Methods

Two successive midclutch eggs were collected from 5 individual green turtles (Chelonia mydas) nesting at Heron Island (23°26′S, 151°55′E), eastern Australia. Eggs (n = 10) were weighed and measured and then incubated at ambient laboratory temperature (Heron Island Research Station) as matched pairs, with each pair in a separate transparent plastic container (16 × 10 × 7 cm) covered in plastic film. They were set on a 2-cm substrate of autoclaved sand collected at a depth of 55 cm (average nest depth) from the nesting beach. A subsurface trickle irrigation of sterile water (see Phillott 2002) maintained sand moisture to a standard, predetermined by the “pinch method” (Blanck and Sawyer 1981).

Eggs were observed until white spot development covered approximately half of the surface, indicating embryonic viability (7–10 days). Eggs were then removed, weighed, and replaced, one in its original orientation, the other inverted. Inversion of green (Parmenter 1980) and loggerhead (Limpus et al. 1979) sea turtle eggs at this time during incubation is known to cause embryo mortality. Control (noninverted) eggs were incubated until hatching; experimental (inverted) eggs were opened at the same time to determine embryonic stage at mortality. During incubation, visual observations of white spot development and other changes in eggshell color were recorded.

Results

Egg diameter and weight at oviposition (Table 1) were within the range previously recorded for green sea turtles at Heron Island (Limpus et al. 1984). During the 7–10 days prior to manipulation, pairs of eggs showed similar signs of development, i.e., progression of white spot and weight change.

Table 1. Dimensions and weight of control and experimental eggs.

View Fullscreen

White spot development ceased immediately after inversion in 4 of the 5 experimental eggs. In the exception, the white spot expanded slightly but at a slower rate than its unrotated control for a further 84 hours. Observations of this egg were subsequently discarded as the exact time of embryonic death was uncertain (it later failed to hatch). Sixteen hours after rotation, the remaining inverted eggs displayed a pale yellowing tinge to the eggshell outside the white spot. This color increased in intensity as time progressed, spreading from the top of the egg downwards. The entire egg displayed a distinct yellowing after 20–24 hours. The chalkiness of the white spot faded significantly after 44 hours. Shell pliability increased slightly. None of the control eggs displayed any alteration in eggshell appearance other than normal expansion of the white spot, and there was no change in shell pliability.

All of the control eggs produced hatchlings, while none of the inverted eggs hatched. When opened, unhatched eggs revealed development consistent with embryonic Stage 18–19 (after Miller 1985), suggesting death synchronous with egg rotation.

Discussion

In the green sea turtle, egg inversion and subsequent embryonic death results in alterations to egg appearance. As eggshell structure is similar in green (Solomon and Baird 1976; Baird and Solomon 1979), leatherback (Dermochelys coriacea; Solomon and Watt 1985; Chan and Solomon 1989), olive ridley (Lepidochelys olivacea; Sahoo et al. 1996a, 1996b), Kemp's ridley (L. kempii; Packard et al. 1982), flatback (Natator depressus; Phillott 2002), and hawksbill (Eretmochelys imbricata; Phillott 2002) sea turtle eggs, postmortem changes in appearance are also likely to be similar. Although Carthy (1992) described pore-like structures on the inner surface of loggerhead (Caretta caretta) turtle shell membrane, these were not detected by Packard et al. (1982), Schleich and Kästle (1988), or Phillott (2002), hence loggerhead eggshell is presumed to display similar deterioration to that described here.

Blanck and Sawyer (1981) described 2 temporary extra-embryonic membranes during the first 2.5 weeks of development in loggerhead turtle eggs at 28°C (approximately the first trimester of the 57-day incubation period). The posterior amniotic tube and the “attachment membrane” (an extension of the amnion that fuses with the chorionic membrane adjacent to the eye region) aid in positioning and maintaining the embryo at the top of the egg. The physical support role of the extra-embryonic membranes is superseded after this time by the thickening of the yolk stalk and increase of the chorion–shell membrane adherence area.

Prior to this time, egg inversion results in the disruption of egg contents and tearing of extra-embryonic membranes and blood vessels (Blanck and Sawyer 1981). Miller (1985) attributed movement-induced early embryo mortality to rupturing of the vitelline membrane. The embryo remains attached to the shell, but is sheared from the embryonic disc (Ferguson 1985). Yolk and fluids within the vitelline sac and subgerminal space are then able to mix with the albumen (Ewert 1979). Postmortem discoloration of eggs in the experiment would likely result from similar mixing. Yellowing was first observed at the inverted northern or nonwhite pole, spreading downwards and finally encompassing the entire egg. This would be due to rearrangement of the egg contents after egg inversion. As the yolk moves through the albumen and rearranges itself to its preferred orientation, maximal leakage of the vitelline sac contents would occur at the new top surface of the egg, hence the initial site of yellowing. As the chalkiness of the white spot relies on shell dehydration (and subsequent opacity) maintained by embryonic metabolic activity, it fades after embryonic death.

The loss of turgor by nonviable eggs occurs as they lose their water holding capacity (Ewert 1979), potentially acting as water donors, with a similar role to that presumed for yolkless leatherback eggs (Hall 1990). Viable eggs maintain a water potential of −900 kPa (Ackerman 1991). Death would result in the loss of metabolic sustenance and a less negative water potential, closer to the −5 to −50 kPa of the substrate (Ackerman 1991), allowing moisture transfer from the egg to the nest environment. Alternatively, water bridges may form between nonviable and adjacent viable eggs for direct exchange.

Unlike nonviable eggs in natural nests examined at full term, the experimentally killed eggs did not exhibit severe flaking of the shell. The slight increase in pliability that occurred was probably due to turgor loss, though it may have been indicative of postmortem degradation of the shell structural integrity. Sea turtle eggshell consists of variable sized aragonite crystals that are not organized into individual shell units (i.e., with intervening discrete pores) but with numerous open spaces allowing gas and water exchange (Solomon and Baird 1976; Baird and Solomon 1979; Packard et al. 1982; Schleich and Kästle 1988; Solomon and Watt 1985; Chan and Solomon 1989; Sahoo et al. 1996a, 1996b). When the underlying shell membrane decays postmortem, it dissociates from the calcareous crystalline eggshell (Hirsch 1983), probably leading to the subsequent degradation of the eggshell. It is possible that the relatively low levels of soil microbiota (when compared to those of a natural nest) were insufficient for complete flaking of the eggshell to be observed in this experiment. Flaking (exfoliation) of viable eggs in the week prior to hatching (Miller 1985) is a result of calcium mobilization from the eggshell by the rapidly maturing embryo (Simkiss 1962).

This experiment caused 2 major alterations to the interior of the egg: yolk displacement and embryonic death. It is unclear whether changes in eggshell appearance were due to these occurrences individually, or in combination. However, other laboratory experiments (to be reported separately) demonstrate that eggs dying without human intervention present similar signs of faded chalkiness, increased yellowing, and increased shell pliability. As yellowing is likely due to mixing of the yolk (and/or subgerminal fluids) with the albumen in combination with embryo autolysis, it would normally occur (in a noninverted egg) with breakdown of the vitelline sac by protein degradation or microbial action and be observed as a gradual stain of the entire egg rather than initiating at either pole. Fading chalkiness (due to rehydration of the shell interstices as metabolic activity ceases) is likely to be a direct result of embryonic death rather than inversion, and hence be the better indicator of egg mortality.

The time lapse between embryonic death and the occurrence of physical alterations in eggshell appearance has relevance to the interpretation of manipulative results derived from the artificial incubation of sea turtle eggs. Unless the white spot has faded substantially within 2 days of a particular incubation event or mishap (e.g., temperature fluctuation or increased microbial load), egg death should not be ascribed to that event with any certainty. Eggs that demonstrate a loss of chalkiness and/or develop a yellow discoloration should be considered nonviable and removed from the turtle nest or incubation container to prevent their becoming a source of microbial infection.