154 COLLECTIONS FROM MELANESIA. 



the interesting A. jukesi, of which Mr. Carpenter has already indi- 

 cated the more essential characters, is indeed represented in this 

 collection, as it is probably in any fair collection of the marine fauna 

 of the Australian coast. 



Set, again, in a paper which will be shortly published in the 

 ' Journal of the Linnean Society ' *, Mr. Carpenter describes eight 

 out of the nine specimens of Antedon from the Hamburg Museum 

 as new, and he speaks in the introduction as estimating the species 

 of Comatulids at something like 400. 



Further, it is of great significance to observe that many of the 

 species here enumerated or described were collected at one station 

 only. 



Lastly, we note that the number of Antedonsis larger than might 

 have been expected ; for in the Moluccas "Antedon seems to be com- 

 paratively rare "f , while of the 29 species here enumerated, 16 belong 

 to that genus. From such material as has passed through my 

 hands, I am inclined to think that on the northern and eastern 

 coasts of Australia we shall find Antedon to be rather more abun- 

 dantly represented in species than Actinometra ; the time, however, 

 for any generalization is still far off. 



In entering into the detailed enumeration of the proportion of new 

 to old species, I had not in view the purpose of apologizing for the 

 presence of so many new forms in this collection, but rather the 

 desire of directing attention to facts which can only be within the 

 knowledge of a limited number of special students ; those who 

 know how few species of Comatulce have been described, and how 

 rich in novelties not only new collections but old museums are, will 

 not think that there is any suspicious wealth of new species in the 

 very valuable and important collection by which Dr. Coppinger has 

 more than doubled the number of specimens and species in the 

 possession of the British Museum j. 



So large a number of new species should be presented in some 

 kind of arrangement, either in the form of a phylogenetic table or 

 of a " key." The former being an impossibility at present, on 

 account of our unsatisfactory knowledge of the ancestry of the 

 Comatulidce, and keys being, of all things, the most unscientific, 

 I propose to give formulas for all the species of Comatulids here 

 described, basing those formulas on the method I proposed to the Zoo- 

 logical Socicty§, as improved by the suggestions of Mr. Carpenter ||. 



* Journ. Linn. Soc. xvi. p, 487. 



t Notes Leyd. Mus. iii. p. 191. 



J [The above is allowed to stand, though written some eighteen months 

 ago, as it puts more forcibly than a briefer and colder statement could, the 

 present tenuity of our knowledge of Orinoid species and the wide area that is 

 opening up to us. — Dec. 4, 1883.] 



§ P.Z. S. L882, p. 530. 



| P. Z. S. 1882, p. 731. I retain A' as the sign for Ac/inomrfra, as a ia used 

 in (ho formula; of the cirri ; and T propose to use br for 1 ho brachials, as h is 

 Likewise used in the formulae of the cirri. Similarly I omit the 10, as A 10 

 followed by A 3 (in such a list as the following) is very apt to mislead. 



