The accurate measurement of insect mortality by parasites is critical in biological control research, both in baseline studies to determine the absence or inadequacy of native parasites and in subsequent efforts to measure the effectiveness of introduced endoparasitic species. Although rearing has been most frequently used to measure parasitism, dissection has been shown to be more accurate in several cases. Selection of the host instar, whether for rearing or dissection, was also found to be important in this study. In two species [Lygus lineolaris (Palisot) and L. hesperus Knight], parasitism by Peristenus digoneutis Loan and P. howardi Shaw, respectively, was highest in instars 3 and 4. Parasitism was underestimated in instars 1 and 2 (because of reduced exposure time) and in instar 5 (because of parasites killing the hosts in instar 4).