This paper is concerned with illustrating some of the problems and potential insights which a Markov chain model can offer to the analysis of inter-regional migration in the countries of tropical Africa. Following a brief discussion of the assumptions underlying the model within the context of migration studies in tropical Africa, an analysis of inter-regional migration in Uganda is offered. The results indicate in a general way some aspects of the spatial organization of Uganda and emphasize the relative importance of periphery-to-core movements in possibly effecting a net redistribution of population over the long term. A testable proposition on the effect of pre-migration social networks upon migrant residential patterns in the capital city is also outlined. The continuing need for an improved and enlarged factual base upon which migration research can proceed overshadows the discussion. A partial evaluation of place of birth data as an indicator of some attributes of an inter-regional migration system is undertaken in the appendix.