Moreover, the juvenile stages of the leatherback, from post- 

 hatchling to a carapace length of over 100 cm, lead an entirely 

 cryptic existence, and virtually never come before human eyes, to 

 be recorded or counted. The adult males, too, are almost as 

 difficult to count; they may be seen on pelagic surveys or as 

 strandings, but formulae are unavailable to convert numbers seen 

 to numbers in existence. 



Thus, the only feasible population estimates will refer to 

 the numbers of nesting females, which may be counted, or at least 

 their numbers estimated, by means of beach patrols or (less 

 accurately) aerial surveys. Different formulae may be utilized 

 to estimate the number of females in a local population from the 

 number seen nesting on an average night. An average inter- 

 nesting interval for individual reproductive females has been 

 established at around ten days for leatherbacks in widely 

 separated populations, but uncertainty enters when one attempts 

 to establish the average number of nestings per female per 

 season. The maximum number is around ten (A. Tucker, pers. 

 comm. , Culebra Island; or from Pritchard's work in Surinam in the 

 1960s) , but the average number is surely considerably less, and 

 some females may nest only once or twice in a season. Pritchard 

 (1982) utilized a formula that assumed that the total number of 

 breeding females in a population was about fifty times the number 

 nesting on a typical, mid-season night. This formula 

 incorporated the assumption that individual turtles had a nesting 

 season that was, on average, half the length of that of the 

 population as a whole; and that remigration occurred after an 

 interval averaging 2.5 years. Mrosovsky (1983a) utilized a 

 different formula, still unpublished, but which was based on 

 studies in Surinam in which a complete record was available for 

 at least one season of the number of nests made each night. 

 Since the turtles were always tagged, a relationship between the 

 total nesting population for the season and the average number 

 nesting per night could be established. Tag loss or shift of 

 turtles to other beaches in the course of a season were 

 complicating factors, but Pritchard's and Mrosovsky 's formulae 

 still gave rather similar results. 



Fitter (1961) estimated that the world population of adult 

 female leatherbacks may be "as few as 1,000," of which 85 percent 

 nested in Trengganu, Malaysia — the only large leatherback rookery 

 known at the time. However, Pritchard (1971) made a much higher 

 estimate of 29,000 to 40,000 breeding females, made up as 

 follows: 15,000 for French Guiana; 8,000 for Pacific Mexico; 

 4,000 for Trengganu; 1,000 for Matina, Costa Rica; and 200-400 

 each for Trinidad, Surinam, Tongaland (South Africa) and Sri 

 Lanka (and adjacent parts of south India) . The higher-range 

 estimate simply assumed that further nesting grounds would be 

 discovered elsewhere. This estimate, it may be noted, gives a 

 population estimate of 16,400-16,800 for the western Atlantic 

 region. 



148 



