THE ROLE OF PREDATION IN THE EVOLUTION OF SOCIAL BEHAVIOR OF NATURAL POPULATIONS OF THE GUPPY, POECILIA RETICULATA (PISCES: POECILIIDAE).

Abstract

The initial task facing an ethologist is to catalog and to describe the repertoire of behaviors exhibited by a particular organism or group of organisms. As this is not usually done in situ, questions often arise as to the relevance of the particular sets of conditions treated in the laboratory. Often an animal behaves differently in the laboratory than in the field, either omitting some aspects of its behavior entirely or modifying others qualitatively or quantitatively; hence, the functional significance of the behavior cannot be completely ascertained. The ethologist must investigate how the behavior studied under controlled laboratory conditions relates to the biotic and abiotic factors comprising the natural environment of the organism. In doing so, one can gain insight into why particular behavior patterns have evolved and how they are modified by different environmental parameters. Predation has been claimed to be one environmental parameter which is a major influence on the biology of the guppy, Poecilia reticulata. Haskins et al. (1961) reported that certain large predators reduce guppy sex ratio by selectively choosing the brightly colored males over the duller females. Also, the mode of inheritance of male secondary sexual coloration (i.e., Xlinked, Y-linked, or autosomal) varies among populations as a response to predation. Seghers (1973) found that sex ratio was reduced not by the large predators as reported by Haskins et al. (1961), but by a small cyprinodontid fish, Rivulus hartii. His evidence indicated that Rivulus selects males because they are less adept behaviorally than females at avoiding predation. In conjunction with this, he found that males coexisting with Rivulus tend to be larger than in the absence of Rivulus. Guppies sympatric with large predators tend to live in cohesive aggregates, while this was not so for those sympatric with Rivulus. Neither of the above studies investigated the effects of predation on social behavior of the guppy. Farr and Herrnkind (1974) in a study of sexual and agonistic behavior of laboratory populations of guppies found that while males display to females at an extremely high rate (seven times per five minutes on the average), the females rarely respond to these displays. No female responses or copulations were observed during that study. One might expect this high rate of sexual display to be disadvantageous under conditions of heavy predation simply because males would be more vulnerable and possibly more visible to a predator when displaying. One would predict, then, that under natural conditions where predation is known to affect other aspects of their biology, display rate would be lower than in populations where predation by larger fish is minimal or nonexistent. This study is a test of the hypothesis that predation is a strong selective force influencing social behavior patterns of the guppy in its natural environment. It will be demonstrated that guppies exhibit patterns of social behavior particular to different predation regimes and ultimately a scheme accounting for the evolution of these patterns will be developed.