BackgroundTsetse flies (Glossina) transmit species ofTrypanosomawhich cause human African trypanosomiasis (HAT) and animal African trypanosomiasis (AAT). Measuring the infection status of wild-caught tsetse is an important part of operations to control HAT. In north-west Uganda, we conducted field studies over a 15-month period to compare classical, microscope-based, and PCR-based methods of detecting trypanosomes in tsetse. We also quantified the age structure and host preferences of tsetse.MethodsUsing Pyramidal traps placed along the Kochi River, in Koboko district, 12512 G. fuscipes fuscipes were caught. A subset of females (¬n= 5051) and males (n= 1221) underwent dissection wherein mouthparts, midguts and salivary glands were screened for trypanosomes. Additionally, the age of the females was estimated using ovarian ageing and the trypanosome-positive status of 1931 females and 438 males was investigated by ITS PCR. Further the bloodmeal sources of 131 tsetse were identified using vertebrate cytochrome b PCR.ResultsInfection rates estimated were significantly greater (1.9-9.3 times) using the PCR-based method compared to the classical dissection-based method. Positive rates forT. bruceisl,T. congolenseandT. vivaxwere 1.6% (1.32-2.24), 2.4% (1.83-3.11and 2.0% (1.46-2.63), respectively by PCR. The abundance and age structure of tsetse populations were relatively stable and the slight seasonal four-fold variation in abundance appeared to be correlated with rainfall. Analyses of age structure suggests a low natural daily mortality of 1.75% (1.62-1.88). The majority of bloodmeals were identified as cattle (39%, 30.5-47.8) and human (37% of meals, 28.4-45.6).ConclusionPCR provides a more sensitive and specific method of estimating the infection rates of all pathogenic species of trypanosome circulating in Koboko. The seasonally stable abundance, low mortality rate and high proportion of bloodmeals from humans may explain, in part, why this district has historically been a focus of sleeping sickness.Author summaryTsetse flies (Glossina) transmit African trypanosomiasis in humans (‘sleeping sickness’) and livestock (‘nagana’). Identifying trypanosome species and quantifying infection rates in tsetse is important in understanding and controlling trypanosomiasis. Traditionally this relied on microscopic examination of tsetse, however, this method is unable to reliably identify different species of trypanosome. We compared microscopy and PCR, trypanosome-detection methods, usingG. fuscipes fuscipescollected from traps deployed in Koboko district, in North-West Uganda.Our results show that the PCR-based method was 1.9-9.3 times more sensitive than the classical approach and able to identify different species ofTrypanosoma, including mixed infections. In collecting tsetse we also obtained data on the seasonal abundance, age and diet of the tsetse population. Analyses of these results showed that natural mortality was low (1.75%/day) among adult females and only a slight seasonal variation in abundance and age structure occurred. The most important hosts were cattle and humans, providing 39% and 37% of bloodmeals, respectively.The PCR method provided a sensitive means of identifying trypanosomes to the species level. The longevity and diet of tsetse in Koboko may explain, why this district was a persistent focus of disease prior to the deployment of Tiny Targets to control tsetse.