STONE AGES OF CEYLON. Lata 
themselves after a considerable interval. Now at Kudremalai 
the red earth is bent over an anticline of sedimentary rocks, 
while plateau beds at Minihagalkanda have suffered in a 
somewhat similar way ; but evidence, so far as I have been 
able to gather it, does not allow one to apply these facts to the 
explanation of the earth domes in general. Indeed, folded 
plateau beds seems to be the exception, not the rule. Owing 
to the density of the vegetation and the thick mantle of red 
earth, and also to the accumulations of detrital matter derived 
from the red earth which flank the domes, the internal struc- 
ture of these features is generally difficult to observe. Small 
streams, however, sometimes afford interesting sections ; 
and from them one gathers that the structure of what I shall 
call the typical red earth dome is as follows. 
The central core is a mass of ancient rock ; over and around 
this is a layer of sedimentary material. Above the whole are 
the plateau deposits. 
There can be no doubt whatever that the cores are the 
remains of ancient turtle backs which once ran in lines across 
the country. They are not in themselves folds, but a product 
of past differential weathering. What is the probable history 
of these remarkable elevations? I think it is this. We 
start from the fact that gravel is deposited in water, and 
water shows a preference for valleys, not for hills. Now the 
plateau beds can be traced from one elevation to another, 
but they are missing in the valleys between ; and so we must 
infer that the valleys have been produced since the plateau 
beds were deposited. All this is plain sailing, but we have to 
account for the curious persistence of the hills. When the 
plateau gravels were left dry, they must have formed a vast 
plain, and in places, such as the north of the Island, where 
there are no permanent rivers (and where, by the way, the 
domes are best developed), there was no @ priori reason why 
valleys should have been produced on particular sites ; and 
apparently none whatever why denudations should choose to 
carve out hills exactly where it had done so before. Mani- 
festly then the cause must be, so to speak, an internal one ; 
and since secular movement cannot be invoked, the cause 
must somehow lie in the buried hills themselves. Now the 
