516 Mr. II. M. Bernard on the 



built up in this way. The original cup is naturally obscured 

 by the continual upward growth of the edges and by the 

 development of tall conical folds (c) and of wings {d) on 

 the faces and sides of the original flaps, which have grown 

 upwards with the growth of the cup, and also by the down- 

 ward streaming of the coenenchyma from within the cup 

 carrying down the lower edge of the flap, so that the stalk is 

 completely obliterated. The lower portion of the stock, 

 especially certain masses of dead corallum, are not easy to 

 understand ; it is possible that the cup shown in the diagram 

 and figure as the foundation of the corallum was not the 

 original cup, but a secondary cup-like growth of parts of an 

 old stock. The question is, Is this strangely symmetrical 

 method of modifying a cup typical or accidental? I would 

 like to invite the attention of those who have access to any 

 collections of Turhinaria to this point. 



Grouping accordincj to Growth probably a natural one. — 

 Accepting these eight types of growth as of practical value in 

 classification, w-e find that they enable us to divide the 

 specimens into more or less well-defined groups. It must, 

 however, remain undecided whether these are natural groups, 

 although this would certainly appear to be the case, in spite of 

 certain somewhat serious objections. I would instance as the 

 greatest difficulty the fact that certain forms of coralla seem 

 to belong to definite localities. There are specimens from the 

 Torres Straits, belonging to at least two species, which show 

 the same form of corallum. This seems to imply that, at 

 least in these cases, the form is due to the environment. But 

 while this fact cautions us against attributing too high a taxo- 

 nomic value to the forms of the coralla, it would be rash to 

 deny them all value. In view of the definiteness of some of 

 these types of growth, more than one coming from the same 

 locality, in view also of the possibility at any time of the 

 living colonies being powerfully modified by exceptional con- 

 ditions of the environment, we are, it seems to me, justified in 

 assuming that these growth-forms are typical developments. 



The 1 axonomic Characters supplied by the Calicles. — These 

 are far more difficult to define than are the methods of growth. 

 The calicles vary in size, shape, depth, and degree of protu- 

 berance according to the part of the corallum they occupy * ; 

 even the number of the septa and the size of the columella 

 vary greatly on one and the same specimen. The only prac- 

 tical course is to select those corallites which appear to be 

 typical, i.e. which appear to be growing normally on the normal 

 fronds or lobes of the special type of growth of the specimens 



* Cf. Bell, " Variations in Turbinaria;' J. R. M. S. 1805, p. 148. 



