Acta Herpetologica 13(2): 191-194, 2018 ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah DOI: 10.13128/Acta_Herpetol-22955 Egg and clutch sizes of western chicken turtles (Deirochelys reticularia miaria) Donald T. McKnight1,2,*, Ethan C. Hollender1, Hunter J. Howell3, John L. Carr4, Kurt A. Buhlmann5, Day B. Ligon1 1 Department of Biology, Missouri State University, Spring field, Missouri, USA. *Corresponding author. Email: donald.mcknight@ my.jcu.edu.au 2 College of Science and Engineering, James Cook University, Townsville, Queensland, Australia 3 Department of Biology, University of Miami, Coral Gables, Florida, USA 4 Department of Biology, University of Louisiana at Monroe, Louisiana, USA 5 Savannah River Ecology Laboratory, University of Georgia, Aiken, South Carolina, USA Submitted on: 2018, 4th February; revised on: 2018, 8th August; accepted on: 2018, 20th September Editor: Aaron M. Bauer Abstract. Chicken turtles (Deirochelys reticularia) are generally characterized as having atypical reproductive charac- teristics relative to other sympatric emydids. However, the comparatively understudied western chicken turtle (D. r. miaria) has been shown to exhibit some reproductive characteristics that differ from the other subspecies. Therefore, we examined clutch and egg sizes from six D. r. miaria (13 clutches) in Oklahoma and compared the results to values that have been reported for the other two subspecies. Females nested up to three times per year, with clutches rang- ing from 8-13 eggs per clutch (mean = 10.9). The eggs were 19.4-25.3 mm wide (mean = 22.2 mm). These values are greater than the means reported for the other subspecies, but the differences were not statistically significant. Keywords. Chelonia, Emydidae, subspecies, reproduction, geographic variation. Chicken turtles (Deirochelys reticularia) occur in shal- low swamps and ponds throughout the southeastern Unit- ed States, and they exhibit several traits that are uncom- mon for North American turtles, the most noteworthy of which relate to their unusual reproductive patterns. The eastern chicken turtle (D. r. reticularia) is the most stud- ied of the three recognized subspecies, and, unlike most North American turtles, females nest in the fall through early spring (Gibbons and Greene, 1990; Buhlmann et al., 2009). Eggs deposited in the ground in the fall enter embryonic diapause and resume development in spring after a period of chilling (Ewert et al., 2006). Addition- ally, when conditions become unfavourable (too cold) in the fall, females can retain eggs, overwinter with them, and lay them the following spring (Buhlmann et al., 1995, 2009). This behavior is uncommon in turtles but has also been reported for the Sonoran Desert tortoise (Gopherus morafkai) in Arizona, USA (Lovich et al., 2017). Florida chicken turtles (D. r. chrysea) extend this strategy by nest- ing throughout the winter and laying up to three clutches (rarely four) per year (Iverson, 1977; Jackson, 1988). In contrast, the western subspecies (D. r. miaria) nests in the late spring and summer (May-July), a pattern that is more consistent with other sympatric emydids (McK- night et al., 2015a; Carr and Tolson, 2018). However, this subspecies is the least-studied of the three, and although current research indicates that it differs from the other subspecies in diet, seasonal activity patterns, and some reproductive patterns (McKnight et al., 2015a, b), there is still a great deal that we do not know about its life history. One of these knowledge gaps is a dearth of information about their clutch and egg sizes. Currently, clutch and egg 192 Donald T. McKnight et alii sizes for D. r. miaria are only reported from five clutches in Louisiana (Carr and Tolson, 2018), three clutches in Arkansas (Dinkelacker and Hilzinger, 2014) and one in Texas (David, 1975). To expand our knowledge of this topic, we captured D. r. miaria in shallow beaver ponds and similar habi- tats in southeastern Oklahoma from 2013-2015 using a variety of hoop nets (detailed trapping methods available in McKnight et al. 2015c). We also attached radio trans- mitters (RI-2B 10 g or RI-2B 15 g, Holohil Systems Ltd., Corp., Ontario, Canada) to the carapaces of three females and recaptured them approximately every two weeks. Upon capture (via trapping or tracking) we used a port- able ultrasound (Echo Camera SSD-500V, Hitachi Aloka Medical, Inc., Tokyo, Japan) to check for the presence of follicles or shelled eggs. If shelled eggs were observed, we immediately transported the turtles to a nearby vet- erinary clinic and used X-radiography to record the size and number of eggs (Gibbons and Greene, 1979). We included a coin or other object of known size to cor- rect for magnification from the X-radiograph (Graham and Petokas, 1989). We counted the eggs and measured their width, but we did not measure their length, because length is only accurate when the eggs are oriented paral- lel with the plastron (Bertolero et al., 2007), a position that cannot be confirmed in a live animal. Additionally, we included data from a female that was captured in southeastern Oklahoma on 16 June 2008 and transported to the Tulsa Zoo. Shelled eggs were observed via X-radiography on 9 July 2008, and 11 eggs were laid on 20 July 2008, following induction using oxy- tocin (Ewert and Legler, 1978; a single egg was laid and crushed on the previous day). These 11 eggs were meas- ured after being laid (including the mass and length), rather than measuring the X-radiograph (they were included with the data from the other clutches for all summary statistics). We measured and recorded clutch sizes for a total of 13 clutches from six females (two additional clutches were observed but could not be X-rayed for quantifi- cation or measurements). The clutch and egg sizes are shown in Fig. 1 and compared to other studies and sub- species in Table 1. The female that laid eggs at the Tulsa Zoo had a mean egg mass of 9.9 g (SD = 1.14; range = 8-12), mean egg length of 31.6 mm (SD = 1.43; range = 30.2-34.4), and mean egg width of 23.6 mm (SD = 0.88; range = 21.9-25.3; the width measurements are included in Figure 1 and Table 1). For the three transmittered females that produced multiple clutches in a year, two produced at least two clutches for each of the two years we monitored them (we were not able to X-ray one of those clutches and it was not included in our summary statistics or the total num- ber of clutches). The third individual produced at least one clutch in the first year, and at least three clutches in the second year (the third clutch was undergoing calcifi- cation when we X-rayed it, and the shells were not suf- ficiently visible to reliably count or measure the eggs; therefore, it was not included in our summary statistics or total number of clutches). This is the first report of D. r. miaria producing up to three clutches in a single year. Deirochelys r. chrysea is known to nest up to four times per year (typically 2-3), but D. r. reticularia and D. r. miaria have only previ- ously been reported nesting twice per annual cycle (Gib- bons, 1969; Gibbons and Greene, 1978; Ewert et al., 2006; McKnight et al., 2015a). Our mean clutch and egg sizes agreed closely with the results reported by Carr and Tolson (2018) and Din- kelacker and Hilzinger (2014) for D. r. miaria in Louisi- ana and Arkansas, respectively. All three studies found large mean egg and clutch sizes relative to most reports for the other subspecies. Indeed, the maximum egg widths in both our study and Carr and Tolson (2018) were the largest egg widths reported for any D. reticularia subspecies. To statistically compare egg and clutch sizes between D. r. miaria and D. r. reticularia, we combined our data with the clutches reported in Carr and Tolson (2018) and compared them with data from D. r. reticularia that were captured in South Carolina between July 1995 and Febru- ary 1996 (previously published in Buhlmann et al., 2009). This resulted in a total of 17 clutches from nine female D. r. miaria and 25 clutches from 16 female D. r. reticularia (one clutch from Carr and Tolson [2018] was not includ- ed because female plastron length was not recorded). We constructed mixed-effects models in R (v3.4.1; R Core Fig. 1. Deirochelys reticularia miaria egg widths in Oklahoma. Each box is the data from a single clutch, and the dashed lines separate individual females. Numbers above the boxes indicate clutch size and numbers below the boxes indicate female plastron length at the time the clutch was observed. Whiskers indicate the 10th and 90th percentiles and all outliers are shown. The center box (clutch size = 12, plastron length = 179) shows the data for the female that laid eggs at the Tulsa Zoo. All other boxes are from X-ray data. 193Egg and clutch sizes of western chicken turtles Team 2017) using the package lme4 (v1.1-15; Bates et al. 2015) with subspecies and plastron length as fixed effect factors (including an interaction). For egg width, we used a linear model via the lmer function, and for clutch size, we used a generalized linear model with a Poisson dis- tribution via the glmer function. To avoid pseudorepli- cation, we included female ID as a random effect factor for the model comparing clutch sizes, and for the model comparing egg widths, we include female ID and clutch ID as random effect factors, with clutch ID nested in female ID. We used quantile-quantile plots and residual plots to check the models’ assumptions, and we assessed the significance of both comparisons using the Anova function in the package car (v2.1-6; Fox and Weisberg, 2011) with type II sums of squares and the chi-square test statistic. We did not find a significant difference in egg width between D. r. miaria and D. r. reticularia (χ2 = 0.8715, P = 0.3505), but egg width and plastron length were correlat- ed (χ2 = 8.1154, P = 0.0044), which is consistent with pre- vious research on D. r. reticularia (Gibbons et al., 1982; Buhlmann et al., 2009). The interaction term was not sig- nificant (χ2 = 1.4155, P = 0.2341). For the comparison of clutch size, there was not a significant difference between subspecies (χ2 = 0.6655, P = 0.4146), or a significant asso- ciation with plastron length (χ2 = 1.4070, P = 0.2355), or a significant interaction (χ2 = 0.2898, P = 0.5904). These results do not suggest a difference in egg or clutch sizes between the subspecies after accounting for female body size, however, our sample size was small, resulting in limited statistical power. It would be useful for future researchers to expand this sample and conduct a more robust comparison. ACKNOWLEDGEMENTS We are indebted to the staff of the All Animal Veteri- nary Hospital, particularly Dr. Dana Harvey, for X-raying our turtles for us. We are also grateful to Dr. Kay Back- ues at the Tulsa Zoo for sharing data on the turtle cap- tured in 2008. Additionally, we would like to thank Jona Tucker for her assistance with this research, as well as the Oklahoma Nature Conservancy and private land owners who allowed us to conduct research on their property. This study was funded by an Oklahoma Department of Wildlife Conservation State Wildlife Grant and the Delta Foundation, and it was approved by the Missouri State University Institutional Animal Care and Use Committee (IACUC protocol no. 10014). KAB was supported by the U.S. Department of Energy under Award Number DE- FC09-07SR22506 to the University of Georgia Research Foundation. We conducted the research under Oklahoma Department of Wildlife Conservation scientific collecting permits #5269, 5950, and 5610 and adhered to the ASIH/ HL/SSAR Guidelines for Use of Live Amphibians and Reptiles. Table 1. Clutch and egg sizes for chicken turtle (Deirochelys reticularia) subspecies. Mean PL = mean plastron length of females. For results from the current study, mean egg width was calculated by averaging the means per clutch, whereas the range is the maximum and mini- mum of any measured eggs. Subspecies State Mean clutch size Clutch size range Mean egg width (mm) Egg width range (mm) # of clutches Mean PL (mm) Source D. r. miaria OK 10.9 9–13 22.2 19.4–25.3 13 174.8 Current study D. r. miaria LA 10.6 8–13 22.9 20.7–24.7 5a 186.1 Carr & Tolson, 2018 D. r. miaria AR 10.7 9–12 22.3 b 21.1– 23.5 b 3b 174.9 Dinkelacker & Hilzinger, 2014 D. r. miaria TX 8 — — 17–23 1 — David, 1975 D. r. reticularia SC 7.2 1–12 — — 57 159 Gibbons et al., 1982 D. r. reticularia SC 8 — 20.8 — 15 — Congdon et al., 1983 D. r. reticularia SC 8 — 20.8 — 13 160 Congdon & Gibbons, 1985 D. r. reticularia SC 9.8 5–17 21.8 17.4–23.4 32 177.5c Buhlmann et al., 2009 D. r. chrysea FL 10.5d 8–12d 21.2 19.9–22.9 4d — Iverson, 1977 D. r. chrysea FL 9.5 2–19 22.4 20.1–23.6 29 176.3 Jackson, 1988 D. r. chrysea FL 10.9 6–18 — — 16 — Ewert et al., 2006 a Plastron length was only recorded for four females. b Egg width was based on only two clutches. c WebPlotDigitizer (Rohatgi 2017) was used to extract measurements from Figure 3 of Buhlmann et al. 2009. d One female contained 5 eggs and “6 ovarian follicles of ovulatory size.” Based on our experience with D. r. miaria, we treated the follicles as unshelled eggs and assumed a clutch size of 11 (otherwise, the mean is 9 and range is 5–12). 194 Donald T. McKnight et alii REFERENCES Bertolero, A., Nougarède, J.P., Cheylan, M., Marín, A. (2007): Breeding traits of Hermann’s tortoise Tes- tudo hermanni hermanni in two western populations. Amphibia-Reptilia 28: 77-85. Bates, D., Maechler, M., Bolker, B., Walker, S. (2015): Fit- ting linear mixed-effects models using lme4. Journal of Statistical Software 67: 1-48. Buhlmann, K.A., Congdon, J.D., Gibbons, J.W., Greene, J.L. (2009): Ecology of chicken turtles (Deirochelys reticularia) in a seasonal wetland ecosystem: exploit- ing resource and refuge environments. Herpetologica 65: 39-53. Buhlmann, K.A., Lynch, T.K., Gibbons, J.W., Greene, J.L. (1995): Prolonged egg retention in the turtle Deiro- chelys reticularia in South Carolina. Herpetologica 51: 457-462. Carr, J.L., Tolson, K.M. (2018): Observations on repro- duction in the western chicken turtle, Deirochelys reticularia miaria, in Louisiana. Southwest. Nat. 62: 280-283. Congdon, J.D., Gibbons, J.W., Greene, J.L. (1983): Paren- tal investment in the chicken turtle (Deirochelys retic- ularia). Ecology 64: 419-425. David, W.D.Jr. (1975): Notes on the egg laying habits of Deirochelys reticularia. Herpetol. Rev. 6: 127. Dinkelacker, S.A., Hilzinger, N.L. (2014): Demographic and reproductive traits of western chicken turtles, Deirochelys reticularia miaria, in Central Arkansas. J. Herpetol. 48: 1-7. Ewert, M.A., Legler, J.M. (1978): Hormonal induction of oviposition in turtles. Herpetologica 34: 314-318. Ewert, M.A., Jackson, D.R., Buhlmann, K.A. (2006): Dei- rochelys reticularia - chicken turtle. Chelonian Res. Monogr. 3: 249-259. Fox, J., Weisberg, S. (2011): An R Companion to Applied Regression, 2nd Edition. Sage, Thousand Oaks, California. Gibbons, J.W. (1969): Ecology and population dynamics of the chicken turtle, Deirochelys reticularia. Copeia 1969: 669-676. Gibbons, J.W., Greene, J.L. (1978): Selected aspects of the ecology of the chicken turtle, Deirochelys reticularia (Latreille) (Reptilia, Testudines, Emydidae). J. Herpe- tol. 12: 237-241. Gibbons, J.W., Greene, J.L. (1979): X-ray photography: a technique to determine reproductive patterns of fresh- water turtles. Herpetologica 35: 86-89. Gibbons, J.W., Greene, J.L. (1990): Reproduction in the slider and other species of turtles. In: Life History and Ecology of the Slider Turtle, pp. 124–134. Gibbons, J.W., Ed, Smithsonian Institution Press, Washington. Gibbons, J.W., Greene, J.L., Patterson, K.K. (1982): Varia- tion in reproductive characteristics of aquatic turtles. Copeia 1982: 776-784. Graham, T., Petokas, P. (1989): Correcting for magnifica- tion when taking measurements directly from radio- graphs. Herpetol. Rev. 20: 46-47. Iverson, J.B. (1977): Reproduction in freshwater and ter- restrial turtles of north Florida. Herpetologica 33: 205-212. Jackson, D.R. (1988): Reproductive strategies of sympat- ric freshwater emydid turtles in northern peninsular Florida. Bull. Florida St. Mus. Biol. Sci. 33: 113-158. McKnight, D.T., Harmon, J.R., McKnight, J.L., Ligon, D.B. (2015a): The spring–summer nesting and activ- ity patterns of the western chicken turtle (Deirochelys reticularia miaria). Copeia 103: 1043-1047. McKnight, D.T., Jones, A.C., Ligon, D.B. (2015b): The omnivorous diet of the western chicken turtle (Deiro- chelys reticularia miaria). Copeia 103: 322-328. McKnight, D.T., Harmon, J.R., McKnight, J.L., Ligon, D.B. (2015c): Taxonomic biases of seven methods used to survey a diverse herpetofaunal community. Herpetol. Conserv. Biol. 10: 666-678. R Core Team (2017): R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-pro- ject.org/. Rohatgi, A. (2017): WebPlotDigitizer, version: 4.0. Avail- able at: https://automeris.io/WebPlotDigitizer. Acta Herpetologica Vol. 13, n. 1 - June 2018 Firenze University Press