Irradiation-attenuated sporozoite vaccinations confer sterile protection against malaria infection in animal models and humans. Persistent, nonreplicating parasite forms in the liver are presumably necessary for the maintenance of sterile immunity. A novel vaccine approach uses genetically attenuated parasites (GAPs) that undergo arrested development during liver infection. The fate of GAPs after immunization, their persistence in vaccinated animals, and the immune mechanisms that mediate protection are unknown. To examine the developmental defects of genetically attenuated liver stages in vivo, we created deletions of the UIS3 and UIS4 loci in the Plasmodium yoelii rodent malaria model (Pyuis3[- ] and Pyuis4[ - ]). The low 50% infectious dose of P. yoelii in BALB/c mice provides the most sensitive infectivity model. We show that P. yoelii GAPs reach the liver, invade hepatocytes, and develop a parasitophorous vacuole but do not significantly persist 40 h after infection. A single dose of Pyuis4(-) sporozoites conferred complete protection, but full protection by Pyuis3(-) sporozoites required at least 2 immunizations. CD8+ T cells were essential for protection, but CD4+ T cells were not. Our results show that genetically distinct GAPs confer different degrees of protective efficacy and that live vaccine persistence in the liver is not necessary to sustain long-lasting protection. These findings have important implications for the development of a P. falciparum GAP malaria vaccine.