LYCODOJDS. 
615 
than it (Lutk., of Ly codes Vahlii). In the first place, 
however, this character is subject, according to Collett’s 
measurements, to considerable variations — even in the 
same specimen, a male 58 cm. long, which has been 
kindly lent me for examination by Professor Collett, 
the row of palatine teeth on one side differs from that 
on the other, the right containing 8 teeth, the left 12. 
In the second place, we have here to deal with a sexual 
character, as both Lutken and Collett have pointed 
out: the males in general have longer rows of teeth 
than the females, both on the intermaxillaries and on the 
palatine bones. Thus, when we remember that in Ly- 
codes Esmarldi the four rows of teeth, or at least three ' 
of them, according to Collett’s measurements (. Nordh . 
Exped., 1. c., p. 90), may be equal in length to each 
other even in specimens 3 dm. long, the character can 
scarcely be regarded as a universally valid distinction 
between the two proposed species. 
The body, which just behind the head is terete, 
about as broad as deep, is gradually compressed later- 
Fig. 150. Ly codes Vahlii , gracilis , twice Ihe natural size. Taken 
at Drobak in 1866 by M. Sars. The property of the Zoological 
Museum of Christiania University. 
ally from this point. The depth at the beginning of 
the anal fin is 10 — 1 3 1 / 2 % of the length of the body 
in old specimens, but only about 8 Vs * thereof in the 
fry, assuming that we are right in referring to this 
species, the fry of which are otherwise unknown, the 
Ly codes gracilis of M. Saks (fig. 150). The type-spe- 
cimen of this proposed species has been placed at our 
disposal for purposes of examination by the kindness of 
Professor Collett. In all essential characters of form 
the relation of this specimen to Ly codes perspicillum and 
L. Rossi — the assumed fry of the preceding species — 
is the same as that of Lycodes Vahlii to L. reticulatus 
in their adult state. The distance between the dorsal 
fin and the tip of the snout is 23 % c \ the distance be- 
tween the anal fin and the latter somewhat more than 
38 % b , and the length of the tail 61 % e of the length 
of the body. The length of the head is only 34 % d 
of that of the tail. 
“ In L. perspicillum and L. Rossi 26 1 / 2 — 29 %. 
b !> H >> !) )! >> 43 1 / 3 — 46 /. 
c ,, ,, ,, ,, ,, ,, 55 ’/ 2 53 %. 
rl „ „ „ „ „ „ 46—40 %. 
e Sars found the same uumber in bis Lycodes gracilis. 
In this species as in the preceding one, the pectoral 
fins are broad and rounded, the membrane being espe- 
cially thick below. Their length from the upper angle 
of the insertion varies between about 14 and 12 1 / s % 
of that of the body. The ventral fins are set just behind 
the line between the lower angles of the gill-openings, 
and so near each other that the distance between them 
is scarcely equal to the breadth of their base. They 
are of an oblong, triangular shape, and their length in 
the male is about equal to the longitudinal diameter of 
the eyes, in the female, according to Lill.jeborg, only 
% thereof. The skin is so thick that it is difficult to 
count the rays without dissection: Brown-Goode and 
Bean state the number at 3 ", Reinhardt and Collett 
at 4, Lilljeborg at 5. 
The vertical fins are of fairly uniform height, but 
ascend anteriorly in a rounded slope and coalesce behind 
into a sharp (in young specimens) or rather obtuse (in 
older ones) point. The anal fin is somewhat lower than 
the dorsal, and its length about i / 5 of that of the latter. 
The longest rays of the dorsal fin, which are situated 
above the anterior part of the anal tin, measure about 
half the depth of the body at the beginning of the latter. 
In the dorsal fin, according to Reinhardt, only the first 
ray is simple, with the exception of the rays that occupy 
the extreme tip of the tail; in the anal fin, according 
to Lilljeborg, the first three rays are simple. 
The scales resemble those of the Eelpout in form 
and structure. They are thin, flexible, and in general 
elliptical, with extremely thin, central nucleus of the 
same form and dense, fine, concentric strhe (ridges), 
which are prettily broken by numerous, straight, radiat- 
ing grooves over the whole surface. When they are 
allowed to dry under the thin skin that covers them, 
we therefore see, in old specimens, small, fine notches 
all round the margin of each scale, the traces of the 
ends of the grooves at the edge. The scales of the 
lateral line are more circular, with the nucleus thickened 
and with a canaliculate duct, contracted and almost 
closed at the middle by the growth of the margins 
towards each other. Scales of this structure appear even 
where the lateral line is externally invisible. The longest 
axis of the scale generally lies transversely across the 
body. In a male 58 cm. long one of the largest scales 
Scandinavian Fishes. 
78 
