g8 SCIENCE IN AFRICA 



levels of hilly country are better for cultivation in wet years, and 

 the lower levels in dry years, the prediction of an abnormally wet 

 or dry year would be of the greatest value to agriculture. It is to 

 be hoped, therefore, that the system adopted in Southern Rhodesia 

 may be extended to other territories. A section on the climate of 

 Southern Rhodesia, prepared in the meteorological office by C. L. 

 Robertson and N. P. Sellick, is included in Koppen and Geiger's 

 great work (1927). 



In Angola there is an important meteorological and magnetic 

 station at Loanda, but little organization in the hinterland. The 

 literature is not extensive, but two valuable papers by Mar- 

 quardsen (191 7) and Roque (1925-26) are cited in the biblio- 

 graphy. 



Mozambique has an official service centred at the Campos 

 Rodrigues Observatory, Lourengo Marques, where upper winds 

 are observed and daily weather charts are plotted. There are 

 three first order stations, all situated on the coast, at Inhambane, 

 Quelimane, and Mozambique, and another at Beira under control 

 of the Mozambique Company. There are a number of second 

 order and rainfall stations, but their distribution inland is some- 

 what irregular. The annual and monthly reports are models which 

 might well be followed elsewhere. The climate is described in 

 Koppen and Geiger (1927) and by Peres (1931). 



Madagascar has an official service and a well-equipped laboratory 

 at Tananarive. Short-term forecasting and the broadcasting of 

 weather reports are highly developed. The island has been studied 

 thoroughly, and practically all that is known about its climate and 

 weather is embodied in the standard work by Poisson (1930). For 

 no country on the mainland of Africa, except Egypt, is there such 

 a complete account. Tropical cyclones and thunderstorms are both 

 formidable manifestations, and are discussed by the author with 

 remarkable acumen. Appendices deal with native weather-lore, 

 and the influence of the moon on barometric pressure and other 

 elements. It is pointed out that Madagascar is exceptionally well 

 situated for the study of minor lunar influences on the atmosphere. 

 This, of course, lends no support to the popular belief that the 

 weather is controlled by the phases of the moon. 



