18 ECHINODERMA OF THE TnDIAN MuSEUM, PART VII. 



Suppose, for instance, we take the type of the centrodorsal in the Comasterida^. 

 This organ differs in the several genera and species composing the family only in 

 the degree of specialization, the development lines being everywhere the same. 

 In some species, as in Comanthus hennetti, the centrodorsal always remains essen- 

 tially as in the young, but increases in size throughout the life of the individual. 

 Usually, however, resorption takes place at the dorsal pole which is gradually 

 planed off, as it were, so that the centrodorsal changes from the primitive hemi- 

 spherical form and becomes discoidal, the rows of cirri dropping off as the 

 sockets are resorbed. In extreme cases the resorption results in reducing the 

 centrodorsal to a thin stellate plate without any traces of cirrus sockets, counter- 

 sunk within the centre of the dorsal surface of the radial pentagon. 



We may arrange all comasterid centrodorsals in a linear series, calling tlie 

 least developed {Comanthus bennetti) type A, and the atrophied stellate disk D, B 

 and C denoting intermediate stages. 



Now the species of the Australian fauna have centrodorsals which run from 

 A to D, but with especial emphasis on the D; the species of the East Indian 

 fauna also run from A to D, but the emphasis is between B and C; the Japanese 

 species run from A to C, with especial emphasis at B ; the West Indian and the 

 East African species are confined between B and C. This holds good regardless 

 of the subfamily or genus to which the species may belong, and exactly the same 

 thing may be worked out in regard to other characters in this family, and with 

 other characters in other families. 



The recent crinoids of the Australian coasts are evidently senescent, unmis- 

 takably indicating very great age. The crinoids of Australia came from the 

 northward, from the great East Indian archipelago; but here continual changes in 

 the distribution of land and sea have constantly rejuvenated the fauna so that none 

 of its component species has been permitted to drift into the peaceful old age so 

 obvious in almost all of the species along the Australian shores. 



The fossil crinoids of Europe (belonging to genera still existing) appear to be 

 senescent ; but they are no more so than, if as much as, the recent crinoids of 

 Australia. Judging from the evidence offered by the recent forms alone the 

 European crinoids reached the European seas by passage from what is now the 

 Bay of Bengal north of what is now India, or at least southern India. It was 

 probably before this that the same genera spread southward from the parent 

 central East Indian region to Australia. 



The crinoids of south-eastern Africa represent a comparatively young fauna ; 

 they must have reached their present habitat by jjassage south-westward from 

 Ceylon along a mon* or less complete land bridge since submerged ; but few of 

 them have as yet entered the Arabian Sea. 



The West Indian fauna is younger again than that of the south-eastern shores 

 of Africa from which it was derived. It must have reached the West Indies by 

 following a land which extended from Madagascar to the Antilles, north of what 

 is now southern Africa. 



