As we pushed on westwards, we found that the river makes a little southing, and some reaches were deeper than any near the sea; but when we had ascended about 140 miles by the river’s course from the sea, soft tufa rocks began to appear; ten miles beyond, the river became more narrow and rocky, and when, according to our measurement, we had ascended 156 miles, our further progress was arrested.  We were rather less than two degrees in a straight line from the Coast.  The incidents worth noticing were but few:  seven canoes with loads of salt and rice kept company with us for some days, and the further we went inland, the more civil the people became.

When we came to a stand, just below the island of Nyamatolo, Long. 38 degrees 36 minutes E., and Lat. 11 degrees 53 minutes, the river was narrow, and full of rocks.  Near the island there is a rocky rapid with narrow passages fit only for native canoes; the fall is small, and the banks quite low; but these rocks were an effectual barrier to all further progress in boats.  Previous reports represented the navigable part of this river as extending to the distance of a month’s sail from its mouth; we found that, at the ordinary heights of the water, a boat might reach the obstructions which seem peculiar to all African rivers in six or eight days.  The Rovuma is remarkable for the high lands that flank it for some eighty miles from the ocean.  The cataracts of other rivers occur in mountains, those of the Rovuma are found in a level part, with hills only in the distance.  Far away in the west and north we could see high blue heights, probably of igneous origin from their forms, rising out of a plain.

The distance from Ngomano, a spot thirty miles further up, to the Arab crossing-places of Lake Nyassa Tsenga or Kotakota was said to be twelve days.  The way we had discovered to Lake Nyassa by Murchison’s Cataracts had so much less land carriage, that we considered it best to take our steamer thither, by the route in which we were well known, instead of working where we were strangers; and accordingly we made up our minds to return.

The natives reported a worse place above our turning-point—­the passage being still narrower than this.  An Arab, they said, once built a boat above the rapids, and sent it down full of slaves; but it was broken to pieces in these upper narrows.  Many still maintained that the Rovuma came from Nyassa, and that it is very narrow as it issues out of the lake.  One man declared that he had seen it with his own eyes as it left the lake, and seemed displeased at being cross-questioned, as if we doubted his veracity.

More satisfactory information, as it appeared to us, was obtained from others.  Two days, or thirty miles, beyond where we turned back, the Rovuma is joined by the Liende, which, coming from the south-west, rises in the mountains on the east side of Nyassa.  The great slave route to Kilwa runs up the banks of this river, which is only ankle-deep at the dry season of the year.  The Rovuma itself comes from the W.N.W., and after the traveller passes the confluence of the Liende at Ngomano or “meeting-place,” the chief of which part is named Ndonde, he finds the river narrow, and the people Ajawa.