SAMPLING AND BIOSTATISTICS Distribution and Sampling of Bemisia argentifolii (Homoptera: Aleyrodidae) and Eretmocerus eremicus (Hymenoptera: Aphelinidae) on Cantaloupe Vines JULI R. GOULD AND STEVEN E. NARANJO 1 Phoenix Plant Protection Center, USDAÐAPHIS, 4125 E. Broadway Road, Phoenix, AZ 85040 J. Econ. Entomol. 92(2): 402Ð408 (1999) ABSTRACT Studies were conducted to examine the spatial distribution of immature Bemisia argentifolii Bellows & Perring and immature Eretmocerus eremicus Rose & Zolnerowich on canta- loupe vines and to develop efÞcient sampling plans. More B. argentifolii eggs were found on the 3rd leaf from the terminal of a cantaloupe vine than on any other leaf. The density of whiteßy nymphs peaked at leaf position 8, whiteßy pupae (large 4th instars) peaked at leaf position 11, and immature E. eremicus peaked at leaf position 14. We looked at 4 parameters to describe the distribution of whiteßy and parasitoid life-stages among the various leaf positions as follows: (1) median leaf position, (2) the leaf position with the highest percentage of a particular stage, (3) the leaf position where insect counts were best correlated with counts on the entire vine, and (4) the coefÞcient of variation (CV). All 4 distribution parameters changed throughout the season. In general the leaf positions described by the 4 distribution parameters increased (i.e., were further from the terminal) until the middle of the season when they began to decline. Across the entire season, the 4 distribution parameters for whiteßy eggs were associated with leaf positions 3, 4, and 5; for whiteßy nymphs with leaf positions 7, 8, and 10; for whiteßy pupae with leaf positions 11 and 12; and for immature parasitoids with leaf positions 11, 13, and 14. Based on considerations of cost and precision, it was most efÞcient to sample leaf 3 for whiteßy eggs, leaf 8 for whiteßy nymphs, leaf 11 for whiteßy pupae, and leaf 14 for immature parasitoids. Using the Taylor power law, density-dependent minimum sample sizes (number of leaves per Þeld) necessary to achieve a predetermined statistical precision (mean  SE) were estimated. Over a broad range of densities, 50 leaves per Þeld are adequate to achieve a precision of 0.20 Ð 0.25 for all life stages. We also provide estimates of the optimum number of leaves to collect per vine based on the within-vine and between-vine variability and the costs (time) associated with counting whiteßies and moving to another vine. KEY WORDS Bemisia argentifolii, Eretmocerus eremicus, cantaloupe, sampling, distribution Bemisia argentifolii Bellows & Perring, the B strain of B. tabaci (Gennadius), is one of the most signiÞcant pests of agricultural production in the southwestern United States (USDA 1995). Early in this decade, B. argentifolii assumed key pest status in several crops in Arizona, California, and Texas, including cotton, Gos- sypium hirsutum L., and cucurbits, Cucurbita spp. (Butler and Henneberry 1994, Brown et al. 1995). Biological control is considered a promising approach toward ameliorating the whiteßy problem, and con- siderable research is underway to examine the impact of native natural enemies (Naranjo and Hagler 1998) and to introduce exotic parasitoids (e.g., Hoelmer 1996, Kirk and Lacey 1996). Work also is ongoing to develop techniques for augmentative releases of para- sitoids in melon crops (Simmons et al. 1996). The development of reliable, cost-effective sampling methods is essential to the study of B. argentifolii population dynamics, making management decisions, and the evaluation of biological control. Estimates of parasitoid activity in melons may affect control treat- ment thresholds, and such information is critical to evaluation of augmentative release programs. The within-plant distribution of Bemisia has been examined on a number of host crops (reviewed by Naranjo 1996). Bemisia females preferentially oviposit on young foliage (Gerling et al. 1980, van Lenteren and Noldus 1990) and distributions of different stages become stratiÞed as the host crop grows as a result of the sessile habit of immatures. Cantaloupe vines grow rapidly, sending out new leaves from the terminal of the vine, and the average number of leaves per vine increases through most of the growing season. White- ßies and parasitoids of different stages would, there- fore, not be expected to be distributed evenly among the leaves on a cantaloupe vine. Thus, the position of leaves sampled from cantaloupe vines will strongly inßuence the stages that are found. Tonhasca et al. (1994) contrasted distributions of whiteßy eggs, nymphs, and pupae (red-eyed nymphs) on crown leaves and terminal portions of vine leaves of canta- loupe, but did not examine distributions within entire vines or distributions of parasitoids. Immature parasi- 1 Western Cotton Research Laboratory, USDAÐARS, 4135 East Broadway Road, Phoenix, AZ 85040. Downloaded from https://academic.oup.com/jee/article/92/2/402/2217005 by guest on 25 May 2022