﻿184 MADEEPORAEIA. 



variable, being here hardly more than reticular threads, there stout rods rising above the surface 

 as tubercles. The tubercles are usually associated with calicles ; but only here and there do 

 they form protuberant rings (of 5 to 6), for the most part one or two rise directly from the 

 calicle margin, either at the side so as faiatly to suggest longitudinal series, or proximaUy so 

 as slightly to tilt the calicle aperture. The suggestion of series is further enhanced by the 

 occasional presence of rows of much smaller tubercles in the depressed interstices, while 

 round the edges, where no tubercles are formed, there is a faint ridging of the surface. 



The type specimen is interesting because it is encrusting a massive almost solid block 

 some 3 cm. thick which has aU the appearance of being formed by the same coral. The section 

 of this block, which progressively solidifies below, suggests the possibility of the trabeculse 

 continuing to grow almost indefinitely. Massive stocks are built up by M. solida, M. rotunda, 

 M. verrucosa, M. alveopora, and II. saxca. Of these, the second and last show tendency to pro- 

 gTessive solidification. This is the only known tuberculate Montipore I can caU to mind 

 which shows the same phenomenon. 



With regard to the affinities of the specimen, I was at first disposed to class it with 

 M. listcri (p. 122). The two specimens have much in common, but the one is a free thin 

 frond, and the other a closely encrusting layer. It seemed at first sight possible to explain 

 most of the differences as due to this diiference in growth, as a comparison between the free 

 fronds of M. plicata with the nodulated crumpled leaves of the same, amply showed. All that 

 I am now prepared to say is, that this specimen may be an encrusting specimen of M. listeri — 

 that is, the range of variation does not preclude the idea — but there is absolutely no evidence. 

 Further, it is clear that we can have no evidence imtil freely expanding fronds of this coral 

 closely associated with encrusting layers, or, what would be equally good, until encrusting 

 layers of 3£. listcri are discovered. Again, then, we must wait for longer series. 



a. Sandal, Lifu, Loyalty Islands. Coll. WiHey. (Type.) 



Postscriptum to first footnote on p. 4. 



Ehrenberg, on p. 121, includes Montipora Q. & G. as one of the many genera and generic 

 names " qnm physiologicis characteribus destituuntm," and " = Madrepora." 



