1 .",1 1 BULLETIN 76, UNITED STATES NATIONAL MUSEUM 



The adambulacral spines stand alternately one and two to a plate, with a 

 variable number of diplacanthid plates following the adoral earina. The spines are 

 similar in form to those of alveolata, but a little longer toward the aetinostome. This 

 character is not of much practical use, owing to variability and the difficulty of 

 finding a method of comparison. In acanthostoma the longest adoral subambulacral 

 spine (of the second, to fourth plate) is about as long as the first eight adambulacral 

 plates measured on the furrow face; and in forma alveolata it is about seven plates in 

 length — perhaps 10 to 14 per cent less. In forma troschelii the spines are usually 

 still shorter. There are generally five pairs of adoral adambulacral plates joined to 

 form a narrow adoral carina and the aetinostome is very contracted, as in alveolata. 



The oral plates are not appreciably different from those of alveolata, there being 

 two (sometimes three) apical, and one suboral spines. 



The madreporic body is large, with fine, very irregular, striae; its outer edge is 

 about mid r. 



The papulae are very numerous and distributed as in alveolata; the abactinal 

 papular areas are often smaller and more numerous, but there is no constant dif- 

 ference. 



The straight pedicellariae are similar to those of troschelii — lanceolate, blunt or 

 pointed, small — and are similarly distributed. 



The crossed pedicellariae are very numerous among the papulae as well as 

 surrounding the spines. They have the distribution characteristic of the genus. 

 The abactinal pedicellariae measure about 0.30 to 0.31 (Unalaska); 0.2 to 0.26 

 (Tongass); 0.23 to 0.24 (Victoria). In profile the upper part of the outer surface of 

 the jaw is flattened or even slightly depressed. Northern specimens have larger 

 pedicellariae. 



Variations. — The principal variations of the type form have already been noted. 

 The abactinal spines vary in robustness as well as in the precise shape. In the largest 

 specimen from Victoria, British Columbia, they are cylindrical or slightly tapered, 

 two or three times as high as thick, blunt or truncate, with a sulcated tip. This is 

 the usual form but a number of spines especially near the end of the ray are thicker 

 and clavate. There is less difference in spine form than in alveolata. The differences 

 in the arrangement of the abactinal spines— by which the papular areas are more or 

 less completely fenced in — constitute one of the major variations. There is also a 

 conspicuous difference in the robustness of the rays. The largest specimen from 

 Victoria, British Columbia, with R 310 to 330 mm., and considerably shrunken in 

 drying has the ray 60 to 65 mm. broad at base, or about one-fifth R. The stoutest 

 specimen has the arm breadth about one-fourth R (Unalaska), while another from 

 the same locality has the arm breadth one-fifth to one-sixth R. 



Tongass, southeast Alaska, Plate 64, Figure 2. The specimens from this locality 

 have been referred to acanthostoma, although they present certain points of difference 

 which at first sight give them a very distinctive appearance. 



The rays are very slender and long, R equalling about Q)i br measured at widest 

 part. The rays taper very gradually from this narrow base. The abactinal spines 

 are unusually numerous and are small. They completely surround the papular areas, 

 dividing the abactinal surface into very numerous irregular meshes. The carinal 

 series is pretty well marked at the base of the ray. The spinelets vary in different 



