Conventional Newcastle disease vaccines are not suitable for application to village chickens in tropical countries of Asia. Trials with food-based vaccines are being initiated and the following experiments were performed to evaluate oral vaccination with Newcastle disease virus. Experimental chickens were vaccinated orally with the avirulent V4 strain of Newcastle disease virus and haemagglutination-inhibition antibody responses were measured. V4 virus was introduced into the crop by tube and total faecal output was collected daily and assayed for Newcastle disease virus. Virus was recovered on Days 5 and 6 after vaccination from most chickens that had received 10(7.4) and 10(6.4) 50% egg-infectious doses (EID50) of virus. There was no recovery of virus from birds receiving a lower dose of vaccine. Groups of chickens kept in cages with wire floors were given various doses of vaccine into the crop. Higher antibody titres were achieved with higher doses of virus. This dose responsiveness was not observed when various doses of vaccine were presented on food pellets and the groups of chickens were kept on concrete floors. Similar antibody responses were then seen with nominal doses of 10(5.2) and 10(8.2) EID50 per bird, possibly as a result of excretion and re-ingestion of the vaccine virus. Spread of the vaccine virus was demonstrated when control chickens and chickens receiving 10(7.7) EID50 of V4 virus on food pellets were housed together on a concrete floor. Similar antibody titres were achieved in both vaccinated and in-contact chickens.