Chelonian Conservation and Biology, 2006, 5(2): 195–209 Ó 2006 Chelonian Research Foundation Reproduction and Nesting of the Endangered Ringed Map Turtle, Graptemys oculifera, in Mississippi

ROBERT L. JONES1

1Mississippi Department of Wildlife, Fisheries, and Parks, Museum of Natural Science, 2148 Riverside Drive, Jackson, Mississippi 39202 USA [[email protected]]

ABSTRACT. – Reproduction and nesting in the ringed map turtle (Graptemys oculifera) were investigated in the Pearl River of west-central Mississippi in 1995 and 1996. Nesting occurred from mid-May until mid-July but peaked in mid-June. Minimum carapace length of females at sexual maturity was 130 mm, but mean size at maturity was between 130 and 140 mm. Mean and modal CSs were 3.7 and 3. Larger females were gravid earlier than smaller ones and both egg and CS declined as the nesting season progressed. CS was positively correlated with both female carapace length and body mass. Mean egg length, width, and mass were 38.8 mm, 22.7 mm, and 11.8 g, respectively. Hatchlings averaged 35.5 mm carapace length and 8.9 g in mass. Annual clutch frequency may range from 0.96 to 1.42, and a minimum of ca. 60% of females reproduced on an annual basis. Predators destroyed an average of 86% of nests each year. Major vertebrate nest predators were armadillos, raccoons, and fish crows. Invertebrate egg predators destroyed an additional 24% of eggs known to be fertile.

KEY WORDS. – Reptilia; Testudines; Emydidae; Graptemys oculifera; turtle; female body size; CS; clutch frequency; egg size; nesting; nest predation; Mississippi; United States

Many turtles of conservation interest are inhabitants of ling size, nesting season, the incubation period, nesting rivers, but as Jackson and Walker (1997) pointed out, we success, and nest predation in the species. know less about the ecology of riverine turtles than of almost any other chelonian group. Graptemys, with 12 METHODS described species, is a genus of small to medium-sized turtles that inhabit rivers and streams of Gulf of Mexico The study was conducted on the Pearl River near drainages in central and eastern North America (Ernst et al. Ratliff Ferry, Madison County, Mississippi (Fig. 1), an 1994). One of these, the ringed map turtle (Graptemys area previously identified as having a high density of G. oculifera), is restricted to the Pearl River watershed of oculifera (Jones and Hartfield 1995). The river at the study Mississippi and Louisiana, where it occurs primarily in the site ranged from 40 m in width in side channels to main channels of the Pearl River and its largest tributary, approximately 200 m in bendways. Water depths ranged the Bogue Chitto River. Although G. oculifera has been from , 2to. 9 m. The 30-year (1962–1992) average discharge of the Pearl River at Jackson (ca. 45 km listed as a threatened species by the US Fish and Wildlife downstream) was 4627 cubic feet per second (Plunkett et Service since 1986 and is listed as endangered by the al. 1993). Highest flows occurred in March and April, and IUCN Red List (IUCN 2004), little information is lowest in September. The highest mean monthly temper- available on its nesting biology and reproduction. atures recorded over a 54-year period (1939–1993) at Females mature at a relatively small size (115–120 Jackson International Airport, ca. 38 km southwest, mm plastron length) and at a relatively late age (10 years) occurred in July and the greatest precipitation occurred (Jones and Hartfield 1995). Cagle (1953) described a in March (Ruffner and Blair 1981; Wood 1996). nesting attempt and provided data from a gravid female Females were captured in open-topped traps, con- indicating that she would have produced 2 clutches in a structed of either 1-inch nylon mesh in a metal frame, 1- season. Anderson (1958) briefly described the locations of inch poultry netting, or 5 3 10–cm wire mesh, which were nests on a sandbar in the Pearl River and provided attached to limbs and logs used as basking sites by the information on the movements of hatchlings at emergence. turtles. Trapping began on 1 May in 1995 and 30 April in Kofron (1991) examined a series of museum specimens 1996 and continued until no gravid females were present and concluded that ovulation and nesting occurred from in weekly samples and/or nesting was no longer observed May through July, that hatching occurred in August, and in the study area. Approximately 15 traps were used each that CS (CS) was 2–3 eggs. Given the paucity of data on day for 4 days during each week of sampling. reproduction in G. oculifera and its conservation status, Captured turtles were permanently marked by drilling the objectives of this study were to acquire information on holes in the marginals using Cagle’s (1939) numbering female body size, CS, clutch frequency, egg size, hatch- system. Straight-line carapace length (CL) and maximum 196 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006

Figure 1. Map of the study area on the Pearl River, Mississippi. The numbered areas are the 11 sandbars monitored for Graptemys oculifera nesting activity. The inset shows the location of the study area in west-central Mississippi. plastron length (PL) were measured to the nearest nests found in the absence of females were from ca. 2 to 12 millimeter with aluminum timber calipers. Midline hours old. Nests were excavated and for each egg the plastron length (MPL) was measured to the nearest 0.1 following was recorded: egg length (EL) and egg width mm with dial calipers. Body mass (BM) was measured to (EW), both measured to the nearest 0.1 mm, and egg mass the nearest gram using a portable electronic balance. (EM), measured to the nearest 0.1 g. Eggs were then Females were palpated and gravid individuals were reburied in the nest cavity. Nest sites were plotted on a transported to a veterinarian for radiography (Gibbons and scale map of the sandbar and keyed to numbered Greene 1979), which was used to determine CS and clutch aluminum tags attached to nearby trees. Both distance frequency. Gravid females found while attempting to nest and compass bearing from the tags were recorded to were measured and marked but were not X-rayed. facilitate relocation of the nests. Eleven sandbars were monitored for nesting activity Physical characteristics of G. oculifera nest sites were by G. oculifera (Fig. 1). Available nesting habitat, measured for intact nests and when females were measured using a hipchain, was considered to be the area discovered while nesting. These included straight-line between the river and the treeline at the top of the bar. distances of the nest site from the river and to the closest Both the percentage of vegetative cover and open sand on vegetation, whether the latter was herbaceous or woody, each bar were estimated by point counts. Most sandbars in the study area were used for camping by recreational the number of both herbaceous and woody stems within a boaters for at least part of the nesting season. This 0.5-m circle centered on the nest site, the percent shade at occurred sporadically during the week and was greatest on 1 m above the nest site as determined with a concave weekends and major holidays. Human presence on the 11 densiometer, soil particle diameter determined using a soil monitored sandbars was recorded daily during the week grain sizing folder (Forestry Suppliers No. 77332), and throughout the 1996 nesting season. substrate type (sand or sand/soil mixture). Nest cavity I surveyed sandbars several times each day for intact depth and diameter were measured when no alteration of nests, which were located either by direct observation of these dimensions had been made while excavating the nest nesting females or by following tracks to nest sites. Given cavity. Substrate temperature at the nest site and air the frequency with which sandbars were surveyed, most temperature 1 m above the substrate in both open and JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 197 shaded conditions were measured when females were plastic trays filled with moistened vermiculite. Trays were encountered while nesting. placed in circulated-air incubators (Hovabator Model Temperatures were monitored for 10 nests in each 1583) at 28.58–29.58C and misted weekly to reduce year of the study with a HOBO-XT7 data logger (Onsett evaporative water loss. CL, PL, and BM were measured Computer Corporation) attached to a 1.8-m cable with a for all hatchlings upon emergence from the nest cavity or thermistor probe. The data logger was sealed in a plastic after yolk absorption in lab-incubated turtles. Hatchlings container and buried ca. 1 m from the nest cavity with the were then released near the sandbar from which the eggs probe inserted at the approximate midpoint of the clutch. originally came. Data loggers were programmed to record nest tempera- Statistical tests were performed using Statistica tures at the same time each day at 48-minute intervals (30 (Statsoft, Inc. 1999). Variables were transformed to their measurements/day) throughout the incubation period. natural logarithms where appropriate to meet the assump- Nests were caged in plastic mesh, which prevented tions of parametric tests. Comparison-wise error rates were predation and permitted an investigation of the incubation set at alpha ¼ 0.05. Sequential Bonferroni adjustments for period. The mesh was folded into an open box and buried multiple comparisons were calculated as in Rice (1989) open-side down over the clutch ca. 3 cm below the soil with an experiment-wise error rate set at alpha ¼ 0.10 surface. All caged nests were checked for evidence of (Chandler 1995). Unadjusted p-values are reported, but hatching at 55 days after oviposition and daily thereafter cases in which adjustment for multiple tests altered until pipping occurred. The incubation period is defined significance are so indicated. Figures were produced in for this report as the time between oviposition and pipping part using AXUM (Trimetrix Inc. 1999). All means are of the eggs, and emergence as the time when hatchlings followed by 1 standard deviation. moved up through the sand from the nest cavity to beneath the plastic mesh. The latter is reported as both days from RESULTS clutch deposition and as days since pipping. Eggs that did Female Body Size Variation. — I captured 705 not pip were returned to the lab and examined under a individual female G. oculifera ranging from 111–215 dissecting microscope to determine if development had mm CL during the 2 years of the study. Using the size of occurred. the smallest gravid female (130 mm CL) as an estimate of Caged nests were checked daily to determine if the minimum size at maturity, 670 (95%) of the captured predators had attempted to excavate them. Predators were females were potentially adults (Table 1). Of these, 175 identified by tracks left in the sand near the nests, and only were recaptured at least once, and several were recaptured the first predation attempt for each nest was included in on multiple occasions, resulting in 896 total captures of subsequent data analyses. Other G. oculifera nests that had adult females during this study. been destroyed by predators (raided nests) were also Jones and Hartfield (1995) estimated from a von counted daily on each of the 11 sandbars. The predator, if Bertalanffy growth curve that female G. oculifera matured identifiable from tracks at the nest site, was recorded for at a MPL of ca. 115–120 mm. Based on the relationship each raided nest. Eggshells left by predators at nest sites between MPL and CL found in this study were removed from the sandbars daily so that raided nests (MPL ¼ 0.8395CL þ 4.9139; F(1, 917) ¼ 21,642.7, were not counted twice. p , 0.0001) an MPL of 115 mm occurs in females with Single eggs from 32 clutches and the entire comple- a CL of approximately 131 mm and an MPL of 120 mm in ment of 19 clutches were incubated in the lab in 1996 to females with a CL of approximately 137 mm. Mean size at investigate the relationship between egg and hatchling maturity, then, based on both the growth curve and the size. Eggs were buried up to approximately two-thirds of reproductive data, which appear to be concordant, their width in separate numbered compartments of covered probably occurs between 130 and 140 mm CL, even

Table 1. Mean carapace length (CL, mm) and body mass (BM, g) of presumed adult ( 130 mm CL) female G. oculifera.

CL BM na Mean Range Mean Range All females 670 156.3 6 15.3 130–215 545.7 6 174.5 269–1457 Gravid females 254 164.6 6 15.8 130–215 627.9 6 204.3 319–1457 Capture dateb Before 20 May 35 174.8 6 19.7 139–215 798.3 6 293.0 378–1457 20 May–8 June 96 162.2 6 14.3 130–210 602.7 6 175.1 362–1276 9 June–28 June 133 163.5 6 14.8 130–211 600.9 6 376.1 319–1297 After 28 June 13 168.7 6 20.9 130–210 655.7 6 248.7 319–1276 a n represents the number of individual females in each category and does not include recaptures of the same turtle. b Temporal groups represent gravid females captured during the first (before May 20), second (20 May–8 June), third (9–28 June), and fourth (after 28 June) quarters of the sampling period. 198 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006 though only approximately 7% of captured females in this June in 1995. The former, however, was based on a size range were gravid (Fig. 2). Larger females were also relatively small sample size (Fig. 4). In 1996, the number more likely to be gravid when captured than were smaller of gravid females captured gradually increased to a individuals (Fig. 2), but it is not clear whether this resulted maximum during the second week of June. The frequency from a relatively greater fecundity of large females or of gravid females in trapping samples declined markedly because larger females were more difficult to capture when in both years after that date (Fig. 4). The first G. oculifera not gravid. nest was found on 12 May in 1995 and on 14 May in Body size and capture date were significantly 1996, and the last on 17 July in 1995 and on 19 July in correlated in gravid females (CL: Spearman’s 1996. The length of the nesting season was therefore 67 r ¼0.124, p ¼ 0.04; BM: r ¼0.205, p ¼ 0.0006). A days in both years. plot of CL and capture date (Fig. 3) indicated that females Female G. oculifera did not appear to retain eggs for found gravid both early and late in the sampling period long periods of time. Thirty-three females first captured were larger than those gravid during the middle part of the while gravid were recaptured 11–43 days later when they period. The sampling period was therefore arbitrarily were no longer gravid, presumably having deposited their divided into quarters (before 20 May, 20 May–8 June, 9 clutches. Eighteen of these females (54.5%) were June–28 June, after 28 June) and gravid female body size recaptured within 15 days of their initial capture. (CL, BM) was compared among them. Females gravid Similarly, 27 females initially captured when not gravid during the first quarter were significantly larger in both CL were recaptured 7–42 days later and were found to be and BM than those gravid during the second and third gravid. Seventeen of these females (63%) were recaptured quarters (Mann-Whitney U-tests, all p , 0.001), but did within 21 days of their initial capture and apparently not differ from those gravid during the fourth quarter became detectably gravid within that 3-week period. Using (p . 0.10). Gravid females collected during the second, the minima of these ranges, female G. oculifera may only third, and fourth quarters, however, did not differ in size be detectably gravid with a particular clutch, as determined (Mann-Whitney U-tests, all p . 0.17). using both palpation and radiography, for a little more than Oviposition Period. — Gravid females were present in 2.5 weeks. trapping samples from 1 May to 12 July in 1995 and from CS. — Mean CS determined from radiography was 30 April to 15 July in 1996. The proportion of gravid 3.78 6 1.48 (range 1–10, mode ¼ 3, n ¼ 86) in 1995 and females in weekly samples ranged to as high as 69% (Fig. 3.59 6 1.02 (range 1–8, mode ¼ 3, n ¼ 159) in 1996. 4). The frequency of gravid females in samples peaked There was no difference in CS between years when body during both the third week of May and the second week of size (CL) was accounted for (Analysis of Covariance

Figure 2. Proportion of gravid and nongravid adult ( 130 mm CL) female Graptemys oculifera in 10-mm CL size classes. Numbers above bars are sample sizes. JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 199

Figure 3. Mean CL of gravid Graptemys oculifera by capture week for both years of the study combined. Numbers are sample sizes for each weekly interval.

[ANCOVA], F(1, 243) ¼ 1.03, p ¼ 0.311). The combined significantly different from that observed for X-rayed mean for both years was 3.66 6 1.20. Mean CS for 134 females (Mann-Whitney U-test; Z ¼1.830, p ¼ 0.067). intact G. oculifera nests found during the study was Both CL and BM were positively and significantly

3.42 6 1.11 (range 1–7, mode ¼ 3), which was not related to CS (CS ¼ 0.054CL 5.143; F(1, 243) ¼ 245.6,

Figure 4. Proportion of gravid Graptemys oculifera in weekly trapping samples for each of the 2 years of the study. Numbers represent adult females captured during each week. 200 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006

p , 0.001; CS ¼ 0.004BM þ 0.882; F(1, 243) ¼ 320.7; capture date, and it is therefore likely that some clutches p , 0.001), with BM explaining a higher proportion of were missed. the variation (r2 ¼ 0.569) in CS than CL (r2 ¼ 0.503). The If the average internesting interval was ca. 17 days CS–body size regressions indicated that an increase of 1 and if females are assumed to have produced clutches additional egg occurred for each approximately 225 g every 17 days, then some females may have produced increase in BM or for each approximately 19 mm increase more than 2 clutches per year. One female, gravid when in CL. captured on 20 May 1996, was found attempting to nest 44 CS declined through the nesting season, even when days later on 3 July. If the 20 May date is assumed to be body size (CL and BM) was accounted for (partial her first clutch of the season, she could have produced a r ¼0.182, p ¼ 0.006) which may have resulted from second clutch by approximately 18 June and a third by the fewer eggs in second or third clutches. However, 10 first week of July. A second female, one of the largest (210 females with 2 measurable clutches in a single season mm CL) captured during the study, was found attempting showed no significant difference in CS between their first to nest on 5 June 1996. Because large females tend to be and second clutches (Wilcoxon matched-pairs test: gravid both earlier and later than smaller individuals, this Z ¼ 0.534, p ¼ 0.593). nesting attempt may have represented her second clutch of Clutch Frequency. — Thirteen females were gravid the season, with the first on approximately 20 May. She twice in either 1995 (5 of 50 [10%] captured at least twice) was recaptured again while attempting to nest on 16 July. or 1996 (8 of 74 [10.8%] captured at least twice) as Assuming a 17-day internesting interval, this female may indicated by radiography and/or capture during nesting have produced a third clutch around 22 June and attempts. Mean annual clutch frequency (Frazer et al. potentially a fourth clutch of the season by 16 July. 1989) for this population was therefore 1.10 and mean Eighty-three females were captured in both 1995 and annual egg production (mean CS 3 mean clutch frequen- 1996, and only 12 (14.4%) were gravid in both years. cy) was 4.03. Mean interclutch interval was 23.3 days Three of these were found at least once on sandbars while gravid but prior to depositing eggs, so their CSs were not (range 12–44 days), but females recaptured more than 21 determined. CSs between years did not differ significantly days apart may have nested during the intervening period. (Wilcoxon matched pairs test: Z ¼ 0.534, p ¼ 0.593) for Mean time between clutches if the longest intervals (28, the 9 females with measurable clutches. The 12 females 33, 41, 44 days) were omitted was 16.7 6 3.8 days. were gravid from 36 days earlier to 19 days later in 1996 Mean annual clutch frequency was probably under- than in 1995, and mean date of capture was 4.00 6 14.46 estimated because females were not recaptured often days earlier in 1996. Dates of capture when gravid were enough to provide a sample size sufficient to adequately not significantly different between the 2 years (Wilcoxon evaluate this trait. A second estimate can be obtained from matched pairs test: Z ¼ 1.098, p ¼ 0.272). females initially captured during the first half of the The timing and frequency of recaptures may have also nesting season (prior to 15 June) and recaptured between precluded a reliable estimate of the proportion of the 12 and 21 days later (range of adjusted interclutch population that was gravid in both years. As indicated interval). This eliminates females initially captured late above, female G. oculifera appeared to nest at approxi- in the nesting season when they may have been mately the same time each year, and the probability of postreproductive and those recaptured at intervals greater capturing a female while she was gravid was higher for than the hypothesized interclutch period given above. females captured prior to 15 June. Therefore, to estimate Mean annual clutch frequency using data only from these the proportion of the female population that was gravid in females (n ¼ 36) was 1.25 and mean annual egg both years, only females that were known to have been production was 4.58. This, however, ignores females that gravid at least once, that were captured prior to 15 June in met the capture–recapture requirements but that were both years and that were recaptured in 1996 within ca. 19 never found gravid (n ¼ 17). Six of these 17 females had days of when they were gravid in 1995 (mean 6 1SDof weight losses of 21–68 g between captures. If these difference in capture dates when gravid in 1995 and 1996) relatively large weight changes are assumed to have were considered. Seventeen females met these require- resulted from clutch deposition, and ignoring the 11 ments, and of these, 10 (58.8%) were gravid in both years. females with no weight changes, then mean annual clutch Some females, as indicated above, reproduced frequency was 1.42 and mean annual egg production was annually. However, recapture data also suggest that some 5.20. If the remaining females that were never found females may not reproduce every year. Two females were gravid are included, then mean clutch frequency was 0.96 captured 4 times in 1996, and were not gravid at any and mean annual egg production was 3.51. Therefore, capture. The first female (154 mm CL) was initially mean annual clutch frequency may range from 0.96–1.42 captured 7 May, was recaptured 13 days later on 20 May, and mean annual egg production from 3.51–5.20. It should was recaptured again on 3 June after a period of 14 days, be emphasized, however, that these are minimum and was last captured 15 days later on 18 June. She may estimates because females that were not gravid when have produced a clutch after her last capture, but because captured may have been so either before or after their females are apparently detectably gravid for ca. 2.5 weeks, JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 201 the clutch would have been a relatively late one that would lab, however, pipped in a significantly shorter period of have been produced sometime after 5 July which is almost time than those incubated in the field in 1996 (Mann- at the end of the nesting season. A second female (149 mm Whitney U-test: Z ¼4.69, p , 0.0001). CL) was first captured on 30 May, was recaptured 12 days Relationships of Egg, Hatchling, and Female Sizes. — later on 11 June, was recaptured again 15 days later on 26 There were no differences in CL, BM, or PL between field June, and was recaptured a final time 14 days later on 9 and lab-reared hatchlings from the same nests (Mann- July. Body size of these 2 females indicated that they were Whitney U-tests: CL, Z ¼ 0.303, p ¼ 0.763; PL, probably mature and their recapture histories suggest that Z ¼ 0.546, p ¼ 0.585; BM, Z ¼ 1.023, p ¼ 0.305), nor they did not produce clutches in 1996. were there any hatchling size differences between years Egg Size. — The 136 intact G. oculifera nests found (CL, Z ¼ 0.320, p ¼ 0.749; PL, Z ¼ 0.713, p ¼ 0.476; during the study contained 456 eggs (Table 2). Mean egg BM, Z ¼0.230, p ¼ 0.818). Data from both years and size per clutch did not differ significantly between years from both lab and field-reared nests were therefore but did among months (Table 2). Eggs deposited in May combined (Table 4). Mean hatchling CL, PL, and BM were significantly larger than June eggs, which were (n ¼ 77) per clutch were all negatively correlated with significantly larger than those from July (Table 2). Mean clutch deposition date (CL: Spearman’s r ¼0.237, EL per clutch in 23 females captured while nesting was p ¼ 0.037; PL: Spearman’s r ¼0.361, p ¼ 0.001; BM: negatively correlated with date of nesting (Spearman’s r ¼0.288, p ¼ 0.011) although there was no significant r ¼0.507, p ¼ 0.014), even when female body size (CL, correlation between CL and clutch deposition date when BM) was accounted for (partial r ¼0.740, p ¼ 0.0001), adjustment for multiple tests was made. When mean egg but neither of the other 2 egg size measures were size (EL, EW, EM) per clutch was accounted for there was significantly related to clutch deposition date (EW: no significant correlation between hatchling size and date Spearman’s r ¼ 0.126, p ¼ 0.568; EM: r ¼0.199, of clutch deposition (PL: partial r ¼ 0.015, p ¼ 0.898; p ¼ 0.361). BM: partial r ¼ 0.061, p ¼ 0.608). Incubation Period. — The first emergence of G. There were significant correlations between female oculifera hatchlings in 1995 was from an unmarked nest body size and mean EW per clutch for the 23 females on 27 July, and in 1996 from a caged nest on 2 August. discovered while nesting, but none for the other measures There was no difference between years (Table 3) in the of egg size (Table 5). Twelve of these clutches produced length of time between egg deposition and pipping (Mann- hatchlings. Mean hatchling CL per clutch was weakly Whitney U-test: Z ¼0.019, p ¼ 0.985) and in time correlated with female CL and PL (Table 5). After between deposition and emergence (Z ¼1.706, adjusting for multiple tests, however, there were no p ¼ 0.88) for caged nests. Data were therefore combined significant correlations between female and hatchling for both years, and mean length of time was 64.4 6 4.7 sizes. days between deposition and pipping, 76.3 6 7.7 days Nest Densities. — I found 403 G. oculifera nests between deposition and emergence, and 12.0 6 5.5 days (intact and depredated) in 1995 and 476 in 1996 (Table 6). between pipping and emergence. Eggs incubated in the The surface area of the 11 monitored sandbars was

Table 2. Mean egg length (EL, mm), egg width (EW, mm), and egg mass (EM, g) of G. oculifera eggs.a

All eggs (n ¼ 456) Mean per clutch (n ¼ 136) Mean Range Mean Range EL 38.9 6 2.1 28.3–47.0 39.0 6 1.9 34.5–44.0 EW 22.7 6 1.3 17.6–26.8 22.5 6 1.2 19.5–25.5 EM 11.9 6 1.5 5.017.4 11.8 6 1.4 9.0–16.9 Means per clutch by year and month 1995 (n ¼ 70) 1996 (n ¼ 66) May (n ¼ 33) June (n ¼ 90) July (n ¼ 13) EL 39.1 6 2.0 39.0 6 2.0 40.3 6 1.6 38.9 6 1.7 36.8 6 1.7 EW 22.5 6 13 22.6 6 1.0 23.1 6 1.1 22.4 6 1.2 22.0 6 1.0 EM 11.8 6 1.5 11.9 6 1.2 12.7 6 1.2 11.7 6 1.3 10.6 6 0.8 Significance tests Months Years May–June May–July June–July Z p Z p Z p Z p EL 0.42 0.67 3.9 , 0.001 4.5 , 0.001 3.6 , 0.001 EW 0.12 0.90 2.5 0.014 2.9 0.004 1.2 0.218 EM 0.56 0.57 3.4 0.001 4.6 , 0.001 3.0 0.003 a Data are for individual eggs and for means per clutch. Significance tests (Z-values) are from Mann-Whitney U-tests examining differences in mean per clutch egg sizes between years and months. 202 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006

Table 3. Intervals (in days) between egg deposition and pipping, egg deposition and emergence, and pipping and emergence for G. oculifera clutches from field nests in 1995 and from both laboratory and field nests in 1996.a

Type Year n Deposition–pipping n Deposition–emergence n Pipping–emergence Field 1995 24 64.5 6 5.6 (56–81) 17 74.1 6 8.4 (63–95) 17 10.3 6 6.2 (2–27) 1996 44 64.4 6 4.3 (55–75) 40 77.2 6 7.3 (64–95) 40 12.7 6 5.1 (4–29) Lab 1996 45 60.6 6 2.8 (54–68) — — a Data are for the first individual in a clutch that pipped or emerged and thus represent minimum values. composed of an average of 39% open sand, 38% optimal nesting habitat. This sandbar had once been part of herbaceous vegetation, and 23% woody vegetation, a lateral and point bar system adjacent to the main channel, primarily black willow (Salix nigra). The number of nests but had become partially isolated as a result of the river per sandbar did not differ significantly between years changing its course. This resulted in less scouring during (Wilcoxon matched pairs test: Z ¼ 1.56, p ¼ 0.120), even floods and an increase of vegetative growth. Although though sandbar 6, which had over 11% of all nests in sandbar 5 no longer had large expanses of open sand, it was 1995, had fewer than 4% in 1996 (Table 6). The density of still one of the largest sandy areas in the lower part of the G. oculifera nests per sandbar ranged from approximately study area, and may have been frequently used by G. 1 nest per 4 m2 to , 1 nest per 200 m2 (Table 6). Most oculifera females because of the absence of better nesting sandbars with an area of less than 1000 m2 had relatively areas in the immediate vicinity. high densities of turtle nests, i.e., 1 nest per 5–20 m2 of Timing of Nesting. — Female G. oculifera were sandbar. Larger sandbars had much lower densities, observed nesting between 0815 and 1700 hrs. Of 27 G. ranging from 1 nest per approximately 70–160 m2. The oculifera observed excavating a nest chamber or laying number of nests per sandbar was correlated with sandbar eggs, 8 (29.6%) were found between 1000 and 1100 hours, area (Spearman’s r ¼ 0.78, p ¼ 0.005) but not with the and 19 (70.4%) were nesting before 1200 hours. Surface amount of open sand nor with the amount of either soil temperatures for nesting females ranged from 238 to herbaceous or woody vegetation present (all p . 0.23). 498C. No evidence of nocturnal nesting was observed in Sandbars 12 and 21 had fewer and sandbar 5 had more either of the 2 field seasons. nests (Table 6) than predicted by the relationship between Nest Sites. — There were no differences in G. area and number of nests per sandbar (nests ¼ oculifera nest site characteristics (Table 7) between years 2 0.0091[area] þ 20.19; r ¼ 0.64, F(1, 9) ¼ 16.23, p , 0.003). or months (Mann-Whitney U-tests, all p . 0.06). Most The relatively few nests on sandbar 12 may have resulted nests were constructed in fine sand approximately 18 m from its frequent use by recreational boaters. During the from the water at locations with a canopy cover averaging 1996 nesting season, boaters were observed at 4 of the 11 approximately 37% and within 1 m of vegetation (Table sandbars on 33 weekdays and were present for almost the 7). There were usually both herbaceous and woody stems entire day on approximately half of those dates. Twenty- within 50 cm of the nest cavity, which averaged seven of these observations (82%) were from sandbar 12. If approximately 6 cm in diameter and 12 cm in depth weekends are included, during which most large sandbars (Table 7). in the study area were occupied by camping boaters, Nest Temperatures. — The mean nest temperature for humans were present on sandbar 12 at least part of the day 10 nests with more than 1000 individual temperature for 46 of the 67 days (69%) that G. oculifera were nesting observations (. 30 days) was 27.26860.848C. Temper- in 1996. Sandbar 21 was infrequently visited by humans, ature extremes for these 10 nests ranged from 13.78 to but had an extensive growth of Salix nigra approximately 5 37.08C (Table 8). Means of each of the 30 daily m from the edge of the river which formed a ring ca. 3 m temperatures were calculated for each nest over the entire wide and which bordered ca. 80% of the sandbar adjacent period during which the data loggers were active (Table 8). to the river. This may have made the sandbar appear Nests reached their average maximum temperatures at unsuitable as a nesting site to female G. oculifera observing 1535 hours and average minimum temperature at 0633 it from the river. Sandbar 5, although relatively large, had hours. The mean difference between highest and lowest relatively little open sand (Table 6) and did not appear to be mean daily temperatures was 6.36861.248C and ranged

Table 4. CL (mm), maximum plastron length (PL, mm), and BM (g) of G. oculifera hatchlings for all specimens and as means per clutch. (See Tables 1 and 2 for definitions of CL and BM.)a

n CL Range PL Range BM Range All turtles 227 35.51 6 2.07 28–40 30.78 6 1.83 24–35 8.91 6 1.28 4.4–11.4 Per clutch 77 35.24 6 1.82 30–40 30.59 6 1.60 26–34 8.73 6 1.12 5.8–11.1 a Means are followed by 6 1 standard deviation. JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 203

Table 5. Spearman rank order correlations between female body size and egg size, hatchling body size and egg size, and between female body size and hatchling body size in G. oculifera. (See Tables 1, 2, and 4 for definitions of abbreviations.)a

Egg size n EL p EW p EM p Female body size CL 23 0.191 0.382 0.598 0.003* 0.314 0.144 PL 23 0.245 0.259 0.648 0.001* 0.294 0.174 BM 23 0.223 0.307 0.546 0.007* 0.240 0.271 Hatchling body size n CL p PL p BM p Egg size EL 70 0.482 , 0.001* 0.621 , 0.001* 0.477 , 0.001* EW 70 0.423 , 0.001* 0.373 0.001* 0.569 , 0.001* EM 70 0.633 , 0.001* 0.650 , 0.001* 0.762 , 0.001* Female body size CL 12 0.607 0.036 0.197 0.539 0.423 0.171 PL 12 0.587 0.045 0.116 0.720 0.491 0.105 BM 12 0.510 0.090 0.077 0.812 0.305 0.335 a Egg–hatchling correlations are for lab-reared specimens and represent individual hatchlings and eggs. Female–hatchling and female–egg correlations are from clutches found while the female was laying where egg and hatchling measures are means per clutch. An asterisk indicates significance after sequential Bonferroni adjustment. from 5.058 to 9.258C. There were significant correlations quencies of known predators were marginally nonsignif- between low mean nest temperature and number of woody icant between the 2 groups of nests (both years combined: stems around the nest site (Spearman’s r ¼ 0.77, H ¼ 3.94, p ¼ 0.05) probably because there were fewer p ¼ 0.007), between mean high nest temperature and predation attempts by fish crows on caged nests. This distance of the nest site from water (r ¼ 0.70, p ¼ 0.025), likely occurred because caging nests disturbed a relatively and between canopy cover and both time of the highest large area of sand and may have made nest sites mean temperature (r ¼0.66, p ¼ 0.039) and time of unrecognizable to fish crows. Predation attempts on caged lowest mean temperature (r ¼0.69, p ¼ 0.026). Howev- nests occurred as late as 69 days after the eggs were er, the latter 3 were not significant after adjustments were deposited, but over 42% of all predation attempts were made for multiple comparisons. made within the first 24 hours, and 81% within the first 14 Nest Predation. — Approximately 86% of the caged days following oviposition. Nests upon which predation G. oculifera nests were attacked by vertebrate predators (Table 9), but none of the attacks were successful. attempts had been made (n ¼ 100) had significantly Armadillos (Dasypus novemcinctus) and raccoons (Pro- greater canopy cover (36.6% vs. 16.3%) than those on cyon lotor) were the most frequent predators identified at which no predation attempts had been made (n ¼ 15) both caged and raided nests. Fish crows (Corvus (Mann-Whitney U-test: U ¼ 365.0, Z ¼3.2, p ¼ 0.001) ossifragus) were significant predators at raided nests but did not differ for the other nest sites characteristics (Table 9). Predator frequencies did not differ between measured (all p . 0.10). This may have been a result of an years for caged (Kruskal–Wallis test: H ¼ 1.11, edge effect on nest predation rates, which has been p ¼ 0.293) or raided nests (H ¼ 0.10, p ¼ 0.753). Fre- observed for other turtle species (Temple 1987).

Table 6. Area of and number of nests found on each sandbar monitored for G. oculifera nesting activity in 1995 and 1996.a

No. nests No. nests Mean no. Area per Predicted nests Bar Area (m2) 1995 (%) 1996 (%) nests (%) nest (m2) (95% CI) 5 912 64 (15.9) 80 (16.8) 72 (16.3) 12.7 28 (12–45) 6 412 45 (11.2) 17 (3.6) 31 (7.4) 13.3 24 (7–41) 11 441 30 (7.4) 25 (5.2) 28 (6.4) 16.0 24 (7–41) 12 2007 10 (2.5) 15 (3.2) 12 (2.8) 160.6 38 (24–53) 15 158 21 (5.2) 37 (7.8) 29 (6.5) 5.4 22 (4–40) 16 135 17 (4.2) 23 (4.8) 20 (4.5) 6.8 21 (3–40) 18 55 8 (2.0) 9 (1.9) 8 (1.9) 6.5 21 (3–40) 19 38 10 (2.5) 7 (1.5) 9 (2.0) 4.5 20 (2–39) 20 4977 63 (15.6) 73 (15.3) 68 (15.5) 73.2 66 (45–86) 21 5641 33 (8.2) 49 (10.3) 41 (9.2) 137.6 72 (48–95) 30 9085 102 (25.3) 141 (29.6) 122 (27.5) 74.8 103 (65–141) a Numbers of nests include both intact and depredated nests as well as those of females interrupted while excavating the nest cavity. Areas per nest are sandbar area divided by the mean number of nests on each sandbar. Predicted nests and 95% confidence interval for each sandbar are from the regression equation for the relationship between sandbar area and mean number of nests per sandbar (See text). 204 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006

Table 7. Characteristics of G. oculifera nest sites.a

n Mean 6 1 SD Range n (%) Canopy cover (%) 133 36.6 6 27.3 0–100 Substrate Distance to water (m) 133 18.3 6 13.9 0.33–61.20 Sand 111 (83) Distance to stem (m) 133 0.7 6 1.0 0.03–4.42 Sand/detritus 22 (22) No. herbaceous stems 133 9.4 6 20.3 0–122 Grain size No. woody stems 133 3.8 6 6.3 0–31 Medium 37 (28) Cavity depth (cm) 106 12.1 6 1.4 9–16 Fine 73 (55) Cavity diameter (cm) 106 5.9 6 1.2 4–9 Very fine 23 ( 17) a Canopy cover is the percentage of shade above the nest site, distances to water and to stem are distances to the river and nearest vegetation, numbers of stems are those within a 0.5-m circle centered on the nest site, cavity depth and diameter are nest cavity dimensions, and grain size is particle diameter of the soil in which the nest cavity was dug.

Egg Development. — Sixty-one clutches containing a 1995, whereas over 80% produced viable hatchlings in total of 206 eggs were caged to exclude vertebrate 1996. The differences in development and emergence of predators in 1995. When these eggs were checked after eggs between 1995 and 1996 did not appear to result from 55 days, only empty egg shells remained for 11 eggs from incubating some eggs in the laboratory in 1996. There 7 clutches, so their developmental status could not be were no differences in 1996 between lab and field-reared ascertained. Thus, 195 eggs, all incubated in the field, were eggs in the proportions of fertile vs. undeveloped eggs examined in 1995. In 1996, 217 eggs from 65 clutches (v2 ¼ 0.24, df ¼ 1, p ¼ 0.626), between pipped vs. fertile were examined. Approximately 37% of these were but unpipped eggs (v2 ¼ 0.01, df ¼ 1, p ¼ 0.936), nor incubated in the laboratory and the remainder incubated between the number of hatchlings that emerged vs. in the nest cavity (Table 10). developed hatchlings that died before emergence The frequencies of fertile and undeveloped eggs (v2 ¼ 1.71, df ¼ 1, p ¼ 0.191). differed significantly between years (v2 ¼ 153.14, df ¼ 1, Egg Predation. — Approximately half of the fertile p , 0.0001), with almost 60% of the 1995 eggs showing eggs that developed but that did not show signs of pipping no evidence of development whereas almost 97% of 1996 were destroyed by Solenopsis molesta, a native species of eggs were fertile, as indicated by the presence of embryos fire ant that appeared to attack eggs just prior to and or live hatchlings (Table 10). In 1995, the entire immediately following pipping (Table 11). Larvae of the complement of 22 clutches showed no evidence of dipteran Tripanurga (¼ Metaposarcophaga) importuna,a development, whereas only 1 clutch in 1996 did not have sarcophagid fly, were also important predators, particularly any eggs that developed. At least 2 of the failed 1995 in 1995. The larvae were present in many of the clutches clutches were inundated by high water in early June, and a that did not show any evidence of development, and were third appeared to have been desiccated by plant rootlets. frequently found infesting eggs that appeared to have been All but 1 of the 19 remaining failed clutches were unpipped. These, however, may have first been attacked deposited in June, and 14 of these were deposited between by S. molesta, which apparently gained entry into eggs by 5 and 16 June during a cool rainy period. Approximately chewing holes through the shell, a behavior which has also 22% of eggs produced viable (emerged) hatchlings in been noted in Solenopsis invicta, the red imported fire ant

Table 8. Temperatures (8C) of G. oculifera nests monitored with data loggers for at least 30 days.a

Temperature means Time of Temperature ranges Nest Date n Overall Minimum Maximum Minimum Maximum Minimum Maximum 29 22 May 1995 1478 27.7 24.9 31.2 0856 1656 17.4 37.0 34 23 May 1995 1418 25.7 22.9 29.0 0711 1511 19.4 32.5 37 23 May 1995 1800 27.2 24.2 30.5 0758 1646 20.1 34.4 46 24 May 1995 1800 27.3 24.4 31.1 0726 1615 19.1 35.0 91 12 June 1995 1800 28.2 25.1 32.1 0810 1658 21.5 35.5 96 14 June 1995 1114 26.0 22.7 27.8 0409 1610 13.7 34.0 147 3 June 1996 1767 27.0 24.9 30.0 0348 1412 19.4 35.0 151 5 June 1996 1140 27.6 24.0 33.3 0617 1329 19.4 37.0 155 12 June 1996 1081 28.1 24.8 32.0 0633 1609 19.8 37.0 165 3 June 1996 1797 27.9 25.6 30.6 0459 1259 22.2 35.0 a Date is when the data logger was installed at the nest, n is the total number of temperatures recorded, and the overall mean is the average of all of the individual temperatures recorded at the nest. Temperatures were recorded at 48-minute intervals at the same time each day (30 times/day). Minimum means are the lowest of the 30 means/day, maximum means are the highest of the 30 means/day, times of minimum and maximum means indicate at which of the 30 times the minimum and maximum means occurred, and temperature ranges are the minimum and maximum individual temperatures recorded at each nest. JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 205

Table 9. Vertebrate predators of G. oculifera nests.a

n Fish crow Raccoon Armadillo Other Unknown None Caged 1995 70 4 (5.7) 13 (18.6) 29 (41.4) 1 (1.4) 13 (18.6) 10 (14.3) 1996 48 0 (0.0) 13 (27.1) 27 (56.2) 1 (2.1) 0 (0.0) 7 (14.6) Raided 1995 328 57 (17.4) 76 (23.2) 65 (19.8) 4 (1.2) 126 (38.4) — 1996 409 51 (12.5) 90 (22.0) 90 (22.0) 2 (0.5) 176 (43.0) — a Caged nests were those covered with plastic mesh and data recorded were predation attempts. Raided nests were those without plastic mesh covering that were destroyed by predators. n is the total number of nests in each category in each year. Other category includes human (3 nests), common crow, Corvus brachyrhynchos (1 nest), speckled kingsnake, Lampropeltis getula (2 nests), and an unidentified canid (2 nests). Percentages are in parentheses.

(Moulis 1997). Approximately 30% of eggs that did not other Graptemys range from 2 to 3 (G. ouachitensis, G. show signs of pipping contained dead embryos. Approx- pseudogeographica; Vogt 1980), to 3 (G. nigrinoda; imately equal numbers of turtles that pipped but did not Lahanas 1982), to as many as 6 (G. ernsti; Shealy 1976). emerge were either killed by Tripanurga or S. molesta or These studies derived their estimates from dissections and died before emerging from the nest (Table 11). Of the 211 therefore are probably maximum estimates of average field-incubated eggs known to be fertile in this study, 50 clutch frequency for those species, whereas my estimate (23.7%) were destroyed by invertebrate predators prior to for G. oculifera, which is derived from radiography and either pipping or emergence. Several clutches when first recapture data, is likely a minimum estimate. The checked for evidence of pipping contained only empty problems inherent in using radiography and multiple shells and were likely destroyed by invertebrates. recaptures of adult females to determine clutch frequency have been discussed elsewhere (Tucker 2001; Horne et al. DISCUSSION 2003) and likely apply equally well here. On the other hand, both Iverson and Smith (1993) and Tucker (2001) CS and Frequency. — Graptemys oculifera is a have used peaks in the distribution of gravid females relatively small species with an apparently low annual captured throughout the nesting season to estimate clutch reproductive potential. Female CL at maturity (130 mm) is frequency. If this is equally applicable to G. oculifera, then less than that of many other species in the genus (e.g., Fig. 4 supports the hypothesis that the majority of females Graptemys ouachitensis, 150–160 mm; Graptemys geo- nest only once during the year. Shelby et al. (2000) graphica, 190 mm; Graptemys ernsti, 212 mm; and suggested that the low reproductive frequency in G. Graptemys flavimaculata, 159 mm; Ernst et al. 1994; flavimaculata might have resulted in part from a disruption Horne et al. 2003), and the mean CL of gravid females of the hormonal system because of chemical pollution. It is found in this study (165 mm) was less than that of G. not known whether a similar problem affects G. oculifera flavimaculata (190 mm; Horne et al. 2003), its sister at Ratliff Ferry or if some unknown selective pressure species (Lamb et al. 1994; Stephens and Wiens 2003). Its favors an extremely low reproductive potential in this CS (3.7) is the smallest yet reported for any Graptemys, species. A study investigating the endocrine system of G. which range from 4.8 for G. flavimaculata (Horne et al. oculifera similar to that of Shelby et al. (2000) would be 2003) and 5.5 for Graptemys nigrinoda (Lahanas 1982) to useful in investigating this question. 14.1 for Graptemys pseudogeographica (Vogt 1980). Some female G. oculifera apparently skip reproduc- Horne et al. (2003) using radiography and Shelby et tion in certain years, a phenomenon that has also been al. (2000) using ultrasound on the same population of G. observed in a number of other turtles (see Congdon et al. flavimaculata found what they termed an abnormally low 1987). The probability of skipping a year of reproduction reproductive frequency of 1.16 and 1.17, respectively. The was apparently size-related in Chrysemys picta, where reproductive frequency of G. oculifera estimated using large individuals reproduced every year but smaller turtles radiography was 1.10 and may have ranged from 0.96– skipped years (Christens and Bider 1986). There were not 1.42 (see Results). The number of clutches produced by enough recapture data to determine whether this also

Table 10. Fertility and hatching success of G. oculifera eggs.a

Eggs that did Hatchlings that Clutches Total eggs not develop Fertile eggs Eggs that pipped emerged 1995 61 195 115 (59.0) 80 (41.0) 55 (28.2) 43 (22.1) 1996 65 217 7 (3.2) 210 (96.8) 187 (86.2) 175 (80.6) Field 136 5 (3.7) 131 (96.3) 117 (86.0) 106 (77.9) Lab 81 2 (2.5) 79 (97.5) 70 (86.4) 69 (85.2) a Percentages (in parentheses) are of total numbers of eggs examined. 206 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006

Table 11. Mortality of fertile unpipped eggs and of pipped G. oculifera eggs.a

Fertile eggs that did not pip Total eggs No. not pipped Tripanurga importuna Solenopsis molesta Solenopsis invicta Unknown 1995 80 25 (31.2) 10 (40.0) 12 (48.0) 1 (4.0) 2 (8.0) 1996 131 14 (10.7) 0 (0.0) 11 (78.6) 0 (0.0) 3 (21.4) Total 211 39 (18.5) 10 (25.6) 23 (59.0) 1 (4.0) 5 (12.8) Pipped eggs from which hatchlings did not emerge Total eggs No. not emerged Tripanurga importuna Solenopsis molesta Solenopsis invicta Unknown 1995 55 12 (21.8) 1 (8.3) 4 (33.3) 1 (8.3) 6 (50.0) 1996 117 11 (9.4) 7 (63.6) 3 (27.3) 0 (0.0) 1 (9.1) Total 172 23 (13.4) 8 (34.8) 7 (30.4) 1 (8.3) 7 (30.4) a Columns indicate number of eggs or pipped hatchlings destroyed by invertebrate predators. Percentages are in parentheses. happened in G. oculifera, but the body size data in Fig. 2 Additional data on the energetics of egg production in G. indicates that larger females were much more likely to oculifera will be necessary to test these hypotheses. In have been gravid when captured than were smaller either case, early nesting may be advantageous because individuals. This suggests that the majority of larger offspring would have longer to grow before the onset of females may be gravid on an annual basis whereas smaller cooler temperatures in the fall, assuming that overwinter turtles are gravid on a more irregular schedule. survival in neonates is positively correlated with body size. Variation in Body Size and Reproduction. — Gravid Both clutch and egg size of G. oculifera declined G. oculifera found early in the nesting season tended to be during the nesting season, even when controlling for the larger than those gravid later in the year, which has also effects of body size. A decrease in CS during the nesting been noted in G. ernsti (Shealy 1976) and in some season has been observed in other turtles (Apalone mutica, populations of Chelydra serpentina (Hammer 1969; Plummer 1977; Trachemys scripta, Pseudemys floridana, Petokas and Alexander 1980). However, this was not Kinosternon subrubrum, Gibbons et al. 1982; Chelydra observed in other snapping turtle populations (Congdon et serpentina, Iverson et al. 1997), both with and without al. 1987; Iverson, et al. 1997), nor in Emydoidea controlling for body-size effects, and a seasonal decline in blandingii (Congdon et al. 1983). Why this occurred is egg size was observed in both Malaclemys terrapin unclear, but may have involved either the relative quantity (Montevecchi and Burger 1975) and Chelydra serpentina of resources available for reproduction or the relative (Iverson et al. 1997). Both egg mass and CS decreased efficiency with which clutch development was completed. seasonally across multiple clutches in individual If the former is true, then larger, early-nesting females may Chrysemys picta (Iverson and Smith 1993), so the have had sufficient fat reserves at the beginning of the seasonal decline in these parameters observed in the summer activity period to complete development of their Ratliff Ferry population of G. oculifera may have occurred clutches, whereas those nesting later in the season may because eggs and clutches found later in the nesting season have been less efficient in acquiring and storing resources were from second or even third clutches. Thus the the previous fall and therefore had to forage in spring in observed seasonal decline in clutch and egg size in G. order to finish clutch development. This implies that larger oculifera may mean that clutch frequency was actually females were more efficient either at foraging or at storing much higher than indicated by the recapture data. lipids than were smaller individuals. However, if most of Alternatively, if these later eggs and clutches were the energy allocated to an initial clutch in the current year primarily from initial (and presumably the only) clutches came from resources harvested the previous year, as has in a season, then the resources available for reproduction been suggested for C. picta (Congdon and Tinkle 1982), in a given year apparently vary greatly among females, then both large and small females should already have had allowing some to produce large eggs and clutches early in sufficient reserves for their initial clutches prior to the the nesting season, but limiting others to smaller eggs and beginning of the nesting season, and would be expected to clutches later in the season. If the latter is true, then the initiate egg-laying at approximately the same time. annual reproductive output in this species may be Although a few small females did nest early in the season controlled largely by extrinsic environmental factors, as (Table 1), most females ovipositing during that period has been suggested for other turtles (Gibbons 1982; were larger individuals. Alternatively, because basking Gibbons et al. 1982). Additional data on clutch frequency may be integral to the maturation of eggs, probably by and reproductive energetics will be necessary to evaluate increasing metabolic rates, larger females may have been these hypotheses. able to maintain higher body temperatures more efficiently Egg and hatchling size were correlated as were EW in early spring and matured clutches more quickly than and female body size, so it was expected that female body smaller females, thus laying eggs earlier in the season. size and hatchling size would also be correlated. However, JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 207 in this study this was not the case. The lack of correlation such a large number of G. oculifera eggs will have to be between female size and hatchling size may have resulted determined in laboratory experiments. from either variation in incubation conditions among The relatively large number of G. oculifera nests clutches and its subsequent effects on hatchling sizes (see destroyed by predators is similar to that found in several Packard et al. 1985; Packard et al. 1987), or from small other studies of North American turtles (e.g., Moll and sample size (n ¼ 12), which included only a limited range Legler 1971; Shealy 1976; Burger 1977; Vogt 1980; of female and egg sizes. Either of these 2 factors could Congdon et al. 1993) with ca. 86% of the nests attacked by easily have overshadowed any variation in hatchling sizes vertebrates and ca. 24% of the remaining eggs destroyed because of differences in adult female sizes. With a larger by invertebrates. Armadillos have not been identified as range in sizes and greater numbers of both females and significant predators of North American turtle nests eggs, and with the latter incubated under identical (Wilbur and Morin 1987), but were one of the principal conditions, hatchling and female sizes may well prove to predators of Trachemys scripta nests in Panama (Moll and be correlated. Legler 1971). The armadillo is a relatively recent addition Nesting and Nest Predation. — All nests of G. to the fauna of Mississippi, first reported east of the oculifera found during the study were on sandbars, in Mississippi River in 1943 (Lowery 1943) and occupying contrast to those of G. flavimaculata, which were found on two-thirds of the state by 1974 (Humphrey 1974). open areas along the riverbank as well as sandbars (Horne Mississippi also has a large population of raccoons (J. et al. 2003). These riverbank areas were used because of Watkins, Mississippi Department of Wildlife, Fisheries, the frequent occupation of sandbars by humans, effectively and Parks, pers. comm.), and the fish crow also appears to preventing their use by nesting females. Female G. be expanding its range (Madge and Burn 1994). Vertebrate oculifera did not use riverbanks as nesting sites because, predation experienced by G. oculifera today may therefore even though the Ratliff Ferry area is heavily used by be significantly higher than it was historically. recreational boaters and human occupation of sandbars Both of the primary invertebrate species implicated in the destruction of G. oculifera eggs and hatchlings have occurred with some frequency, the large number of been previously identified as predators of turtle nests. sandbars in the study area seemed to provide ample Tripanurga importuna preyed upon recently pipped undisturbed nesting habitat. However, in areas of the Pearl hatchlings of G. pseudogeographica (Vogt 1981), de- River where few sandbars are present, as occurs in the stroying 36% of the turtles from 23 clutches. Ants have lower Pascagoula River where Horne et al. (2003) studied also been implicated in the destruction of turtle hatchlings G. flavimaculata, or on individual sandbars in high use in several studies (e.g., Burger 1977; Landers et al. 1980; areas, such as sandbar 12, human disturbance could have a Congdon et al. 1983; Moulis 1997). Solenopsis molesta, significant impact by forcing turtles to nest in suboptimal the primary ant predator found in this study, also preyed habitats. upon Trachemys scripta eggs in Panama (Moll and Legler Production of hatchlings from monitored nests varied 1971). Solenopsis invicta, the red imported fire ant, was significantly during the study, ranging from ca. 22% in expected to be a potential predator of hatchling G. 1995 to ca. 78% in 1996. This resulted in part from the oculifera, because it has been implicated as a predator large percentage of undeveloped eggs in 1995, when on a variety of turtle species (reviewed in Moulis 1997), almost 60% failed to develop. Eggs which show no and it was present in the vicinity of all sandbars. However, evidence of development are usually considered to be it was responsible for the destruction of only 1 clutch and infertile (e.g., Burger 1977; Congdon et al. 1983; Christens was not a significant cause of mortality in the study area. and Bider 1986), but in most of these studies fewer than Mount (1981) noted that S. invicta had difficulty ca. 10% of eggs were in this condition. It seems unlikely establishing colonies in excessively sandy soils, which that infertility was responsible for failure of the large may explain why few G. oculifera nests, which are number of eggs in 1995 because this would require annual constructed primarily in almost pure sand, were depredat- variation in G. oculifera fertility rates ranging from 40% to ed by this species. almost 100%. Eggs were handled in the same manner in Although the Ratliff Ferry population of G. oculifera both years, so it is improbable that the variation was a appears to be stable at present, its future is not entirely result of methodological differences. It seems more likely secure. Turtles are long-lived species, and the demograph- that this resulted from the effects of temperature or ic traits that have evolved with longevity often result in moisture on development. If this were the case, the eggs populations that have a limited ability to adapt to increases were affected soon after oviposition because the contents in mortality, including a chronic increase in the mortality were still bright yellow and none showed evidence of of neonates (Congdon et al. 1993). Predation rates found somatogenesis. A large proportion of the failed clutches in this study, although similar to that reported in other were deposited in June, when mean temperatures were ca. studies, are a reason for concern, particularly because one 1.58C cooler than the long-term average (National Weather of the major egg predators (D. novemcinctus) is a recently Service, Jackson, unpubl. data). Whether this minor arrived component of the fauna, the second (P. lotor) has variation in temperature could have resulted in failure of increased substantially over the state in the last few years, 208 CHELONIAN CONSERVATION AND BIOLOGY, Volume 5, Number 2 – 2006 and the third (C. ossifragus) appears to be expanding both Delayed sexual maturity and demographics of Blanding’s its range and numbers. If G. oculifera cannot adapt to turtles (Emydoidea blandingii): implications for conservation increased nest predation and there is no concomitant and management of long-lived organisms. Conservation Biology 7:826–833. increase in juvenile or adult survivorship, the long-term CONGDON, J.D. AND TINKLE, D.W. 1982. Reproductive energetics prospects for the survival of this species do not appear to of the painted turtle (Chrysemys picta). Herpetologica 38: be high. The low reproductive frequency of this species, if 228–237. it is real, is a serious concern, particularly if it has resulted CONGDON, J.D., TINKLE, D.W., BREITENBACH, G.L., AND VAN LOBEN from an unknown chemical stressor in the watershed. SELS, R.C. 1983. Nesting ecology and hatching success in the Finally, although there did not appear to be a serious turtle Emydoidea blandingi. Herpetologica 39:417–429. conflict between turtle nesting and human usage of ERNST, C.H., LOVICH, J.E., AND BARBOUR, R.W. 1994. Turtles of the United States and Canada. Washington, DC: Smithsonian sandbars in the Ratliff Ferry area, this has the potential Institution Press, 578 pp. to become a serious problem in the future, especially if the GIBBONS, J.W. 1982. Reproductive patterns in freshwater turtles. recent trend toward the establishment of summer-long tent Herpetologica 38:222–227. camps continues. These multitent encampments, which are GIBBONS, J.W. AND GREENE, J.L. 1979. Determination of CS in usually set up in April and remain until early September, turtles with an X-ray technique. Herpetologica 35:86–89. not only disrupt female nesting behavior but also cover GIBBONS, J.W., GREENE, J.L., AND PATTERSON, K.K. 1982. Variation much of the available nesting habitat, especially on small in reproductive characteristics of aquatic turtles. Copeia 1982: 776–784. sandbars. HAMMER, D.A. 1969. Parameters of a marsh snapping turtle population Lacreek Refuge, South Dakota. Journal of Wildlife ACKNOWLEDGMENTS Management 33:995–1005. HORNE, B.D., BRAUMAN, R.J., MOORE, M.J.C., AND SEIGEL, R.A. Partial funding for this study was provided by the US 2003. Reproduction and nesting ecology of the yellow- Fish and Wildlife Service, Project No. E-1, under a Section blotched map turtle, Graptemys flavimaculata: implications 6 Cooperative Agreement with the Mississippi Department for conservation and management. Copeia 2003:729–738. of Wildlife, Fisheries, and Parks. Richard Burton and his HUMPHREY, S.R. 1974. Zoogeography of the nine-banded staff at the Jackson Animal Clinic provided both the armadillo (Dasypus novemcinctus) in the United States. facilities for and expertise in radiography. The Tripanurga BioScience 24:457–462. [IUCN] INTERNATIONAL UNION FOR CONSERVATION OF NATURE AND importuna were identified by N.E. Woodley, and the NATURAL RESOURCES. 2004. 2004 IUCN Red List of Threatened Solenopsis molesta were identified by D.R. Smith, both Species. www.redlist.org (20 March 2005) research entomologists with the Systematic Entomology IVERSON, J.B., HIGGINS, H., SIRULNIK, A., AND GRIFFITHS, C. 1997. Laboratory, US Department of Agriculture. Voucher Local and geographic variation in the reproductive biology of specimens of these insects have been deposited in the the snapping turtle (Chelydra serpentina). Herpetologica 53: US National Collection, Smithsonian Institution, and in 96–117. the Mississippi Entomological Museum, Mississippi State IVERSON, J.B. AND SMITH, G.R. 1993. Reproductive ecology of the painted turtle (Chrysemys picta) in the Nebraska Sandhills and University. The constructive comments by 2 anonymous across its range. Copeia 1993:1–21. reviewers greatly improved an earlier draft of this JACKSON, D.R. AND WALKER, R.N. 1997. Reproduction in the manuscript. Suwannee Cooter, Pseudemys concinna suwanniensis. Bulle- tin of the Florida Museum of Natural History 41:69–167. LITERATURE CITED JONES, R.L. AND HARTFIELD, P.D. 1995. Population size and growth in the turtle Graptemys oculifera. Journal of Herpetology 29: ANDERSON, P.K. 1958. The photic responses and water-approach 426–436. behavior of hatchling turtles. Copeia 1958:211–215. KOFRON, C.P. 1991. Aspects of ecology of the threatened ringed BURGER, J. 1977. Determinants of hatching success in diamond- sawback turtle, Graptemys oculifera. Amphibia–Reptilia 12: back terrapin, Malaclemys terrapin. American Midland 161–168. Naturalist 97:444–464. LAHANAS, P.N. 1982. Aspects of the life history of the southern CAGLE, F.R. 1939. A system for marking turtles for future black-knobbed sawback, Graptemys nigrinoda delticola, identification. Copeia 1939:170–173. Folkerts and Mount. Thesis, Auburn University, Auburn, AL. CAGLE, F.R. 1953. The status of the turtle Graptemys oculifera LAMB, T., LYDEARD, C., WALKER, R.B., AND GIBBONS, J.W. 1994. (Baur). Zoologica 38:137–144. Molecular systematics of map turtles (Graptemys): a compar- CHANDLER, C.R. 1995. Practical considerations in the use of ison of mitochondrial restriction site versus sequence data. simultaneous inference for multiple tests. Animal Behavior 49: Systematic Biology 43:543–559. 524–527. LANDERS, J.L., GARNER, J.A., AND MCRAE, A. 1980. Reproduction CHRISTENS,E.AND BIDER, J.R. 1986. Reproductive ecology of the of gopher tortoises (Gopherus polyphemys) in southwestern painted turtle (Chrysemys picta marginata) in southwestern Georgia. Herpetologica 36:353–361. Quebec. Canadian Journal of Zoology 64:914–920. LOWERY, G.H., JR. 1943. Checklist of the mammals of Louisiana CONGDON, J.D., BREITENBACH, G.L., VAN LOBEN SELS, R.C., AND and adjacent waters. Occasional Papers of the Museum of TINKLE, D.W. 1987. Reproduction and nesting ecology of Zoology, Louisiana State University, 13:213–257. snapping turtles (Chelydra serpentina) in southeastern Mich- MADGE,S.AND BURN, H. 1994. Crows and Jays. A Guide to the igan. Herpetologica 43:39–54. Crows, Jays and Magpies of the World. Boston: Houghton CONGDON, J.D., DUNHAM, A.E., AND VAN LOBEN SELS, R.C. 1993. Mifflin Co., 433 pp. JONES — Reproduction and Nesting of Graptemys oculifera in Mississippi 209

MOLL, E.O. AND LEGLER, J.M. 1971. The life history of a neo- RUFFNER, J.A. AND BLAIR, F.E. 1981. The Weather Almanac. Third tropical slider turtle, Pseudemys scripta (Schroepff) in Panama. Edition. Detroit: Gale Research Co., 527 pp. Bulletin of the Los Angeles County Museum of Natural SHEALY, R.M. 1976. The natural history of the Alabama map History 11:1–102. turtle, Graptemys pulchra (Baur), in Alabama. Bulletin of the MONTEVECCHI, W.A. AND BURGER, J. 1975. Aspects of the Florida State Museum of Biological Science 21:47–111. reproductive biology of the northern diamondback terrapin SHELBY, J.A., MENDONCA, M.T., HORNE, B.D., AND SEIGEL, R.A. Malaclemys terrapin terrapin. American Midland Naturalist 2000. Seasonal variation in reproductive steroids of male and 94:166–178. female yellow-blotched map turtles, Graptemys flavimaculata. MOULIS, R.A. 1997. Predation by the imported fire ant (Solenopsis General and Comparitive Endocrinology 119:43–51. invicta) on loggerhead sea turtle (Caretta caretta) nests on STATSOFT,INC. 1999. Statistica for Windows. Tulsa, OK: Statsoft, Wassaw National Wildlife Refuge, Georgia. Chelonian Inc. Conservation and Biology 2:433–436. STEPHENS, P.R. AND WIENS, J.J. 2003. Ecological diversification MOUNT, R.H. 1981. The red imported fire ant, Solenopsis invicta and phylogeny of emydid turtles. Biological Journal of the (Hymenoptera: Formicidae), as a possible serious predator on Linnean Society 79:577–610. some native southeastern vertebrates: direct observations and TEMPLE, S.A. 1987. Predation on turtle nests increases near subjective impressions. Journal of the Alabama Academy of ecological edges. Copeia 1987:250–252. Science 52:71–78. TRIMETRIX,INC. 1999. Axum Technical Graphics and Data PACKARD, G.C., PACKARD, M.J., AND BOARDMAN, T.J. 1987. Analysis. Seattle: Trimetrix, Inc. Influence of moisture, temperature, and substrate on snapping TUCKER, J.K. 2001. Clutch frequency in the red-eared slider turtle eggs and embryos. Ecology 68:983–993. (Trachemys scripta elegans). Journal of Herpetology 35: PACKARD, G.C., PACKARD, M.J., AND GUTZKE, W.H.N. 1985. 664–668. Influence of hydration of the environment on eggs and embryos VOGT, R.C. 1980. Natural history of the map turtles Graptemys of the terrestrial turtle Terrapene ornata. Physiological pseudogeographica and G. ouachitensis in Wisconsin. Tulane Zoology 58:564–575. Studies in Zoology and Botany 22:17–48. PETOKAS, P.J. AND ALEXANDER, M.M. 1980. The nesting of VOGT, R.C. 1981. Turtle egg (Graptemys: Emydidae) infestation Chelydra serpentina in northern New York. Journal of by fly larvae. Copeia 1981:457–459. Herpetology 14:239–244. WILBUR, H.M. AND MORIN, P.J. 1987. Life history evolution in PLUMMER, M.V. 1977. Reproduction and growth in the turtle turtles. In: Gans, C. and Huey, R. B. (Eds.). Biology of the Trionyx muticus. Copeia 1977:440–447. Reptilia. Volume 16, Ecology B, Defense and Life History. PLUNKETT, M.L., MORRIS, F., III, AND OAKLEY, W.T. 1993. Water New York: Alan R. Liss, Inc., pp. 387–440. resources data Mississippi. Water year 1992. Water Data Rept. WOOD, R.C. 1996. The Weather Almanac. Seventh Edition. New MS-92-1. US Geological Survey, 461 pp. York: Gale Research, Inc., 673 pp. RICE, W.R. 1989. Analyzing tables of statistical tests. Evolution Received: 27 September 2003 43:223–225. Revised and Accepted: 24 March 2005