Abstract. In the present study, we adapted a polymerase chain reaction (PCR) assay, previously shown by us to be very sensitive for detecting cercariae in water, for the sensitive detection of Schistosoma mansoni DNA in infected snails from early prepatency. Polymerase chain reaction primers were designed based on the 121-basepair highly repeated sequence we previously identified in the genome of S. mansoni. The DNA was prepared from the snails by a simple alkaline extraction procedure, and the PCR assay enabled a clear differentiation between infected and normal snails. Infected snails were detected as early as one day after penetration of a single miracidium. The high sensitivity of the test enabled identification of a single infected snail even when its DNA was pooled with material from up to 99 uninfected snails, thus demonstrating the possibility of mass diagnosis in pools of snails. The assay has the potential for large-scale determination of prepatent infection prevalence in snails, thus offering new possibilities for the evaluation of schistosomiasis transmission and for schistosomiais control, as discussed.Schistosomiasis, is caused in humans by three main species of blood flukes (Schistosoma mansoni, S. haematobium, and S. japonicum) and is transmitted by freshwater snails. It continues to plague many developing countries in the tropics 1 with an estimated 200 million infected people worldwide.Freshwater snails, intermediate hosts of schistosomes, become infected by larvae (miracidia) released from schistosome eggs that reach the water with excreta. After several weeks of asexual multiplication within the snail's tissues, larvae infective to humans by active penetration through the skin (cercariae), are released into the water. Snail infection and contact patterns of humans with water infested with cercariae are important factors in the transmission of schistosomiasis, and risk of infection is normally estimated by detecting infected intermediate hosts in sites where humans come in contact with water. 2 Infection rates in field populations of snails are routinely determined by searching for cercariae shed from snails with patent infection. 3 In contrast, prepatent infections, which can constitute a significant proportion of infected snails populations, 4 are not determined routinely because of a lack of a suitable method. Microscopic examination of crushed snails in search of developing cercariae and intramolluscan-multiplying larvae (the sporocysts) 5 is sometimes used for this purpose but is highly inaccurate at early prepatency, and unsuitable for routine large-scale screening of snail populations. 4 Another approach is to examine nonshedding field snails maintained for several weeks in the laboratory for protracted shedding. However, this method is time-consuming, it cannot be done routinely on a large scale, and it can be highly inaccurate if snails mortality is high. The omission of data on prepatent infections leaves out important information for the evaluation and mathematical modeling of trans...