GENERA OF CH^ETETID^E AND MONTICULIPORID^. 289 



from accepting that account. Without, then, going into this 

 question in detail, I may simply say that I see at present no 

 sufficient ground for removing the Monticuliporoids to the 

 Polyzoa. I do not think any weight can be attached to external 

 appearance in a matter of this kind. Thus, we find the com- 

 mon Favosites Canadensis, Billings sp., of the Devonian of 

 Canada, to be so entirely similar in the form and appearance 

 of its colonies to examples of Fistulipora, M'Coy, that it was 

 unhesitatingly referred to this genus (which Dr Lindstrom re- 

 gards as clearly Polyzoan) by Mr Billings and myself. Dr 

 Rominger, however, showed that it has " mural pores " of the 

 regular Favositoid type a discovery which I have myself 

 verified so that in place of being a Fistulipora, and, there- 

 fore, according to Dr Lindstrom, a Polyzoon, it is a true Per- 

 forate Coral. A position among the Favositidce has similarly 

 been now established as the right one for Stenopora, Lonsd., 

 which includes forms so like Chcstetes and Monticulipora in 

 general aspect as to have been commonly included under one 

 or other of the latter heads. Apart from mere superficial 

 appearances which in this case speak at least as strongly for 

 a Ccelenterate as a Polyzoan alliance there is nothing in the 

 actual structure of Monticulipora which would not entirely 

 agree with its being a coral. The only point which could be 

 mentioned which would in any fundamental manner distinguish 

 the internal structure of a Monticulipora from that of, say, 

 Tetradium or Heliolites, is the absence in the former of septa. 

 I do not, however, attach any weight to this, partly because 

 some undoubted corals are equally without septa, partly because 

 the septa in Heliolites and its allies are now known by the 

 researches of Moseley to be only " pseudo-septa," and partly 

 because I do not think that any important change in classifica- 

 tion should be based upon a merely negative character. On 

 the other hand, there are strong resemblances between Mon- 

 ticulipora and its allies and various undoubted corals prin- 

 cipally, perhaps, the Helioporidcz. Thus the " tabulae " of the 

 Monticuliporoids are in all respects similar to those of such 



