12 



of opinion as to the further details of the structure of Stromatopora, and these have led to 

 equally wide differences of opinion as to the affinities and systematic place of the genus. 



According to McCoy (Pal. Foss., p. 12,) the vesicular tissue of Stromatopora is composed 

 of " minute curved calcareous plates," which he compares with the coenenchyma of Palceo- 

 pora and Fistulipora. He also states that the upper surface is occasionally marked " with 

 extremely obscure, distant, quincuneially arranged, small pits," which he appears to think may 

 represent the corallites in the above-mentioned and other allied genera. 



Prof. Hall agrees with McCoy in referring Stromatopora to the Cwlenterata, and in 

 placing it in the neighbourhood ofTubipora (Pal. N. Y., Vol. II. p. 135.) He considers that 

 the fossils of this genus are composed of " minute cylindrical tubes with considerable space 

 between ; and that the laminated structure arises from thin layers of calcareous matter de- 

 posited and filling the spaces between, and enclosing the tubes." 



If the Stromatopora polymorpha of Goldfuss (Petref. PI. Ixiv, figs. 8a, 8/) be a genuine 

 Stromatopora, then this eminent palaeontologist long since recognised the fact that Stromatopora 

 is a genus of Sponges. There is, however, some doubt on this point ; since neither his 

 description, through unusually precise, nor his figures demonstrate the existence in this species 

 of the minute structure peculiar to Stromatopora. Nor is this point cleared up satisfactorily 

 by the description given by McCoy of this same species (Pal. Foss. p. 65) ; whilst D'Orbigny 

 makes Stromatopara polymorpha, Goldfuss, the type of his genus Sparsisponyia. There is, 

 however, much resemblance between S. polymorpha and some of the species of Stromatopora 

 from the Devonian Rocks of Ontario, especially S. granulata, Nich. 



My own investigations of a very extensive series of examples from the Lower and Upper 

 Silurian formations, and from the Devonian Rocks, have led me to the opinion that the genus 

 Stromatopora is clearly referrible to the Spongida, and that it should be placed amongst the 

 Caldspongice, a group represented by many and varied forms both in past time and at the 

 present day. The reasons for this belief may be summed up as follows : 



a. The fundamental structure of Stromatopora is by no means inconsistent with the belief 

 that it belongs to the Caldspongice. It does not consist of reticulated calcareous spicula, as 

 in the more typical members of the group ; but neither 'iocs it consist of a vesicular tissue 

 composed of " minute curved calcareous plates " (McCoy), which could be in any way com- 

 pared with the vesicular coenenchyma of many tabulate corals. On the contrary, it consists of 

 successive calcareous layers, which may be regarded as composed of an amalgamated system 

 of horizontal spicules, separated by intervals, and kept apart by a vertical system of delicate 

 calcareous props or rods, giving rise to a system of more or less quadrangular cells. The hori- 

 zontal laminae are upon the whole continuous, but they sometimes sub-divide and inosculate ; and 

 the vertical pillars are decidedly irregular, being sometimes inclined at various angles, and not 

 being placed at uniform distances in all parts of even the same specimen. Some of the ver- 

 tical rods pass continuously through several laminae and the interspaces between them ; but 

 the greater number are confined entirely to the interval between two successive laminae, and 

 are not continuous, nor correspond with those in the interval immediately above or below. 

 There is no ground, so far as I am aware, for the supposition that these vertical pillars are 

 perforated, or are of the nature of tubes inhabited by the separate zooids of a colony ; indeed 

 in the forms which occur in the Corniferous limestone there is the strongest positive proof 

 that this is not the case, and that they are of the nature of solid rods or dissepiments. There 

 is nothing in the fundamental tissue or groundwork of Stromatopora, as above described, 

 which would necessarily preclude us from referring the genus to the Spongida ; nor can any 

 stress be laid upon McCoy's argument that these organisms cannot be sponges on account 

 of their possessing a rigid and inflexible skeleton, since similar reasoning would compel us to 

 remove from the Spongida a vast number of forms the zoological position of which is beyond 

 doubt. At the same time, if Stromatopora consisted wholly of the laminated and reticulated 

 tissue just described, and possessed none of those openings which are so characteristic of the 

 sponges, then, indeed, the genus might be more properly referred to the Foraminifera, in 

 many respects the close allies of the sponges, but destitute of the canal-system which is pre- 

 sent in the latter. 



ft. Such openings, however, can be shown to exist in certain forms of Stromatopora, and 

 there is strong reason for believing that they will ultimately be found to be present in all. 

 Thus in Stromatopora striatella (D'Orb.), and S. concentrica (Gold.), both typical examples 

 of the genus, Professor McCoy long ago described the existence of vermicular tubes opening 



