Eastern Illinois University The Keep
Masters Theses Student Theses & Publications
1996 Ecology and Community Relationships of the River Cooter, Pseudemys concinna in a Southern Illinois Backwater Michael J. Dreslik Eastern Illinois University This research is a product of the graduate program in Zoology at Eastern Illinois University. Find out more about the program.
Recommended Citation Dreslik, Michael J., "Ecology and Community Relationships of the River Cooter, Pseudemys concinna in a Southern Illinois Backwater" (1996). Masters Theses. 1882. https://thekeep.eiu.edu/theses/1882
This is brought to you for free and open access by the Student Theses & Publications at The Keep. It has been accepted for inclusion in Masters Theses by an authorized administrator of The Keep. For more information, please contact [email protected]. THESIS REPRODUCTION CERTIFICATE
TO: Graduate Degree Candidates (who have written formal theses)
SUBJECT: Permission to Reprodu~e Theses
The University Library is rece1v1ng a number of requests from other institutions asking permission to reproduce dissertations for inclusion in their library holdings. Although no copyright laws are involved, we feel that professional courtesy demands that permission be obtained from the author before we allow theses to be copied. PLEASE SIGN ONE OF THE FOLLOWING STATEMENTS: Booth Library of Eastern Illinois University has my permission to lend my thesis to a reputable college or university for the purpose of copying it for inclusion in that institution's library or research holdings.
Date
I respectfully request Booth Library of Eastern Illinois University not allow my thesis to be reproduced because:
Author Date Ecology and Community Relationships of the River Cooter, Pseudemys concinna
in a Southern Illinois Backwater.
(TITLE)
BY Michael J. Dreslik
THESIS
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Masters of Biology
IN THE GRADUATE SCHOOL, EASTERN ILLINOIS UNIVERSITY CHARLESTON, ILLINOIS
1996 YEAR
I HEREBY RECOMMEND THIS THESIS BE ACCEPTED AS FULFILLING THIS PART OF THE GRADUATE DEGRE_!,..C,ITED ABOVE
5--31-f{r DATE
DATE DEDICATION
This thesis is dedicated to all my past science teachers, Mrs. Parret, Mr. Krueger and
Mr. Gutzler, who all have initiated the spark for my interest in biology. Secondly, this
thesis must be dedicated to my parents, for without their support I would not be here
today.
ii ACKNOWLEDGEMENTS
Funding for this project was provided through the Illinois Endangered Species
Protection Board, Eastern Illinois Council on Faculty Research, Eastern Illinois Honors
Program Undergraduate Research Award, the Linnaeus Fund of the Chelonian
Research Foundation and the E.l.U. Graduate Summer Research Assistantship Award.
I also thank Leon Ezrah Bryant for his assistance in the field (without which much of this study would not have been completed), Edgar Joyner for allowing us to trap on
Long Pond, Edmund Bickett for allowing us to trap on Big Lake, and to Jason Dreslik for his endless hours of fieldwork. I would also like to thank R. U. Fischer, for his review of the drafts of this manuscript and aid in the compilation of the growth curves, E. K. Bollinger and K. C. Kruse for their comprehensive reviews and suggestions, and C. Pederson and J. Ebinger for aid in identification of the plants and algae.
iii TABLE OF CONTENTS
TITLE PAGE ...... i
DEDICATION ...... ii
ACKNOWLEDGEMENTS ...... iii
TABLE OF CONTENTS ...... iv - v
CHAPTER 1. Ecology of the River Cooter, Pseudemys concinna, from a
Southern Illinois Backwater...... 1 - 34
Abstract ...... 1
Introduction ...... 1 - 3
Methods ...... 3 - 6
Results ...... 6 - 8
Discussion ...... 8 - 13
Literature Cited ...... 14 - 18
Tables ...... 19 - 21
Figures 22 - 34
CHAPTER 2. Ecology of a Freshwater Turtle Community in Illinois, with Regional Comparisons...... 35 - 68
Abstract ...... 35
Introduction ...... 36 - 37
Methods ...... 37 - 40
Results ...... 41 - 44
Discussion 44 - 49
iv Literature Cited 50 - 53
Tables ...... 54 - 56
Figures 57 - 68
Key to Abbreviations and Site Loaclities ...... 69
v Ecology of the River Cooter, Pseudemys concinna, from a Southern Illinois
Backwater.
ABSTRACT
In Illinois the River cooter, Pseudemys concinna, is an enigmatic endangered
species. Even throughout its range, ecological studies on the River cooter are rare.
During 1994 and 1995 I quantified the: growth rates and trends, population size and
structure and dietary habits of a population from floodplain lake in Gallatin county,
Illinois. Population estimation (Schnabel method) predicted 136 individuals at a density of 4.6 turtles/ha with a biomass of 3.6 Kg/ha. The sex ratio is slightly male biased (1:1.14). From back-calculated growth data, the von Bertalanffy growth model
(Fabens' method) estimates males with a higher growth rate (k) and lower asymptotic length (a) (k = 0.00062, a= 219.8) than females (k = 0.00023, a= 307.6). Male growth slowed between 5 - 10 years and females between 15 - 23 years. Estimated daily growth regressed to plastral length revealed that growth of both sexes declined as size increased. Diets were predominantly filamentous algae of the genera Cladophora and Oedogonium (98% ).
INTRODUCTION
Ecological studies afford us insight into the dynamics of a natural population through which life history tactics can be surmised and used for the conservation of the population. Many population ecology studies have been compiled for turtles emphasizing: population sizes and structures, home ranges, growth rates, feeding and
1 nesting ecology, reproductive biology and demographics (see Bury, 1979 and Ernst,
1994 et al. for reviews). Some of these studies have been the starting point for long
term life history studies (Congdon et al., 1993). However, a disproportionate share of these studies have concerned a few common generalized species (e.g. Chelydra serpentina, Chrysemys picta, and Trachemys scripta) while many of the rarer species and the riverine specialists have been neglected. I selected the River cooter for study due to its state endangered status and the sparsity of information available concerning its ecology.
The first record of the River cooter in Illinois was based on a specimen collected in the Wabash River near Mt. Carmel in Wabash County (Garman, 1890).
Today the species is known to exist in Alexander, Gallatin, Hardin, Jackson, Massac,
Randolph, Jersey, Union and White counties (Moll and Morris, 1991; Smith, 1961 and
Cahn, 1937). Despite the moderately broad distribution in southern Illinois, the ecology of the River cooter is poorly understood in Illinois (Smith, 1961 and Cahn,
1937) and throughout the species' range (see Ernst et. al. 1994 for review). The dearth of ecological information is due partly to the turtle's wary nature, its relatively inaccessible habitat and its herbivorous diet (which reduces incidental capture with baited traps). However, in Illinois, a greater problem has been locating adequate numbers to study. The species is classified as endangered in Illinois (Herkert, 1992) and even as early as 1937, Cahn wrote that it was too rare to study in Illinois except over extended periods. In 1988 a population of River cooters in a series of backwater lakes near the Ohio River in Gallatin County was discovered (Moll and Morris, 1990).
2 Subsequent trapping indicated that the population was the largest known in the state.
To enhance the ecological data of the river cooter a study commenced at Round Pond,
the most accessible of the lakes, in May 1994 and continues to date. This paper
reports the findings on the community composition, population size and structure,
growth trends and diet.
STUDY SITE
Located about 4 km west of the Ohio River, Round pond is a relatively small
member (ca. 30 ha) of a chain of lakes stretching along the Ohio river flood plain in
southern Gallatin County, Illinois (Figure 1). A fishing camp comprising cabins and
trailers occupies most of the west bank. A sand beach stretches along the southern
shore and the remaining shore is flood plain forest. The scalloped edges of the pond
provide several shallow coves and bays attractive to turtles. Emergent aquatic
vegetation, predominantly spatterdock, Nuphar luteum, lines the shore in places.
During annual floods, the Ohio River connects the lakes directly or via a series of
sloughs.
MATERIALS & METHODS
Individual and Population Attributes
General Methodology. I studied the Round Pond population for 40 days between 17 May 1994 and 30 Sept 1995, collecting turtles with fyke nets, baited hoop traps and trammel nets. Each captured turtle was marked with unique combinations of marginal scute notches (Cagle, 1939), measured using a metric vernier calipers, to the nearest mm, and weighed with pull spring scales (Pesola®) or electronic balances
3 (Ohaus®), to the nearest gram. Routine measurements included: mass (g), carapace
length (CL; mm), carapace width (CW; mm), plastral length (PL; mm), shell height
(SH; mm), and the plastral scute lengths (along the mid-ventral seam). Annuli were measured to the nearest 0.01 mm along the mid-plastral seam on the left pectoral scute for growth studies. Turtles were aged using annuli counts whenever the areola was present (Zug, 1991). Mature males were sexed by their elongated fore-claws and a cloaca! vent which extended beyond the posterior margin of the carapace. Turtles exceeding the carapace length of the smallest determined male but lacking male sex characteristics were categorized as females. Determining the maturity of nongravid females requires dissection, an option not available with a state endangered species.
Therefore, I assumed a minimum plastron length of 200 mm for mature females based on studies of two closely related species that are specific congeners in some classification schemes (see Ward, 1984 and Seidel, 1994). Jackson (1970) reported
190 mm CL as the smallest mature £ .( concinna) suwanniensis found in a Florida population. Gibbons and Cocker (1977) reported adult female f. (concinna) floridana exceed 200 mm PL in South Carolina.
Population Estimation. Using mark/recapture methods I estimated population size of River cooters using the Schnabel method (Schnabel, 1938):
Em (u+r) p = Er (1)
Where P is the estimate of population size, m is the total number of marked individuals in the population, u is the number of unmarked turtles captured each day, and r is the total number of marked turtles captured each day. Biomass was
4 determined in g/ha, for each age/sex category, by multiplying the mean weight of the
age/sex category by the population estimate of that age/sex category. I summed all
four categories for total biomass.
Growth Analysis. The relationship between the plastral seam length of each
plastral scute and the total plastral length in both sexes was determined using simple
correlations. I used a modified Sergeev's formula (Sergeev, 1937) to estimate plastral
lengths at the time an annulus was formed:
(2)
Where PLi is the plastral length at the ith age class, PLc is the current plastral length, ANi is the annulus on the left pectoral scute at the ith age class and ANc is the current length of the left pectoral scute.
I fit the data obtained from the plastral lengths to the Fabens' method of the von Bertalanffy growth equation (Fabens, 1965 and von Bertalanffy, 1957) using the
SAS nonlinear function (SAS, 1982). The equations used were:
(3)
b = 1 - (~) (4)
Where L1 is the greatest length derived from Sergeev's formula, L2 is the smallest length derived from Sergeev's formula, a is the asymptote, d is the estimated number of days between L1 and L2, b is a variable related to hatchling size, e is the base of the natural logarithms and k is the intrinsic rate of growth. To compare the effectiveness of the von Bertalanffy model to actual growth data I combined back- calculated, known-age individuals with those aged by Sexton's method (1959) and
5 plotted them. Linear regressions were used to determine the relatedness between back
calculated daily growth and plastral size in an attempt to determine if growth is
inversely proportional to turtle size.
Dietary Analysis. Turtles were held for 24 hours in buckets of water to
retrieve feces that were then preserved in 10% formalin for later study. Samples were
analyzed under a microscope and the presence of each dietary group was recorded to the lowest identifiable taxon. Specimens were then divided into animal, higher plant and lower plant material and the volummes determined by the water displacement method as described by Moll and Legler (1971).
RESULTS
Individual Attributes
Turtle Size. Cooters ranged from 72 to 292 mm PL (82 - 322 mm CL) with a mode of 30 individuals (46%) falling between 90 and 130 mm PL (Table 1 and Figure
2). Males ranged from 101 - 191 (mean= 158) mm PL while females ranged from
183 - 292 (mean = 236) mm PL. The size difference between male and female ~. concinna is significant (tca1c = 6.68, df = 27, P > 0.00001). A bimodal distribution was observed with one mode encompassing the juveniles, immature females and males while the second described the larger mature females (Fig. 2). Presumably, the absence of individuals below 70 mm PL is due to trapping biases, because the throats of the fyke nets would allow smaller individuals to escape.
Growth. Sergeev's formula back calculates turtle size from growth zones on a specific plastral scute. To determine which scutes best correlated to size I found the
6 correlation coefficients (r) provided the best fit with the gular (r = 0.976) followed by
the abdominal scute (r = 0.953). However, the pectoral scute (r = 0.948) was used for
growth calculations because it possessed the clearest annuli. The curves from known
aged individuals and individuals aged by Sexton's method (1959) revealed rapid
growth for males through year five and a slower steady increase in size for females
through age six (Fig. 3). Based on the von Bertalanffy equations, as modified by
Fabens (1965) for unknown ages, males had a lower asymptotic length (219.7 mm PL)
and a higher rate of growth (k) that decelerated rapidly around ten years of age (Fig.
4). The asymptote of the female growth curve is 307.5 mm PL and the growth rate
declined between 15 and 25 years of age. Growth rate correlated negatively with
turtle size with the female slope (-0.1712, Fig. 5) being steeper than the males (-
0.1186, Fig. 6).
Diet. Filamentous algae, Cladophora and Oedogonium predominate the 16
fecal samples I collected. The numerous species of unicellular algae and diatoms
present, though probably taken incidentally while feeding on algal mats, presumably
augment nutrition. Only three samples (two juveniles and one female) contained higher plants, aquatic grasses (Poaceae). Volummetrically, lower plants comprised
98% of the total food ingested and no significant differences were evident between ages and sex groups (X2 = 1.28, df = 1, P = 0.53).
Population Attributes
Population Estimate, Density and Biomass. Of the 65 £. concinna captured,
15 were recaptured over the two-year study. Based on these data, the Schnabel
7 method estimates the cooter populations at Round Pond to be I36 individuals or a density of 4.6 turtles per hectare of surface water. The estimated total cooter biomass is 3.6 kg/ha comprised of 64.5% (2.3 kg/ha) adult females, I5.4% (0.59 kg/ha) males and 20. I% (0. 73 kg/ha) juveniles and immatures.
Sex and Age Composition. The adult sex ratio was l.I4:1, I6 males to I4 females, with no significant difference (tca1c = 0.13, df= I, P = 0.72). The 33 juveniles and immatures comprised 53.4% and the 30 adults 47.6% of the population, a I. I: I juvenile to adult ratio.
DISCUSSION
Individual Attributes
Turtle Size. At 322 mm CL and 292 mm PL, the River cooter is the largest emydid in the state. In fact the female, described above, is larger than the maximum for Illinois specimens (3I4 mm CL) reported in Smith (196I). Although Cahn (1937) reported two larger, Smith (personal communication) thought these actually came from
Tennessee. Berry and Shine (I980) reported that aquatic, swimming turtles typically are sexually dimorphic with the females attaining a larger size. River cooters exemplify this pattern with males rarely reaching 200 mm PL whereas most females exceeded this size (mean 236).
Growth. Jones and Hartfield (I995) reported that the pectoral scute was inadequate as an indicator of plastron length in Graptemys oculifera and Moll and
Legler (I97I) also found the abdominal scute inadequate for Trachemys scripta. In this study, correlation coefficients for the pectoral and abdominal scute length with
8 plastron length were 0.948 and 0.978, respectively, making either useful to estimate
previous plastron lengths in E,. concinna. This indicates the importance of checking
correlations between plastral scute length and plastral before back-calculations are
conducted.
To construct the von Bertalanffy curves, I used Sergeev's formula to back
calculate from growth rings (Jones and Hartfield, 1995). However, some problems arose with the asymptote of the male curve since I had numerous small individuals and few large ones, the curve failed to asymptote when undergoing successive refinement iterations (Fabens, 1965). Frazer et al. (1990) found that an excess of smaller individuals would offset the true asymptote and their removal had little effect on the curve when compared to the known age curve. Knowing this I removed all individuals less than 126 mm PL and ran the model again. This corrected the problem and I attained an asymptote that was just larger than our largest captured male but this left us with only six individuals. Results derived from known-age and individuals aged by Sexton's (1959) method are similar (Fig. 3). For males and females if the von Bertalanffy trajectory is transposed over the known-age curves the model falls on or very close to the standard errors making it a reliable estimator through the first six years of age for females and first five years for males. This substantiates Jones and
Hartfield's (1995) findings, in that, back-calculated growth trajectories differ little from known-age growth trajectories of turtles.
According to the model, females grew slower than males but continued to grow for a longer period of time with growth slowing between 15 and 23 years as opposed
9 to 8 to 14 years for males. Congdon and van Loben Sels (1993) reported that growth
of female Blanding's turtles similarly slowed between 14 - 20 years. Currently, von
Bertalanffy growth curves exist for Graptemys oculifera (Jones and Hayfield, 1995),
Trachemys scripta (Dunham and Gibbons, 1990), Caretta caretta (Frazer and Ehrhart
1985 and Frazer et al., 1986), Chrysemys picta (Frazer et al., 1991), Kinostemon
flavescens (Iverson, 1991) and Chelonia mydas (Frazer and Erhart, 1985 and Frazer
and Ladner, 1986). All are similar, in that, growth rate is a function of size being
most rapid for juveniles then decreasing for each species with age resulting in the
leveling off of the growth trajectories for each sex sometime following maturity.
Diet. Cooters at my study site appeared to specialize on filamentous algae at
all ages. This is not unique to the Round Pond site, however, as filamentous algae
predominated the diets of cooter populations studied in Florida (Lagueux et al., 1995),
Missouri (Thomas et. al., 1994) and North Carolina (Brimley, 1943). Buhlmann and
Vaughan (1991) reported that New River, WV cooters ate filamentous algae only until
macrophytes appeared in the Spring. Most studies throughout the range have found ~
concinna to be chiefly herbivorous (Allen, 1938; Parker, 1939, Marchand, 1942,
Brimley, 1943 and Lagueux et. al., 1995). This is further substantiated by the finding
of a symbiotic relationship existing with a cellulolytic bacteria that allows for more efficient digestion of filamentous algae and plants (Thomas et al., 1994). However, exceptions exist; Cahn ( 193 7) reported that Illinois cooters were largely carnivorous but did not state the locality or habitat where he made his observations. Buhlmann and Vaughan (1991) found that New River cooter adults were chiefly herbivorous but
10 occassionly took crayfish and that juveniles were highly omnivorous. My data are incomplete because I collected only fecal samples where differential digestion may bias the view of the actual proportions of foods consumed. However, I opted for fecal samples over stomach flushing because this species is state endangered and stomach flushing of other species occasionally caused fatal injuries.
Population Attributes
Density and Biomass. The density estimates of cooters from Round Pond ( 4.6 turtles/ha) fell between those for West Virginia (Buhlmann and Vaughan, 1991) and a
Florida Spring (Iverson, 1982 from data in Marchand, 1942). The distribution of these extreme cases suggests that the low density in the West Virginia population may be due to its location at the northern edge of the range. However, Round Pond is also near the northern limits of the range but densities here are similar to £. floridana populations from Risher Pond and Ellenton Bay in South Carolina (Congdon et. al.,
1986 and Iverson, 1982). This suggests that other factors such as the productivity of the habitat may be more important than latitude in dictating turtle density.
Due to the relatively close proximity of the lakes and their annual connections during flooding of the Ohio River, it is entirely possible that this population is merely a deme of a regional metapopulation. Moll and Morris (1991) and Dreslik and Moll
(unpub. data) have located populations of£. concinna at Big lake, Fehrer lake, Long pond and Running Slough in the same general locality. These sites are no more than
1.5 Km from Round pond and during flooding they are connected allowing for habitat corridors between sites.
11 Iverson (1982) noted that biomass in turtle populations is a neglected subject and summarized the literature available for chelonians. Using data from Marchand
(1942) and Gibbons and Coker (1977), he was able to calculate previously unreported biomass information for populations of the closely related cooters, £. concinna and ;e,. floridana (Table 3). The Round Pond biomass estimate (3.6 kg/ha) is less than a tenth the biomass (384.2 kg/ha) of the population inhabiting the Florida spring studied by
Marchand (Iverson, 1982), but is similar to the estimate for Ellenton Bay, SC (3.8 kg/ha) obtained from data in Gibbons and Coker (1977). Again, the wide range in biomass estimates may correlate with habitat factors such as productivity but no data are available yet to substantiate this.
Sex Ratio: The few sex ratios reported for populations of ;e,. concinna and £. floridana are skewed toward males. The Round Pond sex ratio was slightly, but not significantly, skewed toward males (1.14:1). Round Pond's sex ratio is similar to a
Florida population (1.16:1) of£. concinna reported by Jackson (1970) but much less skewed than the three populations from New River, WV which had an overall 1.9:1 ratio (Buhlmann and Vaughan, 1991) and a population of£. floridana from Ellenton
Bay, SC had a male-biased ratio of 1.31:1 (Gibbons, 1990). Several factors can bias sex ratio calculations, especially season and trapping method (Gibbons, 1990), therefore, long term studies are necessary to determine if the male bias is valid.
Population Vulnerability and Management Considerations
The loss of genetic variability and allelic fixation through genetic drift are important conservation considerations with small, isolated populations (Lacey, 1987).
12 If Round pond's turtle population is indeed a small isolated population and not a
metapopulation, an exchange of animals between lakes could be warranted.
Determination of whether a metapopulation exists in southeastern Gallatin county is
thus significant. Future studies will seek to determine rates of gene flow and dispersal
patterns in this system. Although this species is on its northern range limit here in
Illinois, this study suggests that a lack of suitable habitat may be limiting populations of these organisms. Inasmuch as this chain of backwater lakes in Gallatin Co. currently supports the only sizable population(s) of River cooters known within the state, the possibility of establishing a refuge for the turtle or more stringent environmental controls at the site should also be explored.
13 LITERATURE CITED
Allen, R. E. 1939. Notes on feeding and egg-laying habits of the Pseudemys. Proc.
Florida Acad. Sci. 3:105-108.
Bertalanffy, L. von. 1957. Quantitative laws in metabolism and growth. Quart. Rev.
Biol. 32:217-231.
Buhlmann, K. A. and M. R. Vaughan. 1991. Ecology of the turtle Pseudemys concinna
in the New River, West Virginia. J. Herpetol. 25:72-78.
Cahn, A. R. 1937. Turtles of Illinois. Illinois Biol. Mono. 16:1-218.
Cagle, F. R. 1939. A system of marking turtles for future identification. Copeia
1939:170-173.
Cagle, F. R. 1942. Turtle populations in southern Illinois. Copeia 1942:155-162.
Congdon, J. D., A. E. Dunham and R. C. van Loben Sels. 1993. Delayed sexual
maturity and demographics of Blanding's turtles (Emydoidea blandingii):
implications for conservation and management of long-lived organisms.
Conserv. Biol. 7:826-833.
Congdon, J. D., J. L. Greene and J. W. Gibbons. 1986. Biomass of freshwater turtles:
a geographic comparison. Am. Midi. Nat. 115:165-173.
Dunham, A. E. and J. W. Gibbons. 1990. Growth of the slider turtle. In Gibbons J.
W. (ed.). Life history and ecology of the slider turtle. Smithsonian Inst. Press,
Washington, D. C. pp. 135-145.
Ernst, C. H., J. E. Lovich, and R. W. Barbour. 1994. Turtles of the United States and
Canada. Smithsonian Inst. Press, Washington, D. C.
14 Fabens, A. J. 1965. Properties and fitting of the von Bertalanffy growth curve. Growth
29:265-289.
Frazer, N. B. and L. M. Ehrhart. 1985. Preliminary growth models for green, Chelonia
mydas, and loggerhead, Caretta caretta, turtles in the wild. Copeia 1985:73-79.
Frazer, N. B. and R. C. Ladner. 1986. A growth curve for green sea turtles, Chelonia
mydas, in the U.S. Virgin Islands, 1913-1914. Copeia 1986:798-802.
Frazer, N. B., J. W. Gibbons and J. L. Greene. 1990. Exploring Fabens' growth interval
model with data on a long-lived vertebrate, Trachemys scripta (Reptilia:Testudinata).
Copeia 1990:112-118.
Frazer, N. B., J. W. Gibbons and J. L. Greene. 1991. Growth, survivorship and longevity
of painted turtles Chrysemys picta in a Southwestern Michigan marsh. Am. Midi.
Nat. 125:245-258.
Frazer, N. B., C. J. Limpus, and J. L. Greene. 1994. Growth and estimated age at maturity
of Queensland loggerheads. In K. A. Bjorndal, A. B. Bolten, D. A. Johnson, and P.
J. Eliazar (eds.). Proceedings of the 14th annual symposium on sea turtle biology
and conservation. NOAA Technical Memmorandum NMFS-SEFC-351. Hilton Head,
South Carolina. pp. 42-45.
Garman, S. 1890. Notes on Illinois reptiles and amphibians, including several species not
before recorded in the northern states. Ill. Lab. Nat. Hist. Bull. 3:185-190.
Gibbons, J. W. 1990. Sex ratios and their significance among turtle populations. In J. W.
Gibbons (ed.). Life history and ecology of the slider turtle. Smithsonian Inst. Press,
Washington, D. C. pp. 171-182.
Gibbons, J. W. and J. W. Coker. 1977. Ecological and life history aspects of the cooter
Chrysemys floridana (LeConte). Herpetologica. 33:29-33.
15 Iverson, J. B. 1982. Biomass in turtle populations: a neglected subject. Oecologica
55:69-76.
Jackson, C. G., Jr. 1970. A biometrical study of growth in Pseudemys concinna
suwanniensis. Copeia 1970:528-534.
Jones, R. L. and P. D. Hartfield. 1995. Population size and growth in the turtle Graptemys
oculifera. J. Herpetol. 29:426-436.
Lacey, R. C. 1987. Loss of genetic diversity from managed populations: Interacting
effects of drift, mutation, immigration, selection and population subdivision.
Conservation Biology 1:143-158.
Legler, J. M. 1977. Stomach flushing: a technique for chelonian dietary studies.
Herpetologica 33:281-284.
Lagueux, C. J., K. A. Bjomdal, A. B. Bolten and C. L. Campbell. 1995. Food habits of
Pseudemys concinna suwanniensis in a Florida spring. J. Herpetol. 29:122-126.
Marchand, L. J. 1942. A contribution to a knowledge of the natural history of certain
freshwater turtles. Master's thesis, University of Florida, Gainsville.
Moll, E. 0. and J. M. Legler. 1971. The life history of a neotropical slider turtle,
Pseudemys scripta (Schoepff) in Panama. Bull. L.A. Co. Mus. Nat. Hist. (11):1-
102.
Moll, E. 0. and M.A. Morris. 1991. Status of the river cooter (Pseudemys concinna) in
Illinois. Trans. Illinois Acad. Sci. 84:77-83.
Morris, M.A. and P. W. Smith. 1981. Endangered and threatened amphibians and reptiles.
In Bowles, M. L.(ed). Endangered and threatened vertebrate animals and vascular
plants of Illinois. Illinois Department of Conservation. Springfield, Illinois. pp. 21-
33.
16 Parker, M. V. 1939. The amphibians and reptiles of Reelfoot Lake and vicinity with a key
for the separation of species and subspecies. J. Tenn. Acad. Sci. 14:72-101.
SAS. 1982. SAS User's Guide. Statistical Analysis Systems, Inc., Carey, North Carolina.
Schnabel, Z. 1938. Estimation of the total fish population of a lake. Am. Math. Mon.
45:348-352.
Seidel, M. E. 1994. Morphometric analysis and taxonomy of cooter and red-bellied turtles
in the North American genus Pseudemys (Emydidae). Chelonian Conservation and
Biology. 1:117-130.
Sergeev, A. 193 7. Some materials to the problem of reptilian postembryonic growth.
Zool. Zh. 16:723-735.
Sexton, 0. J. 1959. A method of estimating the age of painted turtles for use in
demographic studies. Ecology 40:716-718.
Smith, P. W. 1961. The amphibians and reptiles of Illinois. Ill. Nat. Hist. Surv. Bull.
28:1-298.
Thomas, R. B., D. Moll and J. Steiert. 1994. Evidence of a symbiotic relationship between
cellulolytic bacteria and a freshwater herbivorous turtle. Southwest Nat. 39:386-388.
Ward, J. P. 1984. Relationships of chrysemyd turtles of North America (Testudines:
Emydidae). Spec. Pub. Mus. Texas Tech Univ. 20:1-50.
Zug, G. R. 1991. Age determination in turtles. S.S.A.R. Hep Circular. 28 p.
17 Table 1: Descriptive statistics for morphometric measurements taken from Pseudemys concinna captured at Round Pond, Gallatin County, Illinois during the summers of 1994 and 1995, as described by sex/age category. SD= Standard deviation.
Measurement Male Female Female Imm. Juveniles
Weight (g) Mean 584.9 1851.3 327.9 155.5 SD ±242.3 ±842.0 ±115.5 ±49.9 Range 160 - 975 950 - 3250 167 - 645 78 - 262
Caragace Length (mm) Mean 175.l 258.7 139.5 106.9 SD ±29.6 ±50.0 ±17.l ±11.5 Range 111 - 220 200 - 322 107 - 181 82 - 124
Caragace Width (mm) Mean 135.3 193.0 112.4 60.7 SD ±18.9 ±44.6 ±10.5 ±8.4 Range 93 - 160 144 - 318 94 - 136 72 - 104
Plastral Leng!h (mm) Mean 158.3 236.3 129.9 97.7 SD ±24.2 ±45.5 ±15.4 ±11.9 Range 101 - 191 183 - 292 102 - 168 75 - 118
Shell Height (mm) Mean 60.1 91.7 51.7 41.9 SD ±8.4 ±17.4 ±5.6 ±3.4 Range 43 - 77 71 - 125 42 - 63 36 - 48
18 Table 2: The frequency of occurrence, percentage of occurrence and percentage volume of dietary items found in 16 Pseudemys concinna fecal samples from specimens taken from Round pond, Gallatin County , Illinois. The number in parenthesis is the percentage of occurrence and the number in brackets is percentage volume. Percentage volume encompasses only the categories of higher and lower plants. The female category includes both immature and mature females. Total n = 16, male n = 4, female n = 6, juvenile n = 6.
Plant Material Plankton
Higher Plants Lower Plants
Poaceae Oedogonium Sagitarria CladoRhora Unicellular Algae Diatoms
Male 0 (0%) [0%] 3 (75%) 0 (0%) 3 (75%) [100%] 4 (100%) 4 (100%) Female 1 (17%) [2%] 4 (67%) 3 (50%) 6 (100%) [98%] 6 (100%) 6 (100%) Juvenile 2 (33%) [3%] 4 (67%) 1 (17%) 6 (100%) [97%] 6 (100%) 6 (100%) Total 3 (19%) [2%] 11 (69%) 4 (25%) 15 (94%) [98%] 16 (100%) 16 (100%)
19 Table 3: Population densities (in turtles per hectare) and biomasses (kilograms per hectare) of _e. concinna and _e. floridana populations from the literature and this study in comparison.
Species Density Biomass Location Reference
_e. concinna 4.57 3.62 Round Pond (IL) This Study
_e. concinna 170 384.2 Florida Spring Iverson, 1982 (calculated from Marchand, 1942).
_e. concinna 2.3 New River (WV) Buhlmann and Vaughan, 1991.
_e. concinna 0.7 New River (WV) Buhlmann and Vaughan, 1991.
_e. concinna 1.2 New River (WV) Buhlmann and Vaughan, 1991.
_e. floridana 154 311.1 Florida Spring Iverson, 1982 (calculated from Marchand 1942)
_e. floridana 7 7.8 Ellenton Bay (SC) Congdon et al., 1986
_e. floridana 4.6 .9 Risher Pond (SC) Congdon et al., 1986
_e. floridana 5.2 3.8 Ellenton Bay (SC) Iverson, 1982 (calculated from Gibbons and Coker, (1977)
20 Figure 1: Map of the research site and surrounding backwater lakes and sloughs in
southeastern Gallatin county, Illinois (37°45'W, 88°14'N). The inset of
Illinois is the present distribution with Gallatin county inside the square.
Solid circles = museum records, open circles = literature records, solid
squares = records from 1985-1988 and opens squares are possible
sightings reviewed by Moll and Morris (1991).
21 Hulda Lake
Black Lake Fehrer Lake
Ohio River
2 Ki lomcter~ Scale I :250000
22 Figure 2: The population structure in terms of plastral lengths for 63 Pseudemys
concinna individuals marked and captured during the two year study.
Only initial captures are included in this graph.
23 10-.----~~~~~~~~~~~~~~~~~~~~~~-,-~~~~~---,---, I I Male en 8 t'ZZ/! Female (I) 1'\'\"'J Female Imm t ::J 6 KXX] Juvenile l-o ..cCi> 4 E ~ 2
0-+---+------f'~"f--"-"'-+--+-~..__-+---+----+~+----+----+----+~-+--+--+-~+-----+-<--<-+---+-'-'--+--i-<-4+<----L-- 24 Figure 3: Psuedemys concinna of known age and those aged by Sexton's (1959) method including hi-low, mean and standard error boxes. Curve a) represents females and curve b) represents males. 25 a) b) 225 i L 200 ,-..-,E 175 . g 150 ..c:: +-I 01) 125 =a.> ~ 100 -~ ~ ~ 75 ~ -~ 50 25 I I O ____ l ___L _J_ __L_ ___ L _____ j __ _j______J______j______L______J______J 0 1 2 3 4 5 6 0 1 2 3 4 5 Age Class 26 Figure 4: von Bertalanffy growth trajectories calculated using Sergeev' s formula for male (n = 6) and female (n = 31) river cooters (Pseudemys concinna) for individuals captured in 1994 and 1995 at Round Pond, Gallatin County, Illinois. Juveniles were included in both curves while female immatures were also included in the female curve. 27 350 ------~~-·- Legend 300 Males 1- Females ,.-._ 250 ~- E ! E I "-" i I ..c:...... 200 t OJ} c::: l ~ ~ 150 ~- ...... ' Vi l ~ Ci: 100 ~; Males Females -I, I / a = 219.75 a= 307.55 II : K = 0.000619 K = 0.000225 50 b = 0.842296 h = 0.88714 r - - ____i I 0 [__1____ !_____1 ______L ___l ______j_ L______l ______J____ _J _ __j__J_ L_____t 0 10 20 30 40 50 60 Year 28 Figure 5: Scatter plot and linear regression for female river cooters captured during the two year study. The data were obtained from annuli using Sergeev's formula (Sergeev, 1939). The x - axis was weighted using the plastral length of individuals. 29 2.5 ~ • • 2.4 • • • 2.3 • • • • .s 2.2 bO i::: ~ • • ca 2.1 ., .b • .. - rJ'l ·1, .. • ~ 2.0 . ., 11. • . . • - • • • j 1.9 • • • •• • • • • • • • • .... 1.8 • • • 1.7 • 1.6 -5.5 -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 Ratio of the Ln Growth in mm/day versus Plastral Length 30 Figure 6: Scatter plot and linear regression for male river cooters captured during the two year study. The data were obtained from annuli using Sergeev's formula (Sergeev, 1939). The x - axis was weighted using plastral length of individuals. 31 2.3 ~ • • • • • 2.21 • • • 2.1 • £ • • • ~ • • • ~ 2.0 • ~ • -.....____ -b • . • C'-l • •• ~ - 1.9 • • • -~ • • • • • i::: .- ~ • • 1.8 1.7 • 1.6 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 Ratio of Ln Growth in mm/day versus Plastral Length 32 Figure 7: The species composition, expressed as a percentage of 376 turtles captures in Round Pond, Gallatin County, Illinois, during the Summers of 1994 and 1995. Abbreviations are TS= Trachemys scripta (66.5%), AS= Apalone spinifera (1.6%), CS= Chelydra serpentina (5.1%), GO = Graptemys ouachitensis (4.2%), SO= Sternotherous odoratus (3.2%), and PC = Pseudemys concinna (17.8%). Chrysemys picta (0.5%), Apalone mutica (0.5%), and Graptemys pseudogeographica (0.5%) compose the MISC category. 33 MISC GO 34 Ecology of a Freshwater Turtle Community in Illinois, with Regional Comparisons. ABSTRACT Relatively little information is available for turtle communities worldwide but the numbers of publications on the topic are slowly increasing. This study looks at an Illinois backwater turtle community and makes comparisons with other such assemblages within the state. At Round pond (Gallatin Co., Illinois) the turtle community comprises nine species: Trachemys scripta, Pseudemys concinna, Chelydra setpentina, Stemotherus odoratus, Graptemys ouachitensis, Apalone spinifera, Graptemys pseudogeographica, Chrysemys picta, and Apalone mutica. The dominant species in terms of biomass and density was I. scripta at 11.6 Kg/ha and 18.0 turtles/ha respectively. Dietary niches ranged from herbivorous specialists (£. concinna, H' = 0.14) to omnivorous generalists (I. scripta, H' = 2.50). Red-eared sliders at Round Pond shifted their diets from chiefly plants in spring to mostly animal matter (i.e. Bryozoans) in late summer. The greatest overlap in diet occurred between _e. concinna and I. scripta in late Spring and early Summer (Ch = 0.32), however most overlap involved an abundant resource, algae. Shannon-Wiener diversity values for all Illinois sites examined ranged from 1.1 - 2.3 with Round pond having an H' of 1.6. Omnivorous generalists predominated in communities occupying the three habitat types examined (Lotic-Mud, Lotic-Sand and Lentic-Mud). 35 INTRODUCTION The concept that organisms occur in assemblages of species interacting to produce distinctive community characteristics was neglected for many years in the herpetological literature. Prior to 1960 most ecological research on amphibians and reptiles was autecological (Scott, 1982). Over the last three decades reptilian community studies have appeared in the literature but the majority have involved squamates, particularly lizards. Bury (1979) observed that studies of interactions within chelonian communities were "virtually unknown." In fact, only two papers on interactions between chelonian species had appeared by 1979 - Mahmoud's 1968 study of feeding ecology in kinosternid turtles and Berry's 1975 study of the interactions between two musk turtles in a Florida spring. More recent publications concerning specific chelonian communities and interactions by Vogt (1981) Vogt and Guzman Guzman (1988), D. Moll (1990), Fuselier and Edds (1994) and Teran et al. (1995) and overviews of species richness and biomass on a global scale by Iverson (1982, 1992 a,b), Congdon et al. (1986) and Congdon and Gibbons (1989) have augmented the limited information on community ecology of turtles. Nevertheless, chelonian community studies still represent a small fraction of the herpetological community ecology literature. In the strict ecological sense community refers to the entire interacting biotic component of a particular ecosystem. Due to the difficulty of studying such a complex system, ecologists have commonly limited their studies to more manageable units which may have prompted Robert MacArthur's (1971) looser definition of 36 community: "any set of related organisms living near each other and about which it is interesting to talk." Herein, "chelonian community" refers to an assemblage of turtles living and interacting in a specific habitat In Illinois, there are few published accounts on turtle communities, however, some information is available in dissertations and reports (Gritters and Maudlin, 1994; Pierce, 1992; Moll, 1977 and Cagle, 1942). To augment the sparse chelonian community data for the state, I examined aspects of community ecology for the chelonian species occupying a pond on the floodplain of the Ohio River in southern Illinois. In this paper I will report on the 1) species richness, abundance, diversity, and dominance within this assemblage, 2) feeding guilds and dietary overlap between dominant species within the community and 3) comparisons with turtle communities in other Illinois freshwater habitats. METHODS Study Site Located about 4 km west of the Ohio River, Round pond is a relatively small member (ca. 30 ha) of a chain of ponds/lakes stretching along the Ohio river flood plain in southern Gallatin Co., Illinois (Fig. 1). During annual floods, the Ohio River connects to the lakes directly or via a series of sloughs. A fishing camp comprising cabins and trailers occupies most of the west bank. A sand beach stretches along Round pond's southern shore and the remaining shore is flood plain forest. The irregular edges of the pond provide several shallow coves and bays affording 37 submergent vegetation and basking sites. Emergent aquatic vegetation, predominantly spatterdock, Nuphar luteum, lines the shore. Routine Measurements I conducted research on Round Pond's turtle community during the Summer months between 17 May 1994 and 30 Sept 1995. Turtles were captured by hand and trapped using fyke nets (Vogt, 1980), baited hoop traps, dip nets, and trammel nets. All turtles were marked permanently by notching marginal scutes (Cagle, 1950), weighed using pull-spring scales (Pesola ®) and electronic balances (Ohaus ®) to the nearest gram, and measured with metric vernier calipers to the nearest mm. Measurements included: carapace length (CL), plastron length (PL), carapace width (CW), and shell height (SH). Finally, I sexed turtles using the secondary sex characteristics of vent extension beyond the posterior margin of the carapace and elongated fore-claws. Finally, I placed them into one of four age/sex categories (juvenile, female immature, male and female). I calculated the population density of the most frequently recaptured species using the Schnabel method (Schnabel, 1938). To calculate densities of the remaining species, I used the following proportional formula: ASP XP AS%= X3 Where ASP is the population estimate of the most frequently recaptured turtle, AS% is the percentage of the most frequently recaptured turtle in the sample, XP is the population estimate of species X and X 3 is the percentage of species X in the sample. To estimate biomass in Kg/ha, I took the mean weight of each species, 38 multiplied it by the population estimate and then divided that by the surface area (ha). Community Attributes Since studies on trapability are lacking I had to assume that all species are equally trapable by each method used during the entire study. In order to indicate and examine variation in species heterogeneity, I first constructed rank abundance curves for 1994 and 1995 based upon the proportion of turtles captured. Second, I plotted the number of fyke net hours required to collect the maximum number of species in Round Pond for 1994 and 1995 in order to determine in which year(s) fyke net trapping provided us with an unbiased estimate (see Vogt, 1980) of species richness and species dominance in terms of both abundance and biomass. I tested the yearly variations in community structure using contingency table analysis. I reviewed the literature (both primary and secondary) for data on community composition and relative abundance in other Illinois turtles communities, assuming that all studies had equal trapping effort and intensity. Studies in which less than 50 turtles were captured at a site were omitted as not being representative of the whole turtle community. A total of 15 communities chosen for this study were divided into three habitats: lentic-mud, lotic-sand and lotic-mud. The Shannon-Wiener function (H), for which eveness values (E) were calculated and used to measure the species diversity of Round Pond and to compare the diversity of 15 other Illinois chelonian communities. The mean diversity values for each category were tested for significance using a Kruskal-Wallis nonparametric test of variance. 39 Niche Attributes To determine diets turtles were held overnight to retrieve feces. Although biased because of differential digestibility of food items, this was deemed to be the safest method for examining the diet of the state endangered species, Pseudemys concinna. (Stomach flushing attempts with other emydids occasionally punctured the esophogas) Since, I had to retrieve feces from one species, I retrieved feces from all species to reduce bias. Feces were preserved in 10 % formalin and brought to the lab for identification and quantification of contents. Food types were divided into nine categories: 1) bryozoans, 2) insects, 3) crustaceans, 4) mollusks, 5) fish, 6) algae, 7) lower plant material, 8) higher plant material and 9) unidentified material/debris. The analysis of food types is expressed in percent volume (i.e. the percent of the total volume consumed represented by a particular food type). To calculate the extent of niche breadth and the degree of niche overlap based on percent volume of foods consumed, I used the Shannon-Wiener function (Shannon, 1949) and the Morisita Hom index (Hom, 1966 and Morisita, 1959a,b) respectively. Finally, a dietary niche dendrogram was constructed partitioning species into guilds. Dietary preferences were defined by a species consuming by volume > 75 % a food type (e.g.,. omnivores with a carnivorous preference consume > 75% animal material). Dietary specialists consume by volume > 90 % of a particular food item (e.g.,. insectivores consume > 90 % insects). 40 RESULTS Community Attributes: Species Richness. Nine species were captured at Round Pond during the study (listed in order of abundance): Trachemys scripta, Pseudemys concinna, Chelydra serpentina, Stemotherus odoratus, Graptemys ouachitensis, Apalone spinifera, Graptemys pseudogeographica, Cln:ysemys picta, and Apalone mutica. The species/ fyke trap hour curve (Fig. 2) for 1994, which levels off at eight species around 790 fyke net trap hours, differs from the actual species richness of that year because the procedure is based on fyke net captures and A. mutica were taken only in trammel nets that year. In 1995 only seven species were captured and the curve leveled off at approximately 590 fyke net trap hours (Fig. 2). In 1994 the curve climbed rapidly (four species were captured by 20.5 hrs. of trapping) and slowly leveled off while the 1995 curve had a gradual incline then leveled. Dominance. Red-eared sliders, I. scripta, composed the greatest percentage of the 376 turtles sampled in both years (65.8% in 1994 and 67.7% in 1995). Rank abundance curves (Fig. 3) revealed a sharp decline between the first- and third-ranked species with a leveling between ranks 3 - 6. These data (Table 1 and Fig. 2) clearly show that I. scripta (17.98 turtles/ha) is the numerically dominant species and~. concinna is the second most dominant (4.7 turtles/ha). Based on biomass (Table 2) the rank order changes: 1) I. scripta (11. 6 Kg/ha), 2) C. serpentina (5 .1 Kg/ha) and 3) ~. concinna (2.9 Kg/ha.). I found no significant difference in the relative numbers of each species captured in 1994 and 1995 (x2ca1c = 5.23, 8 df, P<0.75). 41 Habitat Effects on Species Composition, Richness and Diversity. Shannon Wiener indices for the 16 communities (Tab. 3) ranged from 1.1 - 2.3, with Round Pond having an H' of 1.6. The mean diversity values for lentic-mud, lotic-sand and lotic mud habitats were calculated at 1.5, 1.5 and 1.9 respectively. The Kruskal Wallis test detected a significant difference between the diversity indices of lentic mud, lotic-sand and lotic-mud habitats (H = 8.87, 2 df, P = 0.012). The average number of species for the sixteen sites was 5.2 (3 - 9) with a median of 4.5. The number of species that comprise 95 % of the total individuals averages 3. 9 (2 - 6) with a median of 4. Sixteen Midwestern turtle communities for which data exists (Tab. 3) are lentic, however, comparisons suggest several relationships in concerning species composition. In mud-bottomed habitats of lentic systems and embayments and backwaters of lotic systems, Trachemys or Chrysemys typically dominated chelonian communities. In contrast, Graptemys and Apalone usually dominated more lotic habitats, with A. mutica and G. pseudogeographica or ouachitensis being dominant in lotic-sand habitats whereas G. geographica and A. spinifera are most abundant in lotic-mud habitats. Dietary Niches Niche Relations. Due to small sample size of dietary data on most species, I concentrated on the dietary relationships of the two most common species, T. scripta and ;e,. concinna. Although represented by only two samples, I also analyzed the diet of C. sementina because of it being ranked second highest in biomass. Overall, the 42 diets of I. scripta (n = 27), £. concinna (n = 16) and C. serpentina (n = 2) had H' values of 2.50, 0.14 and 1. 70 respectively. River cooters appeared to be largely algivorous at the study site with no seasonal or ontogenetic variation evident in the sample. The two snapping turtle samples contained 77% animal food (chiefly crustaceans and fish), 18% plant food and 5% unidentified. Seven of the ten food categories were represented in these samples supporting the notion that Chelydra are general omnivores (Ernst et. al. 1994). Red-eared sliders consumed items from all dietary categories and a dietary shift was noted throughout the summer. In early summer I. scripta's niche breadth was 1.19 and the main dietary items consumed were algae (24.1 %) and leaves, stems, roots and bark (68.4%). Later in the summer their diet expanded (H' = 2.72), with 47.8% bryozoans and 8.7% algae being consumed (Fig. 4). Dietary overlap was greatest between river cooters and red-eared sliders (Ch = 0.17) but when broken down between early and late summer the overlaps were 0.32 and 0.14, respectively. Snapping turtle's overlaps with£. concinna and I. scripta were 0.02 and 0.09 respectively. Guild Structure. The dietary guilds used were based partially on dietary data of C. serpentina, £. concinna and I. scripta from Round Pond. Where species lacked sufficient data from Round Pond, I used information provided by Ernst, et al. (1994) to provide the necessary information (Fig. 5). Utilizing a modification of the guild categories proposed by Teran et al. (1995), I recognized five guilds from Round Pond presented in order of relative abundance of individuals: 1) omnivore/generalist 43 (66.63), 2) herbivore/ specialist- algivore (17.83), 3) omnivore/carnivore (9.33), 4) carnivore/generalist (4.83) and 5) carnivore/specialist-insectivore (0.53). From data derived from the literature survey and my site, lotic-mud habitats were primarily composed of species from the omnivore/generalist guild (Fig. 6). In lotic-sand habitats they were in a similar proportion to carnivorous types (carnivorous specialists, and generalists and omnivores with a carnivorous preference). Lentic-mud habitats were also composed mainly of omnivorous generalists but here hebivorous specialists were present as well (Fig. 6). DISCUSSION Community Aspects Of the 16 Illinois turtle assemblages reviewed, Round Pond had the greatest number of species, however, three of the nine species comprised less than 13 of total captures. Global species richness maps (Iverson, 1992) indicated Round Pond could potentially have 12 species, two each of: Chelydridae, Trionychidae and Kinosternidae, and 6 species of Emydidae. The alligator snapping turtle (Macroclemys temminckii), eastern mud turtle (Kinostemon subrubrum) and map turtle, (Graptemys geographica) were not collected at Round Pond. The habitat may be unsuitable for alligator snapping turtles, which typically occur in rivers, and usually do not inhabit small or temporary bodies of water (Pritchard, 1989). Map turtles seemingly are replaced by Ouachita map turtles in this region. According to Minton (1972), Graptemys geographica and Apalone spinifera are the dominant species in the upper Wabash drainage but are replaced in the lower Wabash drainage 44 by members of the G. pseudogeographica /ouachitensis complex and A. mutica. The absence of the eastern mud turtle is more difficult to explain as it occupies a wide variety of lentic and slow moving lotic habitats (Ernst et. al. 1994). Finding the smooth soft-shell at Round Pond was somewhat perplexing since A. mutica, is characteristic of swift moving, sandy-bottomed streams and rivers which have frequent sand bars that provide for basking and nesting (Smith, 1961). However, the smooth softshell's presence is probably explainable by the normal fluctuations of the Ohio river. During floods when the river encompasses Round pond, many riverine species can invade the habitat and as the river recedes some may become stranded in the Round pond. Red-eared sliders were the dominant species in seven of the 11 lentic habitat turtle assemblages reviewed (Table 3) including Round Pond. Painted turtles were dominant in the remaining four. Because I. scripta is an opportunistic species with generalized diet and habitat preferences, it is commonly a dominant species in lentic-mud systems throughout its range. Painted turtles assume the role of the dominant in lentic habitats outside the sliders range to the North and West. In terms of biomass, I. scripta is still the dominant species, however, my estimated total of 11. 6 Kg/ha falls well below those reported for other sites by Iverson (1982b) and Congdon et al. (1986) suggesting that Round Pond may not be the optimal habitat for the species. River cooters are the second most dominant species but the third most dominant in terms of biomass (2.9 Kg/ha), being supplanted by C. serpentina at 5 .1 Kg/ha. This biomass pattern of I. scripta, C. serpentina and 45 £. concinna at Round Pond is similar to that reported for Ellenton bay in South Carolina (Congdon et al., 1982). Density estimates may be low for C. serpentina as snappers may be less vagile compared to the other species. In Canada, male C. serpentina (Galbraith et al., 1987), were found to be territorial and have finite home ranges that did not encompass all regions of the lake. Habitat Effects on Species Composition, Richness and Diversity Overall it appears that both I. scripta and C. picta inhabit mud-bottomed habitats more frequently. Where in Illinois the habitat is located dictates which of the two species will predominate. This corroborates Minton's (1972) assessment in Indiana that both species tend to inhabit slow moving, shallow, soft bottomed bodies of water. Five turtle species were found in relatively high proportions in lotic habitats, the sand inhabiting A. mutica, G. ouachitensis and G. pseudogeographica and the mud inhabiting A. spinifera and G. geographica. Studies on habitat partitioning in Graptemys in Kansas (Fuselier and Edds, 1994) found that G. geographica were restricted to shady streams with a rock or gravel substratum. This is contradictory to my findings in Illinois which are supported from other research throughout its range (Ernst et al., 1994; Johnson, 1987; and Pluto and Bellis, 1986). Interestingly, G. pseudogeographica and G. ouachitensis displayed considerable overlap in habitat preference in Kansas (Fuselier and Edds, 1994) but G. pseudogeographica was never captured on sand substrates. In the Mississippi River it 46 is often the dominant species on sand substrate habitats (e.g., Cape Giradeau; see Moll 1973). I found both Apalone species at Round Pond and Williams and Christiansen (1981) reported that smooth soft-shell turtles in Iowa to chiefly inhabit the downstream portions of rivers but were sometimes found in ponds. Such ponds were always connected to large rivers during floods. Conversely, A. spinifera inhabited large rivers and adjacent ponds but appeared to prefer regions that were dominated with much structure, which is what I found at Round pond. Results indicate that lotic-mud habitats tend to have higher species diversity than either lotic-sand or lentic-mud habitats. This is explainable in part by the fact that lotic adapted species are joined by dietary generalists such as Trachemys, Chrysemys, Chelydra in lotic/mud habitats. Conversely, lentic mud habitats do not attract lotic specialists and lotic sand habitats with their sparse vegetation are usually avoided by generalists. Dietary Niches Niche Relations. Teran et. al (1995) provided a framework for classification of chelonian dietary niches, however, my findings disagree on certain points. The River cooter is primarily herbivorous throughout its range often specializing in filamentous algae particularly early in the season (Lagueux et al., 1995; Thomas et al., 1994 and Buhlmann and Vaughan, 1991). My study supports£. concinna as an herbivore/specialist (H' = 0.14). However, Teran et. al. (1994) also classified red eared sliders as predominantly herbivorous which is contrary to my findings. Sliders 47 at Round Pond were omnivorous consuming a high percentage of plants early in the summe then switching to animal material (bryozoans) in August. This discrepancy is probably in part due to geographic variation and in part due to the opportunistic nature of Red-eared sliders. Moll and Legler (1971) reported that adult Neotropical sliders are chiefly herbivorous but they are also very adaptable. At one site on the west coast of Panama where aquatic vegetation was limited, the turtles mostly consumed organic muck. Dietary studies between the three most dominant species reveal the greatest overlap occurs between I. scripta and~- concinna early in the summer (Ch= 0.32). Both species consume a large percentage of algae and competition for this resource may occur due to the density of T. scripta, however, this is unlikely for two reasons. First, in August the overlap is reduced (Ch= 0.14) and second, algae is probably not a limiting resource. Based on the limited information of the snapping turtles' diet at Round Pond, it does not appear to be an important competitor of either emydids (Ch with T. scripta = 0.09 and Ch with~- concinna = 0.02). This information is possibly significant from a conservation stand point, in that, the River cooter is a state endangered species in Illinois (Herkert, 1992). The degree of dietary overlap between species may indicate the importance of resource competition. If any serious competition for food occurs, it is seems restricted to early in the activity season. Although the diets of the other six species were not examined, any overlap here may have little significance as they only comprise 10.6% of the assemblage. More likely other factors have shaped the structure of this community 48 like rarifaction and competition along another resource gradient. Turtles may not only compete for food but may also compete for breeding and basking sites. Guild Structure. The chelonian food guilds represented in Round pond, range from specialist herbivores to specialist carnivores with the majority being general omnivores. Data from the 16 Illinois assemblages suggests that herbivores may be restricted to lentic-mud habitats in Illinois. This differs from findings in West Virginia (Buhlmann and Vaughan, 1991) where the River cooter inhabits slow moving sandy to gravel bottomed pools of the New River. In Illinois the lotic-sand habitats, which have relatively few aquatic plants, had relatively more carnivore specialists and omnivores with a carnivorous preference inhabiting them. The tendency for omnivorous generalists to inhabit all three systems in about equal percentages is not surprising due to their broad dietary and habitat preferences (e.g. I. scripta and C. picta). 49 LITERATURE CITED Berry, J. F. 1975. The population effects of ecological sympatry on musk turtles in northern Florida. Copeia 1975:692-701. Buhlmann, K. A. and M. R. Vaughan. 1991. Ecology of the turtle Pseudemys concinna in the New River, West Virginia. J. Herpetol. 25:72-78. Bury, R. B. 1979. Population ecology of freshwater turtles. In Harless, M. and H. Morlock, eds. Turtles: Perspectives and research, 571-602. John Wiley & Sons, New York. Cagle, F. R. 1942. Turtle populations in southern Illinois. Copeia 1942:155-162. Cagle, F. R. 1950. The life history of the slider turtle; Pseudemys scripta troostii (Holbrook). Ecol. Monogr. 20:31-54. Congdon, J. D. and J. W. Gibbons. 1989. Biomass productivity of turtles in freshwater wetlands: A geographic comparison. In Sharitz, R. R., and J. W. Gibbons, eds. Freshwater wetlands and wildlife, 583-592. DOE SYmp. Ser. (61). Congdon, J. D., J. L. Greene and J. W. Gibbons. 1986. Biomass of freshwater turtles: A geographic comparison. Amer. Midi. Nat. 115: 165-173. Ernst, C. H., J. E. Lovich and R. W. Barbour. 1994. Turtles of the United States and Canada. Smithsonian Inst. Press, Washington, D.C. Fuselier, L. and D. Edds. 1994. Habitat partitioning among three sympatric species of map turtles, genus Graptemys. J. Herpetol. 23: 154-158. 50 Galbraith, D. A., M. W. Chandler and R. J. Brooks. 1987. The fine structure of home ranges of male Chelydra serpentina: are snapping turtles territorial? Can. J. Zool. 65:2623-29. Gritters, S. A. and L. M. Mauldin. 1994. Four years of turtle collections on Navigation Pool 13 of the Upper Mississippi River. Report by the Iowa Department of Natural Resources, Bellevue, Iowa, for the National Biological Survey, Environmental Management Technical Center, Onalaska, Wisconsin, September 1994. LTRMP 94-SOlO. 7pp. Iverson, J. B. 1992a. Species richness maps of the freshwater and terrestrial turtles of the world. Smithsonian Herpetological Information Service. No. 88. 18p. Iverson, J. B. 1992b. Global correlates of species richness in turtles. Herp. Journal 2:77-81. Johnson, T. R. 1987. The amphibians and reptiles of Missouri. Missouri Dept. Conserv., Jefferson City. 369 pp. Lagueux, C. J., K. A. Bjorndal, A. B. Bolten and C. L. Campbell. 1995. Food habits of Pseudemys concinna suwanniensis in a Florida spring. J. Herpetol. 29: 122-126. MacArthur, R. A. 1971. Patterns in terrestrial bird communities. In Farner, D. S. and J. R. King eds. pp. 189-221. Avian Biology. Vol. 1. Academic Press, New York. Mahamoud, I. Y. 1968. Feeding behavior in kinosternid turtles. Herpetologica 24:300-305. 51 Minton, S. A. Jr. 1972. Amphibians and reptiles of Indiana. Indiana Acad. Sci. Monogr. 3:1-346. Moll, D. L. 1977. Ecological investigations of turtles in a polluted ecosystem: the central Illinois river and adjacent flood plain lakes. Unpubl. PhD. Dissertation. 179 pp. Moll, D. L. 1993. A study of dietary characteristics and growth in turtles incidentally collected during long-term resource monitoring of Mississippi river fishes, July - October, 1992. Unpubl. Technical Report. 23 pp. Moll, E. 0. and J. M. Legler. 1971. The life history of a neotropical slider turtle, Pseudemys scripta (Schoepff), in Panama. Bull. Los Angeles Co. Mus. Nat. Hist. (11)1-102. Pierce, L. 1992. Diet content and overlap of six species of turtles among the Wabash River. Unpubl. Masters Thesis. Eastern Illinois University, Charelston, Illinois. 62 pp. Pluto, T. G. and E. D. Bellis. 1986. Habitat utilization by the turtle, Graptemys geographica, along a river. J. Herpetol. 20:22-31. Pritchard, P. C. H. 1989. Tha alligator snapping turtle: biology and conservation. Milwaukee Public Museum, Milwaukee, Wisconsin. 104 pp. Schnabel, Z. 1938. Estimation of the total fish population of a lake. Am. Math. Mon. 45:348-352. Scott, N. J. Jr. (ed). 1982. Herpetological communities. U.S. Dept. Int. Fish Wild. Ser. Wild. Res. Rep. 13. Washington, D.C. 239 pp. 52 Shannon, C. E. 1949. The mathematical theory of communication. In C. E. Shannon and W. Weaver (eds). The mathematical theory of communication. Univeristy of Illinois Press, Urbana, Illinois. Smith, P. W. 1961. The amphibians and reptiles of Illinois. Ill. Nat. Hist. Surv. Bull. 28: 1-298. Teran, A. F., R. C. Vogt, and M. F. S. Gomez. 1995. Food habits of an assemblage of five species of turtles in the Rio Guapore, Rondonia, Brazil. J. Herpetol. 29:536-47. Thomas, R. B., D. Moll and J. Steiert. 1994. Evidence of a symbiotic relationship between cellulolytic bacteria and a freshwater herbivorous turtle. Southwest Nat. 39:386-388. Williams, T. A. and J. L. Christiansen. 1981. The niches of two sympatric softshell turtles, Trionyx muticus and Trionyx spiniferus, in Iowa. J. Herpetol. 15:303- 308. Vogt, R. C. 1980. New methods for trapping aquatic turtles. Copeia. 1980:368-371. Vogt, R. C. 1988. Food partitioning in three sympatric species of map turtle, genus Graptemys (Testudinata, Emydidae). Amer. Midi. Nat. 105: 102-111. Vogt, R. C. and S. Guzman Guzman. 1988. Food partitioning in a neotropical freshwater turtle community. Copeia 1988: 37-4 7. 53 Table 1: Number captured, percent of capture, population estimate and rank (according to abundance) of trap captures of turtles from Round Pond, Gallatin County, Illinois over the Summers of 1994 and 1995. TRSC = Trachemys scripta, PSCO = Pseudemys concinna, CHSE = Chelydra serpentina, GROV = Graptemys ouachitensis, STOD = Stemotherous odoratus, APSP = Apalone spinifera, GRPS = Graptemys pseudogeographica, CHPI = Chrysemys picta and APMU = Apalone mutica. TRSC PSCO CHSE GROU STOD APSP GRPS CHPI APMU 1994 #Captured 154 46 10 9 7 3 2 2 1 3 in Capture 65.8 19.7 4.3 3.9 2.9 1.3 0.9 0.9 0.4 Pop. Est. 412 123 27 24 19 8 5 5 3 Rank 1 2 3 4 5 6 7.5 7.5 9 1995 #Captured 96 21 9 7 5 3 0 0 1 3 of Capture 67.6 14.8 6.3 4.9 3.5 2.1 0.0 0.0 0.7 Pop. Est. 535 117 50 39 28 17 0 0 6 Rank 1 2 3 4 5 6 ------7 Overall #Captured 250 67 19 16 12 6 2 2 2 3 of Capture 66.5 17.8 5.1 4.3 3.2 1.6 0.5 0.5 0.5 Pop. Est. 535 141 40 34 25 13 4 4 4 Rank 1 2 3 4 5 6 8 8 8 54 Table 2: Mean turtle weight (g), total estimated biomass (Kg), biomass estimates (Kg/ha) and density estimates (turtles/ha) for eight turtle species captured during the summers of 1994-95 at Round Pond, Gallatin county, Illinois. TSSC = Trachemys scripta, PSCO = Pseudemys concinna, CHSE = Chelydra ser_pentina, STOD = Stemotherous odoratus, GRPS = Graptemys pseudogeographica, GROU = Graptemys ouachitensis, APSP = Apalone spinifera and CHPI = Chrysemys picta. TSSC CHSE PSCO GROU STOD APSP GRPS CHPI Mean 647.5 3806.7 618.0 108.3 162 364.8 243 247.5 Total 346.4 152.3 87.1 3.7 4.1 4.7 1.0 1.0 Kg/ha 11.6 5.1 2.9 0.1 0.1 0.2 0.03 0.03 Turtles/Ha 18.0 1.3 4. 7 1.1 0.8 0.4 0.1 0.1 55 Table 3: Relative species abundance for 16 chelonian assemblages compiled from a literature review (Gritters and Maudlin, 1994; Moll, 1993; Pierce, 1992 and Cagle, 1942). Localities are identified in the key to abbreviations on page 69. Lotic-Mud Lo tic-Sand Lentic-Mud Species P13LO WA PL ILRH CG WR P13LE RP ML SL CL MSP EP CR GL WB APMU 3.5 0.0 0.0 0.0 14.0 68.4 1.5 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 APSP 12.0 0.0 22.0 16.0 2.0 5.5 2.5 1.6 5.0 2.0 7.0 0.0 0.0 0.0 0.0 0.0 CHPI 6.0 26.7 38.0 14.0 0.0 1.0 78.0 0.5 20.0 62.0 30.0 55.0 40.0 8.0 24.0 13.0 CHSE 0.5 6.7 5.0 3.0 2.0 0.0 0.5 5.1 2.0 3.0 0.0 11.0 10.0 6.0 0.0 31.0 GRGE 65.0 0.0 13.0 15.0 0.0 0.0 9.5 0.0 5.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 GROU 0.0 2.2 0.0 0.0 0.0 18.0 0.0 4.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 GRPS 13.0 0.0 0.0 8.0 56.0 0.0 8.0 0.5 20.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 PSCO 0.0 0.0 0.0 0.0 0.0 0.0 0.0 17.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 STOD 0.0 0.0 5.0 8.0 0.0 0.0 0.0 3.2 13.0 5.0 12.0 13.0 30.0 6.0 5.0 0.0 TRSC 0.0 64.4 17.0 36.0 27.0 7.0 0.0 66.5 35.0 28.0 50.0 21.0 20.0 80.0 72.0 56.0 H' 1.6 1.3 2.3 2.5 1.6 1.4 1.2 1.6 2.0 1.4 1.7 1.7 1.9 1.0 1.1 1.4 E 0.62 0.65 0.87 0.89 0.68 0.60 0.45 0.51 0.71 0.61 0.74 0.84 0.92 0.52 0.66 0.87 56 Fig. 1: Map of the research site and surrounding backwater lakes and sloughs in southeastern Gallatin county, Illinois (37o45'W, 88ol4'N). The inset of Illinois is the present distribution with Gallatin county inside the square. Solid circles = museum records, open circles = literature records, solid squares = records from 1985-1988 and open squares are possible sightings reviewed by Moll and Morris (1991). 57 "'-.... -, '. (~ . :.--- '\ Hulda Lake I ' --=------__ _, -- \1 , ___.·· --,,.- ,.~ I . ~ i -_j i ,___ ,..'~--4 ~, ~ ---·" ' Big Lake Black Lake Fehrer Lake ~ Fish Lake -To Old Shawneetown Ohio River : Kilometer~ Scale I :250000 58 Fig. 2: Curves depicting the number of trapping hours with fyke nets required to capture the maximum number of species 1994 and 1995 at Round Pond, Gallatin county. 59 10 _..._ 1994 9 13 ···•··· 1995 .a"""' 8 ~ u 7 .. ··································································································································• en Q) ·-~ 6 0.. r.I) Q) 5 'E ~ 4 ~ 0 3 Total Trapping Hours ~ .· 1994 = 830.42 'S 2 ../ - 1995 = 1, 121.00 z 1 .. // Total = 1,951.42 0 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 Trap Hours 60 Fig. 3: Rank-abundance curves for all species captured in 1994 and 1995 at Round Pond, Gallatin county. G. pseudogeographica and C. picta were collected in 1994 but not in 1995. Species ranks are identified in Table 2. 61 6 ...... c 0 1994 _._J 5 (/) B 1995 E ..c ·c+-' 4 co C> 0 _J I-co 3 :::J +-' zm Q) ..c 2 +-' -0 Q) (/)m 1 Ill 0 0 1 2 3 4 5 6 7 8 9 10 Turtle Rank 62 Fig. 4: Overall dietary composition of C. ser.pentina (a, n = 2), £. concinna (c, n = 16) and early (May and June), late (August) and total summer I. scripta (b, n = 7; d, n =20; e = 27, respectively) diets derived from fecal samples collected during 1994 and 1995 at Round Pond, Gallatin county. 63 HPM B) 69.1% A) VERT ~~ INV 13% 64% ------ UM/D LP ~o 1% LPM UM/OVERT INV 17% 24.1% 0.4% 0.7% 5.7% C) LPM 98% HPM 2% E) D) HPM INV 34% 54% HPM 29% INV 49% LPM UM/OVERT LPM UM/OVERT 14% 1% 2% 13% 2% 2% 64 Fig. 5: Dendrogram of guilds of turtles inhabiting Round Pond, Gallatin county, Illinois based on dietary data from this study and a literature review. 65 Herbivore Specialist Algivore Pseudemys concinna & Trachemys scripta Generalist Chrysemys picta Graotemvs oseudo2eowaohica Omnivore Graptemys ouachitensis Carnivorous Preference Chelydra serpentina Sternotherus odoratus Generalist Carnivore Apa/one spinifera Specialist Insectivore Apa/one mutica 66 Fig. 6 Relative percentage of each guild represented in three major chelonian habitats examined in 16 sites from Illinois: herbivore specialist (HERB), omnivore generalist (OMGEN), omnivore with carnivorous preference (OMCAR), carnivore generalist (CARGEN) and carnivore specialist (CARSP). Based on Table 3. 67 Q) "'C a. c >. m I- CJ) +-' I m 0 +-' +:i :.c 0 m 00 _J I \CJ "'C :::J ~ 0 +:i 0 _J z z a:::: w a.. OJ ~ (3 (!) (/) a:::: "o::::: "o::::: a:::: a:::: W..:::..:::<(<( I00(.)(.) o m oo ~ m ~ v ~ N ~ o ~ 0 0 0 0 0 0 0 0 0 0 KEY TO ABBREVIATIONS AND SITE LOCATIONS P13LO = Pool 13 Lotic & P13LE = Pool 13 Lentic, Fulton, Whiteside Co., IL, 41o54'N, 90o09'W. PL =Peoria Lake, Peoria, Tazwell Co., IL, 40o42'N, 89oOl'W. SL = Spring Lake, Sand Ridge State Forest, Tazwell Co., IL, 40o28'N, 90o18'W. CL= Chautauqua Lake, Sand Ridge State Forest, Mason Co., IL, 40o24', 90o14'W. IH = Illinois River at Havanna, Mason Co., IL, 40o18'N, 90o10'W. ML = Meredosia Lake, Meredosia, Cass and Morgan Co., IL, 39°53'N, 90°47'W. WA =Mississippi River near West Alton, St. Charles Co., MO, 38o54'N, 90°03'W. CR = Carbondale Resivoir, Carbondale, Jackson Co., IL, 37o42'N, 88o37'W. GL = Grimsby Lake, Presumably near Grimsby, Jackson Co., IL, 37o46'N, 88o27'W. MSP =Marion Stock Pond, Marion, Williamson Co., IL, 37o41'N, 88o37'W. EP = Elkville Pond, Elkville, Jackson Co. IL, 37o55'N, 88o37'W. WB =Wabash River Backwater & WR =Wabash River, Allendale and Mt. Carmel, Edwards Co., IL, 38o15'N, 87o29'W. RP = Round Pond, Old Shawneetown, Gallatin Co., IL, 37o45'N, 88o14'W. 69