NIAGARA GROUP. 
91 
nearly fluid before the change took place. Masses of the coral often present cavities, from 
which crystals have been dissolved, leaving the spaces with their sides as plane and smooth as 
if cut with some sharp instrument- These examples show how limited maybe the influence 
producing such changes. No effect appears to have been produced in the least beyond the 
edge of the crystal, or the space it once filled ; but by what means solution could take place 
in right lines, it is difficult to explain. The coral in many such cases has scarcely suffered 
any change from its living state, presenting the same porous structure as recent specimens. 
27 . 
The illustration represents a specimen of Porites with linear cavities, from which crystals 
of selenite or some other mineral have been dissolved. The cavities here shown are their 
actual size; in this instance being small, but in others they are several inches long ; and I have 
seen a large hemispheric mass of coral completely divided by a plate of selenite, half an inch 
thick, its limits perfectly defined. Some portions of this selenite still retained the markings 
of the coral, which apparently was not entirely converted to the sulphate. 
An appearance very similar to this occurs in the Water-lime group, and also in the shaly 
limestone of the Onondaga-salt group. Numerous linear cavities present themselves upon the 
surface, generally offering no evidence of having been filled ; but in some instances, where 
there has been no exposure, and moisture has had no access, they are still filled with crystals 
of sulphate of baryta. Specimens have been obtained where the exposed edges were covered 
with cavities, while the fresh surface presented innumerable acicular crystals imbedded in the 
rock. In these cases, we may suppose the crystals to have been segregated during the so¬ 
lidification of the mass. But we cannot account in the same way for the cavities in the coral, 
