254 THE ZAMBESI. Chap. XII. 



Looking from Garden Island, down to the bottom of the 

 abyss, nearly half a mile of water, which has fallen over that 

 portion of the Falls to our right, or west of our point of view, 

 is seen collected in a narrow channel twenty or thirty yards 

 wide, and flowing at exactly right angles to its previous 

 course, to our left; while the other half, or that which fell 

 over the eastern portion of the Falls, is seen in the left of the 

 narrow channel below, coming towards our right. Both waters 

 unite midway, in a fearful boiling whirlpool, and find an 

 outlet by a crack situated at right angles to the fissure of 

 the Falls. This outlet is about 1170 yards from the w r estern 

 end of the chasm, and some 600 from its eastern end ; the 

 whirlpool is at its commencement. The Zambesi, now ap- 

 parently not more than twenty or thirty yards wide, rushes 

 and surges south, through the narrow escape-channel for 

 130 yards; then enters a second chasm somewhat deeper, 

 and nearly parallel with the first. Abandoning the bottom 

 of the eastern half of this second chasm to the growth of 

 large trees, it turns sharply off to the west, and forms a 

 promontory, with the escape-channel at its point, of 1170 

 yards long, and 416 yards broad at the base. After reach- 

 ing this base, the river runs abruptly round the head of 

 another promontory, and flows away to the east, in a third 

 chasm ; then glides round a third promontory, much nar- 

 rower than the rest, and away back to the west, in a fourth 

 chasm; and we could see in the distance that it appeared 

 to round still another promontory, and bend once more in 

 another chasm towards the east. In this gigantic, zigzag, 

 yet narrow trough, the rocks are all so sharply cut and 

 angular, that the idea at once arises that the hard basaltic 

 trap must have been riven into its present shape by a force 

 acting from beneath, and that this probably took place, when 



