440 Scientific Proceedings, Royal Dublin Society. 



globular or«spherulitic bodies. The ruder flakes are mostly made 

 of a material which has the following microscopical aspect: — 

 Imbedded in a dark-brown, non-transparent matrix lie numerous 

 fragments of quartz and a few of felspar. Rounding of the edges 

 by attrition is not visible ; on the contrary, the minerals have pre- 

 served their fracture-lines remarkably well. Their bizarre forms 

 remind the observer of fragmental rocks, like tuffs or breccias, 

 rather than of sandstones. The " burnt Sienna " coloured matrix 

 appears to be formed of highly siliceous limonite ; it is, so to say, 

 the cement which binds the fragments together. Such siliceous 

 limonites are not at all uncommon ; and Walchner has described 

 an occurrence at Kandern, in Baden, with as much as 21 % of 

 silica. 1 The quartz is mostly the quartz characteristic of granitic 

 rocks, and contains numerous fluid cavities, also inclusions of zircon 

 and dark mica (biotite.) But some of the quartzes are evidently 

 derived from the metamorphic areas ; for they show the pheno- 

 menon of granulation in its early stages and the optical anomaly 

 due to compression known as unclulose extinction. The little 

 felspar present is mostly orthoclase ; but traces of plagioclase are 

 observable, though very infrequent. All the fragments are re- 

 markably fresh ; but the geological bearing of this feature will be 

 shown at the end of the paper. 



Rocks similar to this are very common in East Africa. I pos- 

 sess, and have previously described, 2 some specimens from Kahe- 

 Aruscha (a locality situated between the Kilima-Ndjaro massif 

 and Pangani upon the East Coast), which are practically identical 

 with those referred to above. The similarity is, in fact, so striking 

 that we are compelled to conclude that they are derived from one 

 and the same formation. 



. Flakes bearing evidence of human workmanship are not un- 

 common in the Nile valley. The mode of occurrence, and the 

 relations of the flint implements, are well discussed by A. Jukes- 

 Browne in his papers in the Cambridge Antiquarian Society's 

 Communications, and in the Journal of the Anthropological Insti- 



1 Naumann-Zirkel, Mineralogie, 1885, p. 415. 



2 J. S. Hyland, Ueber die Gesteine des Kilimandscharo und dessen Umgebung, 

 Tschermak's Min. u. petr. Mitt, x., Heft iii., p. 266. 



