284 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



and those following gradually become about as long as broad ; the edges of the segments 

 and more or less of the dorsal surface are studded with numerous fine spines. PI and 

 P 2 are similar, 10-11 mm. long with about 21 segments of which the first two or three 

 are short, those following about as long as broad, and those in the outer half of the 

 pinnule elongate; the lower segments have roughened or more or less finely spinous 

 edges, and the distal are never more than twice as long as broad. These three pinnules 

 are moderately slender and taper gradually from the base to a slender and delicate tip. 

 P D is slightly larger than the others. PS is somewhat stouter than P 2 , especially in the 

 distal half, with about 18 segments of which the first three are short, the next three 

 about as long as broad and somewhat broadened, the remainder gradually tapering 

 and becoming elongate. The following pinnules to about Pj are similar but somewhat 

 shorter. The distal pinnules are about 8 mm. long with about 15 segments of which 

 the first is almost triangular, the second is irregularly trapezoidal, those following are 

 about as long as broad, becoming distally from half again to twice as long as broad. 



The disk is completely covered with a pavement of small plates. The brachial 

 and pinnule ambulacra are bordered with well developed side and covering plates. 

 Sacculi are abundant, but small and inconspicuous. 



The color hi alcohol is yellowish white, the division series and first three brachials 

 brown. Presumably in life the color is bright yellow with the division series and arm 

 bases brownish or olive. 



Notes. Although this description covers in a general way a large proportion of the 

 individuals that have been dredged in the Caribbean area, and especially about Cuba, 

 it is by no means adequate for the species as a whole, for this is perhaps the most 

 variable comatulid known. There is no fixed pattern in its ornamentation, which 

 varies from highly developed, coarsely tubercular or spinous, to completely absent; 

 in the form of the division series, which vary from almost flat, with or without water 

 pores, to strongly and evenly convex; in the composition of the division series, which 

 may be all 2 or all 4(3+4), or both in any combination; in the cirri, which vary from 

 short with short segments to more or less elongate with long segments ; in the pinnules ; 

 or in other features usually assumed to be reasonably constant in any single species. 



Many of the forms are so very widely different from others as angusta, gemmata, 

 and insculpta that it seems at first sight impossible to regard them as representing 

 the same species; but they are all interconnected hi such a way that any attempt to 

 separate them specifically is wholly impracticable. 



In his most excellent review of this species published in 1912 Dr. Clemens Hartlaub 

 said that the imposing number of fully, or at least almost fully, developed individuals 

 upon which he based his treatment of brevipinna in part bore specific determinations 

 in Carpenter's handwriting. In addition to these, he referred to brevipinna two speci- 

 mens which Carpenter had determined as representing a distinct species, pourtalesi. 

 The larger of the two Hartlaub regarded as the sole representative of Carpenter's 

 pourtalesi, referring the smaller to his new variety coronata. 



He remarked that since the original description of brevipinna was based upon a 

 single juvenile specimen, it is naturally inadmissible as a diagnosis of the species, for 

 a diagnosis based upon a study of medium sized and large specimens would be entirely 

 different. A comprehensive account of the characteristics of the species seemed to him 

 scarcely possible, and in view of the variability of almost all the characters it seemed 



