POPULATION BIOLOGY OF CHAMAELIRIUM LUTEUM, A DIOECIOUS LILY. II. MECHANISMS GOVERNING SEX RATIOS.

Abstract

The evolution of sex ratios has long been a focus of scientific interest (Darwin, 1871; Fisher, 1930). Many theoretical studies have shown that a one to one primary (zygotic) sex ratio is to be expected in most cases (Fisher, 1930; Shaw and Mohler, 1953; Bodmer and Edwards, 1960; Kolman, 1960; Leigh, 1970; Charnov, 1975; Eshel, 1975). This expectation is based on the fact that in a dioecious sexually reproducing population, each offspring produced will contain equal gene complements from male and female parents. If there is an excess of one sex in the population, the fitness of individuals of the opposite sex will increase, and mutations resulting in an increase in numbers of that opposite sex will be favored by selection. By this mechanism, primary sex ratios are expected to be kept in an evolutionary balance near one to one. However, observations of secondary (adult) sex ratios in many animal and plant species show an excess of one sex or the other (Davey and Gibson, 1917; Correns, 1928; Godley, 1964, 1976; Hamilton, 1967; Harris, 1968; Zarzycki and Rychlewski, 1972; Lloyd, 1973; Richards, 1975; Barlow and Wiens, 1976; Melampy and Howe, 1977; Opler and Bawa, 1978; Werren and Charnov, 1978). Skewed secondary sex ratios could be the result of genetic mechanisms which bring about unequal numbers of male and female progeny (i.e., skewed primary sex ratios); or, alternatively, they could be the result of post-zygotic life history differences between males and females. The dioecious perennial forest floor herb Chamaelirium luteum (L.) Gray (Liliaceae) has been studied extensively with respect to spatial distributions of males and females (Meagher, 1980) and life history differences between the sexes (Meagher and Antonovics, in press). Plants of this species are constant over time with respect to sex expression; there have been no sex changes observed over the seven years of this study (Meagher, 1980). A particularly striking feature of this species, noted in these earlier studies and in Radford et al. (1968), is that sex ratios in natural populations always show a strong excess of males. However, such observations on the sex ratio are limited in that often only a fraction of the plants present will be in flower in a particular year, so that the sex ratio estimate could be strongly influenced by differences between the sexes in flowering schedules (e.g., Vernet, 197 1). In a perennial species such as C. luteum, the sex ratio may be regarded at three levels: 1) the seedling (primary) sex ratio, 2) the adult (secondary) sex ratio, and 3) the sex ratio among plants in flower in any given year (flowering sex ratio). The study presented here was undertaken to answer the following questions. Do the seedling, flowering or adult sex ratios of C. luteum show a bias towards males or females? How are these different sex ratio estimates interrelated? Finally, what influence do various factors, such as genetic mechanisms or ecological and life history differences between the sexes, have on adult sex ratios?