Stage-dependent feeding behavior by western tarnished plant bugs influences flower bud abscission in cotton

Unexplained variability in the relationship between the number of herbivores in a field and the amount of crop damage can arise if there is a large amount of variation among herbivore individuals in the amount of feeding damage each generates. In California, populations of the western tarnished plant bug, Lygus hesperus Knight (Heteroptera: Miridae), produce highly variable levels of damage to cotton plants (Gossypium hirsutum L.) (Malvaceae), even when found at low densities. Because L. hesperus populations are also highly variable in their overall stage structure, we hypothesize that differences in crop damage might result from varying impact by each L. hesperus stage on cotton flower buds (termed squares). Laboratory measurements of L. hesperus mouth‐parts and distance to anther sacs, a preferred feeding site, revealed that 1st?3rd instar L. hesperus nymphs will not be able to feed on anther sacs of larger squares (over 8 mm in length) but will be able to feed on squares that are most sensitive to L. hesperus damage (<7 mm). Because even the 1st instars can feed on the most sensitive ‘pinhead’ squares, size constraints do not rule out damaging effects from the youngest L. hesperus. Laboratory observations revealed that later developmental stages, and adults, spend more time feeding on cotton squares relative to 2nd and 3rd instars. In addition, a field experiment revealed no effect of 2nd instars on square retention (relative to control cages) but did reveal a significant decrease in square retention generated by adult L. hesperus (4th instar L. hesperus resulted in an intermediate level of square retention). In a final study we sampled L. hesperus stage structure and density across 38 cotton fields. Multiple regression revealed that the densities of 1st?3rd instars of L. hesperus are not correlated with anther sac damage or square retention. However, in 2 years 4th and 5th instars were positively correlated with anther sac damage and negatively correlated with square retention. In the a third year, adult L. hesperus showed correlations in the same direction, across fields and across sites within fields. Overall, these results suggest that the adults and the largest nymphs of L. hesperus (4th and 5th instars) are particularly damaging to cotton squares, with the 1st?3rd instars of L. hesperus causing little damage to plants.