CEEXAIK STICHODACTTLINiE TO THE MADEEPORAEIA. 651 



Ehodactis and E-icoedea. — Eetention of a multi-oral disc ; 

 sharp division between tlie peristome and stomodeeura ; stomodaeal 

 wall regularly and deeply ridged ; no gonidial grooves ; mesen- 

 terial filaments simple ; large nematocysts ; weak musculature 

 throughout. 



CoEALLiMOEPHUS should also be mentioned as presenting 

 many coral relationships — e.g. knobbed tentacles, no gonidial 

 i^rooves, weak musculature, no spbincter muscle, large nemato- 

 cysts : indeed all the members of Carlgren's Protaiithese reveal 

 suggestive conditions. 



Odd instances of the above characters may be present in other 

 anemones, but, so far as yet known, no other Actiniae exhibit 

 such a combination of affinities. On the other hand, coral polyps 

 are not alike in the possession of all of them. All that can be 

 claimed is that the tout ensemble is more decidedly Madreporarian 

 than Actiniarian. Morphologically both the Actiniaria and 

 Madreporaria present such diverse features amongst themselves 

 that little importance can be attached to any isolated detail in a 

 species, but where so many are associated in the same species or 

 genus the relationship claimed becomes almost irresistible. 



It is significant that the suggestion which Gosse {ante, p.- 611) 

 made, from external characters alone, of the close resemblance 

 of Corynactis to the coral Garyophyllia should be supported in 

 every important detail by anatomical study. 



In the present state of our knowledge of the morphology of 

 the Madreporarian polyp, it cannot be asserted to which actual 

 coral families the three genera are most nearly related. They 

 differ amongst themselves in fundamental respects, and the allies 

 of Corynactis, at any rate, must be far removed from those of 

 the other two. 



Whether the forms specially studied are to be regarded as 

 representatives of more ancestral anemones from which the 

 skeleton-producing polyps may have taken their origin, or as 

 coral polyps belonging to different families which have for some 

 reason lost the power of secreting a skeleton, must, for the 

 present, also be left an open question. 



