80 
JOURNAL OF MAMMALOGY 
which it was compared, came from the Desert of the Little Colorado. 
Now it is evident from both the maps and the text of the report, that 
the former region occupies a considerably higher altitude than the 
latter, there being an average difference of more than a thousand feet 
between the two. Indeed they are assigned to different Hfe zones. 
While no meteorological records are accessible to me, it seems very 
probable that the precipitation in this belt of pinons and junipers on 
the mountain slopes is considerably greater than on the desert plains 
below. Are we, then, justified in ehminating humidity as the respon- 
sible factor in bringing about the color differences in these two locali- 
ties? Or is it not, indeed, possible that some unknown third factor is 
the one chiefiy concerned? 
Certain other opinions regarding the effect of lava in determining 
the colors of rodents should be referred to before leaving this discussion. 
W. H. Osgood, in his valuable “Revision of the Mice of the American 
Genus Peromyscus”^ tells us (page 16) that “if the range of a given form 
includes a few square miles of lava beds, specimens from that area show 
an appreciably darker color than the normal form occupying the sur- 
rounding region.” Again (p. 70) “in northeastern California, the mice 
of the semidesert lava beds are more like the dark gambeli than the 
pale sonoriensis. Throughout the desert region sonoriensis is the pre- 
vailing form, except on the lava beds.” 
It is unfortunate that more specific instances are not given in sup- 
port of these statements. We should like to know more of the rainfall, 
vegetation, etc., of these lava beds of northeastern Cahfornia; likewise 
(and this is vitally important) their distance from regions in which 
tine gamheli is abundant. Some of Osgood’s other statements (pp. 16, 
70) regarding the effects of narrowly localized environmental differ- 
ences, apparently in the absence of any form of isolation, are not sup- 
ported by the experience of various other collectors. They must, I 
think, merely voice impressions based upon accidental coincidences. 
The careful experiments of H. H. Colhns (not yet pubhshed) show that 
the more marked color differences, occurring in a given locality within 
the range of a single subspecies, are hereditary and not due to any 
immediate environmental influence. 
Goldman^® has recorded observations similar to those of Osgood. 
Of one wood-rat, Neotoma intermedia desertorum, he writes (p. 77): 
® U. S. Department of Agriculture. North American Fauna. No. 28. Wash- 
ington, 1909. 
Revision of the Wood Rats of the Genus Neotoma. U. S. Department of 
Agriculture, North American Fauna,* No. 31. Washington, 1910. 
