ON AN INVESTIGATION OF THE BATOKA GORGE. 299 
east and west over wide areas ; and the basalts are also fractured in the 
same direction by numerous strongly marked vertical planes along 
which the rock is more or less crushed and veined, that evidently mark 
lines of faulting. It is to the presence of these joints and faults that the 
sudden and acute bends of the Zambesi and its tributaries in their gorges 
below Victoria Falls are mainly due, as the erosive agencies work selec- 
tively upon these weak planes, so that the running waters tend to sink 
deeply along them. This result of differential erosion is especially marked 
when the weak planes run transverse to the general direction of the 
stream. Thus at the Falls, the awe-inspiring chasm into which the 
Zambesi is precipitated has been hollowed out along a vertical east-and- 
west fracture, which, as I discovered on descending to the bottom at the 
* Hastern Recess,’ is filled with calcite and other soft vein-material between 
walls of shattered and partly decomposed basalt. 
The ‘ Deka Fault,’ by which, as already mentioned, the Batoka Basalts 
are abruptly truncated along the Deka River, proved in the portion 
examined to strike E.N.E. to W.S.W., and must, I think, have its down- 
throw on the N.N.W. or basalt side. The downthrow was not, however, 
directly deducible from the evidence yielded by the fault itself, but from 
evidence obtained in the Wankie Coalfield and on other grounds that will 
be discussed in a later communication. This fault appears to be the most 
important feature in the structural geology of the region. 
From the Wankie Coal Measures I collected some fragmentary plant- 
remains, and among these Vertebraria has been recognised by Mr. A. C. 
Seward, F.R.S., who kindly examined them for me. This supports the 
suggestion of Mr. Molyneux ! that the Wankie Coal Measures, like other 
Rhodesian coaltields, are of Permo-Carboniferous age. The Batoka 
Basalts, by their position in regard to these rocks, are shown to be newer, 
but their age within narrower limits has not yet been satisfactorily 
determined. 
The surface-deposits of sand, sandy limestone, and cellular quartzite 
which locally overlie the basalts in the traversed part of the Zambesi 
basin are evidently analogous to the superficial formations of the Kalahari 
Desert recently described in great detail by Dr. 8S. Passarge.2 Their 
mode of occurrence in this region does not, however, seem to indicate the 
regular sequence of events deduced by Dr. Passarge from his study of the 
Kalahari formations ; for, although the thick red sand must have been 
accumulated under conditions different from those which now prevail, the 
patches of limestone and quartzite which I examined appeared to me to 
be assignable to local circumstances that still exist. 
I was impressed throughout my traverses with the singular scantiness 
of alluvial deposits in places where the conditions seemed favourable to 
their extensive accumulation by fluviatile agency. It was also remark- 
able that, although the region was tenanted until very recently by extra- 
ordinary numbers of the larger mammals, not a single bone or tooth of 
these animals rewarded my search in the dry river-beds and other likely 
spots. The peculiar conditions of climate that retard the accumulation 
of alluvium seem also very rapidly to destroy even the less perishable 
relics of organic life. 
Fuller information on severa] of the points touched upon in these 
1 Quart. Journ. Geol. Soc., vol. lix. (1903), p. 283. 
? Die Kalahari. Berlin, 1904. 
