F. P. 3[ennell—0n E/iodesimi LaterUe. 351 



regarded as swamp from November to March, though many of them 

 are quite dry for the rest of the year. It is, of course, difficult to 

 ascertain the nature of the changes going on beneath the surface of 

 the soil in a present-day swamp, though by digging in the dry season 

 I have found that deposits of a lateritic nature were disclosed at 

 a depth of a couple of feet under the soil. This last is always some- 

 what peaty, so that there must be abundance of carbonic acid, and 

 possibly other stronger acids of organic origin, to aid in the chemical 

 changes to which the production of laterite is due. 



Mode of Occurrence. — The most conspicuous occurrences of laterite 

 are those in which it may be seen lying about all over the surface or 

 showing up through the soil in large blocks. These are due, of course, 

 to changes in the physical geography of marshy tracts. Erosion along 

 a particular line of drainage has led to the formation of a well-defined 

 channel which has drained the swamp and incidentally carried off 

 most of the soil which originally concealed the laterite. There are 

 some very interesting canon-like valleys cutting through the com- 

 paratively level ground between some of the lines of hills in the 

 Matopos, e.g. to the north of Mount Impu. These little canons are 

 usually about 15 to 30 feet deep, with almost vertical sides, which 

 they owe to the crust of hard laterite overlying the softer material 

 and decomposed granite in which the streams have cut their channels. 

 When the hard bed-rock is reached the process of widening the 

 valleys becomes inevitable, and accordingly most deposits of laterite 

 occur as a capping to terraces on either side of a valley. The slope 

 of these terraces is sometimes considerable. This is not due to erosion, 

 but to the curious feature shown bj" so many swamps in this country 

 (Rhodesia), of occurring on sloping instead of on level ground. Some 

 of the Matopo swamps which show standing water everywhere have 

 quite a considerable inclination. 



The section given shows the general mode of occurrence that we 

 have described. It is based on a sketch made about 10 miles south- 

 east of the AYorld's View in the Matopos. 



Underlying Rochs. — It is perhaps well to emphasize the fact that 

 the accumulation of laterite does not appear to be dependent to any 

 extent on the nature of the underlying rocks. Mr. Kilroe does not 

 seem to realize this important fact ; indeed, the idea seems prevalent 

 that basalt or dolerite is the general source, though it may be noted 

 that Dr. Maclaren's figured example ' is stated as overlying quartz- 

 biotite-schist. Laterite occurs, to my personal knowledge, in Rhodesia, 

 resting on nearly every possible description of rock. The following 

 may be cited, the locality of one such occurrence being given in 

 brackets in each case : — 



Sedimentary : Gravel (Somabula). 



Sandstone (North of Gwelo). 



Limestone (Gwanda). 

 Igneous : Granite (Matopos). 



Basalt (Shiloh). 



Picrite (Selukwe). 



' Geol. Mag., 1906, p. 538. 



