Crinoidea. 449 



monotypic genus) characteristic of the southern coast of Australia. Judging 

 from the exceedingly meager data at hand the range of Comatulella bra- 

 chiolata falls entirely within that of Ptilometra macroncma, not entering 

 that of Ft. mulleri at all; nor does C. brachiolata extend so far up the 

 west coast as does Pt. macronema. 



It is interesting to note that ('. brachiolata has very short and stout 

 arms, all nearly or quite equal in length, and short-segmented cirri. These 

 features appear to be characteristic of species derived from purely tropical 

 stock which have intruded into a permanently or intermittently cold area. 

 In the Comasteridae this is well illustrated by Comanthus Wahlbergii from 

 the southern part of Africa, and in Comactinia meridionalis from the 

 southeastern coast of the United States. It is also brought out in the 

 Antedonidae by Antedon petasus from the coast of Scandinavia, belonging 

 to a genus barely separable from the parent East Indian stock (Mastigo- 

 metra) , though it is entirely lacking in the two Mediterranean species 

 (Anledon mediterranea and A. adriatica) of the same genus. 



If this correlation, that normally tropical forms when in cold water 

 tend to develop short stout arms and stout short-segmented cirri, can be 

 established beyond controversy, it will have a very important bearing upon 

 our interpretation of the ultimate origin of many species, and of the evi- 

 dence concerning many past migration routes. For instance Antedon 

 Hupferi of the west coast of Africa resembles A. petasus in this respect, 

 while the two Mediterranean species are widely different and much more 

 like their East Indian relatives. Judging from what we know we should 

 assume that the west African forms must have come from the north, from 

 beyond the Straits of Gibraltar, and that they have not yet been long 

 enough in their present tropical surroundings to have lost the cold-induced 

 characters of their arms and cirri. Exactly the same thing holds good for 

 the European-African species of Leptometra (L. celtica) as contrasted with 

 that of the Mediterranean (L. phalangiwn) . 



Very warm water appears to have a curiously parallel -effect upon 

 certain forms; but whereas cold appears to delay the metabolic processes 

 so that the formation of the skeleton outstrips the growth of all the other 

 structures, causing the arms to become short and stout, warmth seems to 

 attain the same end by inducing a very rapid and early development of 

 the sexual products which hinders the development of all the other struc- 

 tures, though it has no effect upon the skeleton formation. 



In this connection it is instructive to note that, whereas arm shortening 

 by cold progresses equally on all the arms, arm shortening by warmth- 

 induced sexual maturity affects, in the Comasteridae, the posterior arms 

 first, and is very often confined entirely to them so that, in extreme cases, 



Die Fauna Siidwest-Australiens. III. ^9 



