g2 M.lTlN 76, UNITED STATES NATIONAL MUSEUM 



slender, compressed plates bind the Buperomarginals to the plates above (i) and to the 

 inferomarginals, either directly or by a second similar plate, closely joined in vertical 

 series; while cue or two similar, smaller plates may continue the series across the 

 actinal plates, attached to (heir dorso-ventral lobes, which are themselves somewhat 

 carinate on the coelomic side. These plates thus form ridges separating shallow 

 pockets along the ventrolateral and lateral wall of the ray, into which the viscera are 

 normally crowded. 



The ambulacra! ossicles are not especially crowded, there being 12 in the space 

 of 10 mm. counting from the outer end of the contiguous oral adambulacral plates 

 (oral carina). The form of the pores is shown by the figure. (PI. 34, fig. 1, am.) 



Variations. — The smallest specimen of Gray's type series is "376," or No. 110, 

 \\ hich has R 70 mm. (longest ray), r 16.5 mm. (Perrier gives 17 mm.), breadth of ray 

 at base. 1 ."> mm. It differs from the type and cotype in having a more closely knit 

 skeleton, there being fewer and shorter secondary connecting plates. As a result, 

 the spiniferous primary plates are less widely separated, the papular areas are rela- 

 tively much smaller than in the type, and the spines are very much closer together. 

 The median radial plates are larger than the other dorsal plates and the series is 

 clearly distinguishable both by reason of the larger size of the plates and by their 

 more numerous, close-set, capitate spinelets. In form these might be described as 

 very depressed globose with numerous, fine, rather deep vertical, or meridional, 

 striae, and there are 12 or even more to the larger plates, but only five or six are of 

 major size, the others being crowded on the periphery. This number is greater than 

 is to be found on similar plates of the type. The proximal supermarginal plates 

 carry one or two, sometimes three, longer, clavate spines, while the plates of the outer 

 half of the ray generally have three or four spines subequal to the abactinals, in a 

 transverse, somewhat arcuate, series. The inferomarginal spines, one or two to a 

 plate, are slightly longer and heavier than the superomarginals (especially distally is 

 the difference conspicuous) while the outer actinal spines are subequal to the infero- 

 marginals. The inner series is only starting, and consists of a few smaller spines at the 

 base of the ray. There are generally two adambulacral spines to each plate, or 

 distally two and one alternating. The pedicellariae (pi. 33, figs. 3, 5) are essentially 

 like those of the larger specimens. The subambulacral crossed pedicellariae have the 

 characteristic longer serrulate straight-edged jaw. (Fig. 3a.) 



As previously mentioned, this specimen presents a much more compact appearance 

 than does the type. Apparently increase in stature is due primarily to an increase 

 in the size and number of secondary connecting plates, resulting in a separation of 

 the primary spine-bearing plates and a generally more open skeleton. The Gulf of 

 Georgia specimen is less compact than No. 110, but more closely knit than the type. 



This is the specimen described by Perrier (1875, p. 69) as Asterias douglasi. I 

 have seen a \ i it j Miuilar specimen of L. polaris borealis Perrier from Labrador (U. S. 

 Nat. Mus 1376). The resemblance is uncanny. 



Professor Verrill (1914, p. 112) has described and figured the specimen from the 

 Gulf of Georgia (Museum of Comparative Zoology, No. 1181). Through the courtesy 

 of Dr. II. L. (lark 1 have received this specimen for study. I think it is referable to 

 Leptasterias p. katherinae, although as might be expected from its smaller size and 



