36 
DETERMINATION OF THE 
[CL IV. 
masses has been disturbed, at different periods, by convulsions 
of extraordinary violence,, as in the Alps and other mountain- 
ous districts, there are instances where the original position of 
strata has been reversed ; but such exceptions are rare, and are 
usually on a small scale, and an experienced observer can 
generally ascertain the true relations of the rocks in question, 
by examining some adjoining districts where the derangement 
has been less extensive. 
In regard to volcanic formations, if we find a stratum of tuff 
or ejected matter, or a stream of lava covering sedimentary 
strata, we may infer, with confidence, that the igneous rock is 
the more recent ; but, on the other hand, the superposition of 
aqueous deposits to a volcanic mass does not always prove the 
former to be of newer origin. If, indeed, we discover strata of 
tuff with imbedded shells, or, as in the Vicentine and other 
places, rolled blocks of lava with adhering shells and corals, 
we may then be sure that these masses of volcanic origin covered 
the bottom of the sea, before the superincumbent strata were 
thrown down. But as lava rises from below, and does not 
always reach the surface, it may sometimes penetrate a certain 
number of strata, and then cool down, so as to constitute a 
solid mass of newer origin, although inferior in position. It is, 
for the most part, by the passage of veins proceeding from 
such igneous rocks through contiguous sedimentary strata, or 
by such hardening and other alteration of the overlying bed, as 
might be expected to result from contact with a heated mass, 
that we are enabled to decide whether the volcanic matter was 
previously consolidated, or subsequently introduced. 
Proofs by included fragments of older rocks. 
A geologist is sometimes at a loss, after investigating a dis- 
trict composed of two distinct formations, to determine the 
relative ages of each, from want of sections exhibiting their 
superposition. In such cases, another kind of evidence, of a 
character no less conclusive, can sometimes be obtained. One 
group of strata has frequently been derived from the degrada- 
