littoral region of Gull Lake, I believe the total number of samples taken in this study was sufficient to indicate very limited exchange of individuals between the shallow water and 12-15 meter depths. Only 1 adult and 14 juveniles were collected at depths between 3 and 12 meters during the entire study. Almost 1900 individuals were collected from the 0-3 and 12-15 meter depths. One hundred percent (N = 823) of the ostracods collected from the littoral region during August through November 1965 were immature (Table 16). Adults were found on 8 December, but no record was made of the sex or reproductive condition of these specimens. Only 2 of 434 individuals were immature in 0-9 meter samples collected from 11 February to 1 June 1966. The reproductive cycle ended by early June 1966, and small juveniles were again abundant on 1 July. No adults were found in the littoral areas of the lake after 1 June. Gravid females were abundant in February with a maximum of 40 eggs in some individuals. By late February, 93 percent (N = 185) of the females were gravid; from March through 1 June all females were gravid, but most had released the major portion of their eggs. At depths of 3 and 6 meters gravid females were found only during February and March. One late maturing female had 40 eggs in the uteri on 31 May 1966; however, this Individual was the only specimen collected beyond March which had not released most of its eggs (Table 17). rawsoni was not found in 12 and 15 meter samples until February 1966. Gravid females were present in all samples from February-15 June at 12 meters depth. Femal es which had completed egg laying were present 72 Table 16. Date Percentage of adult and juvenile Candona rawsonl in samples collected from Gull Lake at depths of <3 to 15 meters. No. Samples Transect or Locat ion % Adult N Juvenile <3 meters 1965 8- 2 8- 3 8- 5 8-28 9- 14 9-27 10- 4 10-13 10-18 1966 2-11 2-21 2-24 3-21 3-26 5-31 6- 1 7- 2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2 1 1 2' 3' 11 ' 1' 8' 7’ 39, KL KL AB AB GI, 62 A B 66 , 67 68 GI GI, 74 GI, 80 40 1 189 14 274 238 4 39 10 4 2 7 4 4 5f 1 AB AB AB AB 138-141 GI AB GI 19 3 356 18 19 1 10 1 100.0 0.0 V 89.5 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 10.5 0.0 0.0 0.0 0.0 0.0 0.0 100,0 3 meters 1965 8- 3 11- 4 12- 8 1966 2-24 3-30 7- 1 7-15 1 2 1 MN EF, GH EF 1 5 16 0.0 0.0 75.0 100.0 4 2 2 2 AB AB, IJ AB AB 4 3 200 2 100.0 100.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 25.0 6 meters 1965 8- 3 11- 4 1966 3-30 7- 1 7-15 ' 1 2 MN AB, EF 1 7 0.0 0.0 1 1 1 IJ AB AB 1 3 1 100.0 Aquatic net samples 0.0 0.0 100.0 0.0 100.0 100.0 73 Table 16. Date (cont'd.) No. Samples Transect or Location % N Adult Juvenile 9 meters 1966 7-15 0.0 100.0 1 43 23 8 94 57 22 12 100.0 0.0 67.4 52.2 64 26 22 34 35.9 69.2 27.3 0.0 AB 12 meters 1966 2-21 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 1 1 1 1 2 2 2 2 AB AB AB AB AB AB AB AB 32.6 47.8 0.0 23.4 36.8 13.6 0.0 100.0 76.6 63.2 86.4x 100 .Ox 15 meters 1966 3-30 4-29 5-19 7- 1 x 1 1 1 2 Early instar AB AB AB AB 64.1 30.8 72.7 100.0 74 in 1 July samples, but all of these individuals were feeble and emaciated. Many dead, incompletely deteriorated specimens of rawsoni were present in June and July samples, indicating that physiological depth was a major mortality factor. During the Fall of 1965, immature individuals were particu­ larly abundant in well sorted sandy subtrates and in many samples constituted the only species of Candona present. The early molt­ ing of juveniles to the adult instar at depths of < 3 and 3 meters during February and March, 1966 also was in marked contrast to the development of Immature ostracods in the sublittoral populations. In the latter, late Instar individuals comprised more than 50 percent of the total _C. rawsoni June, 1966. collected from March through 15 Early instars collected on 1 July Indicated the Ini­ tiation of a new generation. This supposition was further supported by the absence of adult and late instar individuals of the parent generation by mid-July. Difficulties of sampling in shallow water, cobblestone sub­ strates, through a thick ice cover, prevented accurate quantification of adult population densities in such areas. Although a large num­ ber of individuals were collected from depths o f < 3 meters, no adequate measure of population density could be made for comparison with that estimated for adult Candona rawsoni at 12-15 meter depths. This species was, nevertheless, one of the conspicuous numerical dominants during February and March at depths of <3 , 3, and 12-15 meters. Density estimates at 12 meters depth Illustrated the rapid re­ duction of adults during June 1966. On 1 June the estimated mean 75 number of adults per square meter was 2034, on 15 June, 1234 and 1 July, 476. By 15 July adults were absent and the estimated number of early Instar juveniles was 260. The latter apparently represented newly hatched offspring of the expired adult population. In the early stages of the reproductive cycle, at both < 3 and 12-15 meter depths, sex ratios were near a 1:1 ratio. Males, in fact, were more numerous than females in all the larger samples. Throughout the remainder of the reproductive cycle, however, females comprised about 70 percent, or more, of the adults. percent of the specimens from May and June were males Less than 15 (Table 17). This corresponds closely to results presented for C^. ohloensis. 76 Table 17. Date Sex ratios of adult Candona rawsoni and average number of eggs present in the uterus of adult females. Number and location of samples corresponding to dates are presented In Table 16 • No. Adults % Male % Female No. females exam. for eggs % Gravid Gravid females Avg. No. eggs/ female Range No. eg] <3 meters 1966 2-11 2-21 2- 24 3-21 3-26 5-31 6- 1 17 3 350 18 19 1 10 52.9 0.0 19.4 16.7 31.7 0.0 10.0 47.1 100.0 80.6 83.3 68.3 100.0 90.0 8 3 185 15 13 1 9 62.5 66.7 93.0 100.0 100.0 100.0 100.0 22.2 11.5 8.0 10.0 9.6 40.0 4.9 0-40 0-17 0-25 2-24 4-20 40 1- 8 100.0 66 .7 8.5 14.3 4-14 0-27 3 meters 12 4 3 0.0 0.0 100.0 100.0 3-30 1 o • o 100.0 6 meters 1 100.0 9.0 9 2-21 3-30 4-29 6- 1 6-15 7- 1 1 14 11 22 21 3 0.0 78.6 18.2 13.6 4.8 0.0 100.0 21.4 81.8 86.4 95.2 100.0 12 meters 1 3 9 19 20 3 100.0 66.7 100.0 100.0 100.0 0.0 4.0 9.5 8.4 6.7 7.8 -- 4 0-14 5-12 2-13 1-14 --- 3-30 4-29 5-19 23 18 6 56.5 27.8 0.0 43.5 72.2 100.0 15 meters 10 13 6 60.0 100.0 100.0 6.7 10.2 10.2 0-17 6-15 3-17 12- 8 * 2-24 3-30 * 1965 4 3 77 Candona acuta Candona acuta was collected from 32 Ekman dredge and 2 fish samples on 10 sampling dates. Results of fish stomach analyses are not Included in the tables below; however, data from one Ekman dredge sample on 23 January 1965 are included to supplement other information on reproductive cycles. This species is similar to Candona rawsoni in size, ranging from about 1.1 to 1.3 ran in length. sexually dimorphic. Juvenile £. The males are larger than females and acuta were difficult to identify; therefore, only the late instar totals are recorded in the following tables. Early instar individuals may have been present in other samples, but could not be identified. positive identification. Adult males were dissected for The number of specimens examined is shown below. Total ____________No. Gull Lake Fish Analyses 454 180 No. Adults 360 180 No. Juveniles 94 -- No. Males No. females No. exam. Females____ for eggs 51 100 309 80 283 --- £. acuta was collected at depths ranging from <3 to 15 meters. The first specimens were found on 4 November 1965. of 844 (3m), 649 (6m), 573 Density estimates (9m) and 390 (15m) ostracods per square meter showed greatest densities at shallower depths. Adults comprised over 75 percent of the individuals at 3 and 6 meters whereas 25 percent, or less, were mature in samples from depths of 9-15 meters (Table 18). No females were gravid. Development rates of C^. acuta appeared to be strongly influenced 78 Table 18. Date Percentage of adult and juvenile Candona acuta in samples collected from depths of <3 to 15 meters in Gull Lake. No. Samples Transect or Location N % Adult % Juvenile <3 meters 1966 2-21 3-26 1 1 AB 138 1 6 100.0 100.0 0.0 0.0 78 76.9 23.1 3 4 100.0 100.0 0.0 0.0 60 28 85.0 100.0 15.0 0.0 1 3 100.0 100.0 0.0 0.0 53 12 13.2 100.0 86.8 0.0 3 meters 1965 11- 4 1966 2-24 3-30 AB, CD EF, GH 4 1 AB AB 6 meters 1965 11- 4 12- 8 1966 2-21 3-30 1 AB, CD EF, GH EF 1 1 AB AB 4 9 meters 1965 11- 4 4 12- 8 1 AB, CD EF, GH EF 12 meters 1965 1-23 12- 8 1966 2-21 1 1 IJ EF 46 60 100.0 100.0 0.0 0.0 1 AB 2 100.0 0.0 36 25.0 75.0 51 9 1 100.0 100.0 100.0 0.0 0.0 0.0 15 meters 1965 11- 4 1966 3-30 4-29 5-19 3 2 1 1 AB, EF CD AB, IJ AB AB 79 by temperature. By 8 December all females (N = 35) from 6-9 meters were gravid, but only 30 percent samples. (N = 54) were oviferous in 12 meter Females from 6 meter samples had 3-11 eggs and averaged 11.8 eggs per individual (N = 24). At 9 meters the number of eggs ranged from 4-16, but the average was 9.1 (N = 11) per female. A maximum of 13 uterine eggs was recorded for 12 meter adults; however, the latter averaged only 4.9 (N «= 14) eggs non-gravid females per individual. in the 12 meter sampleshad just matured had incompletely calcified valves. Forty and most The size and position of eggs within the uteri, condition of ovaries and recently molted appearance of many adults clearly indicated that females from the above depths had not begun laying eggs by early December (Table 19). The number of adults per square meter had decreased markedly by February 1966, andall oviferous females, except one from had released a majority of their eggs. meters, The lag in egg development at lower temperatures again was indicated by the larger number of eggs in females from 12 and 15 meter samples compared to individuals from lesser depths. No adults were found at depths less than 15 meters beyond March 1966. Only 10 adults were collected from 15 meters during April and May. Males were most abundant in the first phase of the reproductive cycle, but never constituted more than 35 percent of the adults in any sample (Table 19). After November 1965, males comprised 10 percent, or less, of the adults sampled. Slightly more than 14 percent (N = 360) of the adults examined were males. In contrast, 56 percent (N * 180) of the adults eaten by the perch, Perea flavescens. were males. These differences in sex ratios strongly indicate si 2e discriminate feeding by the yellow perch. Males are probably more active than females 80 Table 19- Date Sex ratio of adult Candona acuta and average number of eggs present in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 18. No. Adults % Male % Female No. Females exam, for eggs Gravid Gravid Females Avg. No. eggs/Ind. 100.0 100.0 23.0 2.8 23 1-6 % Range No. e; <3 meters 1966 2-21 3-26 1 6 0.0 0.0 100.0 100.0 1 6 3 meters 1965 11- 4 1966 2-24 3-30 60 26.7 73.3 44 0.0 --- --- 3 4 0.0 0.0 100.0 100.0 3 4 100.0 100.0 1,7 2.0 1-2 1-4 6 meters 1965 11- 4 12- 8 1966 2-21 3-30 51 28 35.3 7.1 64.7 92.9 33 24 0.0 100.0 -11.8 -3-11 1 3 0.0 0.0 100.0 100.0 1 3 100.0 100.0 4.0 3.7 4 1-7 0.0 100.0 -9.1 --4- 16 9 meters 1965 11- 4 12- 8 13 12 23.1 8.3 76.9 91. 7 5 11 12 meters 1965 1-23 12- 8 1966 2-21 46 60 2.2 10.0 97.8 90.0 26 54 96.2 25.9 6.0 4.9 0-11 0-13 2 0.0 100.0 2 100.0 8.0 8 -- -- 6.3 6.6 8.0 1-13 1-15 8 15 meters 1965 11- 4 1966 3-30 4-29 5-19 9 33.3 66 . 7 6 0.0 51 9 1 2.0 0.0 0.0 98.0 100.0 100.0 50 9 1 100.0 100.0 100.0 81 during the reproductive period and may be more abundant at the substratewater interface. The reproductive behavior of males coupled with apparent size discriminate feeding by the yellow perch may result in higher predation on male C. acuta. 82 Candona inoplnata Candona inoplnata was one of the most common species of Candona encountered in Gull Lake. and late instars. fication. This ostracod was easily recognized In adult Only the early instars posed a problem of identi­ Females are about 1 mm in length and males somewhat larger. The sexes are strongly dimorphic, but the posteriorly truncate carapace of the female and highly arched dorsum of the male readily distinguished this species from others in Gull Lake. Closely related species in lakes surrounding Gull Lake were not present. Adults show a definite temporal distribution although juveniles or adults are present throughout the year. Early instars, present in sumner and early fall samples, were not included in tables because of difficulties in distinguishing between tiny juveniles of many species. inoplnata is an able climber as well as burrower and often was found in large numbers on highly organic substrates and occasionally on the surfaces of rooted aquatic plants. Fifty-eight Ekman, 34 aquatic net, 1 free-fall core and 2 fish samples contained inoplnata. The number of ostracods examined is presented below. Total No. Gull Lake Fish Analyses * 975* 15 No. Adults No. Juvenile No. Males No, Females 245 188 7 539 8 730 15 Sex of three adults undetermined Mo. females exam, for eggs 478 --- 83 The depth distribution of CL inoplnata ranged from extremely shallow water to 15 meters. Most specimens were collected from depths of 6 meters or less; 95 individuals or 9.7 percent of the total were collected from depths greater than 6 meters. Juveniles from 0-3 meter depths matured earlier than those in deeper waters. The first late instar juveniles were collected from <3 meter depths in mid-October 1965 percent (Table 20). (N * 26) were mature, but non-gravid. 80 percent By 1 November, 11.5 On 2 December, about (N = 46) of the specimens from <3 meters were mature and females contained from 0-26 uterine eggs. From January to May 1966 adults were abundant and constituted between 84 and 100 percent of the individuals sampled on the dates listed (Table 20). The latter juvenile-adult ratios may be misleading because of my inability to identify the smallest instars. Greater than 80 percent of the adult females were gravid, but the average number of eggs per individual de­ clined steadily after mid-February. The maximum number of eggs in a single ostracod was 36. Samples from 3 meter depths had a higher proportion of late instar juveniles than collections from <3 meters during November 1965 through March 1966. The estimated number of juveniles per square meter was 173 on 4 November. Later on 24 February 1966, the estimated number of adults and late instar juveniles was, respectively, 1006 and 368. November estimate represented the mean of four transects; estimate, four replicates from Transect AB. The the latter A two transect sample mean on 30 March gave an estimate of 1234 adults and 390 juveniles per square meter* All ostracods in April-June samples were mature and most females were gravid (Table 21). The number of adult samples decreased rapidly after March 1966. C. inoplnata in 3 meter A single adult was 84 Table 20. Percentage of adult and juvenile Candona inopinata in samples collected from depths of <3 to 18 meters in Gull Lake. Transect No. or % Samples__________Location____________N_________ Adult Date % Juvenile <3 meters 1965 10-18 11- 1 12- 2 1966 1- 4 1-11 2-11 2-21 2-24 3-21 3-26 5-31 7' 7* 4' G I , 80 GI GI 15 26 46 0.0 11.5 80.4 100.0 88.5 19.6 5' 6' 4 2 7 3 3 5' GI GI AB AB AB AB 139-141 GI 58 49 89 40 80 108 32 1 94.8 98.0 95.5 85.0 83.8 95.4 100.0 100.0 5.2 2.0 4.5 15.0 16.2 4.6 0.0 0.0 AB, CD EF, GH EF 17 5 5.9 60.0 94.1 40.0 AB AB, IJ AB AB AB 127 75 17 12 1 73.2 76.0 100.0 100.0 100.0 26.8 24.0 0.0 0.0 0.0 3 meters 1965 11- 4 12- 8 1966 2-24 3-30 4-29 5-19 6 - 4 1 4 2 1 1 1 1 6 meters 1965 10-13 11- 4 12- 8 1966 2-21 3-30 1 4 1 0.0 100.0 1 76 AB, CD EF, GH EF 51 6 0.0 66.7 100.0 33.3 1 2 AB AB, IJ 2 22 50.0 100.0 50.0 0.0 21 13 0.0 92.3 100.0 7.7 9 meters 1965 11- 4 4 12- 8 1 1 Aquatic Net AB, CD EF, GH EF 85 Table 20. Date 1966 2- 5 3-30 4-29 6- 1 (cont'd.) No. Samples Transect or Location N 1 1 1 1 IJ AB AB AB 1 4 1 1 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 14 23 0.0 82.6 100.0 9 2 5 100.0 100.0 0.0 0.0 0.0 0.0 100.0 % % Adul t J u v e n lie 12 meters 1965 11- 4 12- 8 1966 2-21 3-30 1-19 3 1 CD, EF, GH EF 1 1 2* AB IJ IJ 100.0 17.4 15 meters 1965 11- 4 * Free Fall Cores CD 86 collected on 1 June. Density estimates, calculated from means of four transect sam­ ples, on 4 November at depths of 6 , 9, 12, and 15 meters were 551, 227, 151, and 11, respectively. juveniles. All individuals were late instar By 8 December, greater than 50 percent of the ostracods collected from each of these depths were mature and more than one half of the females were gravid. The maximum number of uterine eggs in one female was 15. Females had deposited the majority of their eggs by February, but several eggs remained in the uterus of specimens collected in late March. Only 2 adult inopinata were collected from depths of 6-15 meters after March 1966, Sex ratios of adult (]. inopinata followed the general pattern described for other species. than females. Males began maturing somewhat earlier Eighty percent of the 45 eighth instar juveniles sexed in mid-October-early November were males. In most samples with more than 10 adults, males constituted less than 35 percent of the total. Toward the end of the reproductive cycle the percentage of males had declined further. No males were present in April-August samples, but probably would have been found in early April had samples been taken. 87 Table 21. Sex ratio of adult Candona inopinata and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 20. No. AduIts Date % Male \ Female No. Females exam. for eggs % Gravid Gravid females Avg. No. eggs / Ind. Rang No. eggs <3 meters 1965 11- 1 12- 2 1966 1- 4 1-11 2-11 2-21 2-24 3-21 3-26 5-31 3 37 100.0 5.4 0.0 94.6 --31 -77.4 -- 10.6 ... 0-26 55 48 85 34 67 103 32 1 27.3 10.9 45.9 26.5 29.9 27.2 15.6 0.0 72. 7 89.1 54.1 73.5 70.1 72.8 84.4 100.0 40 34 48 25 44 75 27 1 100.0 97.1 93.8 84.0 84.1 89.3 85.2 100.0 9.8 9.9 11.2 9.7 8.2 7.3 5.3 21 3-19 0-21 0-23 0-22 0-30 0-36 0-13 21 3 meters 1965 11- 4 12- 8 1966 2-24 3-30 4-29 5-19 6- 1 1 3 100.0 -- 0.0 -- ---- 93 57 17 12 1 31.2 24.6 23.5 0.0 0.0 68.8 75.4 76.5 100.0 100.0 24 43 13 7 1 --- ---- --- 79.2 83.7 100.0 71.4 100.0 4.3 5.6 4.5 6.4 6 0-9 0-25 1-10 0-16 6 6 mete rs 1965 12- 8 1966 2-21 3-30 4 25.0 75.0 3 33.3 13 0-13 1 22 100.0 18.2 0.0 81.8 -17 -94.2 --4.4 ^ — 0-11 9 meters 1965 12- 8 1966 2- 5 3- 30 4-29 6- 1 12 9.1 90.9 8 87.5 9.7 0-15 1 4 1 1 0.0 25.0 0.0 0.0 100.0 75.0 100.0 100.0 1 3 1 1 100.0 100.0 100.0 100.0 7 2.0 3 3 7 1-3 3 3 88 Table 21. (cont'd.) No. Date Adults 7As Male % Female No. Females exam, for e R g s Gravid females Avg. no. eggs/ % Gravid Ind. Ran; No egR; 12 meters 1965 12- 8 2-21 3-30 1966 1-19 19 9 2 21.1 0.0 0.0 78.9 100.0 100.0 15 9 2 46.7 100.0 100.0 4.6 4.3 2 0-9 2-6 2 5 0.0 100.0 5 80.0 3.5 0-5 89 Candona elliptica Candona elliptlea Is an elongated, laterally compressed ostracod .8-1.0 an in length. Males are slightly larger than females and do not exhibit the extreme sexual dimorphism characteristic of many other Gull Lake Candona. elliptica was perhaps the most agile of all the burrow­ ing ostracods and often was associated with the phytobenthos. This species is capable of walking or climbing on the smooth glass surfaces of culture dishes and moves much more rapidly than other species of Candona. Consequently, microscopic sorting of ostra­ cods often was delayed to remove C . elliptica from the sides of the sorting tray or from species already sorted. Forty-eight Ekman, 1 free-fall core, 58 aquatic net and 1 fish sample contained C. elliptica. Data from the sediment core and fish samples are not included in the following tables. The number of ostracods investigated is sunmarized below. Total No. ________________ No.____ Adults Gull Lake Fish Analyses * 503* 1 No. Juvenile 374 129 1 No. Males No. Females No. females exam, for eggs 104 1 264 --- 264 -- Sex of 6 adults not recorded ^ . elliptlea was one of the few Candona species whose reproductive cycle spanned the summer months with little overlap into other seasons. This is not surprising when considering the animals' close association with rooted aquatic plants. Locomotive behavior and anatomy of the mouth parts indicate that this ostracod is an eplstrate feeder that probably grazes on the attached periphyton of rooted plants. spatial and depth distributions of The elliptica also support this 90 assumption. Large numbers of adults were present in shallow reed beds and 6 meter depths where rooted plants were abundant. No specimens were collected from depths greater than 9 meters; this depth corresponds generally to the lower limit of the littoral region, depth to which rooted aquatics extend. the I do not infer that food resource is the only factor which controls the spatial distribution of C. elliptica. The life history of elliptlea is not easily interpreted from data presented in Tables 22-23. The reproductive cycle is clearly indicated, but the growth and development of juveniles and consequent changes in the size and age structure of the population are not readily delineated. Higher temperatures and longer growing seasons at shallower depths may account for the major differences in the size structure of the population at the depths sampled (see Table 22). The life history of C. elliptica is discussed more easily by noting that the reproductive cycle begins in May and ends in Sept­ ember. In August of 1965, adult C. elliptica were collected from depths of <3, 3, 6 and 9 meters (Table 22). A majority of the females were gravid and the maximum number of uterine eggs in a given individual was 24. By mid-September, 55 percent (N = 11) of the females from <3 meter samples were oviferous, laying period. but presumably near the end of their egg Most of their eggs were large and located in the posterior portion of the uterus. Fifty-two non-gravid adults collected from late September to mid-October indicated that the reproductive cycle had ended in September (Table 23). I could not determine if the latter adults represented offspring produced during the Sumner of 1965 or ostracods that had completed their first reproductive cycle. Inmature ostracods were found on 1 November, 4 November, and 91 Table 22. Date Percentage of adult and juvenile Candona elliptica in samples collected from Gull Lake. Transect No. or Samples__________ Location % N Adult % Juvenile 1 13 23 6 24 12 2 14 10 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 50.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 50.0 12 12 4 3 46 1 12 36 100.0 100.0 100.0 100.0 100.0 100.0 100.0 30.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 69.4 <3 meters 1965 8- 2 8- 3 9-14 9-22 9-27 10- 4 10-13 10-18 11- 1 ►66 1- 4 1-11 2-21 3-26 5-31 6- 1 6-15 7- 2 1 2 4' 10 f 4f 3T 4' 4' 7' 51 6’ 1 1 5' 1 3' 3’ KL MN, KL GI, 61, 62, GI GI 66 GI GI GI GI GI GI AB AB GI AB GI GI 3 meters 1965 8-3 8-5 11- 4 1966 2-24 3-30 4-29 5-19 6 - 1 6-15 7 - 1 7-15 8 - 1 2 1 2 MN, KL AB AB, EF 9 3 5 100.0 100.0 0.0 0.0 0.0 100.0 4 2 1 1 AB AB, IJ AB AB AB AB AB AB AB 2 5 1 3 8 9 1 8 7 100.0 80.0 100.0 0.0 100.0 88.9 100.0 50.0 71.4 0.0 20.0 0.0 7 2 21 100.0 100.0 23.8 2 2 1 2 2 100.0 0.0 11.1 0.0 50.0 28.6 6 meters 1965 8-3 8-5 11- 4 ' 1 1 1 Aquatic Net KL AB AB 0.0 0.0 76.2 92 Table 22. Date (cont'd.) Transect No. or % % Samples__________ Locat Ion___________N_______ Adult_____ Juvenile 6 meters 1966 2-21 3-30 4-29 5-19 6 - 1 6-15 7 - 1 7-15 8 - 1 1 2 1 1 2 2 1 2 2 AB IJ AB AB AB AB AB AB AB 3 15 7 15 43 45 3 12 33 100.0 13.3 0.0 26.7 97.7 88.9 100.0 33.3 42.4 0.0 86.7 100.0 73.3 2.3 11.1 0.0 66.7 57.6 9 meters 1965 8-3 12- 8 1966 5-19 1 1 KL EF 2 1 100.0 0.0 0.0 100.0 1 AB 2 0.0 100.0 93 8 December, respectively, in samples from 0-3, 6 , and 9 meters. were in some samples but none had eggs. Adults All of the ostracods collected at depths of <3 meters from early January to mid-June 1966 were mature, but gravid females were not found until 31 May (Table 23). Non-gravid females, adult males and late instar juveniles were collected throughout the winter and spring of 1966 from 3-9 meter samples. Gravid females were found on 19 May and 1 June, respectively, in 6 and 3 meter samples. Overlapping age classes of juveniles and adults were impossible to differentiate during the summer of 1966. Some juveniles from the previous fall had not matured before the first young of the summer generation began to appear. Most early to mid- sunner adults were offspring of the preceding year, but toward the end of the study some adults probably represented early maturing individuals of the summer generation. The maximum number of eggs in gravid females was 30. Sex ratios were highly variable in most samples because of the small number of adults. Males were more abundant in late summer and early fall than in other seasons at depths of <3 and 3 meters. samples with over 40 adults less than 35 percent were males. In 94 Table 23. Date Sex ratio of adult Candona elliptlea and average number of eggs In the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 22. No. Adults % Male % Female No. Females exam. for eggs % Gravid Gravid Females Avg. No. eggs/ Ind. Range No. eggs <3 meters 1965 8- 2 8- 3 9-14 9-22 9-27 10- 4 10-13 10-18 11- 1 1966 1- 4 1-11 2-21 3-26 5-31 6- 1 6-15 7- 2 1 13 23 6 24 12 2 14 5 0.0 30.8 52.2 -- 29.2 16.7 50.0 42.9 40.0 100.0 69.2 47.8 -— 70.8 83.3 50.0 57.1 60.0 12 12 4 3 46 1 12 11 16.7 0.0 0.0 0.0 17.4 0.0 16.7 45.4 83.3 100.0 100.0 100.0 82.6 100.0 83.3 54.6 17 10 1 8 3 100.0 77.8 54.5 -0.0 0.0 0.0 0.0 0.0 6.0 3.7 10.7 ------- 6 0-6 0-12 ------- 10 12 4 3 38 1 10 6 0.0 0.0 0.0 0.0 94.7 100.0 90.0 66.7 ----6.1 9.0 5.8 9.0 ----0-23 9 0-13 0-18 1 9 11 3 meters 1965 8- 3 8- 5 1966 2-24 3-30 4-29 6- 1 6-15 7- 1 7-15 8- 1 9 3 55.6 33.3 44.4 66.7 4 2 100.0 50.0 10.0 8.5 4-24 0-17 2 4 1 8 8 1 4 5 0.0 25.0 0.0 12.5 50.0 0.0 75.0 40.0 100.0 75.0 100.0 87.5 50.0 100.0 25.0 60.0 2 3 1 7 4 1 1 3 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 ---6.6 9.8 28.0 16.0 6.3 ---1-13 6-14 28 16 4-9 100.0 50.0 0.0 10.3 8.5 -- 2-18 0-17 -- 6 meters 1965 8- 3 8- 5 11- 4 7 2 5 0.0 0.0 20.0 100.0 100.0 80.0 7 2 4 95 Table 23. Date 1966 2-21 3-30 5-19 6- 1 6-15 7- 1 7-15 8- 1 (cont'd.) Mo. Adult s 3 2 4 42 40 3 4 14 % % Male Female 0.0 50.0 25.0 31.0 32.5 33.3 0.0 35.7 100.0 50.0 75.0 69.0 67.5 66.7 100.0 64.3 No. Females exam. for eggs__ 1 1 3 29 27 2 4 9 % Gravid 0.0 0.0 100.0 69.0 74.1 100.0 100.0 88.9 Gravid Females Avg. No. eggs / Ind. --11.3 10.4 6.3 10.5 10.3 9.0 Range No. eggs --8-14 0-24 0-10 8-13 4-21 0-30 9 meters 5 8- 3 2 50.0 50.0 1 0.0 -- -- 96 Candona porterl An undescrlbed species in the Gull Lake ostracod fauna is Candona porteri. I assigned the name _C. porteri for convenience of discussion; the new species will be named and described in a future report. J3. porteri is one of the larger species of Candona and reaches a maximum length of nearly 2 two. The male is similar in shape and slightly larger than the female. Late instar juveniles resemble the adult, but the first four instars are not easily identified. C. porteri was collected from 28 Ekman dredge samples and 8 free-fall sediment cores. Two adults and 2 juveniles from 4 of the sediment cores are not considered in the tables below. The number of specimens examined is presented in the following summary. No. Total No. ________________ N o ._____ Adults Gull Lake * 921* No. Juveniles No. Males No. Females 505 126 288 416 females exam. for eggs 282 Sex of two adults not recorded The summer reproductive cycle of C. elliptica in the littoral region of Gull Lake porteri. had its closest sublittoral counterpart in . This species essentially was restricted to depths greater than 9 meters. Only 1.2 percent at depths less than 12 meters; (N = 921) of the ostracods were found 3 were adults (Table 24). Early instar juveniles were collected on 19 January 1966 at a depth of 15 meters, but were not encountered again until 19 May. Four gravid females were found in late May and April samples from the above depth. These females presumably were among the earliest 97 Table 24. Date Percentage of adult and juvenile Candona porteri in samples collected from Gull Lake. No. Samples Transect or % Location____________N______ Adult % Juvenile 6 meters 1966 7-15 1 AB 5-19 6-15 1 2 AB AB 1 100.0 0.0 1 9 100.0 11.1 0.0 89.9 49 184 123 90 151 36.7 57.6 94.3 54.4 67.5 63.3 42.4 5.7 45.6 32.5 11 2 2 4 34 28 94 73 0.0 100.0 100.0 50.0 0.0 14.3 6.4 0.0 100.0 0.0 0.0 50.0 100.0 85.7 93.6 100.0 61 6.6 93.4 9 meters 12 meters 5-19 6- 1 6-15 7- 1 7-15 1 2 2 2 2 AB AB AB AB AB 1-19 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 2* 1 1 1 2 2 2 2 IJ AB AB AB AB AB AB AB 6- 1 2 AB 15 meters 18 meters * Free Fall Cores 98 to mature as no specimens were found in samples from other depths until 19 May. Data in Table 24 indicate that the reproductive cycle of began before 19 May. were C. porteri However, most specimens were immature and males more abundant than females at the 12 meter depth. In every other species of Candona studied, the proportion of the males was always highest during the initial phase of the reproductive cycle. Table 25 shows the marked change in adult sex ratios during the early simmer. Of 26 eighth instar juveniles sexed on 19 May, 54 percent were males; 56 percent of the adults collected on this date were males. The latter comprised 51 percent (N = 106) of the adults on 1 June, but the per­ centage of males decreased markedly thereafter. Estimates of the number of males per square meter at 12 meter depths on 1 June-15 June and 1 July-15 July were, respectively, 1169, 693, 238, 346. The percentage of gravid females In 12 meter samples increased during June and July and all females from 15 meter depths were oviferous. The number of adults per square meter, estimated from replicated 12 meter samples, ranged between about 2200-2500 in June and July. A low estimate of 1060 on 1 July may be strongly biased, but this cannot be stated with certainty. No samples were collected from depths of 12 and 15 meters after 15 July 1966. However, the absence of C. porteri in October and November 1965 samples suggests that the 1966 reproductive cycle probably ended by September or early October. 99 Table 25. :e Sex ratio of adult Candona porteri and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 24. No. AduIts % Male % Female No. Females exam. for eggs % Gravid Gravid Females Avg. No. eggs/ Ind. Range No. eggs 6 meters >6 7-15 1 0.0 100.0 1 100.0 3.0 3 100.0 100.0 4.0 14.0 4 14 5-19 6-15 1 1 0.0 0.0 9 meters 1 100.0 100.0 1 5-19 6- 1 6-15 7- 1 7-15 18 106 116 49 102 55.6 50.9 27.6 22.4 15.7 12 meters 44.4 8 49.1 52 72.4 84 77.6 38 80 84.3 50.0 75.0 95.2 97.4 100.0 6.0 4.7 6.5 6.1 7.7 6 0-9 0-11 0-12 1-13 3-30 4-29 5-19 6-15 7- 1 2 2 2 4 6 0.0 0.0 50.0 0.0 33.3 15 meters 2 100.0 2 100.0 50.0 1 100.0 4 66.7 4 100.0 100.0 100.0 100.0 100.0 4.5 3.4 8.0 7.5 3.0 4-5 2-5 8 2-12 2-4 6- 1 4 0.0 18 meters 100.0 4 100.0 3.0 1-6 100 Candona aff. caudata Male C. caudata were not reported in the original species descrip­ tion or in most subsequent papers. In addition, several morphologic features of the carapace in specimens from Gull Lake differ slightly from those described by other Investigators. Assignment of the specific name, therefore, is tentative, but recognition of ostracods that I call C. aff. caudata is positive. strongly dimorphic. Males are larger than females and This species is Intermediate in size between (L acuta and C, porteri. Thirty-two Ekman, 7 free-fall core and 2 fish samples contained C. aff. caudata. Five sediment cores and the fish samples are not considered in Tables 26 and 27; however, data from one Ekman dredge sample collected on 23 January 1965 are included. The number of specimens examined is presented below. Total No. ________________ No.____ Adults Gull Lake Fish Analyses * 880* 476 No. Juveniles No. Males 294 102 398 581 476 No. females No. exam. Females for eggs 479 78 374 -- Sex of 5 individuals undetermined Excluding 4 specimens from 9 meter depths, C. found only in the sublittoral areas of Gull Lake. aff. caudata was This species was first collected on 4 November 1965 from depths of 12 and 15 meters (Table 26). Less than 25 percent (N = 326) of the ostracods were mature, and the females were non-gravid. Almost one-half of the individuals had weakly calcified valves. In addition, males constituted 45 and 39 percent of the adult totals, respectively, in 12 and 15 101 Table 26. Date Percentage of adult and juvenile Candona aff. caudata in samples collected from Gull Lake. No. Samples Transect or Location N % % Adult Juvenile 9 meters 1966 5-19 6- 1 50.0 0.0 2 50.0 2 100.0 134 246 27 100.0 0.0 22.4 81.5 77.6 18.5 32 58 27 7 0.0 0.0 21.2 0.0 5 100.0 100.0 77.8 100.0 100.0 100.0 AB, EF, GH 86 15.1 84.9 109 50 18 14 16 8 1 90.8 84.0 100.0 9.2 16.0 0.0 0.0 2 2 1 AB, IJ AB AB AB AB AB AB 2* 2 IJ AB 2 100.0 100.0 1 1 AB AB 12 meters 1965 1-23 11- 4 12 - 8 1966 2-21 3-30 4-29 5-19 6- 1 6-15 1 3 1 IJ AB, CD, GH EF 1 2 1 1 2 2 AB AB, IJ AB AB AB AB 20 0.0 0.0 15 meters 1965 11- 4 1966 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 2 1 1 2 100.0 100.0 100.0 100.0 0.0 0.0 0.0 18 meters 1-19 6- 1 * Free Fall Cores 11 0.0 0.0 102 meter samples. Based on evidence above and knowledge of the general reproductive patterns of other Candona species, able to conclude that C. I believe It reason­ aff. caudata was just beginning the reproductive cycle in early November. Nearly 82 percent of the ostracods were mature in one 12 meter sample on 8 December and 58 percent of the females were gravid. The average number of uterine eggs per individual was 4.7 and the maximum number was 9 (Table 26). Two adults collected in January 1966 were not sexed. Data from one 12 meter sample collected on 23 January 1965 indicate that the annual reproductive cycle of C. aff. caudata began about the same time in both years. Twenty-five females from the above sample had an average of 7.0 eggs per individual with a range of 3-11 (Table 27). Males comprised only 12 percent (N * 134) of the sample. Oviferous females must have deposited the majority of their eggs between mid-December 1965 and late February 1966. The average number of uterine eggs was less than 6.0 in females collected from February through mid-July. was 18. The maximum number of eggs in a given individual Every gravid female examined from 12 and 15 meter samples had either released most of her eggs or was just beginning to reproduce. Data on the average and maximum number of eggs found in samples of C. aff. caudata accurately represent the reproductive condition of the individuals examined. However, these data reveal almost nothing about the actual number of eggs laid by Individuals of this species. The size of adult females and the space occupied by uterine eggs in those individuals examined indicate that at least 30 eggs could be held in the uteri of a single female. In other species of Candona studied 103 the uterus on each side of the body was greatly distended before the first eggs were released. Most ostracods collected from February to 15 July were mature. The number of individuals and the maximum and average number of eggs in gravid females gradually decreased during this period. No aff. caudata were collected from 12 meter samples after 15 June; 9 adults were found in 15 meter samples from 15 June-15 July. About 30 dead females were sampled from May-July, and in June-July most living adults were inactive, emaciated, and obviously near physiological death. The percentage of males in the population was greatest in the initial stages of the reproductive cycle. On 4 November, 45 per­ cent of the 12 meter adults were m a l e s , but by 8 December they had decreased to about 16 percent. From February through 15 June females constituted a minimum of 87 percent of the adults in each sample. The percentage of males in 15 meter samples decreased less abruptly, but followed the same pattern observed in 12 meter samples. 104 Table 27 Date Sex ratio of adult Candona aff. caudata and average number of eggs in the uterus of adult females. Number and location of samples corresponding to dates are presented in Table 26. No. Adult s X Male X Female No. Females exam. for eggs X Gravid Gravid Females Avg. No. eggs/ Ind. Range No. eggs 9 meters 1966 5-19 6- 1 1 2 0.0 0.0 100.0 100.0 1 2 100.0 100.0 18.0 9.0 18 4-14 12 meters 1965 1-23 11- 4 12- 8 1966 2-21 3-30 4-29 5-19 6- 1 6-15 134 55 22 11.9 45.4 15.8 88.1 54.6 84.2 25 30 19 100.0 0.0 57.9 7.0 3-11 --- --- 4.7 0-9 32 58 21 7 20 5 12.5 8.6 9.5 0.0 5.0 0.0 87.5 91.4 90.5 100.0 95.0 100.0 21 53 19 6 19 5 100.0 96.2 89.5 66.7 84.2 100.0 ■>.8 3.6 4.5 5.8 2.7 2.6 2-11 0-7 0-9 0-6 0-4 1-5 15 meters 1965 11- 4 1966 3-30 4-29 5-19 6- 1 6-15 7- 1 7-15 1-19 6- 1 13 38.5 61.5 8 0.0 --- --- 99 42 18 14 16 8 1 28.3 16.7 11.1 0.0 0.0 12.5 100.0 71.7 83.3 88.9 100.0 100.0 87.5 0.0 71 31 16 14 16 7 98.6 100.0 93.8 100.0 93.8 57.1 4.9 5.9 4.7 4.3 4.4 3.3 0-10 1-11 0-8 1-7 0-7 0-4 --- --- --- 100.0 0.0 18 me te r s — 0.0 100.0 11 --- --- --- 100.0 4.3 2-7 2 11 105 OCher Non-Swliming Species The eight species of ostracods discussed below accounted for 5.9 percent of the total number of non-swimming animals identified to species. A summary of the number of ostracods investigated is fol­ lowed by a brief discussion of each species. No results are presented in tabular form, but certain observations are supported by emperlcal data. Candona scopulosa Total No. ________________ No._____ Adults Gull Lake 377 314 No. No. Juveniles Males 63 No. females No. exam. Females for eggs 112 202 189 Aside from a single juvenile collected on 8 December 1965, C. scopulosa was found only in February and March of 1966, at depths of <3 meters. Seventy-seven percent (N = 167) of the February specimens were mature and about 68 percent (N = 77) of the females were gravid. The latter averaged 10.8 eggs per individual, but the maximum number was 26. In March 89 percent (N = 209) of the ostracods were mature and 94 percent (N * 112) of the females were oviferous. The average number of eggs per gravid individual was 8,2 and the maximum, 21. relatively abundant; Males were in February they constituted slightly more than 35 percent (N * 128) of the adults and in March 36 percent (N = 186). This species may have more than one generation per year as indi­ cated by the presence of several gravid females in <3 meter samples in the summer of 1967. 106 Candona aff. suburbana Total No. ________________ No_._____ Adults Gull Lake 274 No. Juveniles 63 211 No. females No. No. exam. Males____ Females for eggs 30 33 33 This species was found In the littoral areas of Gull Lake. Five ostracods were from 9 meter samples, but over 75 percent came from depths of 0-3 meters. One hundred percent (N = 274) of the ostracods collected from August-December 1965 were juveniles. The reproductive cycle of C. aff. suburbans apparently began by January 1966 as all February specimens (N « 60) were mature and more than one-half of the females were gravid. The average number of eggs in oviferous females was about 12 and the maximum 25. Two females and 1 male were collected in March and April; each female had 10 uterine eggs. The sex ratio was essentially equal in February with 29 males and 31 females. Hundreds of gravid females not considered here were collected from organic rich substrates in March and April of 1967. Candona decora Total No. _________________No._____ Adults Gull Lake Fish Analyses * 131 1 No. Juveniles 131* 1 Sex of 20 adults undetermined 2 No. Males 22 1 No. females No. exam. Females____ for eggs 89 13 107 (L decora was found in 1 fish, 5 aquatic net and 2 Ekman samples. Two specimens were collected from 9 meters, 1 from 12 meters and 130 at depths of <3 meters. Non-reproductive adults were first encountered in early December 1965. By January 1966, females were gravid and all On 11 February, 8 gravid females averaged individuals were mature. 23.8 eggs per individual; the maximum number of eggs was 32. oviferous females were sampled in early April. Several Males constituted less than 35 percent of the adults in all samples. Candona crogmanlana Total No. _______________ No._____ Adults Gull Lake Fish Analyses 64 3 21 3 No. Juveniles 43 No. Males 8 - No. females No. exam. Females for eggs 13 3 13 crogmanlana was collected from depths of 0-15 meters. were found in October. Juveniles By 4 November late instar juveniles, non- gravid adults and a few gravid females were present; six gravid ostra­ cods had a mean of 9.0 eggs and a maximum of 13. female carried 67 eggs and another, 69. In February, 1 Three oviferous females had a mean of 24.3 and a maximum of 41 uterine eggs in late March. Candona sp. Total No. _______________ No._____ Adults Gull Lake 45 2 No. Juveniles 43 No. Males 1 No. Females 1 No. females exam. for eggs -- Thls species was collected in late March 1966 from depths of 1-2 108 meters. Early to late instar juveniles and two recently molted adults indicated that the reproductive cycle of C. sp. had just begun. Candona distincta Total No, N o ._____ Adults Gull Lake 2 No. Juveniles 2 No. Males No. females No. exam. Females for eggs 2 -- Two gravid females were collected from a depth of 9 meters in June of 1966. Llmnocythere verrucosa No. Total No. Gull Lake 26 No. Adults No. Juveniles 4 22 No, Males 1 No. Females females exam. for eggs 3 This species was found in August-September 1965 and June-July 1966, at depths of <1-13 meters. L. verrucosa was more abundant than indicated by the susmary above, but was not easily separated from sed­ iment using the water surface tension technique Empty carapaces were abundant in the lake sediment. Cytherissa lacustris Total No. Gu 11 Lake 3 No. Adults 3 No, Juveniles No. Males No. Females 3 No. females exam. for eggs No living representatives of this species were collected. Three specimens apparently had died shortly before they were collected because all appendages were intact and soft tissues such as muscles had not decomposed. Empty carapaces were abundant in 12 and 15 meter samples. 110 Free-Sw limning Species Cypridopsis vidua Free-swinmlng ostracods were not sampled quantitatively acid in many Instances not in relation to their numerical dominance over other species. Nevertheless, these ostracods often were extremely abundant and surely accounted for a large share (if not most) of the standing crop of ostracods in the littoral areas during the suraner months. Cypridopsis vidua Is perhaps the most cosmopolitan of all fresh­ water ostracods and is found in a great variety of aquatic habitats. This ostracod is easily identified and probably is the best known spe­ cies in the United States. populations in nature. Few investigators, however, have studied An excellent account of the anatomy and biology of £. vidua was given by Keeling in 1951. C. vidua reproduces parthe- nogenetlcally, but young are not retained by the parent as in Darwinula. Seventy-one Ekman, vidua. 66 aquatic net, and 5 sediment cores contained Results of fish stomach analyses are not included in Table 28, but data collected in July 1965 are incorporated. The number of specimens examined is presented below. No. Total No. ________________ N o .______Adults Gull Lake 1143 Fish Analyses 120 742 120 No. Juveniles 401 No. Mal e s --- No. Females 742 120 females exam. for eggs 596 --- C. vidua inhabited the littoral region of Gull Lake; one adult from 12 meters and one juvenile from 18 meters were the only exceptions. Juveniles and adults, month of the study. including gravid females, were collected in every Overlapping age groups were impossible to Ill differentiate. Some juveniles were recruited Into the adult size group each month and, when gravid, could not be distinguished from older adults. There were three rather vaguely defined generations produced during the study period. The first reproductive period was in November and December of 1965. This generation of adults likely re­ presented offspring of the preceding sumner. uterine eggs in one adult was 17. occurred in March and April. A second reproductive period The third and most distinct reproductive cycle was in the summer of 1966. Almost 90 percent ostracods were mature and 86 percent gravid (Table 28). The maximum number of (N = 244) of the (N = 93) of those examined were The largest number of eggs per individual was observed in the summer months with a maximum of 21 eggs in several females. 112 Table 28. Month 1965 July Aug. Sept. Oct. Nov. Dec. 1966 Jan. Feb. Mar. Apr. May June July Aug. Adult-juvenile ratio and reproductive state of adult Cypridopsis vidua in composite monthly samples from Gull Lake. Samples Include individuals collected at depths of from <1 to 9 meters on different dates and from different localities. No. Samples Ekman Net-Core 11 5 1 6 2 12 10 2 2 7 9 4 Total No. % Adult % Juvenile No. exam. for % Range eggs Gravid No. eggs 2 3 12 17 7 4 430 35 43 121 16 19 46.0 51.4 62.8 74.4 93.8 84.2 54.0 48.6 37.2 25.6 6.2 15.8 200 14 37 80 11 4 64.0 50.0 48.6 45.0 36.4 100.0 12 19 19 52 3 41 92 244 9 89.5 78.9 84.6 66.7 75.6 51.1 88.9 66.7 10.5 21 .1 15.4 33.3 24.4 48.9 11.1 33.3 17 15 44 11.8 20.0 65.9 — — — 8 3 3 ------- 29 46 93 6 96.6 37.0 86.0 16.7 0-13 0-4 0-6 0-4 0-5 2-17 0-1 0-2 0-6 ------- 0-9 0-9 0-21 0-10 113 Physocyprla pustulosa Physocypria pustulosa was extremely abundant in or near thick growths of rooted plants such as Najas or Potamogeton. Although density estimates were not calculated for this species, means of quantitative samples would have resulted in estimates of about 100,000 individuals, or more, per square meter. _P. pustulosa was collected from depths of <1-15 meters, but less than 2 percent (N *= 522) were from depths greater number of ostracods examined is presented below. Total No. _________________N o .______Adults Gull Lake * 522* No. Juveniles 464 55 No. Males 215 than 9 meters. No. Females The No. females exam. for eggs 215 146 Sex of 34 adults not recorded Juveniles or adults were present throughout the study. No spe­ cimens were collected in January 1966 (Table 29), but this was a bias in sampling and did not denote the absence of _P. pustulosa in Gull Lake. The major reproductive cycle occurred in the August 1965, gravid females carried an average of individual, suinner months. In nearly 6 eggs per but in October and November all females were non-gravid. The latter possibly represented the early maturing offspring of the Sumner adults (Table 30). By December some females were oviferous. do not know if a second egg laying cycle began before the summer of 1966. The range in the number of eggs observed in February and March (Table 30) may indicate that some females deposited their eggB in March. The maximum number of eggs in July and August adults was 10. The proportion of males was lowest in summer samples and highest I 114 in February and March. Males also were more abundant than females in April samples , but those specimens collected were used in other studies and are not enumerated in Table 30. 115 Table 29. Date 1965 Aug. Sept. Oct. Nov. Dec. i6 Jan. Feb. Mar. Apr. May June July Aug. ' Percentage of adult and juvenile Ifoysocypria pustulosa from Gull Lake. Samples represent composite monthly collections from different localities and depths. No. Total Samples '_________ No ■______ 2 6' 2, 1' 10 4 18 29 23 73 25 6 7 2 2 , 110 7 5 197 38 7 4 54 24 30 — Aquatic Net and Core Samples % % Adult____ Juvenile 100.0 75.9 73.9 87.7 64.0 0.0 24. 1 26. 1 12.3 36.0 ------- ------- 97.5 81.6 2.5 18.4 ------- 100.0 77.8 100.0 90.0 ------ 0.0 22.2 0.0 10.0 116 Table 30. Sex ratio and reproductive cycle of adult Fhysocyprla pustulosa collected from Gull Lake. % Date_______ N 1965 Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. 1 2 18 22 l7l 641 16 192 312 7 4 42 24 27 Male Gravid F ema1e s No, Avg. No, % Females % eggs/ Female____ exam.____ Gravid_____ Ind. 44.4 40.9 47.1 51.4 31.3 55.6 59.1 52.9 48.6 68.7 10 4 7 18 11 70.0 75.0 0.0 0.0 27.3 5.6 3.0 -- 61.5 71.0 38,5 29.0 17 8 --- --- 25.0 23.8 25.0 25.9 75.0 76.2 75.0 74.1 Sex not recorded for 27 adults Sex not recorded for 7 adults — 3 30 18 20 Range No. eggs 0-10 0-6 --- ----- 3.0 0-6 17.6 37.5 3.7 2.0 0-8 0-2 --- --- --- 66.7 86.7 83.3 65.0 5.0 5.8 5.9 3.8 0-6 0-9 0-10 0-6 117 Cypria turnerl This ostracod was one of the most abundant species in Gull Lake. No other ostracod was as uniformly dispersed, geographically or with depth, as turner!. The latter was found at depths of a few milli­ meters to 32 meters and was present in every major substrate sampled. Ten sediment core, 29 aquatic net and 116 Ekman samples contained turner!. tables. Fish stomach analyses are not considered in the following The number of ostracods studied is presented below. Total No. ________________ No._____ Adults Gull Lake 4006 Fish Analyses 29 1 2 No. Juveniles 23251, 2 29 44 -- No. Males No. Females 500 --- 1585 ----- No. females exam. for eggs 1555 - Aliquot of 460 Individuals represents estimate for 2097 animals collected on 4 February. Sex undetermined for 240 adults. The number of C^. turner! collected each month (Table 31) did not reflect its actual abundance. C^. turnerl usually was the numerically dominant species at 9 meters, but: was more numerous at other depths. Juveniles were collected only in the sunnier months (Table 31) which Indicated a single generation per year and possibly an annual turnover of the population. The average and maximim number of eggs in gravid females is not presented in Table 32, but the reproductive state of females is noted. Most females were oviferous in November-January, but a few gravid indi­ viduals were collected in February and early March. Juveniles of the Sumner generation, 1965, matured and reproduced in November-December and laid their eggs before February 1966 (except 118 those noted In Table 32*). began to appear in June. The first juveniles of the next generation These young had not completely replaced the parental stock by August 1966. In late July and early August samples, recently molted adults were less pigmented than older ostracods and were easily detected. recruited adults. About 70 percent of the animals were newly The high percentage of males in October of 1965 also indicated that most were new recruits from the preceding smrnner generation. 119 Table 31. Percentage of adult and juvenile Cypria turner! in com­ posite monthly samples collected from different localities and depths in Gull Lake. No. Total % % Date______ Samples________ No.__________Adult______ Juvenile 1965 Aug. Sept. Oct. Nov. Dec. 1966 Jan. Feb. * Mar. Apr. May June July Aug. ' * 6, l1 A1 2, 31 20, 7 1 A, A* 13 15 A 15 5 3, 7' 20 16 6 235 2 6A 269 301 99.1 100.0 100.0 100.0 100.0 0.9 0.0 0.0 0.0 0.0 21 125 2097 180 A7 27 219 298 121 100.0 100.0 100.0 100.0 100.0 100.0 89.5 9A.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 10.5 6.0 0.0 Aquatic Net and Core Samples Four samples from small shallow man made lagoon with outlets directly to Gull Lake along Transect AB. Total represents aliquots from 4 samples and adult-Juvenile percentages derived from an aliquot of 460 adults from the sample. 120 Table 32. Sex ratio and reproductive state of Cyprla turnerl collected from Gull Lake. No. % % Females Date_____ N_______ Male_____ Female_______ Examined______ With eggs 1965 Aug. Sept. Oct. Nov. Dec. >6 Jan. Feb. * Mar. Apr. May June July Aug. * ^ ^ 233 2 64 268 301 ----- ----- ----- 43.8 17.2 21.6 56.2 82.8 78.4 35 200 236 No. Yes Yes 21 125 460 180 47 27 196 280 121 23.8 14.4 35.4 16.1 21.4 14.8 24.0 17.5 30.6 76.2 85.6 64.6 83.9 78.6 85.2 76.0 82.5 69.4 16 107 297 151 33 16 149 231 84 Yes No Yes No* No No No No No See Table 31 for explanation. Aliquot of 460 adults from larger sample. Three of 180 females with eggs Sex of 5 adults not recorded 121 Other Free-SwIroning Species Cyclocypris ampla Total No. _________________No._____ Adults Gull Lake Fish Analyses 474 4 393 4 No. Juveniles No. Males 81 - No. females No. exam. Females for eggs ---- - ----- 226 --- The sex ratio of Cyclocypris ampla is not recorded in Table 33. Many individuals died with their valves completely closed and not be dissected without destroying the carapace. could Non-reproductive adults were collected from August through December 1965 (Table 33). continued to June. eggs. The reproductive cycle began by February 1966 and Gravid females carried a maximum of 10 uterine Early instar Individuals had hatched by June and passed through successive molts to the adult instar by August 1966. All but ampla were collected from depths of 0-3 meters. This 9 _C. ostracod is a powerful swimmer and capable burrower. Cyprinotus glaucus Total No. _________________N o .______Adults Gull Lake 1569 Fish Analyses 7 842 No. Juveniles 727 7 No. Males 260 -- No. Females___________ 582 --- CL glaucus had the most restricted temporal distribution of any ostracod in Gull Lake and probably the most rapid population growth rate. Individuals of this species passed through 8 successive molts to the adult Instar and began reproducing within about one month 122 Table 33. Percentage of adult and juvenile Cyclocypria ampla in composite monthly samples from Gull Lake. Only 3 individuals from depths greater than 9 meters. No. Total Date______ Samples__________N o . 1965 Aug. Sept. Oc t . Nov. Dec. ,6 Feb. Mar. Apr. May June July Aug. ' % % Adult Juvenile 2, 2' 9* 1' 4 2 18 3 1 6 2 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 15 8 1 L, 1 ' 4 6 2 44 63 4 21 33 184 95 100.0 100.0 100.0 100.0 36.4 67.4 100.0 0.0 0.0 0.0 0.0 63.6 32.6 0.0 Aquatic Net Maximum eggs is _10 during February Females Gravid No No No No No Yes Yes Yes Yes Yes No No 123 from the time of hatching. Juveniles were abundant in May 1966, and had matured and begun to reproduce by 1 June (Table 34). The life cycle ended In September of 1965; this also was true in 1967. maximum number of eggs observed In gravid females was 15. The This ostracod may be the most abundant free-swlnmer in the littoral region during the sunnier period; In biomass, the littoral dominants. I am certain that glaucus is one of Most data concerning _C. glaucus are not yet analyzed. Potamocyprls smaragdlna No. Total No. Gull Lake Fish Analyses 31 1 No. Adults No. Juveniles 30 1 1 No. Males No. Females females exam, for eggB -- -- P. smaragdlna was collected In July and August 1965 and July 1966 from depths less than 3 meters. Some females were gravid in August 1965, but eggs were not counted. Cyclocypris cruciata No. Total No. Gull Lake 15 No. Adults No. Juveniles No. Males No. Females 2 7 15 females exam. for eggs 3 Cyclocypris cruciata was collected from depths of 3-12 meters, but most were found at depths less than 9 meters. the sunnier, but not in October or February. Females were gravid in 124 Table 34. Date 1965 July Aug. Sept. 166 May June July 1 Sex ratio and juvenile - adult percentages of Cyprinotus glaucus collected from Gull Lake. % % No. Samples N Adult 3' 3* 8' 1189 32 16 63.7 71.9 100.0 36.3 28.1 0.0 757 23 16 30.1 13.0 50.0 91 36 205 0.0 47.2 14.1 100.0 52.8 85.9 *«* « 17 29 23.5 58.6 76.5 41.4 5’ 2, 3 ’ 1 , 3' Aquatic Net Samples All less than 6 meters Majority at 3 meters or less No. Juvenile_____ Adult % 7. Male_____ Female 69.9 87.0 50.0 125 Fish Stomach Anaylses Ostracods were found in the stomach and Intestines of three species of fish from Gull Lake. The yellow perch, Perea flavescens (Mitchill), bluegill, Lepomls macrochlrus Rafinesque, and smelt, Osmerus mordax (Mitchill), were the only species examined. Four perch were captured In one Inch gill nets from depths of 6-9 meters, along Transect IJ, on 2 December 1965. They were killed immediately following capture; the standard length of each individual was less than 100 nm. one male Two male and one female Candona ohioensis and decora were found in the stomach of one fish. The stomachs and intestines of the other fish were empty. On 9 January 1966, 15 perch were collected in gill nets from a depth of 15 meters. These specimens were caught near the base of the net shortly before it was retrieved. The perch ranged In length from 91-143 ran; only one was longer than 116 mm. The stomach and gut contents of each individual were carefully examined under a dissecting microscope. A total of 1322 ostracods representing 7 species and 2 genera were collected. remainder Of this number 760 were Identified to species and the to genus. All of the ostracods were mature; most individuals identified only to genus could have been identified to species had time permitted. The concentration of ostracods and other organisms in dis­ crete groups in the stomachs and intestines of most fish clearly indicated selective feeding for each food organism. In addition, the ostracods were completely free of debris even though most were nonswinmers. Males were dominant numerically in those species identified only to genus, but accurate counts of each sex were not recorded. 126 Ostracods were found In 13 perch and were dominant, and biomass, In 7. Cladocera and tendlpedld larvae were dominant food organisms in eight fish. of each species, In number Table 35 shows the number of ostracods sex, and percentage of males. Candona ohioensls. _C. aff. caudata and (2. acuta were the most abundant species; males of the last two species comprised more than 50 percent of the total sample. Three perch were examined on 13 February 1966. These specimens were collected in gill nets along Transect IJ at a depth of 16 meters. Two perch contained ostracods, but the third had only 2 tendipedid larvae In its stomach. 5 species in two genera: C. One fish yielded 136 ostracods representing Candona ohioensls, inopinata and Cypria turner!. with C, acuta second and acuta, C. aff. caudata, (J, aff. caudata was most abundant ohioensls third. Males comprised nearly 64 percent of the 69 _C. aff. caudata,a much higher percentage than was ever found in sediment core or Ekman dredge samples. percent of 28 acuta and 25 percent of the 4 C. Only 14.3 ohioensls were male. Thirty-three specimens were identified only to genus but all belonged to Candona. The second perch ha d 4 male tendipedidlarvae in the intestine. aff. caudata and 40 The fish ranged in length from 91-95 an. Thirteen blueglll (L. macrochirus) were captured and dissected on 19 July 1966. The animals were collected from gill nets in the shallow water littoral areas along Transect AB. Four fish had no organisms in the stomach or gut and only three fish contained ostracods. One hundred sixty-seven ostracods were found in the stomachs of the three fish. ampla, 7 C. There were 120 adult Cypridopsis v i d u a , 4 Cyclocypris c ruciata. 1 Potamocypris smaragdlna. 7 juvenile Cyprinotus 127 glaucus, 27 juvenile Candona ravsonl and 1 elllpttea. The stomach and intestines of most fish contained appreciable quantities of plant debris, small sand grains and other organisms. The latter included amphipods, beetle larvae, tendipedid larvae, tricopteran larvae, water mites, snails, annelids and small minnows. The bluegills obviously were far less discriminate in their feeding behavior than were the perch. Nevertheless, they demonstrate that fish predators contribute to the mortality of juvenile ostracods as well as adults. Three adult Cypria turner! were found in the stomachs of 2 smelt (0. mordax) collected in September and October 1966. These fish were collected in an Issacs Kidd trawl from depths of 12 and 20 meters. 128 Table 35. Sumnary of Che number, sex, and percentage of male ostracods removed from 15 yellow perch (Perea f laves Gens') collected tn Gull Lake on 9 January 1966. Total Percent Species_____________ No ■________ Male______ Female______ Male Candona ohioensls 159 10 149 6.3 Candona acuta 152 96 56 63.2 Candona aff. caudata 403 350 53 86.8 14 6 8 42.9 Candona crogmanlana 3 0 3 0.0 Candona rawsonl 1 1 0 100.0 562 -- -------- — -------- Candona inopinata Candona spp. Cyprla turnerl 28 DISCUSSION Distribution The geographic distribution of many freshwater ostracods in the United States is scarcely known. This is particularly true in regard to the burrowing, non-swlnming species. Several species of the Gull Lake ostracod fauna are widely dispersed or have fossil records which indicate more expansive distribution than currently known. Most of the difficulty in assessing the distribution of a given species results from the limited number of investigations of the Ostracoda in the United States. The ostracod faunas of relatively few states have been studied. Among those works which deserve special note are studies by Furtos (1933) Ohio, (1935) Massachusetts, (1936a) Florida and North Carolina; Hoff (1942) Illinois; Dobbin (1941) Washington; Danforth (1948) Iowa; Tressler (1954) Texas; Ferguson (1954, 1958a) South Carolina, (1957) Michigan, Other authors (1962) Louisiana; and Cole (1965, 1966) Tennessee. (e.g. Moore 1939; Hoff 1943b) have studied the ostracod faunas of more restricted areas. Some problems arise from taxonomic error. the ostracod Candona rawsoni. A good example Is This species was described by Tressler (1957a) from the Great Slave Lake, Canada. Unfortunately, Tressler erred in the description of an important taxonomic feature of the carapace. Thus, Gutentag and Benson (1962) described the same species from Pleistocene deposits in Kansas as (L nyensis. 129 Later, Benson and 130 MacDonald (1963) reported nyenats In sediments of Lake Erie, and Staplln (1963a) described it (C. swalni) from Pleistocene deposits of Illinois. Swain (1947) had reported this species from Pleistocene deposits of Utah and Bronstein (1947) described the ostracod from the U.S.S.R. under the name obtusa. Finally, Delorme (1967c) examined the collections of Tressler and clarified the taxonomic status of CJ. rawaoni. He found (1967b) C^. rawsoni in Canadian lakes and later (1968) in Pleistocene sediments from the Yukon. Most of the free-swimming species in Gull Lake also have a wide geographic range. Cyprlnotus glaucus and Cyclocypris ampla recently were reported from Canada (Delorme 1967b) and from Pleistocene deposits in the Yukon (Delorme 1968). These two species formerly were known only in the midwestern United States. Some distribution records of fossil and living ostracods repre­ senting certain Gull Lake species are sutmiarized below. Candona ohloensis Candona rawsoni Ohio-(Lake Erie) Furtos (1933); (Pleistocene) Winkler (1962) Indiana-(Pretty Lake) McGregor (this thesis) Michigan-(Gull Lake) McGregor and Wetzel (1968) Canada-(Saskatchewan) Delorme (1967a, 1967b); (Manitoba-Pleistocene) Klassen et al. (1967) Utah-(Pleistocene) Swain (1947) Kansas-(Pleistocene) Gutentag and Benson (1962) Illinois-(Pleistocene) Staplin (1963a) Ohio-(Holocene) Benson and MacDonald (1963) Michigan-(Gull Lake) McGregor and Wetzel (1968); (Lake Michigan) McGregor (this thesis) Canada-(Great Slave Lake) Tressler (1957a); (Saskatchewan) Delorme (1967a, 1967b); (Manitoba-Pleistocene) Klassen et al. (1967); (Yukon- Pleistocene) Delorme (1968) Candona inopinata Ohio-Furtos (1933) Michigan-MeGregor and Wetzel (1968) 131 Candona acuta Candona elliptica 111lnols-Hof£ 1942 Michigan-MeGregor (this thesis) (Lake Michigan) McGregor (this thesis) Canada-(Manitoba-Pleistocene) Klassen et al. (1967); (Saskatchewan) Delorme (1967b) Ohio-Furtos (1933) Iowa-Danforth (1948) Michigan-MeGregor (this thesis) Massachusetts-Furtos (1935) Tennessee-Hoff (1943a) S. Carolina-Ferguson (1954, 1958a) Candona crogmanlana Candona caudata Ohio-Furtos (1933) Michlgan-Moore (1939); Kenk (1949); (Rolocene) Winkler (1960); (Lake Michigan) McGregor (this thesis) Illinois-Hoff (1942); (Pleistocene) Staplln (1963a) Iowa-Danforth (1948) Kansas-(Pliocene-Pleistocene) Gutentag and Benson (1962) Georgia-Turner (1895) Canada-(Great Slave Lake) Tressler (1957a) Massachusetts-Furtos (1935) Illinois-Hoff (1942) Ohio-(Pleistocene) Winkler (1962); (Lake ErieHolocene) Benson and MacDonald (1963) Kansas-(Pleistocene) Gutentag and Benson (1962) Washington-Dobbin (1941); Scheffer and Robinson (1939) Canada-(Recent and Pleistocene) Delorme (1967a, 1967b, 1968) Ferguson reviewed the taxonomy and distribution of several species common to Gull Lake. In 1944, he reported Potamocypris smaragdlna from Missouri and later (1954, 1958a) from S. Carolina. The following year Ferguson (1959b) published a synopsis of the genus including description of a new species. from Louisiana Still later he recorded _P. smaragdlna (1962) and Wyoming (1966) and listed it as one of 10 United States species in his (1967b) taxonomic key. Ferguson (1959a, 1967a) also studied the taxonomy and distribution of Cypridopsis and CyprinotuB. Freshwater ostracods are more widely distributed, geographically and ecologically, than most aquatic biologists are aware. Many species tolerate a broad range of chemical and physical conditions and are associated with a diverse array of aquatic organisms. For example, freshwater ostracods have been reported from saline ponds and 132 brackish-water habitats (deVos 1954; Neale 1965; Latour et al. 1966; Hulings 1967), caves and subterranean waters (Kile 1934, 1936; LBffler 1963), hot springs (Moniez 1893; Castenholtz 1967), bromeliad leaves (Tressler 1941, 1956; Furtos 1936b), crayfish burrows (Creaser 1931), and semiterrestrial forest litter habitats (Chapman 1960). Many species are consnon in shallow vernal ponds and other temporary waters whereas some occur in the profundal regions of deep lakes or in streams and rivers. Endemic species are known in some of the older lakes, as Lake Baikal (Kozhov 1963) and Lake Ohrid (Stankovi