Elongate hemlock scale (Fiorinia externa) is an invasive species from Japan that was first detected in the United States in 1908 and has established in most states where eastern hemlock (Tsuga canadensis ) occurs. Scale density is much higher in the US (21 to >400 scales per 100 needles) compared to Japan (<1 scale per 100 needles) despite the presence in both regions of the parasitoid Encarsia citrina that has been assumed to be responsible for low densities in Japan. I have found that phenology of the vulnerable scale stage was asynchronous with E. citrina flight throughout the eastern United States. This asynchrony was the result of overlapping life stages and generations of elongate hemlock scale throughout the growing season. This overlap appears to result from delayed senescence of first generation adult females. To investigate the impact of E. citrina on elongate hemlock scale, natural enemy exclusion experiments were conducted. I found that E. citrina exerts some control on elongate hemlock scale, but insufficient to maintain density at levels seen in its native Japan. Existing records of parasitoids of elongate hemlock scale in Japan are limited to the Kyoto area. I sampled the parasitoid community of elongate hemlock scale and other scales on hemlock throughout Japan. Sequence data from two gene loci, 28SD2 and COI, were used to identify the number of unique parasitoid species attacking scales on Japanese hemlocks. This sequence data was also used to identify possible cryptic species within E. citrina in Japan. Twenty-two genetically distinct parasitoids were identified (seven singletons). Eleven (three singletons) of these were from elongate hemlock scale. With the exception of E. citrina, parasitoids tended to be host species specific. No evidence of cryptic species within E. citrina was found. The parasitoids identified dramatically increase the number of elongate hemlock scale parasitoids previously reported and this study may lead to the identification of several potential biological control agents of elongate hemlock scale for introduction to the United States.

An improved in vitro rearing system, based on a silicone-reinforced Parafilm® M membrane, human hair tufts and reconstituted human blood, enabled the large-scale rearing of pediculicide-susceptible (EC-HL) and resistant (FL-HL and BR-HL) strains of human head lice. Developmental time differed for early instars but differences became synchronized as lice matured. Mean survivorship amongst the three strains reared in vitro were not significantly different when compared to EC-HL and FL-HL reared in vivo. FL-HL and CA-HL were resistant to permethrin compared with susceptible PAHL (5.5- and 3.4-fold, respectively) and EC-HL (8.5- and 5.3-fold, respectively). Permethrin-resistant lice were cross-resistant to pyrethrum and DDT. Permethrin resistance was synergized by piperonyl butoxide (PBO) in CA-HL. Resistance to malathion in FL-HL (1.4- to 2.2-fold) and CA-HL (2.1- to 3.6-fold) was detected. Malathion resistance in FL-HL was synergized by S.S.S-tributylphosphorotrithioate (DEF) and by PBO in CA-HL. Malathion/permethrin-resistant BR-HL lice were synergized by DEF but not synergized by PBO. PBO protected BR-HL from malathion, indicating suppressed desulfuration. Abamectin resistance in FL-HL (1.7- to 2.5-fold) and CA-HL (1.8- to 2.3-fold) was detected. No resistance to lindane was found. Thus, multiple resistance mechanisms against commonly available and widely used pediculicides and insecticides are apparently occurring. Results from linkage analysis substantiate that permethrin resistance in human head louse populations in the U.S. is widespread but occurs at varying frequencies. The log survival time versus logit mortality regression lines of susceptible-homozygotes, resistant-homozygotes and heterozygotes determined that the resistance trait was complete recessive. Permethrin resistance is highly correlated with the presence of the T917I and L920F point mutations, which are suitable for detection by a variety of DNA-based diagnostic techniques. Electrophysiological assessments determined that the three mutations (M815I, T917I and L920F) associated permethrin- and DDT-resistance in head louse caused depolarizing shift of ∼3.2 and ∼8.8 mV in the voltage dependent activation and steadystate inactivation, respectively, and completely abolished permethrin sensitivity when correspondingly inserted into house fly para-orthologous, voltage sensitive sodium channel and expressed in Xenopus oocytes (M827I, T929I and L932F). Amongst these kdr mutations, T929I appears to be the major resistance-causing mutation.

Mutualisms are major structuring forces in biological communities. However, the concept of the trophic cascade has rarely been explicitly applied to mutualisms. Antagonists of one mutualist have the potential to negatively affect the second mutualist through effects on their partner, and the magnitude of such effects should vary with mutualism strength. Bumblebees represent an ecologically and economically important mutualist pollinator group. They are attacked by a range of parasites, and visit a variety of plants that vary in reliance on bumblebees for pollination service. Using bumblebees and their parasites, I investigated whether mutualisms can mediate trophic cascades. I surveyed three parasites (Crithidia bombi, Nosema bombi and conopid flies) in Bombus spp. in Massachusetts and found that parasites are more common in wild bumblebees than previously believed (Chapter 1). To test whether infection by these parasites has top-down effects on pollination service to plants, I examined whether site-level parasitism rates correlated with pollination service to greenhouse-raised plants. I used several plant species which varied in their dependence on bumblebees for pollination. The relationship between parasitism and pollination service differed between plants and parasite species, and there was evidence that plants more reliant on bees for pollination experienced stronger negative indirect effects of parasites (Chapter 2). Finally, I developed an ordinary differential equation model of mutualist population dynamics incorporating a producer, its mutualist, and an antagonist of the mutualist. I varied the mutualism from obligate to facultative, and investigated effects of the antagonist on mutualist coexistence. My results highlight how mutualisms differ from traditional trophic cascades, and the importance of considering obligate, facultative, and asymmetric mutualisms in attempting to understand their interactions with the community (Chapter 3). My contributes to conservation issues and extends our theoretical understanding of basic ecology. I provided valuable data about the incidence and multitrophic effects of parasites in an important native pollinator. By combining mutualisms with the theoretical structure of the trophic cascade, two previously independent areas of ecological research that will benefit from integration, my research improves our understanding of how mutualisms structure the community as a whole.