While developing a simulation model of pink bollworm (PBW) Pectinophora gossypiella (Saunders), population dynamics, we discovered that there was little information available on PBW egg and larval survivorship in insecticide-treated fields. Because we plan to use the model as part of an on-line management program for cotton in the desert southwest, the model must incorporate the impact of insecticides on all susceptible life stages. During the past three growing seasons (1985-1987) we obtained data on PBW egg and larval population dynamics that were amenable to survivorship analysis. Here we report the results of preliminary analyses of these data.

All data were collected in the Palo Verde Valley of southeastern California. In 1985, egg and larval densities were based on random samples of 100 bolls/ field. In 1986, egg and larval densities were based on random samples of 200 and 100 bolls/field, respectively. For the three fields sampled in 1987, all bolls in each of eight 1-meter-row samples/field were examined for eggs and larvae. Data from 1986 were analyzed for both the F2 (typically June 27-July 31) and F3 (July 31-August 31) PBW generations; all other analyses were performed using only F3 generation data.

Survivorship from the egg to larval stages was calculated by first estimating recruitment of eggs and larvae in each generation. We used an algorithm developed by Hogg and Nordheim (1983; Res. Popul. Ecol. 25: 280-297), which is based on the graphical method of Southwood and Jepson (1962; J. Anim. Ecol. 31: 481495). Based on the recruitment estimates, egg-larval survivorship (hereafter referred to as ELS) was calculated by: (larval recruitment/egg recruitment) x 100. ELS refers to the combined survival of eggs, lst, 2nd, 3rd, and 4th-instar larvae.

Average ELS was 21.50% (n=2 fields) and 19.39% (n=9 fields) for generations 2 and 3, respectively. Because we had few generation 2 observations, all data were combined which yielded an average (±SE) ELS of 19.8% (±2.29) with a 95% C.I. of 14.7 to 24.9% (range = 10.2 to 37%). Median ELS was 20.7%. These estimates are about 20% less than previous published data for ELS in untreated cotton, which is probably due to decreased egg hatch and increased 1st-instar larval mortality in insecticide-treated fields (unpublished). Although most eggs, during the boll-producing time of the year, are laid on bolls, ca. 30% of all eggs are still laid on vegetative areas of the plant. Thus, overall ELS is usually less than our estimate of 20%. However, since it is only practical to sample eggs laid on bolls, these results are applicable at the pest management level. For example, we have used these data to refine our insecticide-treatment threshold, which is based on monitoring the egg stage. We have also used these data to refine a simulation model for PBW management. Additional field data will be analyzed to further examine ELS trends between generations and among years.