^{page 468 note 1} Since completing this paper, the writer has referred to an account by F. P. Mennell (Int. Geol. Gong., Compte Rendu, XV Session, South Africa, 1929, vol. ii, pp. 281–2) of sandstone “dykes” in the Sumbadzi River which joins the Rufiji River at the sharp easterly bend about ten miles below the Ruaha-Rufiji confluence in Eastern Tanganyika Territory. The “dykes” occur a few miles above the Sumbadzi-Rufiji junction and exhibit features comparable with those of the “dykes” described here. In width, they range from less than an inch to over two feet. They are usually nearly vertical and contain inclusions of the country rocks which Mennell believes to have fallen from above into the fissures, now occupied by the dyke, at the time they were being filled with sand. The Sumbadzi “dykes” lie in Karrooshales and are associated with irregular, impersistent patches of sandstone.The inclusions are mainly of Karroo shales; the “dykes”, therefore, are atleast post-Karroo. Those of the Dodoma district, on the other hand, are in granite and are believed to be possibly of late Tertiary or early Quaternary age.Google Scholar

^{page 474 note 1} Chamberlin, T. C., and Salisbury, R. D., Geology, London, 1905, p. 490.Google Scholar

^{page 474 note 2} Krige, L. J., “An Interesting Sandstone Dyke in the Old Granite of the Ermelo District,” Trans. Geol. Soc. South Africa, xxxii, 1929, 1930, p. 61.Google Scholar

^{page 474 note 3} Jenkins, O. P., “Sandstone Dikes as Conduits for Oil Migration through Shales,” Bull. Ames. Assoc. Petrol. Geol., xiv, 1930, p. 421.Google Scholar

^{page 474 note 4} Kramer, W., “Dolomite Dikes in the Texas Permian,” Journ. Geol., xlii, 1934, p. 193.CrossRefGoogle Scholar