ESTIMATION OF GREEN TURTLE (CHELONIA-MYDAS) GROWTH-RATES FROM LENGTH-FREQUENCY ANALYSIS

Length-frequency analysis (using the computer program MULTIFAN) was us ed to estimate growth rates and number of age classes in a population of immature green turtles (Chelonia mydas) at Great Inagua in the sout hern Bahamas. Results from MULTIFAN were compared with growth rates es timated from nonlinear regression analysis on capture-recapture data a nd with calculated growth rates for 5-cm carapace length increments fo r the same population of green turtles. MULTIFAN underestimated growth rates of the smaller green turtles but yielded a better estimate of a symptotic size than did the nonlinear regression analysis. Estimates g enerated by MULTIFAN and by nonlinear regression analysis for the numb er of years it takes a green turtle at Great Inagua to grow from 30 to 70 cm carapace length fell within the 95% confidence limits of the ca lculated growth data from capture-recapture data. When the number of l ength-frequency samples was reduced sequentially from 10 years to two years, MULTIFAN was successful in estimating the number of age classes in the population for each of the nine sample sets. Length-frequency analysis shows promise as a method to rapidly estimate the number of y ear classes in a population of immature sea turtles and to estimate vo n Bertalanffy growth parameters.