68 
number of scale rows about the anterior part of the body. 
On the mainland the prevailing formula is 25-23-21-19-17, 
being present in about 66 per cent of the specimens. In the is- 
lands the majority have the count of 25-23-21-19, occurring in 
about 80 per cent. These differences show that in the island 
specimens, associated with the shorter body, there is a ten- 
dency for the VII row to be continued and for the zone of 19 
rows to remain intact to the base of the tail. 
These counts do not indicate such a radical difference as 
is so often the case between valid species. They merely indi- 
cate a difference in the distance down the body that these 
suppressed scale rows extend. 
Neither the number of scale rows, nor the distance to 
which they extend down the body, in any way effects the 
sequence of suppression; this, barring abnormal individuals, 
is definitely fixed for each species. Several hundred species 
have been investigated for this character, which when better 
known will prove of high value in confirming or establishing 
relations between genera. 
The greatest range of variation that is known in the 
number of dorsal scale rows in a single species is found in 
Thamnophis ordinoides (Baird and Girard) 1853. (1) 
There being recorded ten distinct formule : 
21-23-21-19-17 
21-19-17 
19-21-19-17 
19-21-19-17-15 
19-17 
19-17-15 
19-1'7-115-13 
17-19-17-15 
17-15 
15-17-15 
In this Garter-Snake from the Pacific Coast of North 
America the sequence of suppression is (-V-VI-IV-VII)-. It 
is a matter of much interest to note that this sequence is the 
same as that found in Ancistrodon halys. In these two species 
there must be some analogous underlying cause that brings 
about a reduction in the number of scale rows around the body 
by an identical sequence of suppression. 
(1) 1914, Thompson, Proc. U. S. Nat. Mus. Vol. 47, p. 351. 
