ENVIRONMENTAL PREDICTABILITY AND CONTROL OF GONADAL CYCLES IN BIRDS

In most vertebrates, information from the external environment is used to regulate secretion of gonadotropins and thus time gonadal maturation in anticipation of the breeding season. These environmental factors provide "predictive" information in both the long term (initial predictive) and short term (supplementary), are integrated by the central nervous system, and are then transduced into neuroendocrine and endocrine control components. We apply the constancy/contingency model of predictability to breeding data from several avian species to derive a potential unifying hypothesis to explain how these environmental factors are integrated to regulate gonadal maturation. Predictability (Pr) is a function of varying degrees of constancy (C, the habitat is predictable because it is always the same) or contingency (M, the habitat is predictable in the degree of change from season to season). Thus, although variations in Pr provide useful information for timing gonadal maturation and regression, changes in the contributions of C and M to Pr (i.e., the ratio, M/C) provide even more useful information. It is proposed that the degree to which individuals integrate initial predictive and supplementary information to regulate secretion of gonadotropins, and thus time gonadal maturation and onset of breeding, is a function of the ratio of M to C for any given Pr value. When Pr is not significant (i.e., the environment is unpredictable), then only supplementary factors may be useful for timing breeding. Hypotheses generated by this approach are easily testable. Furthermore, this method could be applied to all vertebrates, and perhaps other organisms, and may provide a unifying framework to investigate the mechanisms by which proximate environmental factors regulate reproductive function in general.