﻿200 MESSES. A. E. ANDREW AND T. E. G. BAILEY [May I9IO, 



in thickness of the lower sandstone division. Near Mpata the latter 

 has a thickness of nearly 1000 feet, while west of the INTyika Plateau 

 (D, fig. 2, p. 193) the coal-group rests directly upon the basal conglo- 

 merates, here about 100 or 200 feet thick. 



The divisions into which the Karroo of Northern JNTyasaland has 

 been separated will now be described in more detail, and subsequently 

 a brief account will be given of the various localities in which these 

 beds occur. 



Lower Sandstone and Basal Conglomerate Division 

 (Division 1). — The basal conglomerates are found in various districts, 

 and, in consequence of the prevalent easterly dip of the beds, occur 

 as a rule along the western margins of the isolated troughs and 

 basins rilled in with Karroo rocks. The immediately underlying 

 gneiss or granite is often weathered, and in one locality was seen to 

 be traversed by cracks and fissures infilled with mudstone. 



The conglomerates either lie directly upon the gneiss, or are 

 separated from the same by fine to medium-grained, white or, more 

 rarely, red sandstones. The conglomerates are in some cases well 

 bedded, and consist of subangular to rounded pebbles of gneiss and 

 granite, while in other cases they are represented by lenticular 

 masses of coarse boulder-bed in which large boulders, somewhat 

 rounded at times, lie side by side with smaller stones and pebbles 

 in a fine to coarse-grained sandy matrix. These masses are usually 

 interbeclded with sandstones or finer conglomerates. In the Mwapo 

 area (C, fig. 2, p. 193), and again west of the Nyika, a curious type 

 of bed is found associated with the conglomerates. This consists 

 wholly of angular fragments of vein-quartz scattered through 

 a sand}' matrix, and closely resembles the weathered soil littered 

 with quartz-fragments which covers the hill-slopes of Nyasaland 

 at the present time. It is probably the remanie product of a 

 highly weathered surface-soil formed in early Karroo times. 



The coarse boulder-beds consist of ill-assorted material, great 

 blocks 2 feet or more in length lying side by side with small stones. 

 In some cases, the boulders appear to be completely separated by 

 the sandy matrix. These boulder-beds may, in certain respects, be 

 compared with the glacial beds of the Dwyka. We have, however, 

 failed to note any single instance of a boulder showing facetting or 

 striation ; while, on the other hand, boulders sometimes occur 

 which have a distinctly waterworn appearance. Moreover, the 

 interbedded conglomerates of finer grain were obviously deposited 

 by water. Taking all the facts into consideration, it appears 

 probable that the basal conglomerates of Nyasaland are torrential 

 deposits, and not of direct glacial origin. This accords with the 

 opinion expressed by Dr. Bornhardt as to the origin of the basal 

 Karroo conglomerates in German East Africa. 



Lower Sandstones of Division 1.— The basal conglomerates 

 are usually overlain by a group of beds varying greatly in thickness, 

 but characterized by the presence of fine- to medium-grained sand- 



