196 
ACTINOPTERYGII. 
is then a false appearance of multiplicity of rays. All the rays are 
more or less articulated, and they also usually bifurcate ; but 
there are no traces of fulcra, unless a few short, undivided basal 
rays at the origin of the caudal fin are placed in this category. 
The scales are all quadrangular and somewhat deepened on the 
flank. Their anterior margin is strengthened by a vertical inner 
rib, which forms a peg-and-socket articulation to unite the scales 
of each transverse (vertical) series ; and there is one exceptional 
instance (Mesturus) in which this union is still further strengthened 
by an interlocking jagged suture of the superposed upper and lower 
margins. The series are for the most part regular ; hut in Mesturvs 
there are a few, short, irregular intercalary series both dorsally and 
ventrally, while in this genus, Gyrodus, and Pcdceobalistum there is 
a fine subdivision of the scales at the base of the dorsal and anal 
fins. When the squamation degenerates on the caudal region, it 
does not appear to pass through the stage represented in the 
Dapedioid genus Aetheolepis (see p. 157); the rhombic scales seem 
to disappear directly without first becoming cycloidal, for remnants 
of their anterior thickened rib are often observed on the otherwise 
naked tail. The “ lateral line ” extends along the flank from the outer 
supratemporal to the middle of the caudal pedicle ; and there is also 
always an upper canal passing along the top of the flank to the 
origin of the dorsal fin. The ridge-scales are usually small and 
inconspicuous; but in Pcdceobalistum and Pycnodus those of the 
ventral ridge appear to have considerabl} 7 extended limbs. 
The hinder end of the abdominal cavity is nearly always marked 
by a long, curved bone, extending from a slight expansion at the 
ventral border just in advance of the anal fin, and tapering to a 
point immediately below the notochordal axis. In Gyrodus (PI. XYI. 
fig. 2, x) this has much the aspect of an azygous internal bone (as 
described by B. Tetter, Mittheil. k. mineral.-geol. Mus. Dresden, 
pt. iv. 1881, p. 36) ; while in Coelodus , Pcdceobalistum and Pycnodus 
it appears rather as a great paired dermal scute (as described by J. J. 
Heckel, Denkschr. k. Akad. Wiss., math.-naturw. Cl. vol. xi. 1856, 
pp. 216, 221, 241). The element is certainly not an enlarged support 
of the anal fin, and has no connection with the pelvic fins. It must 
still remain problematical. 
In arranging the genera of Pycnodonts, the presence or absence 
of scales on the caudal region cannot be regarded as of prime 
importance, owing to the uncertainty of the character in some 
degree in Microdon and in a greater degree in Pcdceobalistum. 
Specialization seems to result chiefly (i.) in the strengthening of 
the backbone by the interlocking of its arches ; (ii.) in the reduc- 
