ABSTRACT
The milkweed-oleander aphid, Aphis nerii (Boyer de Fonscolombe) (Homoptera: Aphididae), feeds on different milkweed species in northern California than in Puerto Rico. The hosts vary, primarily between regions, for both identity and quantity of cardenolides that the aphid sequesters for its own defense. In tests for hostassociated fitness trade-offs only one case was found in which host plant and fitness corresponded, but the effect was not significant. However, power to detect fitness trade-offs was limited and the possibility of considerable differences in fitness on a particular host for aphids from different hosts cannot be excluded. On Californian host species, among which migration is common, generalized host use could result from selection for general-purpose genotypes. However, this explanation cannot apply to generalized host use of Californian and Puerto Rican milkweeds because the regions are isolated by distance. A cardenolide sequestration mechanism that is free of substantial energy costs could provide the basis for fitness homeostasis on variable host plants that makes trade-offs unlikely even on hosts from different regions.
ABSTRACT
From southeastern Queensland to southern Victoria, over a transect of 11° latitude, the Moreton taxon of the Australian grasshopper Caledia captiva exhibits a cline in chromosome structure that involves change from a metacentric to an acrocentric genome. In this study, we show that embryonic development time covaries with chromosome structure along the transect. Both development time and chromosome short arm length exhibit an overall negative correlation with latitude, but with maxima just south of the northern limit of the taxon's distribution. Selection for such a pattern appears to arise from changes in voltinism along the cline in season length that exists along the transect. Populations with the highest temperature thresholds for avoidance of embryonic diapause also have the slowest development time and probably represent the northern extreme of a primarily univoltine life cycle. North of this region bivoltinism increases in frequency and, as expected from a split of the season length, development time decreases. Maximum chromosome short arm length occurs in the vicinity of the northern univoltine populations, rather than at the limit of distribution where bivoltinism prevails. We conclude that variation in chromosome structure could be contributing to the heritable variation for development time that forms the basis for adaptive change in this trait. These results provide justification for investigating causal relationships between chromosome structure and development time, with an ultimate aim of understanding the adaptive significance of chromosomal variation in C. captiva.
