Diplostomiasis, or eyefluke is caused by the metacercarial stage of the digenean parasite <i>Diplostomum spathaceum</i> which selectively invades the lenses of freshwater fish and can cause partial and total blindness. This study investigates the host-parasite relationship of <i>D. spathaceum</i> infections in rainbow trout with particular reference to the diplostomule or post-penetration stage of the parasite. The point of cercarial penetration influenced the speed with which diplostomules reached the lens and the length of time they were exposed to the fish defence mechanisms. The further away the point of cercarial penetration from the eye, the longer diplostomules take to reach the lens and the fewer establish as metacercariae. Total body immersion using cercariae resulted in significantly more metacercariae establishing in the lens compared with entry via intramuscular, intraperitoneal or intracardial injection of either cercariae or diplostomules. Cercariae may be drawn in by the respiratory current and enter the fish via the gills and buccal cavity. The gills provide an immediate source of entry into the circulatory and lymphatic system of the fish. There did not appear to be any preferred route of migration. The optimum conditions for the <i>in vitro</i> collection of diplostomules were developed. The optimum conditions for <i>in vitro</i> transformation of <i>D. spathaceum</i> cercariae were established. The optimum conditions for the short-term <i>in vitro</i> maintenance of diplostomules were investigated. L-15 medium supplemented with 5% heat-inactivated foetal calf serum maintained approximately 80% survival for the first 24-48 h of culture. Rainbow trout were immunized with cercariae and diplostomules of <i>D. spathaceum</i>. Circulating antibody was detected by ELISA to both cercariae and diplostomules by 6-7 weeks post-immunization. Cross-reactivity occurred between anti-cercarial serum and diplostomule and metacercarial antigens, and anti-diplostomule serum cross-reacted with cercarial and metacercarial antigens in ELISA. Immunofluoresence studies indicated that the tail region of the cercriae was strongly antigenic; a differential fluorescence was observed for both anti-cercarial and anti-diplostomule sera with the cercarial tails fluorescing more strongly than the bodies. No cross-reactivity was observed in fluorescence between anti-cercarial or anti-diplostomule sera and metacercarial antigen although anti-cercariae serum cross-reacted with diplostomule antigen and <i>vice versa</i>. Normal trout serum and anti-diplostomule serum were cytotoxic to diplostomules of <i>D. spathaceum in vitro</i>. Immunized fish exhibited a significant degree of protection against further infection by <i>D. spathaceum</i>. However, there was no correlation between protection and the level of serum antibody. Isolated normal trout macrophages kill significant numbers of diplostomules <i>in vitro</i> at effector: parasite ratios of 150:1 or greater. <i>In vitro</i> killing was not increased using diplostomules opsonized with anti-serum or <i>in vivo</i>-activated macrophages individually, but when combined, increased killing occurred. Diplostomules were capable of eliciting respiratory burst activity from macrophages <i>in vitro</i>, suggesting that reactive oxygen species may have a role to play in the killing mechanism. Diplostomule antigen was capable of activating macrophages <i>in vivo</i>.