Clark — A Revision of Thalassometridse. and Himerometridse . 17 



Ends of the basal rays visible as tubercles in the interradial angles. 

 Eadials only visible in the angles of the calyx, sometimes entirely hidden; 

 I Brj very short, liandlike or more or less crescentic ; i Br.^ rhombic, over 

 twice as broad as long; ii Br 2, always present in the full series; iii Br 2, 

 developed 2, 1, 1, 2, not always present. The i Br, division series, and 

 lower bachials are in very close apposition and very sharply flattened 

 against each other; these joints also have the edges all around slightly 

 everted; synarthrial tubercles broad and rounded, not prominent; i Br 

 and division series with a low broadly rounded more or less linear tubercle 

 on their component joints. 



Arms 20 to 30, moderately deep and compressed, but rounded dorsally 

 and never carinate; first ten brachials o1)long, about twice as broad as 

 long (the first two wedge-shaped), then triangular, broader than long, 

 soon becoming as long as broad, and in the terminal portion of the arm 

 wedge-sliaped, and more or less elongate; the brachials have more or less 

 prominently overlapping and finely spinous distal edges. 



Pi considerable longer, and stouter, than P.^, though not especially 

 enlarged ; following pinnules decreasing to about two-thirds tlie length of 

 P], then slowly increasing in length distally, the distal pinnules being 

 rather longer than Pj. 



Color ( in spirits ) . — ' ' White with faint patches of brown here and there," 

 to uniform dark brown. 



Distribution. — Sahul Bank, north Australia, northward and northeast- 

 ward to Japan and the Hawaiian Islands. 



Depth. — The only records are for Hawaiian species, which were taken 

 between 319 and 355 fathoms. 



Included species: 



Cosmiometra crassicirra (A. H. Clark) 

 delicata (A. H. Clark) 

 " komachi (A. H. Clark) 



" woodmasoni (Bell).* 



I am unable to properly place the following species belonging to this 

 sub-family, liecause of a lack of material upon which to liase comparisons, 

 and inability to grasp the characters ni toto from the published diagnoses. 



* It is possible that "Antedon adriani," which was brought back by the " Discovery," 

 belongs to this sub-family, tliough there are grounds for believing it to be a member of 

 the Tropiometridre. The proved occurrence of either of these families so far soutli would 

 be of the greatest interest in its bearing on zoogeography. Unfortunately, neither the 

 diagnosis nor the figure (which differ radically from each other) affords any clue to the 

 family, generic, or specific relationships of tlie form; it is certii in, however, that it can 

 not liave nuich in connnon with Heliometra glacialis, with whicli it is compared ; it does 

 not belong in the same family. 



It is to be hoped that " Antedon adriani " will soon be adequately described and 

 figured and that " Promachocrinus kerguelenensis " {sic) and " Antedon au.stralis " will 

 be reidentifled, especially the " young " of tlie latter, whicli possibly belong to a differ- 

 ent genus from the larger ones. Some clue sliould ha ve Ijoeu given as to which of the two 

 " Antedon australis " described by Carpenter is maint, though the supposition is tliat it 

 is the later one. 



