Fitness consequences of hibernal diapause in the pitcher-plant mosquito, Wyeomyia smithii

Ecology, June, 1998 by W.E. Bradshaw, P.A. Armbruster, C.M. Holzapfel

INTRODUCTION

At temperate latitudes, insects exploit the favorable seasons with active development and reproduction and avoid or endure the unfavorable season through dormancy (diapause) or migration, which provide "escape in time and space" (Slobodkin 1963). The concept of escape implies that insects effectively can avoid the exigencies of the unfavorable season and, when favorable conditions return, resume normal development. Yet, we know that migration involves not only risks to survivorship, but also allocation of resources to wings, flight muscles, and even thoracic space that otherwise might go to reproduction (Southwood 1977, Dingle 1984, Solbreck et al. 1990). Similarly, diapause may (or may not) involve allocation of resources to cold-hardiness or somatic maintenance and thus incur costs to survivorship and/or reproductive performance. In the few cases where it has been examined, post-diapause individuals usually exhibit reduced fecundity relative to non-diapausing individuals; however, in some insects that feed while in diapause, females can actually exhibit increased fecundity or can live longer as adults than non-diapausing females and compensate for energy loss during diapause by post-diapause feeding (Tauber et al. 1986:63-64, 266-271, Danks 1987:37-44, Leather et al. 1993:148-173, Ishihara and Shimada 1995, Chang et al. 1996).

While previous studies have considered the effects of dormancy on single traits such as survivorship and fecundity that are clearly related to fitness, no one to our knowledge has yet considered how performance of individual traits relates to other potentially compensating traits or to more inclusive composite indices of fitness. Herein we consider the cost of diapause in terms of all of the components of a composite index of fitness, the overwintering cohort replacement rate ([R.sub.0]) in the pitcher-plant mosquito, Wyeomyia smithii (Coq.). We compare fitness and its components among three environments: near-optimal summer conditions ("optimal"), conditions intentionally designed to impose severe environmental stress ("stressful"), and a 22-wk simulated winter ("winter"). We have specifically designed our experiments to mimic as closely as possible the natural environment likely to be encountered by W. smithii in nature, and we show that there is a substantive cost of overwintering in diapause. We then relate our findings to theoretical models for the optimal timing of diapause and to tradeoffs in life histories.

Wyeomyia smithii is distributed in North America from the Gulf of Mexico to Labrador and northeastern Saskatchewan (30-54 [degrees] N). Throughout its range, this mosquito completes its preadult development only in the water-filled leaves of its carnivorous host, the purple pitcher plant, Sarracenia purpurea L. The leaves of S. purpurea persist through the winter and W. smithii overwinter in these leaves in a larval diapause that is initiated, maintained, and terminated by photoperiod (Bradshaw and Lounibos 1977).

METHODS

Collection

We collected the mosquitoes (Wyeomyia smithii) in October 1993, as diapausing larvae from two localities ("MM" and "PB" of earlier publications from this laboratory, e.g., Bradshaw and Lounibos 1977, Bradshaw and Holzapfel 1992) at 40 [degrees] N in the New Jersey Pine Barrens. The two localities were separated by 15 km. Within each locality, we identified three clusters of pitcher plants, 100-300 m apart. We collected 15002000 larvae from within each cluster and maintained them as separate populations. We reared each population on long days to adulthood and placed their offspring ([F.sub.1]) on short days to induce diapause. After all populations were synchronized in diapause, we returned them to long days to induce development, reared them to adults, and used their offspring ([F.sub.2]) in experiments.

Conditions common to all experiments

We run all of our experiments in the mosquito's natural habitat, the water-filled leaves of intact pitcher plants. In each environmental treatment, 40 larvae were reared from day of hatch in 30 mL distilled water in separate leaves on intact pitcher plants. To simulate natural prey capture by the host plant, experimental leaves were provided with rations of individually counted, freeze-dried, adult Drosophila melanogaster. Plants were placed in terraria in controlled-environment rooms programmed to simulate natural temperature cycles with smooth, sine-wave thermoperiods that lagged the photoperiod by 3 h. The photic environment was programmed for long- (L:D = 17:7) or short-day (L:D = 8:16) photoperiod that included 0.5 h dim "twilight" at each end of the photophase.

Upon the start of pupation, leaves were checked 3 times per week and the pupae removed, weighed, and transferred to adult-cohort cages. Each cage was provided with pesticide-free raisins for adult nutrition, an open jar with 50 mL distilled water for pupae and adult eclosion, and a single 10-20 mL cut leaf of Sarracenia purpurea for oviposition. The bottom of the cage was covered with adsorbent paper and this paper was soaked with distilled water 3 times per week. These cages were checked 3 times per week for adult eclosion, adult death, and eggs. The eggs were removed, placed in 75 mL distilled water, and the number of first-instar larvae hatching over a 10-d period was recorded. Fertility was then calculated as percentage hatch = [(number of larvae hatching from a cohort) / (number of eggs produced by the cohort)]. Finally, we calculated cohort replacement rate,