408 THE AMERICAN NATURALIST.  [Vou. XXXII. 
at the Laboratory of the United States Fish Commission, I 
examined several hundred winter flatfish (Pleuronectes ameri- 
canus) with the following results : 
Of 100 flatfish collected at Woods Holl, only one had 62 
dorsal fin-rays, seven had 63, twelve had 64, twenty-two had 
5 TI tT Trt 
Coceceecet ct CT Cocco 
RAT |T Mi PEL. 
ecco ry Cl i 
a ct H Coe 
Cocco TT T TTET 
ananasu H TT | 
ananass im a 
Cocco Hi ti 
ELIITTI Hi ct 
PETTITT mi mi 
= ct H 
S tt ti RTT 
tT tH | a 
Į ul | ae 
rT T REITT 
tt ti eC 
H REN 
EEE aeeaaesir tone fue 
an TZ ETT 
Wi | -A 
MAET PC = 
= -—- os 
z TRH 
eT == 
we 
PETT] @ 
Coot coco 
ct Choy 
FETTI 
E 
Too a 
Cort Cor 
Coo 
Toot coo 
= TI HHH 
T] EN 
bAi 
H HHE HHHH 
H heer 
Toor Coot 
Coot co 
eee crt as 
ae 
a 
= PEET] 
a sane 
se -a 
HH HEH 
Coote coo 
HHHH aani 
EET FE 
Coo ESSE 
pasma snnsa n: BARAS JAN HH 
Cocco IETT ect 
Coot Ere a 
Sn we CPCB CCT 
CCC WoO Boor = 
teeth TELLEREN 
§9 6o 6i Ge 63 64 65 66 G GB © m 7 F P 
Fic. r. 
65, etc., as represented by the ordinates of curve 4 in Fig. I. 
The amplitude of variation is between 62 and 72, the arith- 
metical mean being 66, represented by the vertical line at the 
right of 66.1, curve A. The curve drawn through the upper 
ends of the ordinates represents graphically the distribution 
of the 100 variants around this mean. 
