Conflicting conclusions from long-term versus short-term studies on growth and reproduction of a tropical snake

A ten-year (1989-1998) mark-recapture study on long-lived aquatic filesnakes (Acrochordus arafurae) challenges conclusions from two earlier short-term (1982-1983 and 1985-1988) studies on the same population. Because of methodological shortcomings and substantial annual variation in most demographic traits in this population, the earlier work (1) underestimated mean growth rates (and thus, overestimated age at maturation) and (2) underestimated the mean frequency of reproduction by adult females. Based on our more extensive data, we estimate that most filesnakes mature at about three years of age (versus 5-7 years in earlier studies), and that adult females reproduce (on average) once every 3-4 years (versus 8-10 years in earlier studies). These data support two general conclusions. The first is that long-term studies are necessary before we can accurately characterize the mean values for demographic traits of long-lived organisms in variable environments. The second conclusion is that the mean value of a trait may be less important than its variance. Even when we have enough data to quantify "mean" values for traits such as age at maturation or reproductive frequency, few individual females in the population may display those values. Instead, a female's life-history depends upon the conditions that she encounters during her lifetime. Biological insight into the life-histories of organisms in such systems will come from consideration of the variance, rather than the mean.