69 



pairs) at the basis of the arm or farther out, sometimes only 

 on one side of the arm. — At Angmagsalik are taken some 

 small specimens (8) together, whose habitus is so much the 

 same, that it would be quite absurd to make two species of 

 them. Now most of them have the adambulacral spines dispo- 

 sed in a single series or alternately one and two, but in one 

 specimen they form a regular double series in the inner half 

 of the arms, also in a few of the other specimens 2 — 3 paired 



« 



spines may be found together in some part of the arms. The 

 same may be found in the specimens from Henry Land. This 

 character is thus not to be relied upon. 



As distinguishing characteristics the following features are 

 further pointed out: in A.. groenlandica there is an impression 

 across the basis of the arms, the disc thus forming a <ipolygon 

 with rounded corners»; this feature is not found in A. Müllerin 

 »vel rarissime indistincte». (Levinsen p. 392). Though mostly 

 very distinct this character is neither to be relied upon, as it 

 may be found well developed also in typical A. Müllerin as well 

 as in other species (rubens). 



In A. groenlandicd there are no pedicellaries on the disc 

 (Lütken: Grønlands Echinodermata p. 29. Levinsen: loc. 

 cit.), whereas pedicellaries are found here in A. Mülleri. But 

 in many specimens of Mülleri there are found only very few 

 pedicellaries on the disc, and in groenlandica a few pedicella- 

 ries may also be found there in some specimens. — In A. Mülleri 

 there are always pedicellaries on the adambulacral spines, in 

 groenlandica pedicellaries are not found here. This character 

 evidently is the best one, but neither this is to be relied upon 

 with certainty, as there may sometimes be found pedicellaries 

 on the adambulacral spines also in groenlandica. 



Further the madreporite is said to be encircled and almost 

 covered by spines in groenlandica, which is not the case in 

 Mülleri (Sars. Op. cit. p. 91). This seems to be of very little 

 specific value, as well because this circle of spines is very 



