1907.1 AND SUPPOSED SPECIES IN CORALS. 523 



or even as localised patches in otherwise normal colonies, pther 

 corals may divide actively and yet form no calcareous partition- 

 walls. 



The meaning of these assumptions of an entirely foreign method 

 of division is somewhat difficult to understand, but when an 

 Astrea exhibits Meandrine fission, it is usually a sign that the 

 colony is not flourishing and is in an unsuitable site ; it would 

 almost appear that the building of unnecessary partition -walls 

 was too great an effort for the unhealthy zooids (text-fig. 151 and 

 PL XXIX. fig. 2). 



Text-fig. 151. 



Vegetative reproduction in Corals. 



Diagram showing an Astrea assuming (as a sport) a Meandrine form of 

 '° ° division. From an actual specimen. 



Ao-ain, the thickness of the intervening partition between two 

 adjacent zooids is subject to endless variation, and in consequence 

 the surface-pattern and the density of the coral may vary widely 

 within the limits of a species. The same variation is to be seen 

 in the level of the site of origin of the lateral buds, and this will 

 have to be referred to later on, when the cause of the variations 

 is considered. 



Another feature that is subject to an excess of variation, and 

 must therefore be considered, is the amount of raising from the 

 o-eneral surface of each individual corallite. Every corallite of 

 the colony may be flush with the general surface, or it may be 

 raised from it in varying degrees, and the degree of raising gives 

 very characteristic appearances to the colonies. The degree of 

 elevation of the corallite is no safe criterion for determining 

 specific rank, for it is a variable factor depending altogether on 

 the reactions of the coral to its environment. The portions of 

 the coral body that lie between the actual corallites are also im- 

 portant in this connection, for many characters that may have 

 undue importance attached to them are displayed here. The 

 interspaces may be smooth or rough, they may be sculptured in 

 various fashions, and they may be elevated or depressed; but 

 very great caution, and a very long study of the possibilities of 

 variation, must be used by anyone who would assign specific rank 

 to any of their forms. 



Althouah a very large number of corals have the normal habit of 

 reproducing equally from all parts of their surface, still evidences 



