SUMNER — DESERT AND LAVA-DWELLING MICE 85 
Reduced to terms of black, white, and color (in this case a shade of 
yellow) we have 
BLACK . 
WHITE 
COLOR 
Lava 
79.8±0.19 
16.0 
4.2 
Oro Grande 
79.6=t0.18 
15.9 
4,5 
The trivial excess of black in the lava series is of no statistical signifi- 
cance, as appears from the fact that this difference is less than its 
probable error. It is to be noted that the lava series likewise shows a 
trivial excess of white, which is further evidence of the ^^accidentaF’ 
character of all these shght differences between the two sets of figures. 
Various objections may perhaps be brought forward by critics to 
the propriety of drawing any general conclusions from this single 
instance which yielded negative results. It will perhaps be pointed 
out that I have dealt with a wholly nocturnal animal. I should, it 
may be urged, have chosen some diurnal species, for which concealing 
coloration would be more necessary. In reply, I need only remark 
that two of Doctor Merriam’s species (the Onychomys and the Perog- 
nathus) are as completely nocturnal in their habits as is Peromyscus 
crinitus, while the remarks of both Osgood and Goldman likewise apply 
to nocturnal species. 
It may be urged, too, that the lava fields on which I have trapped my 
specimens are of unknown age and that they are perhaps much more 
recent than those of the San Francisco Mountain district. I cannot, 
I confess, meet this argument directly. It is certainly pertinent to 
point out, however, that the pale crinitus mice of this Mojave Desert 
lava field still succeed in maintaining themselves in great abundance, 
after a period which is certainly to be measured in centuries, and per- 
haps by even greater periods of time. This fact does not harmonize 
well with the assumed rigid selection on the basis of color-tone. 
Another objector may insist that Peromyscus crinitus may, by reason 
of special habits, have no need for protective coloration, whereas the 
same may not be true of various other rodents. This, indeed, is quite 
possible. If it be true, however, we may well query why this species 
Those who have had no experience in color analysis may be surprised by 
my characterizing as “pale” a shade containing nearly 80 per cent of black. I 
will merely point out that a piece of faded khaki which I tested gave photometer 
readings very close to those for these mouse skins, but that the former was 
slightly darker. 
JOTTENAIj of mammalogy, VOL. 2, NO. 2 
