94 BULLETIN 82, UNITED STATES NATIONAL MUSEUM VOLUME 1 



very different, and was at once distinguishable from all other Japanese comatulids by 

 its characteristic cirri of relatively short segments. 



Further investigation showed this species to be most nearly related to the South 

 African species originally described, through a misconception of its characters, under 

 the name of Cominia occidentalis. 



In ignorance of this decision, Gislen in 1927 included minuta in the synonymy of 

 Iridometra adrestine. In the two specimens for which he cited the proportions of the 

 longest cirrus segments both had them from two to three times as long as wide. Pre- 

 sumably the rest were similar in this respect and none could have been referable to 

 minuta. 



ANNAMETRA OCCIDENTALIS (A. H. Clark) 



FIGURE 6 



Cominia occidentalis A. H. CLARK, Die Crinoiden der Antarktis, 1915, p. 104 (collected by the Gauss 

 at Simon's Bay), p. 164 (detailed description; localities; parasitized by Eulima capensis), p. 165 

 (description with figure of Eulima capensis, by Thiele), pi. 10, figs. 1-5. H. L. CLARK, Ann. 

 South African Mus., vol. 13, 1923, p. 228 (in key), p. 231 (station; remarks). 



'Comium' PELSENEER, Bull. Soc. Zool. France, vol. 53, 1928, p. 172 (parasitized by Eulima capensis). 



'Capillaster capensis' PELSENEER, as above. 



Annametra occidentalis A. H. CLARK, U.S. Nat. Mus. Bull. 82, vol. 1, pt. 2, 1921, pp. 647, 648, 681, 

 723; The Danish /nsoZ/-Exped., vol. 4, No. 5, Crinoidea, 1923, p. 41 (range); Journ. Linn. Soc. 

 (Zool), vol. 36, 1929, p. 659 (locality); Proc. U.S. Nat. Mus., vol. 83, 1936, p. 247 (type species 

 of Annametra'); John Murray Exped. 1933-34, Sci. Reports, vol. 4, No. 4, 1937, p. 102 (range), 

 p. 105. GISLEN, Kungl. Svenska Vet-Akad. Handl., ser. 3, vol. 17, No. 2, 1938, p. 4 (listed), 

 pp. 20-21 (supplementary description; history); text-figs. 18-26, p. 19. A. M. CLARK, Trans. 

 Roy. Soc. South Africa, vol. 33, pt. 2, 1952, p. 203 (Cape Town University stations). 



Diagnostic features. PI and P 2 have up to 35 very short segments. The arms are 

 about 50 mm. long in the adult and the longest cirri measure up to 15 mm. 



There is little danger of mistaking this species for any other member of the family, 

 for its general appearance and the character of the first two pinnules give it a very 

 strong resemblance to one of the 10-armed species of the Comasteridae with which it 

 was at first confused. 



Description. The centrodorsal is small, thin discoidal. Its bare dorsal pole is 

 flat or slightly convex and measures 1 mm. in diameter. The cirrus sockets are ar- 

 ranged in three closely crowded, irregular marginal rows. 



The cirri are XXXV-XL, 14-19, and the longer ones measure up to 15 mm. in 

 length. The first segment is shorter than the second, which is half again as broad as 

 long, the third is about as long as broad or slightly longer than broad, the fourth or 

 fourth and fifth are the longest, one third to half again as long as the median diameter, 

 and the following ones gradually decrease in relative length so that the six outermost 

 are about as long as the median diameter. The longer proximal segments are slightly 

 constricted centrally and in lateral view appear comparatively slender; as the segments 

 distally become shorter and laterally compressed they also become broader in lateral 

 view so that the short outer segments may be nearly or quite twice as broad as the 

 longer proximal. All the segments have a perfectly straight and smooth dorsal profile, 

 and the outer also have a straight ventral profile. The opposing spine is minute and 

 terminal in position. The terminal claw is somewhat longer than the penultimate 

 segment, moderately stout and moderately curved. The cirri in general suggest those 

 of Antedon petasus. 



