HERITABILITY AND SELECTION ON CLUTCH SIZE IN DARWIN'S MEDIUM GROUND FINCHES (GEOSPIZA FORTIS )

Abstract

Abstract.-I studied the causes ofvariation and selection on clutch size in a population ofDarwin's Medium Ground Finches (Geospizafortis) on Isla Daphne Major, using data collected over a nineyear period (1976)(1977)(1978)(1979)(1980)(1981)(1982)(1983)(1984). Quantitative-genetic analyses were carried out using the first clutch laid by a female in a given year. I used both unadjusted clutch-size values and values adjusted for between-year differences in mean clutch size for repeatability and regression analyses. Repeatability ofclutch size was small (:oS 8%) and nonsignificant in all cases. Sib-sib analyses and parent-offspring regressions gave no evidence of a significant additive genetic component to clutch-size variation. Slopes of mother-daughter regressions were actually negative, suggesting possible maternal effects of mother's clutch size on daughter's clutch size. There was a small positive relationship between female age and clutch size but no effect of male or female body size or oflarge-scale differences in habitat quality on clutch size. Selection on clutch size was generally directional and positive: in almost all years in which successful breeding occurred, large clutches tended to fledge more chicks and produce more young surviving to the following year, possibly because there was no trade-off between clutch size and the weights of individual chicks at fledging. Thus, sustained directional selection for large clutch size may have reduced additive genetic variation in clutch size to low levels in this population. The size ofa female's clutch may be primarily determined by unidentified proximate environmental factors which vary from year to year, rather than by any long-term optimization of clutch size with respect to adult survival.

Genetic Variation for Clutch Size in Natural Populations of Birds From a Reaction Norm Perspective

“…Apart from the technical aspect of violating the assumption of reaction norm linearity more seriously when the among-year component of V P is greater, there is also the biological aspect that there may exist stronger selection for plasticity in fluctuating and unpredictable environments. Across the (limited number of) studies in Table 1, there is, however, no evidence for a negative correlation between heritability and the amount of between-year phenotypic variance (including Gibbs (1988), Pearson r ϭ Ϫ0.60, P ϭ 0.16; excluding Gibbs (1988), Pearson r ϭ 0.34, P ϭ 0.50).…”

“…Of the 11 studies that had sufficient statistical power, 10 found significant heritabilities for clutch size, ranging from 0.21 to 0.48. Only Gibbs (1988) found very small, nonsignificant repeatabilities and heritabilities for first clutch sizes in Darwin's Medium Ground Finches (Geospiza fortis). Except for Van Noordwijk et al (1981), Lessells et al (1989), and the more recent studies mentioned, all of these studies use fewer than 10 years of data.…”

“…Clutch size has been shown to be under directional selection to increase in several populations of birds, based on persistently positive correlations between components of fitness and clutch size, or differences between the most productive clutch size and the average clutch size (references in Price and Liou [1989], Gibbs [1988], and Larsson et al [1998]; but see Van Noordwijk et al [1981] and Sheldon et al [2003]). Taking this and the fact that there is general agreement that clutch size has a heritable component into account, clutch size is generally expected to increase with time.…”

“…Column heads are defined as follows: Years, the years that were used for the heritability estimate; V within and V among , the amount of phenotypic variance within and among years, respectively; N, the number of data points in the mother-daughter regression; Mean, the mean clutch size across all years on which the heritability estimate was based. References are: 1, Lessells et al (1989); 2, Sheldon et al (2003);3, McCleery et al (2004);4, Van Noordwijk et al (1981); 5, Gibbs (1988);6, Flux and Flux (1982 formation to construct such mapping functions for clutch size, as well as many other traits.…”

“…Since then, a total of 17 publications have reported heritability estimates for clutch size estimated in natural populations (see the Appendix). Some, however, contain heritability estimates for more than one population, different subsets of the same population, or make use of different methods (e.g., Van Noordwijk et al 1981, Gibbs 1988.…”