86 SCIENCE IN AFRICA 



where earth movements occur at frequent intervals, and served as 

 the only link in the records between Cairo and the Cape. During 

 1 930 and 1 93 1 two seismometers were working at Entebbe, oriented 

 at right angles to one another. Two epicentres for the numerous 

 local shocks of Uganda were discovered, one in the Ruwenzori 

 Mountains and one at the north-west end of Lake Albert. This 

 branch of research has its practical application, since the Fort Hall 

 earthquake of 1928 was a fairly serious matter, and a somewhat 

 stronger shock from this epicentre might damage buildings in 

 both Kenya and Uganda. 



The variation in gravity over different parts of the earth's surface 

 has received attention from geophysicists in connection with the 

 structure of the earth. In Africa it is interesting chiefly in the 

 region of the rift valleys. Previous to the war, observations were 

 made in Tanganyika by the Germans, and in 1933-4 Dr. E. C. 

 Bullard, from Cambridge, made a special expedition to the rifts 

 of Kenya and Uganda. A summary of his results is published by 

 the geological department of Uganda (Bullard 1935), and E. J. 

 Wayland remarks in his preface that the work is a valuable con- 

 tribution to the controversy with regard to the nature and origin 

 of rift valleys. Its significance is more than academic, for it has 

 certain bearings on the petroleum-winning possibilities of the 

 Albertine depression. The full results of this investigation have 

 been published by Bullard (1936), and on the same subject the 

 veteran American geologist, Bailey Willis, who visited East Africa 

 recently, has written an important book (1936). This work, which 

 is discussed by Simmons (1937), consists of two main parts, the 

 first giving the general picture and the second detailing characters 

 of the several areas. An hypothesis is advanced for the formation 

 of the whole plateau of eastern Africa and of the rift valleys. The 

 measurement of gravity, if carried out more or less evenly over the 

 whole surface of the earth, would lead to practical results in an- 

 other way. Numerous determinations would enable the true form 

 of mean sea-level to be computed with reference to an adopted 

 mathematical figure. If this form were known, then astronomical 

 determinations of latitude and longitude could be used with con- 

 fidence for the control of certain classes of survey. 



Passing to the subject of Palaontology, it may be fairly stated that 



