THE NATURE OF CAUNOPORA. 129 



III. Caunopora and Diapora as " states " of Stromatopora and Stromatoporella. 



The only other hypothesis which seems worth a moment's consideration as an 

 alternative to the theory of commensalism, is that the ordinary " Caunopora^ " and 

 " Diapora, " are states of certain species of Stromatopora and Stromatoporella. 

 The fact that the " Caunopora-tubes " are, as a rule, only found in particular 

 species, belonging to particular genera, affords prima facie ground for supposing 

 that they belong to the species in which they are found. We have seen, however, 

 that all the species which exhibit these tubes also exist without the tubes. It is 

 therefore clear that if the " Caunopora-tubes " belong to the organism with which 

 they are associated, they can only represent structures which are developed in 

 certain individuals and not in others. It would therefore be a not unnatural 

 hypothesis to suppose that the " tubes" of " Caunopora" and "Diapora" repre- 

 sent the cavities in which the reproductive zooids were lodged. I was at one time 

 strongly tempted to take this view, and there are certain facts which would go a 

 considerable way in its support. Thus, there is an undoubted resemblance between 

 the " Caunopora-tubes " and the tabulate axial tubes of Idiostroma, Stachyodes, and 

 Amphipora, these structures belonging unquestionably to the organism in which 

 they are found. Again, there is a still more striking resemblance between the tubes 

 of " Caunopora " and " Diapora " on the one hand and the large round-mouthed 

 tubes of Idiostroma oculatum, Nich., on the other hand. These resemblances do 

 not, however, go far enough. Thus, the tabulate axial tubes of Idiostroma and its 

 allies have no proper walls, and communicate freely with the ccenosarcal canals of 

 the general skeleton. In Idiostroma oculatum, also, the large round-mouthed and 

 tabulate tubes, though furnished with proper walls near their mouths, appear to 

 lose these walls internally, and also seem to communicate freely at their bases with 

 the interlaminar spaces of the general skeleton. Until, however, we obtain some- 

 thing like positive proof of the existence of a free communication between the 

 cavities of the tubes of " Caunopora " and " Diapora " on the one hand and the 

 ccenosarcal canals of the surrounding Stromatoporoid on the other hand, it seems 

 impossible to accept any hypothesis which would treat these tubes as being 

 constituent parts of the Stromatoporoid in which they are found. Moreover, it is 

 now certain that " Caunoporce " and " Diaporos " are not exclusively referable, as 

 regards the tissue of the enveloping Stromatoporoid, to the two genera Stromato- 

 pora and Stromatoporella. Had this held good, there would have been strong 

 ground for regarding the embedded " Caunopora-tubes " as belonging to the 

 investing Stromatoporoid. We now know, however, that species of other genera 

 than the two first mentioned occur occasionally as " Caunopora}." Upon the whole, 



