Fen, 1893. THE UNDERGROUND WASTE OF THE LAND. 125 
large amount of mineral ingredients compared with some other 
saline waters, daily carry away between three and four tons of solid 
matter, and the removal of this must cause cavities. Perhaps the 
landslips that have occurred from time to time at Bath may be to 
some extent influenced by subterranean movements, caused by the 
filling-in of cavities by the overlying strata. The waters of Burton-on- 
Trent may be cited in reference to chemical erosion, and it has been 
estimated that about 150 tons of gypsum are annually imbibed in 
potations of Burton beer. 
In the more hilly and mountainous regions, where bare rocks 
frequently jut out at the surface, the evidences of destruction are 
apparent. Among the gentler hills and vales of the midland and 
southern counties, the evidences of erosion are to be judged mainly 
by the solid matter carried away by the streams. The matter held 
in solution is due chiefly to subterranean erosion; that held in 
suspension is due almost entirely to superficial erosion. The 
hills are, for the most part, composed of limestones, sands, and sand- 
stones, through which the rain-waters may percolate, until arrested 
by a band of clay, when they issue as springs. Hence the hills are 
not so much subject to superficial denudation as are the vales, for 
these lie mainly in tracts of clay that are directly acted upon by the 
streams that flow across them. The main features of hills have, 
therefore, remained permanent for long ages, though their general 
level may have been continuously, if imperceptibly, lowered. 
The question arises whether the underground erosion may not 
be partly mechanical, even if it be mainly chemical. Underground 
waters that flow on the top of amass of clay must form channels in 
that material. The permanence of springs that issue here and there 
along the foot of escarpments, indicates plainly that the underground 
waters follow definite courses; and the actual outlet of such springs 
may be found at varying levels beneath the plane of division that 
separates the porous and impervious strata, because the springs 
have eroded their channels in places, perhaps, to depths of five or six 
feet. 
Referring to the Central Himalayan region, Mr. C. L. Griesbach 
(4) mentions a limestone which rests conformably on calcareous 
shales, and this, like most limestones, is much jointed ; consequently, 
all the drainage finds its way through the joints into the underlying 
shales. These become disintegrated, and are gradually carried away, 
while the thick limestone band sinks down to the level which was 
formerly occupied by the shales. 
In an article in which I discussed the origin of the Scenery of 
Norfolk, I remarked that on the clayey strata the rainfall must 
accumulate or flow away at once towards lower levels; on the sandy 
and gravelly strata it will sink down until arrested by impervious 
beds beneath. Hence the earliest exposed channels no doubt com- 
menced on the clayey areas that formed the surface. Beneath the 
