SUMNEK — DESERT AND LAVA-DWELLING MICE 
83 
the present case, if anywhere, the darkening effect of the lava would 
manifest itself. 
Five hundred and thirteen trap nights on the lava field yielded me 
157 specimens of crinitus, along with 23 specimens of mice of other 
species. From the former rather more than fifty skins were prepared. 
No selection whatever was made in choosing the individuals for skin- 
ning, save that mice with immature or with badly damaged pelage were 
rejected, while many others were unavailable owing to early decom- 
position. 
The first results of this trapping made it plain that these mice belonged 
to no specially modified lava-dwelling race. They had retained the 
pale brown-gray hue, sprinkled dorsally with black, so familiar to us in 
desert rodents generally. When laid upon the dark lava rocks, they 
certainly could not be called inconspicuous, as the accompanying 
figure shows. (Plate 6.) 
The fact that this seemingly conspicuous rodent flourishes in such 
numbers upon these lava fields gives us good ground for skepticism 
as to the need for concealing coloration among closely related species. 
And it surely justifies us in challenging those who assume without 
evidence the existence of a selection so rigid that trifling differences 
in shade are of frequent survival value. 
It should also be mentioned here that none of the other species of 
mice and rats which were trapped upon this lava field by Mr. Hunt and 
myself gave any evidence of modification in the direction of deeper 
pigmentation. The species taken comprised one other Peromyscus, 
two species of pocket mice and a wood-rat. While no extensive series 
of skins was prepared for these other species, any very pronounced 
darkening would doubtless have been observed. With a single possible 
exception, the only noticeably dark pelages were those of juvenile 
specimens. 
Although it was evident that no considerable degree of darkening 
had occurred among these lava-dwelling representatives of Peromyscus 
crinitus, the question still remained whether there had been any modi- 
fication whatever, which might be revealed by careful quantitative 
methods. For this purpose a control set of specimens was necessary. 
The control series I should naturally have collected in the Bullion 
Mountains, a rocky range, a few miles to the southwest, in which I 
knew that these mice occurred. But the presence of another extensive 
lava field upon the nearby slopes of this range, was regarded as com- 
