Glycosylphosphatidylinositol-anchored proteins on the egg surface have been proposed to play a role in gamete fusion on the basis of in vitro experiments. We tested this hypothesis by asking if oocyte GPI-anchored proteins are required for fertilization in vivo. Oocyte-specific knockout mice were created using the Cre/loxP system to delete a portion of the Pig-agene, which encodes an enzyme involved in GPI anchor biosynthesis. Conditional Pig-a-knockout females are infertile, and eggs recovered from the females after mating are unfertilized. In in vitro assays, the knockout eggs are severely deficient in their ability to fuse with sperm. These results demonstrate that GPI-anchored proteins are required for gamete fusion. Loss of the GPI-anchored complement of plasma membrane proteins could prevent fusion by altering the organization and function of GPI-anchored protein-containing lipid domains. Alternatively, a single GPI-anchored protein may be required in the fusion process. To distinguish between these possibilities, we have begun to identify the GPI-anchored proteins on the egg surface. We have identified one egg GPI-anchored protein as CD55, an ∼70 kDa complement regulatory protein. It has previously been found that CD55-knockout mice are fertile,demonstrating that CD55 is not essential for fertilization. This finding also means that the presence of the full complement of egg GPI-anchored proteins is not necessary for gamete fusion. Other egg GPI-anchored proteins acting in the fusion process can now be investigated, with the goal of understanding the mechanism of their function in sperm-egg fusion.