INTRODUCTION

It is generally assumed (e.g. Maynard Smith, 1978; 1982) that ancestrally, gametes were small and isogamous (monomorphic). The evolution of anisogamy (gamete dimorphism) is a crucial transition in evolution (Maynard Smith and Szathmáry, 1995): it represents the evolution of the two sexes, males and females. Following Parker et al. (1972), I favor defining a sex in relation to the type of gamete a sexual phenotype carries. A sex is thus an adult phenotype defined in terms of the size of (haploid) gamete it produces: in an anisogamous population, males produce microgametes and females produce macrogametes. A simultaneous hermaphrodite is thus both male and female simultaneously, and a sequential hermaphrodite transforms sequentially from male to female (or vice versa). This definition of a sex differs from one that defines a sex in terms of gamete mating types (e.g. Wiese, 1981; Hoekstra, 1990). Under the Parker et al. definition of a sex in terms of gamete size, a mating type is not considered to be a sex, but simply a gametic type (that may or may not be related to gamete size) that shows a preference for fusion with certain other gamete types. In isogamous populations, there is thus one sex (though there may be several mating types). Retaining the definition of a sex for an adult phenotype that produces a given gamete size, and a mating type for a gamete phenotype that has a given characteristic for selective fusion may serve to remove some of the confusions that have arisen in the literature.