It has been the purpose of this thesis to demonstrate the logical development of a programme of work on the swollen - shoot disease of cacao in pest Africa. In the introduction the history of the swollen -shoot disease is discussed and its effect on the economy of the Gold Coast is described. The research work accomplished up to January 1947 by Posnette, Strickland, Todd and others is also described in the introduction. ;hen the writer became a member of the virus research team in January 1947 work by Posnette and Todd on various aspects of the disease was being actively pursued and the writer was allotted the task of improving the technique of insect transmission of the viruses. The method used was to study the fundamental relationship between the virus, Theobroma virus 1h and its vector Pseudococcus ralensis. The results of this study are shown in section I of the thesis where it is shown that:- 1) Insects which have previously fed on an infected plant are able to cause infection twenty minutes after being placed on the test plants and give their highest rate of transmission after 60 minutes or longer on the test plants. 2) Pre-starved insects which have not previously fed on infected plants show their greatest infectivity when given an infection feed of ten hours, on leaves; after twelve hours infection feeding there is a significant drop in the infectivity of the insects. Unstarved insects show a much slower rate of rise of infectivity with time, do not reach the same high peak of infectivity at ten hours and do not show the significant drop after twelve hours infection feeding. 3) When the further development of the curve of infectivity was studied and infected cotyledons were used as the virus source the infectivity of the insects feeding on this source was lower than on leaves and the peak of infectivity was not reached after eighteen hours continuous feeding. However the low rate of transmission and the large experimental error prevented any detailed analysis of the curve of infectivity the data being best fitted by a straight line with the regression equal to 1.3. In these experiments there was no great difference between pre -starved and unstarved insects. 4) Insects starved after an infection feed retain their infectivity at a high level for twenty hours and lose it completely after 36 hours starvation. 5) Insects fed for 30 minutes on a healthy plant after an infection feed will occasionally infect a second plant but there is no evidence for extended persistence. In section II of the thesis some of the factors leading to an increase in the experimental error are discussed and it is shown: - 1) That in Pseudococcus citri there is a biological race which is incapable of transmitting Theobroma virus 1C. 2) That test plants of 0 - 3 weeks old are twice as susceptible to insect inoculation as plants of 4 weeks and older. 3) That when embryos are used as test plants, plants with one cotyledon removed give a better rate of transmission than plants with two cotyledons. 4) That best transmission is obtained from young flush leaves showing symptoms but that the actual age of the plant on which they are borne does not matter. There is evidence however that in transmissions from the stems there is definite localisation of the virus especially in older plants In section III unsuccessful attempts at sap transmission are described and it is shown that the tannins of the cacao plant do not inactivate tobacco mosaic virus. In section IV the knowledge of vectors obtained in sections I and II are used in an attempt to find alternative host plants of Theobroma viruses 1C and 1M. Inbryos and seedlings of various members of the Sterculiaceae and Bombacaceae were infested with viruliferous mealybugs under conditions which would ensure at least 90.3 infection in the cacao controls. .;hen no infection was found in the suspected alternative host after extensive testing it was assumed to be incapable of being infected.