One of the most fundamental problems to the physical geographer is the study of the energy and water exchanges at the surface. Of major importance to these exchanges is the knowledge of the radiation. In spite of the significance of this parameter, no work has been done in presenting the spatial distribution of solar radiation, neither has there been much done in determining or presenting the spatial distribution of net radiation in East Africa in general and Uganda in particular. The only direct studies available are studies which tend to relate cloudiness and sunshine to incoming insolation for particular locations see for example, Rijks and Hudley, 1964, Rijks, 1968; Woodhead, 1968a, 1968b.
It has always been acknowledged that monsoons have a pronounced influence on the movement of dhows between the East African coast and the peripheral lands of the Arabian Sea. On the other hand, it has invariably been assumed that monsoons have little, if any, effect on the movement of dhows along the East African coast itself. This article sets out to test this common belief that monsoons have no influence on the movement of local dhows and thus complements the already published analysis on the impact of monsoons on seasonal dhows B.A. Datoo, 1970.
Burundi lies by the western arm of the Great Rift Valley, at the head of Lake Tanganyika, and shares frontiers with Zaire, Rwanda and Tanzania. This study centres on that region of Burundi known as Bututsi, an undulating plateau at an altitude of approximately 1800 to 2200 metres. An outline will be given of the development of markets from traditional times, through the colonial experience to the present. A discussion is held on the nature of the market system of this area in the light of present theory and with reference to the information gathered during fieldwork.
The ""core-periphery"" model that has been recently coined by John Friedmann1 and amplified by Edward Soja2 appears to have come as a reaction against earlier theoretical approaches to regional development. For while former theories of regional development complement each other in many ways, and may each be relevant for specific kinds of planning analysis, none of them can be, as cogently argued by Friedmann,3 accepted as a sharp tool for regional development planning in its comprehensive form. The primary purpose of this article is to outline the major themes of Friedmann's ""core-periphery"" model, as incorporated with the amplifications and modifications of it by Soja, and to present it as a basis for discussion on Patterns and Processes of Spatial Development in Uganda.
Uganda, like most countries of Africa, is educationally under-developed, but the education system has expanded rapidly since Independence in 1962. In that year there were 7,400 pupils in aided secondary schools but by 1970 this figure had risen to over 40,000! This enrolment, however, represents less than 5% of the relevant age group.
The study of place-names typonymy is now well-established as a field of interest to linguists historians and geographers, as well as to the informed layman. In England, for example, the English Place Name Society has published or is in the process of publishing separate volumes for every county listing the names of physical features, settlements, fields and woods, and often even individual streets, and suggesting their origins and tracing their historical development.
Temporary labour migration from Malawi has long been a significant feature of the economy of the country. It has been estimated that by about 1910 approximately 10 per cent of the adult male population of the country were working abroad and by 1921 this figure had probably reached 20 per cent1. Estimates for subsequent dates indicate that the proportion of male absence may have risen to 25 per cent and that for the whole period between 1921 and 1966 probably did not fall below 20 per cent. These figures are significant in themselves but they assume major proportions when broken down on a sub-regional basis. In certain localities the level of male absence reached 75 per cent and even on a District-wide level may have exceeded 60 per cent2.
The steady improvement in data sources and the growing number of University computers with large storage capacity has allowed something of a tide of multivariate methods to wash over geographers in recent years. It is surprising, therefore, that when techniques such as principal components analysis, multiple regression, multi-dimensional scaling, etc., are in frequent use in many departments, one technique of great potential use - covariance analysis - seem; to have been largely excluded from the geographer's portfolio.