Temporal and Spatial Patterns of Parasitoid Attack on a Root-Galling Cynipid
MetadataShow full metadata
Here in I assess for the host-specific gall former Belonocnema treatae (1) the effects of temporal variation in oviposition timing on (a) gall growth, and (b) parasitism; (2) the effects of spatial variation m gall-former densities on (a) gall discovery, and (b) gall parasitism; (3) ecological parameters related to (a) host-tree selection, (b) gall growth, (c) gall occupancy, (d) parasitism, and ( e) sex. (1) To assess the effects of variation m oviposition timing on gall growth and parasitism rates, I created three B. treatae cohorts, and exposed them to parasitoid attack during each of four intervals. Oviposition timing significantly affected developmental gall sizes (p < 0.001), but not the sizes of galls at maturity (p > 0.10). Both host-tree selection (p < 0.001) and oviposition timing (p < 0.05) significantly affected the rates at which parasitoids discovered galls, and an interaction between these two variables had a significant effect on gall parasitism rates (p < 0.05). Gall discovery (p > 0.90) and parasitism rates (p > 0.10) were not size dependent. Analyses of temporal patterns of parasitoid attack suggest that parasitoids were attacking eggs and not mature galls. (2) To test for density dependent parasitism, I collected four 1/8-m2 samples of galls from unmanipulated root regions of 10 host-trees and lab reared their occupants. Gall densities were significantly different among host-trees (P < 0.005), yet these differences did not correlate with gall discovery (R2 = 0.003), or gall parasitism rates (R2 = 0.003). Coefficient estimates of both gall discovery and parasitism rates suggest that the probability of each increase by only 1 % for every 50 galls/m2 increase. Density independent patterns of parasitoid attack are explained by visual limitations and high travel costs associated with searching for prey in a subterranean environment, or by small sample sizes that were possibly insufficient for detecting subtle patterns of density dependant foraging. (3) To describe root-galling B. treatae ecological parameters, I collected and lab reared the occupants of 771 galls taken from unmanipulated root regions of 10 host-trees. Analyses of these galls showed that significantly more females emerged from root galls than did males (p < 0.001), yet the number of female and male producing galls did not differ within (p > 0.50) or among (p > 0.10) trees. Analysis of gall sizes between the sexes showed that on average females emerged from larger galls than did males (p < 0.05), yet their mean gall occupancy levels did not differ (p > 0.60). Neither gall discovery (p > 0.05) nor gall parasitism rates (p > 0.20) differed between the sexes. Analysis of host tree regulated fitness parameters suggests that host-trees exert a maJor selective force on root-galling B. treatae. These analyses show that host-tree selection influences everything from gall size (p < 0.01) and gall occupancy levels (p < 0.0001) to gall discovery (p < 0.0001) and gall parasitism rates (p < 0.00001). When considered together, my results provide evidence that root-gall parasitism rates are both temporally and spatially influenced, and that bottom-up and top-down regulatory processes are experienced similarly between the sexes.