associated with higher embryonic mortality. The greater 

 embryonic mortality and lower overall hatch rate (35.8%) of 

 leatherbacks is partly attributable to their tendency to nest 

 closer to the water than green turtles and so leave their eggs 

 more liable to wetting by exceptionally high seas; 40% of 

 leatherback nests laid above the high tide mark are nevertheless 

 washed over. The idea that greater infertility is an important 

 cause for lower success of leatherback nests is refuted. Once 

 again, the obvious predators such as the numerous ghost crabs are 

 not the main agent of destruction. The apparently imperfect nest 

 site selection by the turtles themselves results in a hatching 

 failure of more than twice the number of eggs that predators 

 take. 



The Whitmore and Dutton study cited above contains a wealth 

 of other information, including data on incubation in styrofoam 

 boxes (greater embryonic mortality than natural nests laid above 

 the high tide lines) and on nests reburied higher up on the beach 

 (lower predation, cf. Stancyk et al. 1980). SomQ potential 

 problems are that determining fate of eggs by counts of emerged 

 hatchlings do not always match up with those obtained by 

 examining the broken eggshells, and whether or not sampling of 

 nests was random is unclear. However, the more serious 

 limitations of such studies are that they stop at hatching, and 

 that they cannot be generalized to other beaches, as comparison 

 of the green turtle data to those in the nest example will 

 illustrate (Tables 1 and 2) . 



Example 3: Green turtles at Tortuquero. Costa Rica 



A useful study by Fowler (1979) shows the fate of 350 green 

 turtle nests at Tortuguero, Costa Rica (Table 2) . Fowler also 

 gives the hatching rate for the undisturbed successful nests, 

 83.1%. Since only 42.6% of the nests were in this category, with 

 another 4.9% of the nests giving some lesser hatch rates, a 

 hatching percent for the total sample of all nests, whatever 

 their fate, is <40%. This means that an artificial hatchery 

 removing predation by humans and animals, and loss to tides, only 

 has to achieve a 4 0% hatch rate to be more successful in terms of 

 hatchling production than leaving the eggs where they were laid. 

 Of course other considerations, such as sex ratios in artificial 

 hatcheries (Dutton et al. 1985), also have to be taken into 

 account. 



A potential problem with this study is that the sample may 

 not have been representative. In addition to the 350 nests whose 

 fate was followed, a further 86 nest sites originally marked were 

 excluded because of failure to locate the eggs. Some of these 

 may have been false crawls but eggs may have been missed in other 

 cases. 



253 



