Plant Impressions or Incrustations . 381 
botanists have used this method extensively. So far Schulze’s method 
has been used almost exclusively in the study of Mesozoic plants ; Ruth, 1 
however, has shown that it can be employed in the investigation of Carboni¬ 
ferous incrustations. 
Now most fossil plants of this encrusted type occur in rocks which 
have been formed by deposition of sediment in some lake or estuary and 
have a layered structure, or, in other words, exhibit planes of bedding. The 
rock splits or cleaves more readily in a direction parallel to these planes 
than in any other direction. 
As the plant fragment is drifted and sinks to the bottom it will 
generally, if it is a leaf, tend to lie in the plane of the bedding, and that is 
what is usually found. Occasionally, however, it may settle down and be 
fixed in a position oblique to the plane of bedding. The presence of a leaf 
or frond often determines the plane along which the rock cleaves ; there 
, seems to be a surface of weakness due to the discontinuity at the surface of 
the fossil. Very often when a leaf occurs embedded at an angle with the 
plane of the bedding the rock cleaves along the surface of the fossil and not 
along the bedding plane of the rock. The split will occur over either the 
upper or lower surface of the plant, whichever offers least resistance to 
cleavage; but, other conditions being equal, it will occur over the most even 
of the two surfaces. 
On the whole the upper surface of a leaf, in the mature state 
usually convex, is the most uniform. The under surface may have promi¬ 
nent veins, hairs, or scales, and the majority of the stomata. On this 
supposition, therefore, we should expect to find that in the majority of 
specimens the upper surface of the leaf is exposed on cleavage of the 
rock in which it is embedded. On the whole this seems to be borne out 
after examination of a large number of such fossils. If the plane of 
cleavage did by any chance pass over the lower surface of a leaf which had 
hairs, scales, or other appendages, these would probably be shorn off and 
remain embedded in the other half of the block, and would not be repre¬ 
sented on the surface of the specimen except by very minute scars, or 
possibly depressions, so that some of the lower surface features would 
escape notice even though the lower surface was exposed. 
By a method of transferring the fossil on to a transparent base I have 
been able to examine both surfaces. 
There is a very common fern-like frond occurring in Carboniferous 
strata, Dactylotheca plmnosa , Artis sp., the specific name having been given 
to it presumably on account of the feathery appearance of the multipinnate 
frond. In the commonest type of specimen of this plant the surfaces of the 
pinnae and pinnules are convex and smooth. On examining the under 
surface by the transfer method I found that it was covered with long hairs 
1 Huth, W. (1913). 
