362 BULLETIN 82, UNITED STATES NATIONAL MUSEUM VOLUME 1 



and are usually disposed quite irregularly. The calcareous network extends up into 

 the edges of the food groove, but it does not form regular side- or covering-plates. 

 There are gaps in it for the reception of the sacculi. Where these plates are absent, 

 as at the ends of the ovarian pinnules and in the later pinnules of the arm, the cal- 

 careous network is still largely developed at the sides of the ambulacra, though without 

 any regular arrangement into plates. 



In 1903 Mortensen described and figured well-developed covering plates in glacialis 

 from eastern Greenland, and showed that barentsi is merely a variety of glacialis with 

 these plates more developed than is commonly the case. 



In 1904 Grieg confirmed Mortensen's disposition of barentsi. 



In 1905 Doderlein also confirmed the identity of barentsi with glacialis, and re- 

 marked that covering plates, more or less developed, are always found in glacialis 

 except in the youngest specimens. 



Michailovskij in the same year also assented to this interpretation of barentsi; he 

 says further that glacialis is subject to great variation which apparently to a large 

 degree is correlated with local ecological conditions. 



In 1908 I described under the name of Heliometra juvenalis some very curious 

 specimens from off Cape Raper, Davis Strait, in 109 meters (Esquimaux collection). 

 The arms are 75 mm. long with the middle and outer brachials wedge-shaped and 

 longer than broad. The cirri are XX-XXX, 15-20, from 10 to 12 mm. long. Pj is 

 2 mm. long, with 4 or 5 squarish segments. P 2 is similar, but slightly longer. P 3 is 

 still longer, with 8 segments. P 4 is 4 mm. long, with about 12 segments. P 5 is 6 mm. 

 long, with 15 segments, mostly rather longer than broad, of which the third to fifth 

 bear a large rounded gonad which is repeated on the following 14. 



Mr. Rodger, in his account of the collection which he made while on the Esquimaux, 

 says that at this locality he obtained crinoids in great numbers, including a species 

 which he was assured by Canon A. M. Norman was new. Undoubtedly this was the 

 form referred to. 



These specimens scarcely resemble Heliometra glacialis in any detail of their struc- 

 ture, yet I am convinced that they are nothing but aberrant examples of this species. 

 In H. glacialis of this size the ossicles of the IBr series and the lower brachials are spiny, 

 whereas in these they are smooth. But they are also smooth in the variety quadrata. 

 The lower pinnules are obviously stunted and functionless; the outer segments of these 

 pinnules in glacialis being exceedingly small and delicate, it is possible that in these 

 they have been destroyed by some acidity in the preserving fluid. The cirri, however, 

 agree with the cirri of small glacialis, and t his feature is sufficiently diagnostic to indicate 

 their true identity. 



Abnormal specimen. Levinsen (1886) described a 9-armed specimen from the 

 Kara Sea. In this (Part 2, pi. 37, fig. 1231) the first to fourth brachials on each of the 

 two derivatives from one of the axularies are approximately normal, though only one 

 of the fourth brachials bears a pinnule, which is rudimentary. Beyond the two ap- 

 posed fourth brachials is a single ossicle which covers the entire distal surface of both; 

 following this is a series of five similar ossicles of diinishing size beyond which is a series 

 of normal brachials. 



Localities. Northeastern part of Georges Bank ; schr. Martha and Susan; Glouces- 

 ter Donation 931 (3, U.S.N.M., 4912). 



Western Bank; schr. Mystic; Gloucester Donation 806 (1, U.S.N.M., 35882). 



