The external shape and limits to the size of the individual Eozoön have not been determined with certainty, on account of its indefinite mode of growth, and the manner in which the fossilized masses are connected with the highly crystalline matrix in which they are imbedded; there is no doubt, however, that they spread over an area of a foot or even more, and attained a thickness of several inches. As they seem to have increased laterally by buds which never fell off, they formed extensive reefs; at the same time they had a vertical growth, for in some of the reefs the older portions appear to have been fossilized before the newer were built up on them as a base, exactly like the coral reefs in the tropical ocean of the present day,[10] with this difference however, that shells and other crustaceans are associated with the corals, while no organic body has been found in the Eozoön reefs; nevertheless the Eozoön must have had food. It may therefore be inferred that parts at least of that primeval ocean swarmed with animal life, whose remains have been obliterated by metamorphic action. Carbon (which in the form of graphite both constitutes distinct beds, and is disseminated through the siliceous and calcareous strata of the Laurentian series, as well in Norway as in Canada), may indicate the existence of vegetation in the Eozoön period. The Eozoön is by no means confined to Canada and central Europe. The serpentine marble of Tyree which forms part of the Laurentian system on the west of Scotland, and a similar rock in Skye, when subjected to minute examination, are found to present a structure clearly identical with that of the Canadian Eozoön. And the like structure has been discovered by Mr. Sanford in the serpentine marble of Connemara, known as Irish green. The age of that rock however, is doubtful: for when it was discovered to contain Eozoön, Sir Roderick Murchison who had previously studied its relations was at first inclined to believe it belonged to the Laurentian series; now however, he considers the Connemara marble to be of the Silurian age. ‘If this be the case it proves that the Eozoön was not confined to the Laurentian period, but that it had a vast range in time, as well as in geographical distribution; in this respect corresponding to many later forms of Foraminifera which have been shown by Messrs. Parker and Rupert Jones to range from the Trias to the present epoch.’[11] The Carpenteria found in the Indian seas forms a link between the Foraminifera and Sponges. The shell is a minute cone adhering to the surface of corals and shells, by its wide base which spreads in broad lobes. Double-walled chambers and canals form a spiral within it, and are filled with a spongy sarcode of a more consistent texture than the sarcode of the Foraminifera, which in the larger chambers is supported by siliceous spicules similar to those which form the skeletons in sponges. CLASS III.—SPONGES.