Effects of nutrient availability on primary sexual traits and their response to selection in Spergularia marina (Caryophyllaceae)

• Published in: Journal of evolutionary biology Vol. 16; no. 4; pp. 767 - 778
• Main Authors: Mazer, S. J., Lowry, D. E., Hansen, T.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Science Ltd 01.07.2003
• Subjects: artificial selection; Caryophyllaceae - physiology; Flowers; Genetic Drift; genetic variation; Genetics, Population; G × E; Models, Theoretical; Nutritional Status; Phenotype; phenotypic plasticity response to selection; Reproduction; Selection, Genetic; sex allocation; Spergularia marina
• ISSN: 1010-061X; 1420-9101
• DOI: 10.1046/j.1420-9101.2003.00555.x

Theoretical models of the evolution of sex allocation generally assume a negative genetic correlation between components of male and female investment or function. To test this assumption, and to determine whether the expression of the correlation is sensitive to environmental conditions, we conducted an artificial selection experiment targeting primary sexual traits in the autogamous herb, Spergularia marina. Mass selection favouring individuals with high (or low) pollen and ovule production per flower had previously been conducted for two generations under uniform conditions. Following a third episode of selection (reported here) within the high‐pollen‐, high‐ovule‐, low‐pollen‐ and low‐ovule‐producing lineages, selected maternal seed families were replicated and cultivated in three nutrient treatments. With this design, we observed the effects of nutrient availability on: floral phenotype, the realized response to selection and the expression of genetic variation in, and covariation between, male and female investment. The first two episodes of selection Mazer et al. 1999 (Evolution 53:717–731) detected evidence for a genetically based negative correlation between ovule and anther production. Following the third episode of selection, phenotypic differences among the selected lines were maintained in all treatments, but evidence for a negative correlation between male and female investment nearly disappeared. Only under low‐nutrient conditions, in which plants selected for low ovule production exhibited elevated anther production, was a genetic trade‐off expressed. Either genetic drift or the combination of novel growing conditions (relative to previous generations) and G × E interactions may have caused the change in the expression of the genetic correlation. This experiment also allowed us to test the prediction that autogamous selfers should exhibit higher canalization of the ratio of male to female investment than of its components. Supporting this prediction, the A : O ratio was more strongly buffered against environmental variation than either anther or ovule production per flower.