The albumen gland, a female accessory sex gland of pulmonate snails, produces the perivitelline fluid. The ultrastructure of the albumen glands of control and infected specimens of Lymnaea stagnalis and Biomphalaria glabrata was studied. The albumen gland of L. stagnalis contains two types of secretory cells--light (active) and dark (inactive)--and two types of supporting cells--centroacinar and myoepithelial. The secretory cells apparently represent two activity stages of one type of cell. The gland B. glabrata possesses only one secretory cell type, which alternates with one type of supporting cell. The albumen glands of L. stagnalis and B. glabrata infected at a juvenile stage were studied 4 and 14 weeks (L. stagnalis) and 4 and 9 weeks (B. glabrata) after exposure. After four weeks' infection, B. glabrata produced some egg masses, but in subsequent stages egg mass production completely coased. Infected L. stagnalis never produced eggs. B. glabrata was apparently infected at a "physiologically" more mature stage than L. stagnalis. The morphology of the albumen glands four weeks after exposure (the daughter sporocyst stage) is in agreement with this hypothesis. At this interval the secretory cells of L. stagnalis appeared to be much more severely affected (inactive Golgi bodies and rough endoplasmic reticulum, crinophagy of the secretory granules) than the cells of B. glabrata. In the later stages studied (shedding of the cercariae), the glands of both species appeared to be completely inactive (reduced height of the epithelium, inactive organelles, crinophagy, absence of secretory granules). At this stage of infection, daughter sporocysts containing cercaria embryos were seen in the connective tissue of the albumen gland of B. glabrata, but not of L. stagnalis. The results thus indicate that the development and synthetic activity of the albumen gland are seriously affected by infection. These processes are known to be under the endocrine control of the female gonadotrophic hormones. Since it has been established that these hormones are normally present in the haemolymph of infected snails, the findings can be explained by assuming that the parasite interferes in some way or other with the snail's endocrine system.