512 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



with their bases in apposition. Po is 5 mm. long and is composed of 20 segments, 

 resenibhng Pi but weaker and more slender. P3 is 3.5 mm. long with 15 segments, 

 resembling P2. P4 and the foUo^ving pinnules are stouter than those preceding. 

 They are 6 mm. long and are composed of about 20 segments, of which the first 2 are 

 short and the remainder about as long as broad, becoming longer than broad distally. 

 The distal ends of the segments are spinous and strongly overlapping, and the more 

 proximal segments are usually furnished with more or less prominent dorsal spines in 

 addition. Distally the pinnules gradually become shorter and more slender, the 

 distal pinnules being 6 mm. long, very slender, composed of about 20 segments, of 

 which the first 2 are short, the third is longer than broad, and the remainder are 

 elongated with slightly expanded ends, becoming about twace as long as broad distally. 

 Terminal combs usually occur on most of the genital pinnules, and at intervals on 

 those in the distal portion of the arms. 



The disk is covered with rather coarse papillae. The mouth is submarginal and 

 the anus subcentral. 



Localihj. — Albatross station 5153; Tawi Tawi group, Sxdu (Jolo) Archipelago; 

 Tocanhi Point, bearing S. 27° E., 2.1 miles distant (lat. 5° 18' 10" N., long. 120° 

 02' 55" E.); 89 meters; coral sand and shells; February 19, 1908 [A. H. Clark, 1909, 

 1911, 1912, 1918] (1, U.S.N.M., 35243). PI. 63, figs. i76, 177. 



History. — This specimen was originally identified as representing Mtiller's Aledo 

 multifida and was so referred to in 1909. The identification was made by means of 

 the keys in the Challenger report, where, curiously enough, it runs nicely to Adino- 

 mefra multifida, which was placed by Carpenter in the Parvicirra group. 



In another paper published in 1909 it was mentioned as Phanogenia carpenleri, 

 and the locality at which it was taken, together with notes on its arm division, were 

 given. 



The name Comaster carpenteri had been proposed in 1908 as a new name with the 

 following explanation: 



Doctor Carpenter in 1888 described a species of Comaster as AdinoJiietra 

 multifida, referring the name to Johannes Miiller; but Miiller, according to his 

 own statement, proposed the name multifida merely as a sid:)stitute for La- 

 marck's multiradiata. His idea was that, two quite diflerent species having 

 been called multiradiata — one by Lamarck and one by Goldfuss, but Lamarck's 

 description being quite worthless, whereas Goldfuss's is accompanied by an 

 excellent figure — the name sho^dd hold for the form with the recognizable 

 diagnosis, so he restricted it to Goldfuss's form. Subsequently he examined 

 Lamarck's original types, and from them drew up his diagnosis of multifida. 

 Now, thanks to Doctor Carpenter's investigations, we know what Lamarck's 

 multiradiata really is, and (as it is the same as the Linnaean and Retzian 

 Asterias multiradiata) of co\irse the specific name multiradiata must be retained 

 for it. Doctor Carpenter realized this, but he resurrected multifida for a 

 specimen which was among Lamarck's types, and difTered both from the 

 multiradiata of Lamarck and the multiradiata of Goldfuss; but multifida is a 

 pure synonym of multiradiata Lamarck and can not be used for any other 

 species. The form may be renamed Comaster carpenteri. 



