138 
Proceedings of the Royal Irish Academy. 
tached to one another nncler the extremely varied and dissimilar forms 
constituting the "canal system." Even the most branching of the 
latter show by their occasional angularities, excavated and rounded 
surfaces, and beaded character, that they are entirely made up of crys- 
talline grains — skeletons of larger groups — reduced and fashioned into 
their present remarkable shapes by the wasting action of a solvent. 
Moreover, isolated grains occur with a thin white crust enclosing their 
translucent substance : others are seen with a portion of their sub- 
stance removed, but the crust remaining and intact ; while close by 
are hollow spheroidal cases identical with the crusts. These facts, 
which are, of course, best revealed by decalcifying the specimens,^' 
prove beyond doubt that the vacancies in the grains were occupied by 
calcite. 
Clearly the grains at one time were altogether composed of mala- 
colite ; and it is equally clear that the calcite now occupying the 
vacancies has replaced the malacolite. The substitution of a siliceous 
mineral by a calcareous one is seen in all its stages ; and it is as self- 
evident a case of pseudomorphism as any that have been recorded. 
The spheroidal grains of malacolite, with their external crust pre- 
served, and enclosing an interior of calcite, are precisely analogous to 
the " crystals of garnet from Tvedestrand, which are wholly calcite 
within, there being but a thin crust of garnet. "f 
Again, the crust itself exhibits most obviously the final stage of 
its waste. Unmistakable portions are seen fixed in the undissolved 
part of the calcareous intercalations ; and when a number of such, 
belonging to different grains, are attached to one another, they give 
rise to irregularly undulating leaf-like expansions, some of which 
strikingly resemble the " curiously curved" configurations detected by 
Dr. Giimbel in a Bavarian ophite, and considered by him to represent 
the canal system" of his so-called Uozoon Bavaricumy% These 
examples are demonstratively fragments of branching varieties of the 
canal system ;" and they must be accepted as completely confirming 
our view of the origin of this feature. Ocasionally, crusts may be 
seen entirely riddled, and approximately simulating in this respect the 
perforated shell-case of a globular poly cystine. § 
Now, considering that the grains of malacolite show themselves in 
every stage of decretion, it clearly follows that in numerous instances 
they have disappeared altogether; and it is equally to be inferred 
that the interstitial calcite, or dolomite— even that forming the layers 
— has replaced a corresponding amount of malacolite. In both cases 
the change may have been effected by the rock having been permeated 
by heated water holding a carbonate in solution. 
* By this process the calcite of the grains is removed, as well as that forming 
the adjacent calcareous layers and interstices, 
f " Dana's Syst. Mineralogy," 5 ed., p. 272. 
J " Canadian ^' aturalist," vol. iii., plate i., fig. 7. 
§ Skeletons of apparently the grains of the amphibole-like mineral also occa- 
sionally occur as thin porous or rudely reticulated fragments. 
