Summary and Conclusions. 



That part of British East Africa which borders on the Athi Plains, 

 and extends westwards from Machakos up to the eastern edge of 

 the Rift Valley, is an undulating country composed of foliated 

 rocks of ancient appearance, among which biotite-gneisses are con- 

 spicuous. Pegmatite-veins cross the foliation, and are not associated 

 with intrusive granites or dyke-rocks, so far as mj- observations go. 



In the neighbourhood of the headwaters of the Turoka Paver, a 

 group of altered sediments has been traced over an area of about 

 100 square miles, and these it is proposed to designate by the name 

 of the Turoka Series. 



A large number of rock-types are found, varying from one 

 purely arenaceous, forming the locally conspicuous hills from which 

 the majority of the headwater streams take their rise, to beds 

 of marble of considerable thickness. Other varieties have been 

 mentioned above, the most important being a series of kyanite- 

 gneisses or schists and a scapolite-garnet rock. A river-bank 

 section has also been described, showing the local succession and the 

 passage, as I believe, between a calc-mica-rock and a biotite-gneiss. 



Near the headwaters of the Turoka River, the dips of the planes 

 of foliation of the various gneisses were consistently north-eastward 

 or east-north-eastward. 



Por permission to publish these notes, I am indebted to the 

 kindness of the Magadi Soda Company, to whom also my thanks 

 are due for allowing much information, incorporated in the 

 accompanying map (p. 536), to appear. The locations of the fault- 

 scarps, the principal of which are shown, have been obtained from 

 the Survey maps of the Magadi Branch Railway, produced under 

 the direction of Mr. T. A. Ross, Resident Engineer, to whom I 

 tender my thanks for much kindness and sympathetic help rendered 

 while I was in East Africa. I am also greatly indebted to 

 Mr. Herbert H. Thomas, Sec.G.S., for kindly preparing the micro- 

 photographs reproduced in the plate. 



EXPLANATION OF PLATE LI. 



Fig. 1. Section showing the relationship of the hornblende-schist to the 

 common type of impure marble. The hornblende-schist is shown on 

 the extreme left, and the marble on the right. The intermediate zone 

 is mainly composed of hornblende after malacolite, calcite, and 

 sphene. Ordinary light. Magnified 10 diameters. 



2. Hornblende-schist. The photograph shows irregular plates of common 



green hornblende, locally replaced by almost colourless diopside, in a 

 clear granular mosaic of plagioclase and quartz. Ordinary light. 

 Magnified 9 diameters. 



3. Eyanite-schist. The rock consists of kyanite, muscovite, biotite, and 



quartz, with subsidiary microcline. The grains are elongated parallel 

 to the direction of foliation. Ordinary light. Magnified 14 diameters. 



4. Garnet-schist related to the kinzigites. The section shows pink garnets 



lapped about by a red-brown mica, and set in a base of clear felspar 

 and quartz. Ordinary light. Magnified 10 diameters. 



5. Epidote-hornblende-gneiss. The rock consists of yellowish epidote 



and dark-green hornblende, often closely associated, in a matrix of 

 quartz and felspar. Sphene is a common accessoi'y. Ordinary 

 light. Magnified 16 diameters. 



