The fishes of the Danmark Expedition. 657 



are found exclusively on this, being the most characteristic fish on 

 this otherwise so poor region. 



As soon as I put the young (ca. 25 mm) in a glass with water, 

 they swam about vigorously in it (by means of rapid, wriggling 

 movements of the tail) and then soon attached themselves to the 

 sides by means of their sucker; here they could remain for a 

 surprisingly long time. The young specimen of 18 mm was taken 

 on 19.— 9.— 06, all the others in July and August 1907 and 1908; of 

 these the 5 largest measured ca. 65, ca. 70, ca. 75, ca. 85 and 115 mm. 

 A cluster of eggs of the size of an acorn (probably a piece of one 

 much larger) was fished up in Danmarks Havn from 20— 30 m (hard 

 bottom) on 29.— 8.— 07. 



In the external appearance of thç fish the spinous equipment 

 is most characteristic. As is known, it is extremely variable with 

 regard to extension, development and time for its appearance. The 

 condition in my specimens is as follows. At 18 mm the four 

 nostrils are almost longer than the spines; the latter are best devel- 

 oped above and behind the eyes, as also under D^ and D^. The 

 sides of the tail and abdomen are uneven from projecting spines. 

 Rays are apparent in D^, that is, not covered by the fold of skin 

 which appears later; their points have no small shields, like A. 



A specimen of 26 mm differs from the foregoing in that there 

 is no indication of spines along the sides of the tail, whereas some 

 small spines are present on the gill-cover. The dorsal spines are 

 also found here, but have increased to a double row from above 

 the eyes to below the anterior edge of D- ; there are ca. 8 scattered 

 spines on the abdomen laterally; like the dorsal spines they are 

 well-developed with several points. The rays of D^ are now apparent, 

 the margin of the fin is still without shields, like A; nor are there 

 any spines between the sucker and the anal fin or on the forehead 

 between the eyes. 



A specimen of 27 mm differs from the above by the lateral 

 region of the tail and the interorbital space having smaller or larger 

 spines and by the abdominal spines being arranged in 2 parallel, 

 curved rows from the gill-cover to the anus; further, from the 

 sucker on both sides of the anus there is a row of spines which 

 continues along A. The dorsal spines are as in the previous 

 specimen, but now reach still farther back. The gill-cover likewise 

 has small spines and of the D^ rays (which are still apparent) the 

 first 5 each have a small, shield-shaped spine at the tip. 



At a length of ca. 70 mm we have a specimen which differs 

 greatly in regard to spines from two others almost of the same size 

 (ca. 65 and ca. 75 mm). All the spines are smaller and almost of 



