Theileria-free waterbuck (Kobus defassa) born in captivity were successfully infected with Theileria parva sporozoites derived from ticks infected by feeding on African buffalo (Syncerus caffer). All waterbuck underwent mild infections with the development of sporadic schizont and piroplasm parasitosis when inoculated with sporozoite doses lethal to cattle. A carrier state of T. parva was demonstrated by feeding clean R. appendiculatus nymphs on two of these infected waterbuck. Tick batches from these waterbuck on 2 of 5 occasions transmitted lethal Theileria infections to cattle. In a separate experiment, waterbuck cells were infected and transformed in vitro by T. parva sporozoites derived from buffalo but not by cattle-derived T. parva (Muguga) sporozoites. Waterbuck cells infected in vitro with T. parva isolated from buffalo were inoculated into autologous waterbuck but no infections developed. Theileria parva isolates generated in this study from various sources were characterized using anti-T. parva schizont monoclonal antibodies (MAbs), and it was found that buffalo-derived and waterbuck-passaged isolates had different profiles. Species-specific synthetic oligonucleotide probes, restriction fragment length polymorphism (RFLP) analysis with cloned T. parva DNA probes, and DNA sequence analysis of the p67 sporozoite antigen gene confirmed that the waterbuck-passaged parasite was T. parva. The Tpr repetitive probe hybridization patterns from the waterbuck-passaged parasites were different from the other samples tested. The ribosomal genotype of the waterbuck-passaged T. parva was similar to that of cattle-derived T. parva Muguga. Analyses with both probes and MAbs suggested that a minor parasite population present within the T. parva 7014 buffalo- derived stock had been selected during waterbuck passage. A variable region of the p67 sporozoite antigen gene of the waterbuck-passaged T. parva was similar to that of cattle-derived T. parva stocks and different from that of buffalo- derived parasites. Based on these results, methods were suggested to confirm and quantitate the involvement of waterbuck in the epidemiology of cattle theileriosis.