1901.] Through the Heart of Africa from South to North. 535 



prominent a position as it does in South Africa — and, in fact, the 

 countries themselves bear a close resemblance one to the other. Nor 

 is this resemblance confined to vegetation. I noticed many small 

 birds in the North which had counterparts in South Africa. So, too, 

 the giraffe, whose habitat in the South is limited by the Zambesi, 

 once more appears on the Nile ; so does the secretary-bird. The 

 North and South African ostrich are identical, though a different 

 species intervenes in East Africa. It has been assumed that the so- 

 called white rhinoceros, JR. simus, never existed north of the Zambesi, 

 and in the early nineties naturalists and sportsmen lamented his 

 total extinction, though it has since transpired that a very limited 

 number still exist ; yet I killed and brought home the skin and skull 

 of one of these animals, which I chanced on near Lado, on the 

 Nile. His measurements and outer appearance coincided in all 

 respects with B. simus, though I understand he differs slightly in 

 minor points, affecting teeth and skull bones. 



In April, 1899, 1 was travelling along the well-defined bed of a 

 river about 100 yards wide and 20 feet deep. A cursory examination 

 showed that this bed was not habitually waterless ; yet the fact that 

 at the end of the wet season, when vleys and streams were full, its 

 deepest depressions were perfectly dry, seemed to suggest that 

 normal conditions did not apply in this case. In two days' time the 

 mystery was solved, though in the process of solution I had perforce 

 to wade knee-deep for three-and-a-half days. 



It appears that at the end of the wet season the Okavango over- 

 flows its eastern bank in 19° S. latitude, and I was meeting the first 

 rush of water. The flat country becomes inundated for miles until 

 the many subsidiary streams thus formed converge into one and 

 escape along the bed, already described, into the Kwando Kiver. 

 Thus the Okavango for some three or four months in the year is 

 actually connected with the Zambesi system by this considerable 

 stream of water, which is known among the natives as Mag' wekwana. 

 It is conceivable that this river has at one time been a Zambesi 

 affluent, and that, on the old bed being choked by drift and sand, the 

 river has made a new one in its present course. However, be this as 

 it may, the diversion of the Okavango through the. Mag' wekwana into 

 the Kwando would call for no greater engineering skill than the 

 cutting of a new, or the re-cutting of the old channel, as the case 

 may be, through the few miles which separate the Okavango from the 

 well-defined portion of the Mag' wekwana bed. The Okavango, a 

 large river some 300 yards wide in places, flows southward into the 

 Kalahari under present conditions. Here the whole of that strong, 

 deep stream disappears, and is wasted. It may be deemed worth 

 while at no distant date to partially or entirely divert its course to 

 the Kwando, and thus, by water communication, tap the extensive 

 rubber-fields contiguous to the upper Kwito, whose first products are 

 now finding an outlet through Portuguese territory. This overflow 

 occurs in the northern confines of the Maiye country — a South 



