The Journal of Cotton Science 21:1–7 (2017) 1 http://journal.cotton.org, © The Cotton Foundation 2017 ARTHROPOD MANAGEMENT Accuracy of Lygus hesperus Knight (Hemiptera: Miridae) Egg Counts Improves with Egg Development Dale W. Spurgeon*, Evelien VanEkert, and LeAnne Elhoff ABSTRACT mproved management and availability of Ireduced-risk insecticides have diminished the The western tarnished plant bug, Lygus hespe- economic importance of lygus bugs in some rus Knight, is a key cotton (Gossypium spp.) pest, western cotton (Gossypium spp.) production regions managed primarily by application of insecticides (Anonymous, 2013). Despite these advances, lygus according to nominal thresholds. Efforts to reduce bugs (primarily Lygus hesperus Knight) remain reliance on insecticide-based tactics will require the most important pest complex in western cotton a more astute understanding of the physiological (Williams, 2016). However, as pest management ecology of L. hesperus than is currently available. improves and economic impact declines, additional A key biological parameter that may be manipu- improvement becomes more difficult using the lated through cultural or genetic means is repro- traditional model of pest management. For this duction. Estimates of L. hesperus oviposition are reason, the Pest Management and Biocontrol commonly obtained by visual inspection, but the Research Unit, Maricopa, AZ, is focused on accuracy of those estimates has been questioned. developing ecologically-based management tactics Because the eggs swell during development, we that maximize environmental or genetic-based hypothesized that delaying counts of eggs to permit resistance to reduce lygus reproduction, survival, some development might improve sampling accu- or crop injury. This approach will require improved racy. Estimates of L. hesperus egg numbers were understanding of the molecular biology of L. obtained immediately after oviposition and again hesperus, and more detailed knowledge of its basic at three days after oviposition, and counts at both biology and physiological ecology. times were regressed with total eggs determined Detailed knowledge of the impacts on L. hespe- by the sum of either hatched eggs or the number rus reproduction of environmental stressors, such as of nymphs, and unhatched eggs. Sampler experi- those induced by high temperatures typical of the ence was a major factor influencing fidelity of egg arid and semi-arid production regions of the West, counts with total eggs, but a single repetition of is lacking. Design of experiments with sufficient the experiment accompanied by knowledge of the statistical power to discern subtle, but ecologically results was sufficient to optimize sampler effec- important responses to environmental conditions tiveness. Regressions relating total eggs to counts, would be enhanced if accuracy of lygus egg counts whether the counts were taken immediately after could be maximized. Many species of mirid, includ- oviposition or after three days, were significant. ing Lygus spp., insert their eggs within host tissue However, compared with counts of newly-laid eggs, and often only the operculum is visible (Stewart the regression relating egg counts at three days and Gaylor, 1993; Wheeler 2001). In some stud- with total eggs was more stable and showed better ies of Lygus and similar species, the detection of agreement. These results can be used to maximize eggs was presumably enhanced by staining with statistical power and minimize sampling errors ethanolic Safranin O (Alvarado-Rodriquez et al., in future examinations of L. hesperus oviposition. 1986; Benedict et al., 1981; Ferran et al., 1996). However, this method would likely influence egg survival if subsequent observations are needed of 1 D.W. Spurgeon*, E. VanEkert , and L. Elhoff, USDA, the same cohort of individuals. In at least one study Agricultural Research Service, Pest Management and Biocontrol Research Unit, 21881 N. Cardon Lane, Maricopa, of L. hesperus parasitism, eggs were counted after AZ 85138, USA. 1Current address: Bayer CropScience, West hatching (Jackson, 2003). In most studies of Lygus Sacramento, CA 95605, USA spp. oviposition, eggs in a wide variety of host plants *Corresponding author: [email protected] have been detected by visual observation, often SPURGEON ET.AL.: IMPROVING ACCURACY OF LYGUS EGG COUNTS 2 with the aid of a dissecting microscope (Abel et al., by a number on the pod using black permanent ink 2010; Balachandran et al., 2014; Brent et al., 2011; (Pigma Micron 005, Sakura Color Products, Osaka, Conti et al., 2012; Elmore, 1955; Noma and Strick- Japan). Marking of the eggs facilitated later exami- ler, 2000; Ugine, 2011, 2012). During studies of L. nations for hatching, and minimized the potential to hesperus egg development (Cooper and Spurgeon, count the same egg more than once. 2012, 2013), it was observed that, despite careful Each sampler was randomly assigned five bean examination, the numbers of eggs detected visually pods for egg counts. After recording the initial (zero- in pods of green bean (Phaseolus vulgaris L.) were d) egg counts, each bean was returned to its dish often lower than the numbers of resultant nymphs which was wrapped in Parafilm (Pachiney Plastic (W.R.C. and D.W.S., unpublished data). This ob- Packaging, Chicago, IL) to prevent desiccation of servation raises questions regarding the accuracy of the developing eggs. Dishes with eggs were held in visual egg counts of Lygus spp. an environmental chamber maintained at 26.7±0.5°C Cooper and Spurgeon (2012) reported that with a 14:10 (L:D) h photoperiod. When the eggs fertilized eggs of L. hesperus swell during the early were three days old, the beans were reexamined by stages of development. Steward and Gaylor (1993) the same samplers and any previously undiscovered reported similar observations for L. lineolaris. It eggs were numbered and recorded. Sampling times seems plausible that after swelling, partially devel- were not recorded but were generally less than one oped eggs may be more apparent and more accurately h to sample five beans. The longest sampling times sampled than are newly deposited eggs. Our objec- were associated with higher egg counts primarily tives were to estimate the relationships between total because of the time necessary to number eggs. L. hesperus eggs and counts of newly hatched and After the three-d sampling, dishes were again three-d-old eggs, respectively, and to examine the wrapped in Parafilm and returned to the environmental effects of sampler experience on these relationships. chamber. Dishes were examined daily for excess con- densation during the egg development period. When MATERIALS AND METHODS condensation was evident, the dishes were opened, the condensate was removed by blotting, and the dishes The experiment used gravid adult female L. hes- were resealed with Parafilm. Removal of condensate perus originating from fields of alfalfa (Medicago reduced the potential for data loss from mold growth. sativa L.) near Maricopa, AZ, or the F2 progeny of Beginning on the fifth day after oviposition, the Petri the field-collected females. The field-collected fe- dishes were examined at least twice daily for hatching males were used in four experimental repetitions, and nymphs. At each inspection, newly-hatched nymphs their F2 progeny were used in two repetitions. Prior were tallied and removed to reduce the potential for to the experiments the insects were maintained in the egg predation by the nymphs. Inspections of the dishes laboratory on pods of green bean and raw seeds of continued until no additional hatch was observed, typi- sunflower (Helianthus annuus L.) at 27±2°C with a cally through the eighth day after oviposition. At the 14:11 (L:D) h photoperiod as described by Spurgeon completion of hatching, each bean was inspected for and Brent (2015). marked or unmarked eggs that remained unhatched. In each repetition of the experiment, each female The numbers of hatched and unhatched eggs were L. hesperus was held individually within a 150 × 15 recorded to permit estimation of hatch rates. Egg mm Petri dish. A 150-mm diameter disk of Whatman hatch was identified by an empty chorion and exer- No. 2 filter paper was centered in the Petri dish lid and tion of the serosal cuticle (Cooper and Spurgeon, the perimeter of the disk was pleated so it fit the lid 2012). Eggs that were collapsed or desiccated were tightly. When the Petri dish bottom was in place the typically amber in color and were recorded as dead. filter paper inhibited the escape of hatching nymphs. Total eggs were estimated by adding the numbers of Each female was provided an intact green bean pod nymphs to the numbers of unhatched eggs, except on for feeding and oviposition. The Petri dishes with fe- the few occasions when the number of hatched eggs males were held on the laboratory counter (~26–28°C) was higher than the number of nymphs detected. In under fluorescent lighting for six–eight h before the this latter case, total eggs equaled the sum of hatched females were removed and the initial count of eggs and unhatched eggs. was conducted. Eggs were counted with the aid of a Three samplers participated in each of six repeti- dissecting microscope, and each egg was identified tions of the experiment, except for the fifth repetition JOURNAL OF COTTON SCIENCE, Volume 21, Issue 1, 2017 3 when only two samplers were available. Samplers of freedom (the DDFM=KR option of the MODEL ranged from no prior experience in sampling Lygus statement).