ESTIMATING EFFECT OF MEGASTIGMUS-SPERMOTROPHUS (HYMENOPTERA, TORYMIDAE) ON DOUGLAS-FIR SEED PRODUCTION - THE NEW PARADIGM

In a pollen exclusion experiment performed on the cones of five Dougla s-fir (Pseudotsuga menziesii [Mirbel] Franco) trees, the number of see ds infested by a seed chalcid, Megastigmus spermotrophus Wachtl, did n ot differ significantly between pollinated and unpollinated cones from the same tree. This finding led to a revision of the formula used to calculate M. spermotrophus effect on Douglas-fir seed production becau se the traditional formula, which is based on the assumption that only pollinated seeds are infested by these chalcids, exaggerates their ef fect. The relationship between the new formula and the traditional for mula is nonlinear, varying with both pollination rate and infestation level. To assist other researchers in estimating the error in past cha lcid studies, the discrepancies for a range of pollination rates were calculated. Past assessments were strongly biased only where pollinati on rates were <70% and chalcid attack rates were from 50 to 85% using the traditional formula. For low pollination rates, the discrepancy ca n exceed 50%. Evidence is presented to explain how the unfertilized fe male gametophyte, which is small and is normally resorbed in the absen ce of pollination, can support the development of chalcid larvae. In t his study, there was a strong correlation (-0.93) between cone length and chalcid attack rate.