Key points are not available for this paper at this time.
Hypotheses on the direction of evolution in interspecific interactions generally rely upon the probability of interaction between the species. Attack by Depresssaria multifidae (Lepidoptera, Oecophoridae) on tagged individuals of the perennial herb lomatium grayi ( Umbelliferae) was studied for five (Smoot Hill population) or three (Little Tucannon population) years to analyze how plant demographic characteristics affect the probability that a plant is attacked at least once in its lifetime. The plant populations and the pattern of attack by the moths were analyzed based upon a model of size—specific demographic characteristics. Both plant populations were in steppe in eastern Washington, but plants at Smoot Hill had a higher probability of flowering eacy year within the larger size classes, spent fewer years in the largest size class, and had a lower probability of reaching the larger size classes than plants at Little Tucannon. The populations were similar, however, in the probability of attack each year within each size class. The probability that an established plant was attacked at least once in its lifetime was 0.12—0.17 at Smooth Hill and 0.30—0.55 at Little Tucannon. The results suggest that small to moderate differences in plant demographic characteristics can have large effects on the probability that a plant will be attacked at least once during its lifetime by a particular herbivore species; that is, small shifts in demographic characteristics can have large effects on “apparency” (sensu Feeny). The evolution of interactions between species may occur through slight shifts in life histories as often as it occurs through shifts in chemsity and morphology. The model developed here is applicable to other interactions involving size—structured populations and is a measure of probability of interaction that is useful in assessing how shifts in life histories and demographic variables can affect interactions between species.
Thompson et al. (Tue,) studied this question.