RAPID POPULATION DIFFERENTIATION IN A MOSAIC ENVIRONMENT. II. MORPHOLOGICAL VARIATION IN ANTHOXANTHUM ODORATUM.

Abstract

Population differentiation is most likely to occur where environmental factors vary greatly in space, but are relatively stable in time (Levins, 1962, 1963; Maynard Smith, 1966). Populations may differentiate rapidly in time (Dobzhansky, 1947; Kettlewell, 1961; Charles, 1964), and in space (Creed et al., 1959; Briggs, 1962; Aston and Bradshaw, 1966; McNeilly, 1968; Antonovics and Bradshaw, 1970), if selection differentials are sufficiently large. The Park Grass Experiment at Rothamsted, U.K., is an ideal field location to study rapid population differentiation, in response to well defined, small-scale environmental variation, in an area with a known history. Various fertilizer treatments have been imposed annually since 1856, and liming treatments have been imposed at four year intervals since 1903; apart from these treatments, the plots have been managed identically. The various fertiliser and liming treatments on the Park Grass Experiment have produced a mosaic of contrasting environments with sharply defined borders. The plots vary in size, but the average size is approximately 17 m X 36 m. The large spatial variation in environmental factors, together with relatively stable conditions in time for 50-100 years, are similar to the conditions specified by Levins (1962, 1963) and Maynard Smith (1966) as most likely to induce population differentiation. A previous study (Snaydon, 1970) has shown that populations of the outcrossing grass species Anthoxanthum odoratum L, collected from the Park Grass plots, differed in response to contrasting soils, and