▪ Abstract Behavioral responses of insect herbivores to toxins are examined in managed and natural systems with reference to two important but largely ignored factors: heterogeneity in toxin distributions and the nature of the relationship between behavioral responses and physiological adaptation to the same toxins. Heterogeneous toxin distributions, which provide the opportunity for behavioral responses, are ubiquitous in managed and natural systems. Insect herbivores have evolved a wide variety of behavioral responses to such toxins. The nature of behavioral responses reflects toxin apparency, mode of action, and the extent to which sublethal effects influence behavior. The interaction between these behavioral responses to heterogeneously distributed toxins and physiological mechanisms of tolerance has influenced the evolution of insecticide resistance in managed systems and the evolution of plant defensive strategies in natural systems. An understanding of this interaction could lead to more evolutionarily stable methods of crop protection.

Seasonal activity of carrot weevil,Listronotus oregonensis(LeConte), adults, eggs, larvae, and root feeding was studied in plots of parsley,Petroselinum crispum(Mill.) Nyman ex A.W. Hill, by sampling weekly from mid-May to the first week of September in 1998 and 1999. The initial population of overwintering adults, one complete generation, and at least a partial second generation were observed. The majority of eggs were deposited in June and most of the larvae were counted in July; the greatest damage was noticed from July onward. In both 1998 and 1999, the overall seasonal means of adults, eggs, larvae, and root feeding were significantly different among plots that were planted on different dates. Root feeding and all carrot weevil life stages sampled steadily decreased from the earliest to the latest planting. Most of the damage occurred in parsley plantings sown from late February until the third week of April. The number of eggs removed from parsley plants during the first cutting was significantly different among planting dates, but the impact of this removal may not be significant in infestation reduction. Results of this study will help parsley growers to target when control measures for carrot weevil will be most effective and which plantings are most likely to be damaged.

Indicative value of nematode food web indices and trophic group abundance was assessed for differentiating habitats in two soil types. A total of 604 soil samples were collected from croplands, grassy borders, turfgrass lawns, shrublands and forests in Huron County, Ohio characterized by muck soils. Also, a total of 242 soil samples were collected from two habitats, turfgrass lawns and croplands, in Wayne County, Ohio characterized by mineral soils. All nematodes extracted from soil samples were identified to genus level, categorized to trophic groups and various nematode community indices were calculated. Multivariate analysis of variance using composite of nematode measures revealed that croplands were significantly different from forests, shrublands, turfgrass lawns and grassy borders in the muck soils. Turfgrass lawns in muck soils were not different from forests but were different from shrublands. No differences were observed between forests and shrublands. Turfgrass lawns in mineral soils differed significantly from croplands in the mineral soil. No differences were observed between the soil types when nematode data from croplands and turfgrass lawns from each county were lumped together. All habitats showed moderately enriched food webs [enrichment index (EI) 69–78%] in both soil types. Forests, shrublands and turfgrass lawns in muck soils had high maturity (MI) and structure index (SI) whereas grassy borders and turfgrass lawns in mineral soils had intermediate and croplands in both muck and mineral soils had the lowest MI and SI values. Principal component analysis (PCA) indicated that MI and EI contributed the most to differences between habitats in the PC1 while SI and channel index significantly influenced PC2 axis. Forests and shrublands had highly structured food webs as expected but their high enrichment may be unique to the muck soils. Ordination of habitats using nematode indices revealed that grassy borders and turfgrass lawns hold an intermediary position between undisturbed habitats (forests and shrublands) and highly disturbed habitats (croplands). Although overall contribution of plant-parasitic nematode index (PPI) was less compared to the other indices, PPI explained the variation among the habitats in the PC3 axis. Ordination using trophic group abundance showed lower resolution compared to the food web indices.