Search

1. Crown architecture remains one of the least studied plant traits that influence plant–herbivore interactions. The hypotheses that dense crown architecture of mountain birches from open habitats favours leaf-tying caterpillars through bottom-up and/or top-down effects associated with high leaf connectivity were tested. 2. Population densities of leaf-tying herbivores in open (industrial barren and seashore) habitats were three times as high as in the shaded (forest) habitats. An experimental increase in leaf density by branch binding did not affect foliar consumption by free-living herbivores but increased consumption by leaf-tiers. 3. The specific leaf weight was lower in shaded habitats and in bound branches, but branch binding did not influence either the foliar concentrations of carbon and nitrogen or the pupal weight of the most abundant leaf-tier, Carpatolechia proximella Hbn. (Lepidoptera: Gelechiidae). 4. Caterpillars of C. proximella build several shelters during their lifetime and spend a considerable amount of time outside the shelter, where they excrete most of their faeces. In bound branches, caterpillars built new shelters more frequently than in control branches, and consumed less foliar biomass per shelter. 5. Mortality from parasitoids in bound branches was half that in the control, presumably because the complex environment disrupted parasitoid searching behaviour and/or because of lower damage to leaves from which the shelters were built. 6. It is concluded that the crown architecture associated with high leaf connectivity decreases mortality risks from natural enemies both outside and inside the shelter. Compact and dense crowns of host plants may at least partly explain high population densities of leaf-tiers in open habitats.

1. The leaf beetle, Chrysomela lapponica, originally uses the salicyl glucosides (SGs) of its host plants to sequester salicylaldehyde, which serves as a defence against generalist enemies but attracts specialist enemies. However, some populations of C. lapponica have shifted to SG-poor hosts, and their secretions do not contain salicylaldehyde. 2. In was suggested that beetles shift to SG-poor hosts to escape from specialist enemies. To test this hypothesis, we compared field mortality between two populations of C. lapponica that were associated with SG-rich willow, Salix myrsinifolia (Kola Peninsula and Finland) and two populations that fed on SG-poor willows, S. glauca (Ural) and S. caprea (Belarus). 3. Mortality from generalist enemies was significantly higher in Belarus than in three other populations, whereas mortality from specialists did not differ among populations. A specialist predator (syrphid fly larvae, Parasyrphus nigritarsis) and specialist parasitoids (phorid flies, Megaselia spp.) were attracted to the secretions of larvae reared on both SG-rich and SG-poor hosts. 4. Feeding on leaves of S. caprea and S. myrsinifolia both previously damaged by leaf puncturing and by the larvae of potentially competing species Chrysomela vigintipunctata, decreased the weight and prolonged the development of C. lapponica. 5. Thus, populations of C. lapponica that have shifted to SG-poor willow species did not obtain enemy-free space because specialist enemies have developed adaptations to herbivores that switched to a novel host plant. We suggest that in some populations host plant shift was favoured by interspecific competition with the early season SG-using specialist, C. vigintipunctata.