43 A NATURALIST IN MID-AFRICA. 



down, or a corduroy road of palmstems will be 

 made along them. 



The formation of canals will probably be the 

 best method. If a channel is cut through the 

 centre of the marsh, and if the material (chiefly 

 papyrus roots) be piled upon either side, the cur- 

 rent, although sluggish, will probably keep it 

 clear ; and with suitable rough machinery, I do 

 not think there would be very much difficulty in 

 cutting it. 



The papyrus has a very curious growth. The 

 knotted roots crawl horizontally over the surface, 

 and these are so close together that dead leaves 

 and mud soon begin to form a soil. A number of 

 marsh-loving plants, chiefly Convolvulus (Ijjomoea), 

 and grass grow on this, and soon turn it into nearly 

 dry ground, on which, in uncultivated districts, 

 fine forests spring rapidly into existence. Thus in 

 such places one passes out of the dismal papyrus, 

 first over a narrow strip of grass and then into a 

 magnificent shady forest which usually covers the 

 level bottom of the valley and climbs a little way 

 up the hill. The trees are unable to resist tires 

 after a short distance from the water, and become 

 scattered or altogether absent, so that the tops of 

 the hills are covered with grass. The hills them- 

 selves are usually granite, and the summits are 

 often covered with a curious hard reddish-brown 

 crust of rock — u laterite ' -which appears to be a 

 product of the intense sun-heat and carbonic acid 



