A Nosema-like microsporidian pathogen of the bumble bee Bombus terrestris (provisionally referred to as "NBT") is described in terms of its occurrence, its developmental cycle and morphology. Variation in Nosema infection levels in B. terrestris queens from four sites in the central North Island suggested seasonal and climatic effects, with greater infection of queens from milder and/or urban habitats. Experimental infection of bumble bee workers provided details of the morphology and life cycle of the microsporidian. Features of the somatic stages of NBT observed from stained, smear preparations of tissues were characteristic of the genus Nosema (e.g. predominantly binucleate somatic stages comprising meronts, sporonts and sporoblasts). Each group of stages were variable in size and had from 1-4 nuclei. The timing of the appearance of somatic stages was investigated and a life-cycle for NBT was proposed. The minimum generation time of NBT spores at 30°C was approx. 4 days, with 50% of spores appearing within 7 days. Transmission electron microscopy indicated two spore forms were present, one Nosema-like with a polaroplast, diplokaryon, and 14-18 polar filament coils, the other having the same basic features but lacking the polar filament. NBT spores and somatic stages were found primarily in the Malpighian tubules and occasionally in the tissues of the midgut. The remaining tissues were free from infection. In a cross-infectivity test NBT was found to be slightly infective to Plutella xylostella, a Lepidopteran species that feeds on brassica plants. As bumble bees visit brassica flowers it is possible that natural cross-infection of NBT may occur, with other insect species acting as reservoirs for the disease. Comparisons of the morphology and life history of NBT with other Nosema species known from Hymenoptera indicated that NBT was different from all but N. bombi Fantham and Porter 1914. It is concluded that the NBT of this study is N. bombi. Experiments were undertaken to examine the transmissibility of NBT and its maintenance in individual bumble bees. Fecal material from naturally and experimentally infected queens was examined and numbers of spores estimated with a haemocytometer. The time taken from the ingestion of heat-killed spores by queens, to their complete elimination in the faeces, was approximately 5 days. Contamination of the nectar wick with faeces containing NBT spores caused the spore loads of individual bees to increase. Similar results were achieved by feeding bees directly with faeces (mixed with sugar syrup) containing a known number of spores. No significant reduction in spore numbers was observed from infected bees treated with Fumidil-B (an anti-Nosema drug). NBT infection did not appear to affect the nest initiation or egg-laying behaviour of bumble bee queens nor was there evidence for Nosema-linked mortalities.