During the six days of delay at Konduchi we occupied a neat hut under the care of the Diwan, ‘Mtu Mku Wambele’ and the good Banyan Premji. We strengthened ourselves by high living, by sea-baths, and by shower-baths in the heavy rain; and we had another hard tussle with the hippopotamus. The nights were remarkably fresh and comfortable; the day-sky was milky white, and a glance at the cool celadon colour of the islet-studded sea was itself a refrigerant. We found nothing remarkable in the village, another Sa’adani: its site is the usual glaring white sand-strip, setting off tall cocoas, that wave in the fine sea-breeze, and mangroves bathed by the flood tides. The coast-colours contrast well with the red ochreous earth of the Mrima, two steep raised beaches which back the jungly flat, and command a view of Zanzibar Island.

One day we were surprised by the abrupt entrance of a youth, eminently North German in aspect, with sandy hair, smooth face, and protruding eyes, flat occiput and projecting ears. He announced himself as Dr Albrecht Roscher, of Hamburg, and he made himself doubly welcome by bringing from Zanzibar the wished-for supplies, letters, and newspapers—for 18 months we had not looked upon a white face save the Albino, or a new print in any form but that of a Low Church tract.

The traveller, who appeared at most 22, applied himself forthwith to the magnetic survey, for which he had been engaged by the Prussian Government. A visit to Mozambique, and a run up coast, had taught him everything learnable about East Africa. He despised the dangers of climate, against which he was cautioned: having hitherto escaped fever, he held himself malaria-proof, and he especially derided our advice about not wandering over the country unarmed. He lauded to the skies his fellow-townsman Dr Barth. He severely criticised Dr Livingstone; he patronized, in a comical way, Herr Petermann; he highly extolled his own book;[[61]] and he wrote to Zanzibar—so we afterwards heard—a far from flattering estimate of our qualifications as travellers. He stayed with us two days, and then departed northwards, intending to make Mbweni, the Booamy or Bovamee of Dr Krapf, the village at the mouth of the Panga-ni river. Thence he crossed to Zanzibar Island, and, after scant preparations, he landed at Kilwa: in 1859 he marched through Uhiao upon the Nyassa water. He reached it after long delays, caused by almost constant illness, on November 19, 1859, about two months after Dr Livingstone, who first saw it on Sept. 16. As he was walking without weapons, two of his four Africans shot their arrows into his back. This happened in S. lat. 12° 40′, and at a short distance from the lake’s eastern shore. The assassins were sent in irons to Zanzibar by the chief of the country, who feared retribution, and on August 23rd, 1860, Captain Grant saw them beheaded outside the Fort.

On February 10, 1859, we set sail in the shabbiest of Batelas with a cabin like a large drawer, hot as a native hut, and full of vermin. The skipper had neglected to lay in wood and ballast, we heeled over unpleasantly, and the drinking water stood in an open cask, no joke, considering that the action of a special infectant was to be feared, and that the germs of cholera poison are so easily conveyed in liquids and in dust. Two of the ‘sons of water’ at once died of the disease, two others were taken ill, and Caetano appeared to be sickening. The latter recovered, but after three days our crew of seven was reduced to three, including one, Taufiki, who survived the attack, and who regained health at Kilwa. We could do nothing but bury the unfortunates, so sudden and foudroyant were the attacks, and the scanty personnel was not good for much amongst bad reefs.

Our course lay past the settlements of Msásáni and Mágogoni and the little Mbwezi river to Mbuámáji, ‘rain water,’ in the Mission map called Mburomaji, and vulgarly Boromaji. The little port-village with jungle rolling up to the walls, and anchorage defended by the Sinda Islets, is a favourite entrance to the East African interior. South of this point the coast people are called Watu wa Rufiji, or Rufiji folk. The next night was passed in an open roadstead off Rási ya Ndege—Bird Point—the S. Western portal of the Zanzibar channel, a well-wooded red rise. We then coasted along a low and forested shore sighting Ra’as Kimbizi and Point Puna, which can hardly be called Point or Cape. Khwale (partridge) Island, a link in the long chain of little ‘inches’[[62]] which runs parallel with the coast to Kilwa, showed the usual physiognomy, coralline ledges, yellow sands, and tufted verdant trees: the pretty little patch is said to abound in hippos. Koma, the next inhabited islet, gave us a few cocoas, but no game; the people, serviles from Kilwa, would not answer our questions without bakhshish. The next day saw us fighting against a strong northerly current, and a sharp struggle was required to make the Kisima-ni (the well of) Máfiyah.[[63]]

The watering-place lies on the westernmost point of Mafiyah, in our maps Monfia, and not to be pronounced with Mr Cooley ‘Mofiji;’ it is the longest island in the Southern Archipelago of the Zangian Seas, and ranking after Zanzibar and Pemba. The anchorage was smooth and deep, allowing the largest ships to ride in safety, and the abysses around it are as usual unfathomed. Pits a few hundred yards from the sea supply the casks with water of a quality somewhat better than usual. The tree-clad island is flat and sandy; its growth is by no means so luxuriant as that of its greater sisters, and the population appears to be scanty. We saw no wild animals but a black monkey and a guinea-fowl. The mean breadth of the Manche is here 9 miles,[[64]] and the bottom is said to be very foul.

Opposite Mafiyah lies the Delta of the Rufiji, Lufiji, or Ufiji river, the Loffih, Luffia, or Loffia of older maps, which was made by them to issue from a great lake: it is a reduced copy of the Zambeze farther south, and a waterway worth exploring, as possibly the future high road of nations into Eastern Africa. The people declare that boats can ascend it for a month, and larger craft for a week—this appears, however, doubtful. The stream, then flooding, overflowed its banks, and its line was marked by heavy purple nimbi with hangings and curtains of grey rain. We anchored off Sumanga, an open roadstead, about four miles south of the embouchure: here the land is low, and the village, on account of the high tides, is built a good mile from the water. It contains some large huts, and the people supplied us with milk, rice, sugar, and custard apples. Cattle, though plentiful, is subject, they say, to murrain, and must often change air. Here probably the Tsetse fly is common, as at Kilwa, where I found a fine specimen, afterwards deposited in the British Museum. At that time its habitat was unduly limited northward to the Valley of the Zambeze river: in after years I met with it upon the coast of Guinea, and MM. Antinori and Piaggia observed it amongst the Jurs of the Upper Nile, whilst Sir Samuel Baker saw it in the country of the ‘Latookas,’ 110 miles east of Gondokoro (N. lat. 4° 55′). It will probably be found scattered in patches, especially of lowland virgin-forest, throughout Intertropical Africa.

M. Guillain, again by solely considering distance, would place Rhapta, ‘the last mart of Azania,’ at the ‘embouchure de l’Oufidji;’ while the older geographers prefer Kilwa. Ptolemy, I have said, mentions three places of that name, to the north a river in E. long. 72° and S. lat. 7°; a city in E. long. 71° and S. lat. 7°, therefore lying up stream and one degree to the west, and lastly the Rhapta Promontory, in E. long. 73° 30′ 20″ and S. lat. 8° 20′ 5″. I believe them, for reasons given in vol. I. chap. v., to be the Rufiji river, Old Kilwa, and Cape Delgado. Local tradition preserves no trace of an emporium lying up the stream, nor would so exposed a locality have been chosen by the older traders, who invariably preferred the shelter of islands. Dr Livingstone (near Lake Bangweolo, South Central Africa, July, 1868) proposes the Rovuma—so he writes the word Rufuma—as the probable position of Ptolemy’s river Rhapta. This has the same disadvantage as the Rufiji—it places an important point or points at an unimportant site.

We had no sweeps to make head against the river, even for a few miles, and all dissuaded us from attempting exploration at this season. According to the Banyan Jetha, who declared that he had lived 20 years hereabouts, the stream takes its name from the Rufiji village, a little way up its course. He moreover asserted that some 15 days ago a Banyan had been plundered when travelling to the interior, that the Washenzi (savages) were dangerous, and nowise under the authority of Zanzibar; and, finally, that white men would want letters from the Wali of Kilwa, addressed to three Diwans in the Rufiji village B’ánás Hasi, Kangayya and Furiyya, with two up-country sultans, Monga and Dumbo.

The next feature was the low islet of Chole, rich in cattle and hippos: here the Mtepe-craft is superiorly made, as are also the Chinese-like dish covers (Káwá) of dyed and plaited straw. It was followed by the comparatively large and inhabited island, Songo-Songo—the Songa-Songa of M. Rebmann. Here I heard one of the men use a Persian phrase with Kisawahili termination—‘Tumbak nísti’ (for níst), there is no tobacco: it reminded me of a Kentish woman threatening to ‘frap’ her child. Thence about noon (Feb. 15) we sighted Kilwa Kivinjya. It lies at bottom of a broad shallow bay broken by juttings from the land, and backed by high rolling ground, cleared for mashamba and orange orchards. The mangrove-belted sea ebbs about half a mile, and flows right up to the buildings: we ran close in, and before the tide was out we propped ourselves, like our neighbours, with strong poles.