covered with clambering plants in precarious
positions upon a denuded base, or lying bare
in the beds o f the stream, exposed to the action
o f the running water. Pebbles also, lodged on
small shelves o f rock in the streams, borne
thither b y their force during rainy seasons,
attest the nature o f the formations higher up
their course. Among these, we saw varieties o f
quartz, p ro p h y ry , greenstone, dark g re y shale,
granite, hematite, and purple jasper, chalcedony,
and other gravels.
T h e ro ck -sa lt discovered has a large mass
exposed to the action o f the stream. In its
neighbourhood is a greyish tufa, also exposed,
with a brown mossy parasite running in threads
over its face.
W o od is abundant in large clumps soon after
passing Kikoka, and this feature o f the landscape
obtains as far as Congorido. T h e Wami has a
narrow fringe o f palms on either bank; while,
thinly scattered in the plains and less fertile parts,
a low scrubby brushwood, o f the acacia species,
is also seen, but nowhere dense. A lon g the
base and slopes o f the mountains, and in its
deep v a lle y s , large trees are v e ry numerous,
massing, at times, even into forests. T h e extreme
summits, however, are clothed with only grass
and small herbage.
Mpwapwa has also some fine trees, but no
forest; the largest being the tamarind, sycamore,
cottonwood, and boabab. The collection o f
villages denominated b y this title lies widely
scattered on either side o f the Mpwapwa stream,
at the base o f the southern slope o f a range
o f mountains that extends in a sinuous line from
Chunyu to Ugombo, I call it a range because
it appeared to be one from Mpwapwa; but in
reality it is simply the northern flank o f a deep
indentation in the great mountain-chain that e x tends
from Abyssinia, or even Suez, down to
the Cape o f Good Hope. A t the extreme eastern
point o f this indentation from the western side
lies L a k e Ugombo, just twenty-four miles from
Mpwapwa.
Desertions from the expedition had been frequent.
A t first, Kacheché, the chief detective,
and his gang o f four men, who had received
their instructions to follow us a d a y ’s journey
behind, enabled me to recapture sixteen o f the
deserters ; but the cunning Wangwana and Wa-
nyamwezi soon discovered this resource o f mine
against their well-known freaks, and, instead o f
striking east in their departure, absconded either
south or north o f the track. W e then had detectives
posted long before dawn, several hundred
yards away from thè camp, who were bidden
to lie in wait in the bush, until the expedition
had started, and in this manner we succeeded
in repressing to some extent the disposition to
desert, and arrested v e ry many men on the point