THE NATURE OF CAUNOPORA. 119 



II. The Theory of the Commensalism of Caunopora and Diapora. 



Tu discussing Prof. Roemer's theory of the " commensalism" of " Caunopora" 

 and " Diapora," I shall, in the first place, review generally the arguments against 

 the theory and those in favour of it. In the second place, it will be necessary to 

 discuss the question whether, if the theory of commensalism be accepted, the 

 *' Caunopora-tubes " are referable to Syringopora, or whether they belong to 

 Aulopora. 



(a) General Arguments against Commensalism. — The following, stated briefly and 

 in a summary form, are the principal facts and considerations which tell against 

 any theory of the commensalism of" Caunopora!'' It should be premised that all 

 those cases where Stromatoporoids demonstrably envelop different kinds of Corals 

 are here left out of sight. All we have to deal with here are the typical " Cauno- 

 pora?" and " Diapora?," in which we cannot at present demonstrate commen- 

 salism. "With regard to all such specimens — and they are very numerous — it may 

 be taken for granted, with our present knowledge, that if the organism be the 

 result of the commensalism of a Coral and Stromatoporoid, the foimier must belong 

 to Syringopora or Aulopora, or to some closely allied type. We are not, at any 

 rate, acquainted with any Paleozoic Corals, except the species of these two genera 

 or of closely related types, the internal structure of which is such as to permit of 

 our supposing that the " Caunopora-tubes " might belong to them. 



1. In the first place, colonies of Aulopora are often found associated in different 

 ways with Stromatoporoids, and yet not giving rise to " Caunoporce " or " Diaporce." 

 Thus, in the Wenlock Limestone of Britain nothing is commoner than to find 

 Aulopora colonies spreading over the upper or under surfaces of Stromatoporoids, 

 and even sometimes in part enveloped in these ; but in an examination of two or 

 three hundred of such specimens I have not detected a single one in which the 

 Stromatoporoid had completely enclosed the Coral, or in which the latter had been 

 induced to lengthen its tubes or to alter in any way its normal mode of growth. 

 On the other hand, in the Devonian Strata I have often noticed tubes apparently 

 belonging to Auloporce, or to some of the types which have been placed under 

 Syringopora, completely immersed in Stromatoporoids, and nevertheless not giving 

 rise to " Caunopora? ;" the growth* of the embedded Coral being altogether irregular 

 and showing none of the peculiar characters of the latter. 



2. Again, there are extensive groups of rocks in which all the conditions required, 

 on the theory of commensalism, for the production of " Caunopora? " are present, 

 and yet the fossils so called are unknown, or are extremely rare. Thus, as just 

 noted, the "Wenlock Limestone of Britain contains a vast abundance of Stromato- 

 poroids (including three species of the genus Stromatopora itself) along with 



