GRENADIER-FISHES. 
591 
the body, but also that in the dorsal margin, in front 
of the first distinct ray of the second dorsal fin, there 
lies a row of soft, whitish, dermal protuberances, while 
from each of the posterior among these protuberances, 
forward to a point 3 /jo of the way along the body from 
the tip of the snout, there projects the tip of a ray, 
extremely small and scarcely perceptible. There can 
be no other explanation of this than that the second 
dorsal fin originally extended much further forward 
even than it does in adult specimens, but that its an- 
terior part undergoes a reduction. Still, it is a rule 
— though with considerable individual variations — 
as Collett has pointed out, that the number of rays 
in this fin as well as in the anal fin increases with acre 
during the latter period of growth, even while the length 
of the body increases from half a metre to a metre. 
This increase — from about 100 to about 200 in the 
former fin" — is in its magnitude an extremely un- 
common phenomenon in the class of fishes. Collett 
also connects it with another great abnormity, the in- 
crease in the number of the vertebras — 13 — 14 ab- 
dominal and 49 — 75 caudal vertebras — which proceeds 
according to the same rule and with the same exceptions 6 . 
The first dorsal fin is set further forward than in 
the rest of the Scandinavian Macruroids, its beginning 
lying in front of the end of the first fifth of the body. 
It moves somewhat further back, however, with age. 
The length of its base is less than i / 5 (79 — 58 %) of 
the least breadth of the interorbital space. The first 
ray is as usual rudimentary, the second ray both the 
longest and the thickest, nnarticulated, and measuring 
rather more than half (50 — 55 %) the length of the 
head. The other rays — except the last, which is also 
simple — decrease in length uniformly, but sharply, 
the fin thus • acquiring a more triangular form than in 
Macrurus Fabricii. 
The anal fin begins at a distance from the tip of 
the snout which in young specimens measures about J / 4 
of the length of the body, in older ones about 1 / 3 thereof. 
It is of fairly uniform height, with slightly concave 
margin: the first ray is slightly shorter than the next 
ones, the length of the rays increases or decreases ac- 
cording to the curvature of the belly, and behind the 
middle of its length the height of the fin is greater than 
the depth of the body at the^same point. Round the 
tip of the tail the anal fin coalesces with the second 
dorsal fin into a symmetrical tip. The vent is situated 
just in front of the beginning of the anal fin. 
The pectoral fins are set just in front of the per- 
pendicular from the beginning of the first dorsal fin. 
They are evenly rounded at the tip, and their length 
is perceptibly more than l / 3 (58 — 6l7 2 %) of that of 
the head. The first ray is extremely short, and the 
ninth (8 — 10) is the longest. 
■ The ventral fins are situated in old specimens in 
front of the perpendicular from the beginning of the 
first dorsal fin; but during youth they lie considerably 
behind it. The usual removal in a forward direction of 
these fins thus takes place in this species too, the distance 
between the foremost point in the insertion of the ventral 
fins and the beginning of the anal fin in young speci- 
mens (12 cm. long) being only slightly more than Vs 
(22'8 %) of the distance between the anal fin and the 
tip of the snout, but in old specimens nearly 1 / 3 (30 or 
31 %) thereof. The first ray is elongated to a long tip, 
hairlike at the extreme end, which extends to the lie- 
ginning of the anal fin; the second ray is about V 3 as 
long; and the length of the other rays is such that the 
margin of the fin is somewhat convex. 
The jaw-teeth form a kind of transition from the 
preceding species to the following one. On the inter- 
maxillary bones (in adult specimens) they are set in a 
card containing 5 or 6 rows, but behind this card, for 
the greater part of the length of the bones, they are set 
in only 2 rows. In the lower jaw the case is similar, 
but here there is only one row behind. In front the two 
intermaxillary cards have an empty space between them. 
The pharyngeal teeth — 2 or 3 rows on the two anterior 
upper pharyugeals, 4 rows on the posterior upper pha- 
ryngeal; 6 rows on the oblong, curved lower pharyugeals, 
with the largest teeth in the innermost (hindmost) row 
— are straight and conical, without the sharp break at 
the apex which characterizes them in Macrurus Fabricii. 
The anterior nostrils are as usual round, the posterior 
oblong and set transversely just in front of the orbital 
margin; but the latter are generally remarkably narrow, 
this being due to the projection throughout the whole 
of their length of the marginal membrane in front. 
O o 
° Still, it is not a superfluous remark that mistakes may easily be made in the counting, partly on account of the magnitude of the 
numbers, and partly because the two lateral halves of the otherwise simple rays easily fall apart, and the rays are thus counted twice over. 
b Both the rule and the exceptions should be kept in mind, when one has to judge of the systematic significance of the varieties or 
races which, especially among the Salmonidce and Clupeidce , have been based upon the variations in the number of the vertebra;. 
Scandinavian Fishes. 
75 
