F. Helen Rodd

Natural populations of guppies were subjected to an episode of directional selection that mimicked natural processes. The resulting rate of evolution of age and size at maturity was similar to rates typically obtained for traits subjected to artificial selection in laboratory settings and up to seven orders of magnitude greater than rates inferred from the paleontological record. Male traits evolved more rapidly than female traits largely because males had more genetic variation upon which natural selection could act. These results are considered in light of the ongoing debate about the importance of natural selection versus other processes in the paleontological record of evolution.

In most animals, the origins of mating preferences are not clear. The "sensory-bias" hypothesis proposes that biases in female sensory or neural systems are important in triggering sexual selection and in determining which male traits will become elaborated into sexual ornaments. Subsequently, other mechanisms can evolve for discriminating between high- and low-quality mates. Female guppies (Poecilia reticulata) generally show a preference for males with larger, more chromatic orange spots. It has been proposed that this preference originated because it enabled females to obtain high-quality mates. We present evidence for an alternative hypothesis, that the origin of the preference is a pleiotropic effect of a sensory bias for the colour orange, which might have arisen in the context of food detection. In field and laboratory experiments, adult guppies of both sexes were more responsive to orange-coloured objects than to objects of other colours, even outside a mating context. Across populations, variation in attraction to orange objects explained 94% of the inter-population variation in female mate preference for orange coloration on males. This is one of the first studies to show both an association between a potential trigger of a mate-choice preference and a sexually selected trait, and also that an innate attraction to a coloured inanimate object explains almost all of the observed variation in female mate choice. These results support the "sensory-bias" hypothesis for the evolution of mating preferences.

We have previously reported a correlation between the life-history patterns of guppies and the types of predators with which they coexist. Guppies from localities with an abundance of large predators (high predation localities) mature at an earlier age and devote more resources to reproduction than those found in localities with only a single, small species of predator (low predation localities). We also found that when guppies were introduced from a high to low predation locality, the guppy life history evolved to resemble what was normally found in this low predation locality. The presumed mechanism of natural selection is differences among localities in age/size-specific mortality (the age/size-specific mortality hypothesis); in high predation localities we assumed that guppies experienced high adult mortality rates while in the low predation localities we assumed that guppies experienced high juvenile mortality rates. These assumptions were based on stomach content analyses of wild-caught predators and on laboratory experiments. Here, we evaluate these assumptions by directly estimating the mortality rates of guppies in natural populations. We found that guppies from high predation localities experience significantly higher mortality rates than their counterparts from low predation localities, but that these higher mortality rates are uniformly distributed across all size classes, rather than being concentrated in the larger size classes. This result appears to contradict the predictions of the age/size-specific predation hypothesis. However, we argue, using additional data on growth rates and the probabilities of survival to maturity in each type of locality, that the age-specific mortality hypothesis remains plausible. This is because the probability of survival to first reproduction is very similar in each type of locality, but the guppies from high predation localities have a much lower probability of survival per unit time after maturity. We also argue for the plausibility of two other mechanisms of natural selection. These results thus reveal mortality patterns that provide a potential cause of natural selection, but expand, rather than narrow, the number of possible mechanisms responsible for life-history evolution in guppies.