RESUMEN
We demonstrate that natural heat stress on wild larval Drosophila melanogaster results in severe developmental defects in >10% of eclosing adults, and that increased copy number of the gene encoding the major inducible heat shock protein of D. melanogaster, Hsp70, is sufficient to reduce the incidence of such abnormalities. Specifically, non-adult D. melanogaster inhabiting necrotic fruit experienced severe, often lethal heat stress in natural settings. Adult flies eclosing from wild larvae that had survived natural heat stress exhibited severe developmental anomalies of wing and abdominal morphology, which should dramatically affect fitness. The frequency of developmental abnormalities varied along two independent natural thermal gradients, exceeding 10% in adults eclosing from larvae developing in warm, sunlit fruit. When exposed to natural heat stress, D. melanogaster larvae with the wild-type number of hsp70 genes (n=10) developed abnormal wings significantly more frequently than a transgenic sister strain with 22 copies of the hsp70 gene.
RESUMEN
We tested alternative developmental hypotheses describing when during an insect oviposition cycle reproductive tactics are determined. Newly eclosed adult females of the grasshopper Romalea guttata were raised on eight different feeding treatments consisting of a low food diet, a high food diet, and changes from high to low food, or low to high food, at different times during the first oviposition cycle. When initial food availability was high, a decline in food availability >7 days after adult eclosion produced no significant increase in time to oviposition compared to constant high food. In contrast, when initial food availability was low, an increase in food availability as late as day 14 produced a significant decrease in time to oviposition compared to constant low food. Thus, time to oviposition is determined by feeding rate early in the oviposition cycle, but the time of this determination is later when food availability is lower. Masses of individual eggs were unaffected by these treatments. When initial food availability was high, a decrease in food availability on day 21 produced no significant change in numbers of eggs in a clutch compared to constant high food. In contrast, when initial food availability was low, an increase in food availability after day 7 produced no significant change in number of eggs in a clutch compared to constant low food. Changes in egg production resulted from oocyte resorption, which appeared to become unresponsive to food availability between day 14 and day 21. Our results refute the hypothesis that reproductive tactics are continuously flexible. Development toward oviposition seems to be structured so that reproductive tactics become independent of feeding late during the first oviposition cycle. Reproductive tactics become unresponsive to food at different times for groups initially receiving low or high food, suggesting that a particular developmental state, rather than some absolute time, marks the shift to development that is unresponsive toâ£food. Plasticity in reproductive tactics appears to beâ£controlled by hormones in a manner similar to the hormonal control of plasticity of metamorphosis in other insects.