THE ASTRORHIZiE. 53 



It is also almost certain that the large tabulate axial tubes of such genera as 

 Idiostroma, Winch., Staehyodcs, Barg., and Amphipora, Schulz, with their lateral 

 branches, served for the lodgment of a special series of zooids ; but we have at 

 present no absolutely final evidence on this point. If, moreover, it were possible 

 to show that the large, thick- walled, tabulate tubes which characterise the so-called 

 genera Caunopora, Phil!., and Diapora, Barg., really formed a constituent portion of 

 the Stromatoporoids in which they are found, we should have had in these an 

 admirable example of dimorphism. Indeed, the comparisons which have been made 

 by earlier observers between the Stromatoporoids and the recent Millepora have 

 usually been based upon specimens of " Caunopora." The real nature of the tubes 

 in question in Caunopora and Diapora is, however, a subject involved in such 

 difficulty, that I shall consider it in a separate section. 



(f) TJie Astrorhizce. — One of the most prominent features in many Stromato- 

 poroids is the presence on the surface, and also at all deeper levels in the skeleton, 

 of numerous shallow grooves arranged in definite stellate systems upon the surfaces 

 of the concentric lamina? (Plate IV, figs. 2 and 6). For these stellate canal-systems 

 Mr. Carter's apt name of " astrorhizaa " may be employed with advantage. There 

 is, also, no reason to doubt that Mr. Carter has decided correctly in his determina- 

 tion of these structures as the homologues of the branching coenosarcal grooves on 

 the surface of the skeleton of many Eydractinice (Plate VI, figs 3 and 9). They 

 may also be compared with the branching and inosculating coenosarcal canals of 

 the ccenosteum of Millepora (Plate IV, fig. 5). The correctness of this view seems 

 to be sufficiently proved by a consideration of various other facts which are now 

 known as to the structure of the skeleton in various Stromatoporoids, and especially 

 by the fact that many of them can be proved to have possessed tabulate zooidal 

 tubes. At the same time, it should be remembered that, in the absence of this 

 confirmatory evidence, earlier observers were not without justification in comparing 

 the astrorhizEe of the Stromatoporoids, as many have done, with the dermal canals 

 of certain of the Sponges. 



The size of the astrorhizee is very variable in different types of Stromatoporoids, 

 but, when present at all, they are always visible to the eye, and they are often 

 extremely conspicuous objects (Plate IV, fig. 2). Whatever their size may be, 

 their general form is tolerably constant, each astrorhiza consisting of a stellate 

 group of comparatively large-sized shallow gutters, which spring from a central point 

 and branch as they radiate outwards, diminishing at the same time in diameter, and 

 giving off more or less numerous lateral branches. These branchlets communicate 

 freely, and they finally inosculate with the terminal twigs of adjoining astrorhiza? 

 (Plate III, fig. 3). The entire series of astrorhizas thus forms a system of shallow, 

 open, anastomosing grooves on the surface of the ccenosteum, and doubtless served 

 for the lodgment of corresponding coenosarcal stolons. 



