1. Bumblebees are important pollinators in North America and are attacked by a range of parasites that impact their fitness; however, few studies have investigated the extent or causes of parasitism in North America.2. This study used a 2-year multi-site survey of bumblebee parasitism to ask: (i) how common are parasitoid conopid flies and the internal parasites Crithidia bombi and Nosema bombi in Massachusetts; and (ii) what factors are correlated with parasitism? 3. Infection rates by all three parasites were higher in this study than previously documented in North America. Overall, conopids infected 0-73% of bees in each sample, C. bombi infected 0-82% of bees, and N. bombi infected 0-32%.4. Conopid flies infected female bees more than males and intermediate-sized bees more than large or small bees. Crithidia bombi infection rates were higher in certain bee species and sites, and exhibited a unimodal pattern of prevalence over time. Nosema bombi parasitism was higher in male than female bees. 5. Infection by N. bombi in two rare bumblebee species was higher than expected based on parasitism rates of common bee species but C. bombi infection was lower.If high prevalence of N. bombi in these bumblebee species is common, parasitism may be a potential cause of their decline.6. Given the documented effects of these parasites, the high levels of infection may affect bee populations in Massachusetts and threaten the stability of their valuable ecosystem services.

• Premise of the study: Early reproductive maturity is common in dry and ephemeral habitats and often associated with smaller fl owers with increased potential for within-fl ower (autonomous) self-pollination. We investigated whether populations from locations that differ in moisture availability, known to vary for whole-plant development rate, also varied in the timing of autonomous selfi ng. This timing is of interest because the modes of selfi ng (prior, competing, and delayed) have different fi tness consequences.• Methods: We measured timing of anther dehiscence, stigma receptivity, and herkogamy under pollinator-free conditions for plants from three populations of Collinsia parvifl ora that differed in annual precipitation, fl ower size, and time to sexual maturity. Using a manipulative experiment, we determined potential seed production via prior, competing, and delayed autonomous selfi ng for each population.• Key results: Stigma receptivity, anther dehiscence, and selfi ng ability covaried with whole-plant development and climate.Plants from the driest site, which reached sexual maturity earliest, had receptive stigmas and dehiscent anthers in bud. Most seeds were produced via prior selfi ng. The population from the wettest site with slowest development was not receptive until after fl owers opened. Although competing selfi ng was possible, all selfi ng was delayed. The intermediate population was between these extremes, with signifi cant contributions from both competing and delayed selfi ng.• Conclusions: Our results demonstrate that within-species variation in the timing of selfi ng occurs and is related to both environmental conditions and whole-plant development rates. We suggest that, if these results can be generalized to other species, mating systems may evolve in response to ongoing climatic change.

Surprisingly, floral sesquiterpenoid volatiles, which are associated with direct defense, indirect defense, and attraction, rather than defense traits such as cucurbitacins, appeared to drive interactions with both pollinators and floral herbivores across cucurbit taxa. Identifying the relevant plant traits for attraction and deterrence is important in this economically valuable crop, particularly if pollinators and floral herbivores use the same plant traits as cues.

Researchers increasingly recognize the important role of mutualisms in structuring communities and view positive interactions in a community context rather than as simple pairwise interactions. Indirect effects, such as those that predators have on lower trophic levels, are a key process in community ecology. However, such top-down indirect effects have rarely been extended to mutualisms. Antagonists of one mutualist have the potential to negatively affect the second mutualist through negative effects on their partner, and the magnitude of such effects should vary with mutualism strength. Bumble bees are ecologically and economically important pollinators that are an ideal system to determine if such indirect effects play an important role in mutualisms. Bumble bees are attacked by an array of parasites and predators, and they interact with a range of plants that vary in their dependence on bumble bees for reproduction. We tested whether variation in parasitism rates by Nosema bombi, Crithidia bombi, and conopid flies correlated with reproduction of greenhouse-raised plants placed in the field. At multiple sites over two years, we studied four plant species that varied in reliance on bumble bees as pollinators. We found a consistent negative relationship between Nosema parasitism and measures of pollination for Trifolium pratense and Solanum carolinense, plant species with high bumble bee visitation, whereas Rudbeckia hirta and Daucus carota, plant species with generalized pollination, experienced no impacts of Nosema. However, both Crithidia and conopids showed inconsistent relationships with pollination service. Although these patterns are correlational, they provide evidence that parasites of bumble bees may have negative indirect effects on plants, and that mutualism strength can moderate the magnitude of such effects.

North American tiger beetles (Cicindela spp. L.) have been reared in the laboratory for more than a century, and here we summarize the relevant literature to develop a general rearing protocol. We used this protocol to experimentally overwinter adults in the laboratory and observe variation in oviposition and fecundity among several species. Overwintering experiments, involving Þve North East North American Cicindela species with spring-fall life historiesÑCicindela repanda (Dejean), Cicindela hirticollis (Say), Cicindela purpurea (Olivier), Cicindela scutellaris (Say), and Cicindela tranquebarica (Herbst)Ñ demonstrated that both a long cooldown (20 to 4ЊC by a degree a day) and a short photoperiod (8:16 [L:D] h) maximized survival and minimized overwintering weight loss, which varied between species and sex. Observations of oviposition, larval abundance and larval development involving Þve Cicindela species with summer life histories revealed that Cicindela punctulata (Olivier) produced more Þrst-instar larvae than Cicindela abdominalis (F.), Cicindela dorsalis dorsalis (Say), Cicindela puritana (Horn), or Cicindela unipunctata (F.) and that high mortality due to accidental desiccation may be overcome by rearing larvae individually in tubes rather than in bins. We also present a Þrst account of larval rearing of the federally threatened species C. puritana and the northern MarthaÕs Vineyard population of the federally threatened species C. d. dorsalis. Notably, C. d. dorsalis produced fewer larvae than more common species reared in this study. We conclude that rearing large numbers of larvae is feasible with endangered as well as common species and we propose future improvements for rearing as part of conservation efforts.