THE CYPRINOIDS OF CENTRAL NEW JERSEY. 329 
examine the figure of Abramis versicolor, we find a fish varying in 
the size and shape of its fins also. 
n studying the very large collection of specimens of the 
“roach,” from streams of different character, made during the 
past summer, we think we have traced a uniform variation in 
the size and shape of the fins, more especially of the dorsal and 
ventrals; and with it, a constant difference of the color of the 
Stilbes taken in small, rapidly running streams, and the ordinary 
“roach” of our mill-ponds and quiet creeks. These differences, 
in some respects, agree with the distinctions drawn by DeKay be- 
tween the Stilbe chrysoleucas and Abramis versicolor, but not in all. 
Indeed, we have never met with a “roach” that was strictly iden- 
tical with the fish described by DeKay as a “‘ variegated bream.” 
The variations we have traced out in a large series, and which 
we believe to be constant, are as follows :— 
Delaware River specimens. 
Adult, total length,* 74 in. 
Dorsal fin, depressed, reach- 
point opposite the 
line drawn through the pupil. 
Scales with 7 to 12 radii. 
olor as described by DeKay . 
yellow, with black lines on the 
dorsal and caudal fins. 
Shabbaconk Creek specimens. 
Adult (?), total length 5 in. 
Dorsal fin, depressed, reach- 
ing to a point opposite the last 
ral depressed, 
iiia to the anus. 
Mouth more oblique; when 
closed, the front of the lips on 
a line with the upper edge of 
the orbit. 
Scales with 4 to 5 radii. 
Color uniformly blue, with 
no shade of green or golden; 
lighter on belly, but scarcely 
silvery. Fins pale yellow, but 
at no time crimson or golden. 
In order not to mislead the student, it must be clearly under- 
stood, that while the specimens of Stilbe from the Shabbaconk 
reek are uniformly different, as here pointed out, we do find 
“river” specimens, which are partially grown individuals of the 
n the 4th yol. NATURALIST we have referred to the pe of having gathered many 
es eight and nine inabot, long. Compared with the whole number of individ- 
i i easurements are iet Eee and the length above 
given is about that of the psn adult fish. 
