Factors which may determine trematode–snail interactions were assessed in the present study. Compatibility was examined using a bacterial clearance assay to detect the modulatory effects of both compatible and incompatible trematode infections on the activity of haemocytes from Lymnaea stagnalis, during the early stages of infection. Exposure to and injection with Trichobilharzia ocellata, a compatible trematode, or the incompatible Schistosoma mansoni, resulted in modulation of haemocyte activity. However, T. ocellata activated haemocytes 1·5 h post-infection (p.i.) and then suppressed activity 24–72 h p.i. whereas with S. mansoni no suppression, only activation of haemocytes was observed throughout the test period (1·5–72 h p.i.). In previous studies, modulation of the haemocyte clearance activity by T. ocellata was found to be mediated by 2 E–S fractions, an activating fraction and a suppressing one. Investigations to assess whether the lack of suppression of haemocyte activity, observed in the S. mansoni–L. stagnalis incompatible trematode–snail interaction studied, was due to either the absence or ineffectiveness of the suppressing E–S fraction, were performed on a second incompatible combination, T. ocellata–Planorbis corneus. Using this combination it was revealed that only the activating E–S fraction had modulatory effects on P. corneus haemocytes, indicating that the suppressing E–S fraction, which actively interferes with the clearance activity of haemocytes from L. stagnalis, appears to act in a host-specific manner. In conclusion, the suppressing E–S fraction determines, at least in part, compatibility in the trematode–snail association studied. This is also probably likely in other trematode–snail combinations.