346 



BULLETIN 100, UNITED STATES NATIONAL MUSEUM. 



with numerous intermediate sizes which I give in the following 

 table: 



Whatever the size may be, the general form of the individuals 

 remains the same; in the dorsal median line the arms are raised up 

 into a very pronounced dihedral angle, and their cross section is 

 triangular; the spines are almost confined to the ventral surface. 

 In the smallest specimens the triangular form of the arm is naturally 

 less marked, though in them the dorsal median line is already 

 strongly elevated. 



In none of the Abatross specimens have I found the ventral pores 

 as developed as in the Investigator specimens which served me as 

 the type series. For comparison I show on plate 77, as figure 1, one 

 of the latter in which at least five pairs of pores can be recognized, 

 and even the traces of a sixth pair. In describing the species I re- 

 marked that the number of these pores was not constant, and that in 

 certain specimens there were three or four pairs only (Koehler '96, 

 p. 324). The Albatross specimens in which these pores are most 

 developed are that from station 5492, in which the disk measures 

 45 mm. in diameter, and that from station 5494, in which the 

 diameter of the disk is 41 mm. ; I have shown the latter in figure 2 ; 

 two pairs of pores are visible on each arm, and the third is usually 

 indicated. The specimens from station 5274. which have the disk 43 

 mm. in diameter, also show two pairs of pores at the base of each 

 arm, but the second pair is only slightly evident. On the other large 

 specimens these pores are not generally visible; but I find one pair 

 well marked, though on only three arms, in the specimen from sta- 

 tion 5697 (fig. 2). On other smaller specimens the pores of the first 

 pair are often present ; I find them for instance in the three individu- 

 als from station 5495, in which the diameter of the disk measures, 

 respectively, 27, 28, and 33 mm.; these pores are especially well 

 shown on two specimens of which I show the ventral surface in fig- 

 ures 9 and 11. I also find them on a specimen from station 5216 with 

 a disk diameter of from 27 mm. to 28 mm., and on another (fig. 6) 

 from station 5587 with a disk diameter of 19 mm., in which the pores 

 are visible on two arms only. On the small specimen from station 

 5114 (fig. 3), in which the disk is only 11 mm. in diameter, this first 



