i'ki-:sii)i:\i"r.\i, addki-.ss — sivcxiov n. 45 



one time have run over tlie top uf a tiat region connecting the high 

 groimd that lies to the north of the Dam Hotel and the Alatopos 

 — otherwise along the revealed pre-Karroo plains. 



Other illustrations occur in great numbers. The ]\Iolun- 

 gwane Mountains beyond Essexvale are of an altitude corres- 

 ponding with tlie Alatopos and Crombie's Ridge, with the Mssex- 

 vale granite basin in between. The rapid and recent reduction 

 of tliis circular basin is shown by the rapid fall of Fern Creek — 

 almost a " hanging valley." The banded ironstone ranges of 

 llelingwe. Tabas In\;oga, Victoria, (jhoka, Insiza, Iron Mask at 

 .Mazoe, Simoona, Relingwe, and elsewhere, and the granite of 

 Hillside, AJatopos, Tabas, Inyoni', Shangani, and Victoria suggest 

 the same general principle. 



The exceptions are those mountains of elevation tliat lie 

 along the axis of the spinal movements of the continent, such as 

 near Melsetter, Umtali and Inyanga and Fort Hill in North- 

 ^Vestern Rhodesia. 



The existence of the peneplane is supported by the courses 

 of many of the rivers. The Umniati passes in its middle course 

 over ground that still retains patches of Karroo rock ; beyond, 

 it cuts through ridges of quartzite ; and finally, before its con- 

 liuence with the Zam1)esi, cuts through a high range — a corner 

 of the Urungwe plateau — of gneiss and granite. Yet by a 

 course to the west the Sanyati (as it is then called) could have 

 passed round the mountains in softer Karroo sediments. Here 

 is evidence that its course was decided at a time that the Karroo 

 deposits situate in the region of its middle section were at a level 

 with the arch?ean ranges of its last section, the covering on the 

 latter being earliest removed, and thus, when the river cut down 

 into the underlying gneiss, it gradually drew off and eroderl. and 

 is eroding, the Karroo covering of its centre reaches. 



Cases <uch as this are paralleled by the Gwai, Lubu, and 

 Sengwe and the Limpopo. Somewhat different, but showing 

 how they were directed in their courses over a higher surface, 

 are the Hunyani. which passes through a poort in the range of 

 hills next the railway, the Oueque poort. Sebakwe poort, the 

 Mazoe through the Iron Mask range, the Umfuli. the Kafue 

 through the Niga Nega hills and others, the Zambesi itself drawn 

 through the Kariba, Kebrabasa ,and Lupata gorges. The 

 Lusenfwa is another case from North-\\'estern Rhodesia. 



If. then, the pre-Karroo landscape as now revealed was a 

 l)lain of denudation of remarkable evenness, and if the end of 

 the Karroo age saw another levelling up of all the depressions 

 caused by the partial tilting of the strata of the plateau, we have 

 other problems to face. Was the first the result of glaciation in 

 Rhodesia, as in the Cape Colony, by which the pre-Karroo high- 

 lands, where our plateaux now are, were ground down and their 

 debris carried to fill the depressions of the south? And in the 

 last phase were desert conditions so persistent that they too 

 exercised a levelling influence, and as effectively brought the 

 great final Karroo landscape to another plain of remarkaljle 

 uniformity? 



D 



