_5th September, 1866._--Our march is along the shore to Ngombo promontory, which approaches so near to Senga or Tsenga opposite, as to narrow the Lake to some sixteen or eighteen miles. It is a low sandy point, the edge fringed on the north-west and part of the south with a belt of papyrus and reeds; the central parts wooded. Part of the south side has high sandy dunes, blown up by the south wind, which strikes it at right angles there. One was blowing as we marched along the southern side eastwards, and was very tiresome. We reached Panthunda's village by a brook called Lilole. Another we crossed before coming to it is named Libesa: these brooks form the favourite spawning grounds of the sanjika and mpasa, two of the best fishes of the Lake. The sanjika is very like our herring in shape and taste and size; the mpasa larger every way: both live on green herbage formed at the bottom of the Lake and rivers. _7th September, 1866._--Chirumba's village being on the south side of a long lagoon, we preferred sleeping on the mainland, though they offered their cranky canoes to ferry us over. This lagoon is called Pansangwa. _8th September, 1866._--In coming along the southern side of Ngombo promontory we look eastwards, but when we leave it we turn southwards, having a double range of lofty mountains on our left. These are granitic in form, the nearer range being generally the lowest, and covered with scraggy trees; the second, or more easterly, is some 6000 feet above the sea, bare and rugged, with jagged peaks shooting high into the air. This is probably the newest range. The oldest people have felt no earthquake, but some say that they have heard of such things from their elders. We passed very many sites of old villages, which are easily known by the tree euphorbia planted round an umbelliferous one, and the sacred fig. One species here throws out strong buttresses in the manner of some mangroves instead of sending down twiners which take root, as is usually the ease with the tropical fig. These, with millstones--stones for holding the pots in cooking--and upraised clay benches, which have been turned into brick by fire in the destruction of the huts, show what were once the "pleasant haunts of men." No stone implements ever appear. If they existed they could not escape notice, since the eyes in walking are almost always directed to the ground to avoid stumbling on stones or stumps. In some parts of the world stone implements are so common they seem to have been often made and discarded as soon as formed, possibly by getting better tools; if, indeed, the manufacture is not as modern as that found by Mr. Waller. Passing some navvies in the City who were digging for the foundation of a house, he observed a very antique-looking vase, wet from the clay, standing on the bank. He gave ten shillings for it, and subsequently, by the aid of a scrubbing brush and some water, detected the hieroglyphics "Copeland late Spode" on the bottom of it!