DISCUSSION OF SPECIES A^B THEIR DISTRIBUTION. 201 



readies to below tlie anterior margin of the eye; preoperculuui with two spines at the angle; 

 operculum terminating in two obtuse points, separated by a notch. The dorsal spines are of 

 moderate strength, and tubercular; the second to the fifth are the longest, and as high as 

 the body; the posterior <mes gradually decrease in length; caudal forked; anal spines 

 short; pectoral of moderate length. Each ventral spine is inserted in advance of tlie pec- 

 toral; it is tubercular, very long, its length being equal to tlie height of the body. The 

 vent is situated far backwards, and its distance from the head is more than the length of 

 the latter. Scales apparently none; lateral line bent downwards anteriorly. Uniform sil- 

 very. 



Radial formula: D. 18 | IS; A. -,%; V. 1. 



The specimen described by Giinther is only 2;^ inches long, and apparently young. 

 DICROTUS PARVIPINNIS, Goode and Bkan, n. s. (Finmo 212.) 



The steamer Albatross captured several examples of a species of Dicrotus in the western 

 Atlantic. One at station 2.>>7, N. lat. 39^ m' 4.5", W. Ion. 70° 50' 30"; another at station 

 2542, K lat. 40° 00' 15", W. lou. 70=42' 20", and a third at station 2G'J1, N. lat. 34° 39' 

 15", W.lou. 75° 33' 30" off Cape Hatteras. The first two were obtained iu excursions off 

 Newport, R. I. The example from station 2C01, measuring nearly 1 inch in length, is 

 referred to in the description, the other two not being at the present time accessible to us. 

 We are led to refer this individual to the genus Dicrotus because of the absence of a dagger- 

 shaped spine behind the vent and because of the similarity in the number of dorsal spines, 

 we assume tliat Dicrotus of tiiinther is the young of Promethichthys atlanticus. Our spe- 

 cies resembles very strongly the illustration by Dr. Liitken, in "Si)olia Atlantica'* of Nca- 

 lotus tri2)es, and as there is no indication of the presence of a dagger-shaped spine even in 

 an example of nearly 1 inch in length we can not believe that our individual belongs to Nea- 

 lotus; it possesses the characters o{ Dicrotus much more nearly and we shall refer to it under 

 that name. The length of the individual described is 21 millimeters to the base of the 

 caudal. The greatest height of the body(3A millimeters)iscontained6 times in the standard 

 length. The least height of the tail is one-half the length of the eye. The length of the 

 eye is one half the length of the postorbital part of head and somewhat more than one-sixth 

 of the total length of the head. The length of the head (9 millimeters) is contained 2J times 

 in the standard length. The upper jaw reaches to the vertical from the front of the orbit, 

 and the lower jaw to below the beginning of the postorbital third of the orbit. The nostril is 

 placed ill front of the eye a distance eipial to two-thirds the length of the eye. Three large 

 fangs in the upper jaw, a large fang near the tip of the lower jaw, and 8 smaller teeth. 

 Three weak diverging spines on the border of the preoperculum. The dorsal originates at a 

 distance behind the i\ve about etpial to one and one-half times the length of the eye, or about 

 over the middle of the operculum. The spines an; all serrated; the first five are about 

 equal, their length nearly equal to that of the postorliital part of the head; they diminish 

 gradually in size from the fifth, and the last is only about one-half as long as the eye. Tlie 

 longest ray is scarcely more than two-thirds as long as the eye. The anal r)rigin is under 

 that of the soft dorsal. The length of the anal base is about one-half that of the snout. 

 The soft dorsal base is not much longer tliau the anal base. Tlie first anal si)ine is onetliird 

 as long as the eye; the second s])ine is one-third as long as the soft dorsal base. TIk^ long- 

 est anal ray is two-thirds as long as the eye. The ventral origin is under the sixth si)iue 

 of the dorsal; the spine is strongly serrated; its lengtli ei|ual to that of tlie ninth dorsal 

 spine, slightly more than one-eighth of the standard lengtli. The ventral contains, also, 

 either a single bifid ray or two simple rays. The pectoral origin is under the third spine 

 of the dorsal. The length of the fin is onefourth the length of the head. Tlie cand.il is 

 moderately forked; the middle rays one-half as long as the snout and about two-thirds as 

 long as the external rays. 



Radial formula: D. xxi, 11; A. ir, S; P. 12; V. i, 1, or i, 2. 



Color silvery; caudal peduncle and top of back at base of dorsal brownish. 



