242 
SCANDINAVIAN FISHES. 
partly as spermatic vesicuke. In the females no cor- 
responding structure appears, except that the posterior 
part of the left ovary is often more or less distinctly 
separated from the remainder by a fold, being bent 
downwards and slightly forwards. In the female, how- 
Fig. 64. Ovicapsule from the ovary of a Gobius niger , taken on the 
19th of Joly, 1877, in the island-belt of Gothenburg; together 
with the egg enclosed in the capsule. Magn. 33 diam. 
ever, there appears another peculiarity: the eggs are 
enveloped in a special covering, which at the same time 
serves as an instrument of adhesion. This covering 
consists of an oval capsule (fig. 64), which, as long as 
it is retained in the ovary at least, is transparent; and 
within it the egg lies in a perfectly transparent fluid, 
in which, however, there float a number of more solid, 
refractive granules which collect in a girdle round the 
egg in the middle of the capsule. To the thick end 
of the capsule is attached the adhesive apparatus, which 
is shaped like a frill or cap, formed by a great number 
of contiguous, transparent filaments concentrically ar- 
ranged round this pole. In the ovary we have always 
found this frill hanging down over the capsule; but 
when the capsule leaves the fish, the frill is turned up, 
and the filaments spread, attaching the capsule to some 
object, with the thick pole as fastening -point". 
The Scandinavian Gobiince, which are all without 
palatine or vomerine teeth, belong to three genera, 
which may lie distinguished as follows: 
A : First dorsal fin with at least 5 rays. 
a: Jaw-teeth in several rows Genus Gobius. 
b: Jaw-teeth in one row, with single 
canines within this row _ Genus Aphya. 
B: First dorsal fin with two rays or rudi- 
mentary Genus Crystal l o<j obius. 
Genus GOBIUS. 
The fairly elongated body for the most part covered 'with ctenoid scales, but 
just behind it), during youth at least, naked or with cycloid scales, 
card or in several rows in the lower jaw and on the intermaxillary bones. 
Ventral funnel united to the ventral side only at the base, 
least depth of the tail is generally more than 2 / 3 
species is less than V 2 the length of the head. A relatively broad part 
the isthmus or to the corresponding membrane on the other side, the gill-openings 
Branchiostegal rays 5. Where the air-bladder is present, 
It is the giants and pigmies among the vertebrates 
that generally cause the systernatizer most trouble. A 
great part of the genus Gobius must be ascribed to the 
latter class. However various may lie the opinions as 
to the limitation of the species, in any case this genus, 
which was first established by Arteih 6 , in the form 
in which it has been adopted by Gunther 1 , with 
anteriorly {on the head and the parts 
Simple, but firmly set, pointed teeth in a 
First dorsal fin with at least 5 rays. 
Gape relatively small, and the jaws so short that the 
of the length of the lower jaw , which in all the Scandinavian 
of the branchiostegal membrane united to 
being thus more or less vertical, 
it is usually small. 
nearly 300 described species, is sufficiently large to 
require division, if only on the score of convenience. 
An attempt to accomplish this has been most thorough- 
ly made by BleekeiG. Whether the 36 genera which 
he adopted, all have a right to their position, is, how- 
ever, a matter of doubt, and in the Scandinavian fauna 
of little importance. 
“ Cf. Malm, Gbgs, Boh. Fn ., p. 442, tab. II, figs. 2 and 3. 
b Gen. Pise., p. 28. 
c Introd. t. Study of Fish., p. 486; Handl. d. Tchthyol., p. 345. 
d Arch. Neerb, 1. c. 
