In many respects, the behavior of bumblebees is similar to that of the closely related honeybees, a long-standing model system for learning and memory research. Living in smaller and less regulated colonies, bumblebees are physiologically more robust and thus have advantages in particular for indoor experiments. Here, we report results on Pavlovian odor conditioning of bumblebees using the proboscis extension reflex (PER) that has been successfully used in honeybee learning research. We examine the effect of age, body size, and experience on learning and memory performance. We find that age does not affect learning and memory ability, while body size positively correlates with memory performance. Foraging experience seems not to be necessary for learning to occur, but it may contribute to learning performance as bumblebees with more foraging experience on average were better learners. The PER represents a reliable tool for learning and memory research in bumblebees and allows examining interspecific similarities and differences of honeybee and bumblebee behavior, which we discuss in the context of social organization.

We censused butterflies flying across the Panama Canal at Barro Colorado Island (BCI) for 16 years and butterfly hostplants for 8 years to address the question: What environmental factors influence the timing and magnitude of migrating Aphrissa statira butterflies? The peak migration date was earlier when the wet season began earlier and when soil moisture content in the dry season preceding the migration was higher. The peak migration date was also positively associated with peak leaf flushing of one hostplant (Callichlamys latifolia) but not another (Xylophragma seemannianum). The quantity of migrants was correlated with the El Niñ o Southern Oscillation, which influenced April soil moisture on BCI and total rainfall in the dry season. Both hostplant species responded to El Niñ o with greater leaf flushing, and the number of adults deriving from or laying eggs on those new leaves was greatest during El Niñ o years. The year 1993 was exceptional in that the number of butterflies migrating was lower than predicted by the El Niñ o event, yet the dry season was unusually wet for an El Niñ o year as well. Thus, dry season rainfall appears to be a primary driver of larval food production and population outbreaks for A. statira. Understanding how global climate cycles and local weather influence tropical insect migrations improves the predictability of ecological effects of climate change.