Lumpy skin disease (LSD) is a disease of significant economic importance in Africa. It causes considerable production losses and its presence in semen is a constraint to international trade. Recent findings that LSDV viral DNA can be found in the semen of artificially infected bulls for up to five months, while viable virus could be isolated 42 days after infection, indicated the need for studies into the mechanism by which this protracted shedding occurs. Six healthy, seronegative, postpubertal Dexter bulls were housed in vector-free stables and challenged with LSD virus by intravenous injection. Sheath washes, vesicular fluid and semen collection was performed every other day and subjected to PCR. On these days, blood was collected for serum neutralization tests and virus isolation, and ultrasonography of the reproductive tracts performed. Semen was centrifuged to separate cell-rich and seminal plasma fractions, and tested by PCR. Clinical parameters were recorded twice daily. Bulls shedding viral DNA 28 days after challenge were slaughtered, their reproductive tracts were harvested and diagnostic post mortem was performed. Histopathology, immunoperoxidase staining, electron microscopy, virus isolation and PCR were done on tissue samples. Of the six bulls, two showed no clinical signs, two showed mild and two showed severe clinical signs. Fever appeared five to seven days and lesions eight to ten days post challenge. Bulls were viraemic and febrile during the same time. Viral DNA was detected in all semen fractions of all bulls, but mostly from the cell-rich fraction and from the bulls showing the most severe clinical signs. Ultrasonography showed infarction in the testes and epididymides of the two bulls that were most severely affected. Necropsy of the two bulls that were still shedding after 28 days showed testicular degeneration and infarction, as well as epididymal granuloma formation. None of the accessory sex organs showed significant pathology. Histopathological changes seen were necrogranulomata in testes and peididymides. IMP staining of reproductive tissues showed that staining was restricted to areas in the testes and epididymides that were associated with necrosis. Virus could be seen on negative staining EM of sections of the testes. Our results show that LSDV is not limited to specific fractions of the ejaculate and that the testes and epididymides are most profoundly affected. Blood contamination is not responsible for the presence of viral DNA in semen, and it is unlikely that the virus is sperm-associated. Results suggest that the ejaculate is contaminated with viral DNA as it is shed from necrotic lesions in the genital tract. Further research is indicated into the ability of infected semen to produce disease as well as treatment protocols that could render semen free of viral DNA.