DEVELOPMENT OF SOME STARFISHES. 557 



stalk in question. The two forks thus ride saddlewise on the 

 preoral lobe-stalk, but when this stalk is obliterated, the saddle- 

 cavity simply fills out to form the coelom within arm-rudiment I. 

 Further, in general, during the growth of the larva, as the 

 epigastric coelom remains relatively small, the larval posterior 

 ctelom has to encroach somewhat, on the right or aboi'al side, and 

 in particular sections the encroaching shelf may give an impression 

 of bifidity. 



Masterman described indications of early doubling of the pos- 

 terior coelom, and considered these as supporting his view that the 

 coelom in question is made up of right and left morphological 

 elements, namely, a right posterior and a left posterior coelomic 

 cavity. This view, although it might suit the ontogeny of 

 Cribrella, is out of harmony with the data from other starfishes 

 and from echinoderms generally, including double-hydrocoele 

 specimens. In discussing the subject elsewhere (2, p. 234 ; 3, 

 p. 32) I have put forward the view that the larval posterior 

 (future hypogastric) coelom of Solaster is morphologically a left 

 posterior coelom originating from the gut by a separate, meta- 

 merically posterior, outgrowtii, and that the corresponding cavity 

 on the right side, namely the epigastric coeloxn, originates, as in 

 AsteiHna, Asterias, Echinus, etc., by backward extension of the 

 anterior coelom. 



(8) Sequence in formation of hydrocmle pouches. — The first 

 thickenings for the hydrocoele pouches make their appearance at 

 about the ninth day, pouch I. being slightly the latest. There- 

 after, pouches III. and IV. diflierentiate a little more quickly 

 than the rest; pouch V, and pouch II. are next in order; while 

 pouch I. is the slowest. According to Masterman the growth 

 series usually begins with pouch Y. (his pouch I.) and gradually 

 works round to pouch I. (his pouch Y.). The facts are of interest 

 in connection with the question as to what is the proper number- 

 ing of the rays in Cribrella and other starfishes (3, p. 276). 



(9) Enantiomorphic and double-hydroccele larva;. — Masterman 

 has stated (9, p. 403) that enantiomorphic specimens, i. e. speci- 

 mens in which the hydrocoele develops on the right instead of the 

 left side of the larvae, are probably not uncommon in Cribrella. 

 None, however, was observed in my series. Indeed, in very large 

 numbers of starfish larvae examined from time to time (Asterias 

 rubens, Solaster papposus, S. endeca, Porania) I have not come 

 across a single example of this abnormality, and only one 

 specimen {Porania) in which a right instead of a left hydropore 

 was present. 



No double-hydrocoele or double-hydropore larvae were found in 

 Cribrella, although examples of these abnormalities are not 

 infrequent in other starfishes (4, p. 69). 



(10) The brachiolarian notches. — A special gap or notch (oral 

 brachiolarian, or hydrocoelic, notch) between pouches I. and Y., 

 such as occurs so markedly in Asterina, Asterias rubens, and 

 Solaster, is not characteristic of Cribrella. Nor is there at any 



