158 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



internally a rudimentary arm with syzygies between brachials 1+2 and 16 + 17. In 

 describing this specimen in his report upon Dr. Mortensen's collection, Dr. Gisl6n 

 said that P had developed into a new arm; but hi the diagram published in 1924 the 

 new arm is shown to arise from the division of the fourth brachial, normally the epizy- 

 gal of the first syzygial pair, into two ossicles. 



In a fourth specimen from Mortensen's station 10 one postraclial series bears to 

 the right an undivided arm and to the left an arm which branches on the eighth 

 brachial; on the outer branch the first syzygy is between brachials 6+7, and on the 

 inner branch there are syzygies between brachials 1+2, 6+7, and 11 + 12. 



Remarks. Dr. P. H. Carpenter in 1888 described Antedon manca and A. disci- 

 formis. In his key to the species in the Palmata group he differentiated these species 

 from the others in the group by the absence of P. He gave manca as having two 

 axillaries following the IBr series and the inner arms of each postradial series usually 

 without PI, whereas disciformis has but one axillary following the IBr series, and 

 P t is present on all the arms. These supposed differences have since proved to be 

 of no significance. 



In 1910 I examined the types of manca and of disciformis at the British Museum 

 in London and found that hi manca the longest cirrus segments are scarcely longer 

 than broad and the dorsal spines on the outer segments are not nearly so long as hi 

 disciformis, in which the longest cirrus segments are nearly twice as long as broad, and 

 the dorsal spines on the outer segments are long and sharp. 



In 1918 I reviewed the relationships of disciformis and manca on the basis of a 

 detailed examination of 56 specimens referred to disciformis from Albatross stations 

 5212, 5213, 5154, 5356, 5367, and 5369, and from the Challenger collection. I wrote 

 that the cirri of typical disciformis and typical manca are very different. In both the 

 dorsal spines are well developed, but in disciformis they are much longer than they are 

 in manca, while furthermore the earlier cirrus segments hi disciformis are elongated, 

 twice as long as broad instead of only very slightly, if at all, longer than broad as in 

 manca. However, I found that in deep and cold water the elongated earlier cirrus 

 segments of disciformis as a rule rapidly shorten, so that the character of the cirri 

 approaches that of typical manca. 



In the key to the species of the genus Cyllometra published hi 1918 I separated 

 disciformis from manca on the basis of the relative length of the earlier cirrus seg- 

 ments, saying that in disciformis these are longer than broad and may be twice as 

 long as broad or even longer, whereas in manca they are not longer than the width 

 of their distal ends. 



The differences between the cirri of disciformis and those of manca are by no 

 means so considerable as this would appear to indicate. In the entire absence of 

 correlated characters hi other structures they are certainly not sufficient to justify 

 maintaining disciformis and manca as distinct species. At the most disciformis may 

 possibly deserve recognition as a form of manca inhabiting shallow and warm water. 



In 1907 I mentioned indefinitely, under the name Antedon manca, specimens of 

 a species of Cyllometra from Japan, and in April 1908 Dr. Hubert Lyman Clark 

 definitely recorded Cyllometra manca from the Uraga Channel. On May 14, 1908, I 

 described Cyllometra albopurpurea, which I said represents C. manca in the waters 

 about southern Japan, differing strikingly from that species in having all the cirrus 



