38 



THE GENUS MONTICULIPORA. 



like M. petropolitana, Pand. (fig. i, a), and M. tindulata, Nich. 

 (fig. 2, a), the walls are so thin that the partitions between the 



Fig. 2. — A, Tangential section of a single corallite of Monticulipora tindulata, Nich.; F, Tan- 

 gential section of a single corallite of Monticulipora petasiformis, Nich., showing the 

 duplex character of the wall ; C, Tangential section of a corallite oi Moniiailipora ritgosn, 

 Edw. and Haimes ; D, Tangential section of a corallite of Alonticitlipora Aininnvsii, 

 Nich. All the sections are enlarged fifty times. 



visceral chambers of contiguous corallites appear to be abso- 

 lutely indivisible and without structure, presenting themselves 

 in thin sections merely as delicate dark lines. Dybowski (Die 

 Chsetetiden) lays great stress upon this feature, and uses it as 

 a basis for subdividing the Monticuliporoids into two primary 

 sections. For my own part, I believe that in such a complex 

 group as the present, no single character, such as this is, is of 

 itself sufficient for the establishment of primary subdivisions ; 

 and I feel satisfied that this apparent amalgamation of the 

 walls of contiguous corallites must be due to our imperfect 

 methods of observation. That this is the case seems suffi- 

 ciently proved by the consideration — which I shall have to 

 speak of again — that even in forms like these (viz., HI. lui- 

 didata, Nich.), rough fractures will demonstrate what thin 

 sections fail to show — namely, that the apparently structureless 

 wall is really double. In rough fractures, that is to say, we 

 find that the corallites always separate cleanly from one an- 



