XX.] I LEAVE NYANGWfi. 271 



On the 26tli of August, having bid farewell to Muinyi Du- August, 

 gumbi, I set about getting my men across the river in readiness i^'^^. 

 for starting with Tipo-tipo early the following day, Tangan- 

 yika provided canoes, and assisted me much ; but in the after- 

 noon a bad attack of fever laid him up, and I was thrown upon 

 my own resources. I saw nearly every man away from the 

 l^yangwe side, and then, being very tired, left Bombay with a 

 canoe containing a portion of my kit, to bring the remaining 

 men across after me. 



On landing on the other side, I found the village where we 

 had to camp situated on the bank of a stagnant, muddy back- 

 water, reeking under the sun's rays. The place was inhabited 

 only in the dry season by the fever-proof Wagenya, owing to 

 its being flooded for four or five months of the year. 



In vain that night did I look for Bombay and the remainder 

 of the stores and men ; and when he joined me at noon the 

 next day, Asmani, his chum Mabruki, and another pagazi had 

 deserted, taking with them guns and ammunition, I heard that, 

 the moment I was out of sight, Bombay unloaded the canoe, 

 and coolly returned to the settlement to indulge in a big drink. 

 My bed, cooking gear, provisions, and medicine-chest were all in 

 that canoe, and to the want of them may, in a great measure, be 

 attributed the heavy attack of fever I had after sleeping on the 

 low left bank of the river. 



Fever or no fever, I determined to go on ; and at one o'clock 

 started to meet Tipo-tipo, who had crossed the river rather low- 

 er down. Our road led through many villages, the inhabitants 

 of which were employed either in catching fish in the backwa- 

 ters, or making large egg-shaped pots used for storing palm-oil. 

 JS^early every hut had a pig tied to the door-post, and its odor, 

 combined with that of mud, rotten fish, etc, made a houquet 

 cFAfrique not to be imagined. 



Soon after joining Tipo-tipo we left the river, and began to 

 ascend a gentle slope ; and, passing a market in full swing, ar- 

 rived, after four hours' marching, at the river Kovubu, a large 

 stream, which we crossed on a gigantic fishing -weir bridge. 

 The weir was composed of poles, in many instances over forty 

 feet in length, and from the number used it was evident that a 

 great amount of patient and well-directed labor must have been 



