Variation in oviposition timing, adult emergence, and susceptibility to natural enemies: Prelude to population divergence in a specialized insect herbivore

Background/Question/Methods Gall-former life cycle events are tied to host plant phenology but may be altered by the effects of other herbivores on host plants. Here we describe variation in adult emergence and oviposition phenology and associated natural enemy mortality in the gall-former, Belonocnema treatae (Cynipidae: Hymenoptera), host specific to live oak, Quercus fusiformis, as a prelude to testing whether defoliation of the host plant by lepidopteran species alters the timing of life cycle events in this gall former. B. treatae exhibits alternation of generations on live oak with an asexual generation developing within leaf galls followed by a sexual generation developing within root galls. Up to 24 species of natural enemies attack B. treatae within leaf galls. Collections from 1995–2000 in Hays County, Texas, showed that B. treatae emerged from leaf galls from October–December with mortality ?99%. In 2001 in adjacent Travis County, B. treatae emerged from leaf galls in January–March with reduced mortality (40–60%). Consequently, from November 2006–February 2007 galls were collected monthly from 8 trees in Hays and 22 trees in Travis counties and monitored weekly to estimate emergence phenology and mortality per tree. Results/Conclusions A region-wide shift in emergence phenology was detected as all trees exhibited delayed adult emergence to January–March. Mortality due to natural enemies ranged from 80–99% at both sites. Live oaks have recently experienced region-wide defoliation by lepidopteran species. Defoliators consume B. treatae eggs deposited during the initial leaf flush, and defoliation results in a second leaf flush. To determine if observed shifts in adult emergence are due to shifts in oviposition phenology driven by defoliation, oviposition phenology and degree of defoliation were monitored from March–July 2007 on the aforementioned trees. Oviposition timing varied among trees but occurred later compared with historical records for all trees surveyed, while defoliation ranged from low to near complete. The correlation between extent of defoliation and emergence phenology will be examined when the 2007–2008 emergence season ends to determine the relationship between defoliation, oviposition timing, adult emergence and exposure to natural enemies. We hypothesize that (a) the second leaf flush resulting from defoliators is responsible for the observed shift in life cycle events and (b) B. treatae formed during the second leaf flush are “out of phase” with natural enemies and experience reduced mortality. Manipulative defoliation experiments have been implemented to determine whether defoliation creates temporally separated “broods” potentially limiting gene flow among B. treatae populations.