652 The American Naturalist. [August, 
Table VII indicates the number of specimens, the average 
number of dorsal spines, and the number of specimens with 
thirteen, fourteen, fifteen, sixteen and seventeen spines from 
each of the localities from which specimens were examined. 
The localities are arranged as they occur, from north to south. 
It will be seen that the prevailing numbers occurring in the 
more northern streams are fourteen and fifteen. As we go 
farther south the usual number is fifteen and sixteen, and in 
the most southern streams the numbers are fifteen, sixteen and 
seventeen spines, the specimens from Texas are peculiarly poor 
in the number of spines. 
TABLE VII. 
28 | 23 | ov] $9) 3S) 35 
si | 54 [SLES | 83/2 | 25 
cg agn cid a R SEL OE 
- B oa H Sait ie 
LOCALITY. 2$ FI. EE a3 EE FE E 
LEBCSEHEHEHEHEE 
Z < Z zZ Z * zZ 
Torch Lake, Mich......... ................- 7 144 3| 4 
Cadar Rapidi, T. coss isasce tenes daii] A 14 1 
hoa River, at Indianapolis Nep 007 E 1 14 1 
5 | 14 Il a4 
San Blossom, Ind 17 BN B91 7 
Rushville, 1 14 1 
Wild Cat ot Ind 1 15 1 
Pike Creek, 2 143 i11 
Illinois 1 15 1 
Nipisink Lak 2 144 PL 
Monongahela Rives 1 15 1 
4 5 1 2 1 
Green River Greensburg, E —— pe 15 8 
Barren River, Osceola, Ky......... 1 15 TL) 
Little South Fork Ciebetland River, 
Eagle e dedi y Ky 1 16 1 i 
2 63 1 
ay aie Lr loma, Tenn «oid Pete 13 o 2| 81 $ 
eer 2 15 1 
North Pork Holsten n River, Saltville; Vero 16 1 
1 16 1 
Chest Cel, Oxf Oxford, Ala 4 15 21 3 
San Marcos Springs, Nek. 2 isi II-1 
