Consequences of stabilising selection for polygenic variation

Genetic analysis of an avoidance-avoidance response in Mus domesticus

Latency to leave a lighted platform and enter a novel chamber in which other mice had received shock was measured in 2592 mice from eight inbred strains and all 56 F1 crosses. An analysis of the diallel matrix indicated a clear genetic architecture, although genetic effects accounted for only 10% of the total phenotypic variance. Dominance favoring a slight delay in chamber entry suggested a selective advantage in spending a longer time on some elements in the behavioral chain involved in avoidance-avoidance responses. Inbred strains showed greater litter variance than F1 hybrids, suggesting greater developmental buffering of heterozygotes. Both the genetic architecture and the strain rankings differ from those typically found in open field and similar tests of locomotor activity. The results illustrate the problem of interpreting behavior genetic results in terms of proportions of total phenotypic variance and difficulties in generalizing to ancestral or other populations in an attempt to interpret genetic results in an evolutionary context when reliability is low.

Stabilizing selection and metabolism

This paper attempts to provide a biochemical explanation for selection against extreme phenotypes. From current ideas on the analysis of metabolic control, it is argued that, on average, most mutations will reduce flux through metabolic pathways and change substrate pool concentrations in an arbitrary direction. Currently available experimental data suggests that there may be a positive relationship between flux and fitness. It is also argued that there may be a relationship between phenotypic traits and substrate pool concentrations. Given these premises, it is shown that intrinsic selective constraints occur against the production of extreme phenotypic variants.