There is increasing evidence that ecological variables involving stress are important in determining evolutionary rates. This paper incorporates recombination into this scenario.
In Drosophila melanogaster, recombination increases at developmental temperatures above and below normal culture temperatures, giving a U-shaped curve which is most pronounced in centromeric regions; however, at near lethal temperature extremes there is some evidence for a fall in recombination. More limited data from other organisms are generally consistent with this conclusion. Nutritional stress in the form of starvation increases recombination in D. melanogaster, and behavioural stress has been found to increase recombination in male mice.
In natural populations recombination is under complex genetic control analogous to other quantitative traits. In D. melanogaster in a novel environment, there is evidence that additive genetic variability for recombination is higher than in a standard laboratory environment. During selection in populations exposed to extreme stress increased recombination may occur; this implies that in marginal (stressful) habitats, variability generated by recombination may increase.
In D. melanogaster, structural heterozygosity due to inversions in one part of the genome tends to increase recombination in the remainder of the genome in a qualitatively similar manner to, and cumulative with, direct environmental effects especially temperature. Substantial recombination should be inducible under combinations of karyotypes and environments deviating from existing circumstances, especially if the suggestion that effects are often synergistic due to a dependence upon available energy levels can be confirmed.