THREE-SPI N ED STICKLEBACK . 
651 
well as the other plates and bones that appear at the 
surface of the body, are rough with grooves alternating 
with granulated ridges, corresponding to the radiating 
grooves and striae in the structure of an ordinary scale. 
This marking is most handsome and most regular on 
the surface of the operculum, where it radiates from the 
articulation (the upper anterior corner) of the bone, with 
the grooves, in their even and handsome curves, and the 
almost smooth ridges terminating at the hind inferior 
margin thereof. In the gymnurus form, where the ossified 
dermal growths are generally more or less reduced, the 
external bones of the head may also be covered to a great 
extent by the skin, thus losing their Ganoid appearance. 
Of the structure of the fins we know already how 
the spinous dorsal rays vary in number and position. 
The ordinary place of the first spinous ray lies at about 
the end of the first third of the length of the body, of 
the second spinous ray at the end of its second fifth, 
and of the last spinous ray a little behind the middle 
of the body. The most usual supernumerary ray lies 
a little in front of the middle of the body. On this 
head we must remark, however, that in later life the 
first spinous ray is generally removed distinctly in a 
forward direction, its average distance from the tip of 
the snout thus sinking from 32‘8 to 30‘8 % of the length 
of the body while this length increases, on an average, 
from 55'8 to 69'7 mm. The spinous rays are of a 
more or less broad, triangular form, with concave or 
nearly straight sides, and with the front convex and 
the back concave. The front is rough and lacunose, like 
the above named parts of the dermoskeleton; the lateral 
margins are spiniferous for a greater or less distance 
from the base out towards the subulate tip. The rough- 
ness and the spines partially disappear, however, under 
the same conditions as the plate-armour of the body, 
in the gymnurus forms. These spinous rays also vary 
considerably in length. As a rule the second is the 
longest, though the first may sometimes equal it in 
length; the maximum length we have found in the latter 
case, is 42 l / 2 % of the distance between the first spinous 
ray and the tip of the snout, while the minimum lengths 
(in percent of the same distance) have proved to be 
16 in the case of the first ray, 17 in that of the second. 
The length of the last spinous ray varies between 7 */ 2 
and 17 % of the said distance; and this ray 
forms a sharp, almost geniculate curve. 
these spinous rays stand straight up; when depressed, 
they lie in a straight line with each other, the tip of 
each ray sometimes (when the length of the rays is so 
great) resting on the base of the ray behind it. The 
interspinal plates on which the rays are set, are fur- 
nished with a groove in which the latter are partly re- 
ceived when depressed. These spinous rays, as well as 
the others (those of the anal and ventral fins), are ar- 
ticulated in such a manner that they can be locked fast 
by the fish at any angle whatever — though generally 
and at most in an upright position — and cannot be 
depressed by force, unless they are simultaneously lifted 
a little way. This result is attained by the presence 
at the middle of the base of each spinous ray, in ad- 
dition to its true articulation — a trochlea articulation 
between each lateral half of the base of the ray and a 
transverse hollow in the interspinal plate — of a deep 
fissure (articular cavity), into which is fitted a disk- 
like, circular process of the interspinal plate. The po- 
sition of the former articulation is a little eccentric 
relatively to this disk, and still more so relatively to 
the articular pivots of the spinous ray, which bound 
the said fissure, and which are more elliptical in out- 
line. Thus, when the spinous ray is erected, these 
pivots with their elliptical friction-surfaces fall into the 
circular articular cavity on each side of the central 
disk of the interspinal plate. The lower lever of each 
pivot attains a greater size than the radius of the disk 
and of the said articular cavities. Any attempt to bend 
the spinous ray results merely in a pressure partly of 
the pivots against the walls of the articular cavities, 
partly of the bottom of the middle fissure of the spi- 
nous ray against the top of the disk. As soon as the 
spinous ray is lifted, however, and the said level* thus 
shortened, the pressure ceases, and the ray falls. Each 
of the spinous rays has a small, triangular fin-mem- 
brane behind it. By means of this membrane the last 
spinous ray is united to the base of the first ray of the 
soft-rayed dorsal fin, which in this species begins dis- 
tinctly in front of the perpendicular from the vent. 
This fin is of an elongated triangular form, sloping even- 
ly backwards from the first rays, and finally passing 
almost imperceptibly into the dorsal margin of the pe- 
duncle of the tail. Its length (base) is about equal to 
the length of the head in old specimens, in young ones 
less than the same"; and its height (longest ray) is 
generally 
When erected, 
The specimens we have received from Archangel, are marked in this respect, as in several others, by the persistency of the juvenile 
characters. 
