The results of valid yellow fever protection tests on the sera of 1,069 wild Uganda monkeys are reported. The genetal immunity rate in this sample, which covers about half of Uganda and includes 16 different kinds of monkey, was very high—over 40 per cent. being protective. Immune monkeys were found in all the areas adequately sampled, with the single exception of the Ruwenzori Mountain Forest. The incidence of immunity varied considerably from district to district. The highest rates were found in Bwamba County, Toro District, and in parts of Masaka and Mengo Districts. The reasons for variation in immunity rate from area to area remain obscure. Four species were sufficiently numerous for detailed discussion. These are Cercocebus albigena johnstoni, Cercopithecus aethiops centralis, Cercopithecus ascanius schmidti and Colobus abyssinicus ituricus. No significant difference was found between the immunity rates in the case of C. albigena johnstoni, C. ascanius schmidti and C. abyssinicus ituricus. The rate in C. aethiops centralis, however, was significantly lower than those in the other three. It may be noted that C. aethiops centralis is to a large extent terrestrial, while the others are either exclusively or mainly arboreal. The entire sample was divided into three groups based on habits and behaviour. Two of these groups, the exclusively arboreal and the mainly arboreal species, showed no significant difference in the incidence of immunity. The third group, consisting of monkeys which are largely terrestrial, showed an immunity rate significantly lower than did the other two. Confining the discussion to a single area (Bwamba County) it was found that there is no significant difference between the immunity rates in monkeys collected in the middle of a large uninhabited rain forest, monkeys collected round the forest edge and monkeys collected in the inhabited area adjoining the forest. It was found that the immunity rate is high just after birth, that it falls sharply in the juvenile age-grade, and thereafter begins to rise, reaching its highest level in the old age-grade. This is in accordance with the results of previous work. It was also found that the main rise in immunity does not occur in all species at the same time of life. In the case, for example, of C. abyssinicus ituricus the most notable rise occurs in the adult age-grade. In C. ascanius schmidti it occurs earlier, in the subadult age-grade. The reasons for this difference are not understood. The role of monkeys in yellow fever epidemiology is discussed. On theoretical grounds it is postulated that (a) There may be a diurnal arboreal mosquito vector of the monkey disease in addition to the known nocturnal arboreal vector (Aedes (Stegomyia) africanus Theo.). (b) In certain areas it does not seem possible that continuous monkey-to-monkey transmission can be maintained. It is concluded that in such areas new infection must either be introduced by wind-borne mosquitoes coming from long distances or, alternatively, that there is an arthropod reservoir of the virus (perhaps associated with an unknown mammal cycle) of which, as yet, nothing is known. Emphasis is laid on the importance of obtaining adequate monkey samples in all forms of yellow fever survey work, and it is pointed out that the results of surveys confined to human sera may be misleading.