Progress Report (February 2011)
Funding provided by the Lucille F. Stickel Box Turtle Research Award has supported my work on the phylogenetics of the Terrapene genus. It has provided for funding to acquire tissue samples, extract DNA from those tissue samples, amplify and sequence genes, and analyze the sequence data. To date, via a vast network of volunteers, we have collected ~1500 tissue samples from every currently recognized subspecies/species of Terrapene with the exception of T. nelsoni; more samples continue to be received on a regular basis. All of the samples in our possession have been inventoried and most have had their DNA extracted. In addition, we are very close to having mtDNA sequence data for every species and subspecies, and most of the states within the North American box turtle distribution are already represented in the mtDNA dataset.

The current mtDNA dataset based on the cytochrome b gene indicates a variety of relationships. To begin, it is clear that there are two main groups (an eastern and a western). However, relationships within those groups appear a bit more muddied. To begin, the data indicate that the three-toed box turtle (T. carolina triunguis) is more distantly related to the eastern box turtle (T. c. carolina) than previously believed (Figure 1). In fact, they may be distinct enough that T. c. triunguis may need to be elevated to a separate species. In addition, our mtDNA dataset indicates that the ornate box turtle (T. o. ornata) are less distantly related to the desert box turtles (T. o. luteola) than previously believed (Figure 1) and that the two may be safely folded into one single species rather than two separate subspecies. Additional evolutionary relationships that appear to be muddled include the placement of the gulf coast, Mexican, and Florida box turtles, as their distribution in the dataset is not straight forward.

Since the current phylogeny is based solely on mtDNA data, we are in the process of acquiring additional data from nuclear sequences to confirm the relationships revealed so far. We expect to have a full data set on one nuclear gene completed within the next month and will sequence at least one more prior to preparing our data for publication. At that time, we will be sure to acknowledge funding from the Lucille F. Stickel Box Turtle Research Award and will forward a copy of the accepted manuscript to the board. Thank you so much for your support!!!

Figure 1: A phylogram representing 79 cytochrome b haplotypes generated from 178 gene sequences. Bootstrap values are indicated by their corresponding branches. The phylogram is color coded (with T. o. luteola in light blue, T. o. ornata in dark purple, T. c. carolina in orange, T. c. major in pink, T. c. triunguis in green, T. c. mexicana in red, T. c. yucatana in brown, and T. c. bauri in yellow).

INTRODUCTION

The most researched aspects of Terrapene reproductive ecology cover mating behaviors and egg production (Dodd 2001). Nesting ecology and nest success, however, are largely understudied and past studies have focused mostly on the Ornate Box Turtle, Terrapene ornata (Converse et al . 2002). It is likely that the lack of box turtle nesting data is a result of unpredictable nesting behaviors.

The nesting of T. carolina can occur between early May and the middle of July (Legler 1960; Dodd 2001). The time of day at which nesting occurs varies depending on temperature and other weather conditions. Terrapene carolina will generally nest just after, before, or during a rainfall (Congello 1978; Dodd 2001). The duration of nesting activities is highly variable; sometimes lasting up to five hours, and likely depends on environmental conditions (Ewing 1935; Stickel 1950; Legler 1960; Dodd 2001).

Nest site selection influences the development and survival of young turtles (Dodd 2001; Kolbe and Janzen 2001). Substrate type, temperature, and moisture dictate hatching success. Prior to oviposition, female T. carolina move sporadically around the general nesting site (Stickel 1989). Between one and seven eggs per clutch are deposited, depending on the population's geographic location (Dodd 2001). For example, northern populations deposit four or more eggs and southern populations deposit one to three eggs (Ernst et al . 1994; Dodd 2001). Geographic location and temperature will also influence the emergence time of young turtles. The field incubation period is approximately 65 days, however warmer temperatures may increase development rate, causing eggs to hatch sooner (Dodd 2001).

MATERIALS AND METHODS

The study site was located in East Central Illinois, along the Collison Branch of the Middle Fork Vermilion River . Beginning in the spring of 2009, box turtles were located and captured through visual encounter surveys. Each individual was uniquely marked by notching marginal scutes (Cagle 1939). Individuals were weighed on an electronic balance and aged by counting the annuli on their left pectoral scute (Sexton 1959). Carapace length and width, shell height, left pectoral scute length, anterior plastron length and width, and posterior plastron length and width were measured. Adult females were x-rayed to determine clutch sizes (Gibbons and Greene 1979). A small radio transmitter was glued to the carapace of each individual. Individuals were tracked daily and the following environmental data was measured: relative humidity, wind speed, air temperature, substrate temperature, and percent canopy cover. GPS coordinates were also taken.

RESULTS

In the spring of 2009, nine female Terrapene carolina were found to be gravid. The mean clutch size was 5.6 eggs (Table 1). Though females were monitored daily, none were observed nesting. The lack of nesting data from my recent findings is disappointing; however, a summary of my results are as follows:

Search Effort

The search effort consisted of 205 hours between the dates April 23 and June 10. The 47 individuals found between these dates (21 females, 23 males, 3 immature), resulted in an overall catch per unit effort of 0.23 turtles/search hour. The catch per unit effort for gravid females was not determined; only 12 adult females were x-rayed. Searches for individuals continued until October 28 and a total of 95 individuals were found at the study site (36 females, 54 males, 5 immature). The resulting density estimate for 2009 was approximately 0.89 turtles/ha.

Clutch Sizes

In 2009, 12 female Terrapene carolina found before June 9th were x-rayed. Only nine of these individuals were gravid. Three individuals had a clutch size of 7 eggs (1L 2R, 3L 3R, 9L 8R), two had a clutch size of 6 eggs (10L 11L, 3L 11L), one individual had a clutch size of 5 eggs (1L 1R), and three individuals had a clutch size of 4 eggs (1L 12L, 1L 11L, 9L 9R).

Nesting

The x-rayed individuals were never observed nesting. A single, unmarked female; however, was observed covering a nest at 19:24 on June 13. The female nested on a gravel road near the entrance to the field site. The nest was approximately 2 meters from a wooded edge. The air temperature was 24.5 degrees Celsius and the relative humidity was 85%. The substrate was a mix of dirt and gravel. At the time of nesting, the substrate was 22.2 degrees Celsius (Table 2). The nest was never depredated and hatching was not observed. Nesting movements were not analyzed because the nesting observation was incidental and the female was not fit with a radio transmitter.

Table 1. Carapace length (CL), carapace width (CW), weight, shell height (SH), and clutch size for x-rayed females. CL, CW, and SH given in millimeters, weight given in grams.

Table 2. Observed nesting event of an unknown female. Percent vegetation represents the amount of ground covered by vegetation and percent bare represents the amount of bare ground.

DISCUSSION

My disappointing results obviously preclude in depth analysis of the original objectives of this project. Several factors contributed to these substandard results:

Although the totaling number of female Terrapene carolina was 36 individuals, the number of mature females found before the start of the nesting season indicated that obtaining an adequate sample size of gravid females may not be possible. Females found 1-2 days before the scheduled x-ray session (12 individuals) were photographed. After nine gravid females were returned to their habitat, they immediately moved long distances within the site. It took several days to re-locate these individuals and it is likely that the nesting events took place before the individuals were found.

Terrapene carolina utilized a much larger portion of the study site than expected. The active area of the site was approximately 107 hectares. The habitat was very diverse and difficult to navigate. Private land , ravines, and old fence rows divided the site into four main habitats; grasslands, wet meadows, upland forests, and floodplain forests. It was concluded that navigating through the site during evening hours would be too dangerous.

Field help was a limiting factor for this project. The daily locating of radioed individuals took much longer than expected. Originally, I had planned on monitoring gravid females twice daily (once in the morning and once in the evening). Tracking individuals once during the day would sometimes take up to 10 hours even with field help.

Although the study was not carried out according to the original project outline, I was able to adjust the objectives and managed to retrieve some interesting and valuable data. Remaining funds were used to monitor both male and female Terrapene carolina to study the home range, movement, and habitat use of these individuals. Data from 2009 was presented in the form of a written thesis and submitted to the graduate school at the University of Illinois.

This project was the start of a long-term study in which home range, density estimates, habitat use, and hibernation will be considered. Upon the completion of my graduate work, I continued to locate radioed individuals until hibernation. Each hibernation site was set with a data logger which prepared for the continuation of this study.

LITERATURE CITED

Cagle, F.R. 1939. A system for marking turtles for future identification. Copeia 1939:170-173.

Congello, K. 1978. Nesting and egg laying behavior in Terrapene carolina . Proceedings of the Pennsylvania Academy of Science 52:51-56.

Converse, S.J., Iverson J.B., and J.A. Savidge. 2002. Activity, reproduction and overwintering behavior of ornate box turtles ( Terrapene Ornata Ornata ) in the Nebraska sandhills. American Midland Naturalist. 148 (2):416-422.

Dodd, C.K., Jr. 2001. North American box turtles: A natural history. Norman : University of Oklahoma Press, 1-231

Ernst, C.H., Lovich, J.E., and R.W. Barbour. 1994 Turtles of the United States and Canada . Washington and London : Smithsonian Institute Press, 1-578.

Ewing , H.E. 1935. Further notes on the reproduction of the eastern box turtle, Terrapene carolina . Copeia 1935: 102.

Gibbons, J. W. and J.L. Greene. 1979. X-ray photography: A technique to determine reproductive patterns of freshwater turtles. Herpetologica. 35 (1):86-89.

Kolbe, J.J., and F.J. Janzen. 2001. The influence of propagule size and maternal nest-site selection on survival and behaviour of neonate turtles. Functional Ecology. 15 (6):772-781.

Legler, J.M. 1960. Natural history of the ornate box turtle, Terrapene ornata ornata Agassiz . University of Kansas Publications Museum of Natural History. 11: 527-669.

Sexton, O.J. 1959. Spatial and temporal movements of a population of the painted turtle, Chrysemys picta marginata ( Agassiz ). Ecological Monographs. 29:113-140.

Stickel, L.F. 1950. Populations and home range relationships of the box turtle, Terrapene c. carolina (Linnaeus). Ecological Monographs. 20:351-378.

Stickel, L.F. 1989. Home range behavior among box turtles ( Terrapene c. carolina ) of a bottomland forest in Maryland . Journal of Herpetology. 23:40-41.