. Metabolic heating caused by physiological processes during the development of oviparous embryos can raise nest temperatures above those of the surrounding substrate and may be sufficient to increase embryonic growth rates, influence sex ratios of hatchlings with temperature-dependent sex determination, and increase hatching success in seasonal environments. In sea turtles with large clutch sizes, metabolic heating can raise nest temperatures by as much as 6°C. However, no studies have directly investigated metabolic heating in any species of freshwater turtle. We investigated whether metabolic heating occurs in nests of snapping turtles ( Chelydra serpentina ) from southeastern Michigan, United States. A temperature logger was placed in the center of 8 unaltered snapping turtle nests. A second temperature logger was placed at the same depth in the surrounding substrate 5 cm from the side of the nest chamber. Metabolic heating is more pronounced in nests with larger clutches, so we artificially increased the size of 2 additional nests using donor clutches of 11 and 21 eggs, respectively. Temperatures were recorded at 2-hr intervals until after the presumptive hatch date of all nests. We found that there was a significant increase both in mean nest temperature and accumulated heat units for natural and experimental treatment nests during the last third of incubation. Further, in nests with experimentally increased clutch sizes, mean nest temperature was significantly greater than substrate temperature throughout incubation, suggesting that large nests also exhibit a thermal inertia that results in positive heat balance throughout development, at least in the soils studied.
Hatchling river turtles (Smooth Softshell Turtles, Apalone mutica ; Spiny Softshell Turtles, Apalone spinifera ; Northern Map Turtles, Graptemys geographica ; False Map Turtles, Graptemys pseudogeographica ; and Ouachita Map Turtles, Graptemys ouachitensis ) were released in a variety of settings to help place orientation and dispersal from nests in the context of nest site selection by females and juvenile recruitment habitat. Visual cues associated with near open or far dark horizons were the primary environmental cues used during initial orientation and dispersal of the hatchling river turtles. On a river beach, hatchlings of both species of Softshell Turtles dispersed toward the open horizon of the nearby river. In contrast, hatchlings of all 3 species of Map Turtles on the same beach dispersed toward the near dark horizons of a forest that led them away from the river. Hatchling Map Turtles of all 3 species released in autumn and Northern Map Turtles also released in spring at a field arena dispersed toward near dark horizons of pine ( Pinus spp.) and deciduous trees to the north and south of the arena (directions that were parallel to the Mississippi River). At a site in upland prairie habitat with no nearby wetlands, hatchlings of all 3 species of Map Turtles dispersed to the north toward near dark horizons of an oak ( Quercus spp.) forest rather than toward more distant dark horizons of mixed oak and pine trees. At a lowland prairie site where no near dark horizon was visible, Northern Map Turtles dispersed toward 2 far dark horizons that were ∼ 230 m to the north (a pond surrounded with trees) and to the south (an area of large deciduous trees at the west end of a windrow of pine trees), but not toward the large riparian wetland 280 m to the west. The bimodal dispersal pattern toward 2 equidistant dark horizons within 230 m but not toward the riparian area 280 m to the west suggests that the perception distance for hatchling Northern Map Turtles is between 230 and 280 m. Dispersal of a combined sample of naïve hatchling False Map and Ouachita Map Turtles released in a mature corn field was not different from random, but the directions taken by the majority of hatchlings were to the north and south across corn rows that may be the closest match to dark horizons used for dispersal in typical habitats.
We examined the orientation of 76 naïve painted turtles ( Chrysemys picta belli ) and 746 snapping turtles ( Chelydra serpentina ) during initial dispersal from experimental nests in the Weaver Dunes area of southeastern Minnesota. We conducted 15 releases into large circular arenas in 4 natural nesting areas and 2 atypical areas. Hatchling orientation and dispersal for both species were 1) all nonrandom, 2) appeared to be based on vision (i.e., nonpolarized light), and 3) toward nearby, open, and highly illuminated horizons, regardless of whether or not they were associated with the wetlands. A first-order estimate of hatchling snapping turtle perception distance was 55–90 m. We found no evidence that suggests that specular light from the wetlands, olfaction, or humidity gradients were important in orientation. At 2 of 3 locations, substantial changes in orientation direction occurred when hatchling snapping turtles were released in morning vs. late afternoon. Changes in dispersal directions in the morning and afternoon indicated that hatchlings were not orienting toward the sun per se but toward different highly illuminated nearby prairie areas. At one site, hatchling orientation in the afternoon (but not in the morning) was toward a nearby wetland and was consistent with either dispersal toward highly illuminated near horizon or with the perception and use of reflected polarized light from the wetland. Collectively, the results from our study also indicate that 1) hatchlings disperse toward open horizons rather than toward wetlands themselves (i.e., open areas that are not necessarily associated with wetlands), 2) dispersal direction is influenced by time of day, apparently because of changes in the degree of illumination of different horizons, and 3) far horizons apparently were not used because they were beyond the perception distance of hatchlings. The most parsimonious evolutionary explanation of solutions to orientation problems is that, for each species, both adults and hatchlings have similar perception distances and use the same sensory modes and types of environmental cues during terrestrial movements. Comprehensive conservation and management plans for aquatic turtles should include consideration of how habitat changes in nesting areas might alter the environmental cues that determine the initial orientation and successful dispersal of hatchlings. We compared the results from this study with the dispersal patterns of naïve hatchling Blanding's turtles ( Emydoidea blandingii ) that emerge from nests located much farther from wetlands.