314 Mr. H. J. Carter on the Strom atoporidse. 



Goldf., furtlier than Lonsdale has done by stating that it does 

 not present " tubes " like those of his " Coscinopora placenta j'"' 

 = our Caup'^vora placenta — which is totally inadequate to our 

 present puipose. There are, however, other features which 

 are common to many forms of Stromatopora-'&trvictvLX&j which 

 I have never seen in Caunopora \ and if we combine these with 

 the absence of the " tubes " and a more or less gnarly struc- 

 ture (like that seen in knotty wood) in the section, already re- 

 presented by Phillips {op. cit.) in liis figure of 8. concentrica^ 

 we might then get a typical form to which we might still apply 

 the name given to it by Goldfuss, and then know what we 

 were writing about. To these " other features " I shall now 

 allude. 



The " gnarly " or undulating character in the general struc- 

 ture of Stromatopora, where the bend of the undulation may 

 vary from a few lines to as many inches in diameter more or 

 less, should not be confounded with the curvilinear fibre of the 

 coenenchyma in Caunopora already described, nor with the 

 rectilinear fibre of Stromatopora about to be mentioned. The 

 term " curvilinear " has already been explained j and that of 

 ^^rectilinear'''' means that the lines representing the elemen- 

 tary part or fibre of the coenenchyma are all more or less 

 straight. 



As with Caunojpora^ so with Stromatopora ; my observations 

 have been manifold ; but not having had the opportunity of 

 dissecting a large block of the latter, as witii the former, they 

 have been made on fragments of weathered or unpolished and 

 polished specimens respectively, the largest not exceeding nine 

 by four and a half inches in diameter. 



Looking at the vertical section of Stromatopora concentrica^ 

 Goldf., as presenting the typical structure of the family, the 

 coenenchyma may be observed to consist of more or less straight 

 lines arranged horizontally, cut at right angles by vertical ones 

 or rods, which are the thickest of the two; while the Jwrizon- 

 tal section presents a number of white points, which are the 

 ends of the " rods," united together by a rectilinear structure 

 consisting of straight lines, which extend between the points 

 and between each other, so as to produce a cribriform lamina 

 with triangular or multiangular spaces. For the term " rec- 

 tilinear " I have hitherto used that of " hexactinellid," which 

 was evidently a misnomer. Further it should be remembered 

 that in some specimens or species the coenenchymal structure 

 is extremely fine, and in others, especially the rods, compara- 

 tively coarse. 



Now, as the vertical section shows no trace of the structure 

 seen in the horizontal one, saving the margin of the lamina, 



