642 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



The specimen from Siboga station 273 is small, with 10 arms. The large perisomic 

 interradials of the young persist just above the radials in the interradial angles. 



The specimen from Siboga station 162 is medium sized, with 14 arms. 



In Bennett's specimen from the East Indies, according to Carpenter, the overlap 

 of the brachials is nearly as marked as in the types of timorensis, but the spines on the 

 outer cirrus segments are less developed, and fewer of the lower pinnules have a 

 terminal comb. Carpenter said that this individual is remarkable for having dimorphic 

 arms which are all grooved. The posterior arms taper much more rapidly than the 

 anterior ones and are composed of fewer segments, but their ambulacra are nearly as 

 well developed. 



The specimen from between Fremantle and Geraldton is typical, with 22 arms 

 70 mm. long. Of the 10 IIBr series, 7 are 4 (3 + 4) and 3 are 2. There are 2 IIIBr 

 series, one 2, developed internally, and one 4 (3 + 4), developed externally. The cirri 

 are VIII, 14, 7 mm. long. 



The 7 specimens from the Abrolhos as described by Dr. H. L. Clark have from 

 12 to 23 arms which are from 35 or 40 mm. to more than 100 mm. in length. The 

 cirri are few and weak and have the characteristic form and proportions. The color 

 as preserved is yellow brown. 



The example from Cape Baudin has 13 arms. Of the IIBr series, 3 are 4 (3 + 4) 

 and 7 are 2. The elements of the division series and the proximal brachials have the 

 distal edges thickened and very prominent, though smooth, quite as in certain species 

 of Cosmiometra. Some of the axillaries and the second segments following the axil- 

 laries have indistinct and blunt median keels, such as occur in some species of Cosmio- 

 metra. There are no cirri, the centrodorsal being small, flattened, and only slightly 

 raised above the dorsal surface of the radial pentagon. The interradial perisome is 

 filled with a large number of small calcareous plates. 



In the specimen from Mjoberg's station 5 identified by Gisle"n as Comaster 

 typica the centrodorsal does not cover the radials. 



The cirri are rudimentary, VII, from 0.5 to 1.5 mm. long, indistinctly segmented, 

 not reaching beyond the IBr^ 



Eight of the IIIBr series are 2, and 2 are 4 (3 + 4). The arm divisions are 

 1, 2, 2, 1 or 2, 2 or 2, 2, 2, 1. 



The 30 arms are from 45 to 70 mm. in length. There are no swollen genital pinnules. 

 The anal tube is subcentral and not inflated. The color in alcohol is yellowish red. 



In the description of this small specimen as published, only one of the IIIBr 

 series is given as 4 (3 + 4), and the number of arms is said to be 28. I am indebted 

 to Doctor Gisle"n for making the necessary corrections. 



It is quite possible that this is in reality an example of Comanthus timorensis. 



In 1 of the 2 specimens from Mjoberg's station 5 identified by Gisle"n as Comaster 

 multifida the centrodorsal is 3 mm. in diameter, slightly concave in the center and 

 with a swollen margin. 



The cirri are VII, 12-13, 5 mm. long. The first segment is three times as broad 

 as long, the second and third are longer, and the fourth and following are about as 

 long as broad. From the sixth onward the segments are provided with obtuse dorsal 



