130 A NATVBALIST IN CELEBES ch. vi 



inside of the reef, where there is plenty of room for it, and, 

 comparatively speaking, little competition for the floating 

 food, the tubes are large and long. If, on the contrary, it 

 is found amongst a crowd of other corals on the outer edge 

 of the reef, the tubes are small and short. Every possible 

 variety exists upon the shore intermediate between the 

 extremes of largeness and smallness ; and I feel perfectly 

 certain that had I devoted myself exclusively to this one 

 problem, I could have brought home with me from the reefs 

 within a few miles from my hut in Talisse examples of 

 every species of Tubipora that have yet been described. My 

 conclusion is, then, that there is only one species of 

 Tubipora at present known, and this it is best to call T. 

 viusica. 



The varieties of Clavularia found, as I have said, usually 

 on the inside of the reef, were of considerable interest to 

 me. 



Clavularia (fig. 16) differs from Tubipora principally in the 

 character of its hard or skeletal parts. Instead of having 

 a skeleton composed of fused spicules as Tubipora has, 

 its walls are strengthened by a few long isolated spicules 

 imbedded in a hard horny substance. The polypes re- 

 semble those of Tubipora in shape and size, but the colour 

 rarely varies from the deep brown previously mentioned. I 

 discovered that some varieties of Clavularia possess a curious 

 mode of branching and budding resembling that of the ex- 

 tinct Syringopora (33, 34, 53). Now, this genus of fossil 

 corals has been tossed about from the group Zoantharia to 

 the Alcyonaria by different authorities, but this last observa- 

 tion of mine supports the view Professor Moseley has main- 

 tained, that Syringopora was undoubtedly an Alcyonarian. 



Besides the Alcyonarians, however, a few living Zoan- 

 tharian corals may be found upon the inside of the reef. 

 These are chiefly the brain corals, fungias, and perhaps a 



