188 THE AMERICAN NATURALIST [Vol. LII 



sentative collections of any two adjacent races belonging 

 to this series, there is found to be a broad zone of over- 

 lapping. That is to say, there are many individuals in 

 each set which, so far as color goes, could be equally well 

 placed in either. I have, for example, laid out in par- 

 allel rows considerable numbers of sonoriensis and the 

 La Jolla form of gambeli, and found that the darker half 

 of the former set completely overlapped the paler half of 

 the latter. While no confusion would be possible between 

 the paler sonoriensis and the darker gambeli, there were 

 a large number of specimens which could only arbitrarily 

 be assigned to either " subspecies." Indeed, it is freely 

 admitted by systematists that in many cases they can 

 assign a given specimen to its proper subspecies only if 

 they know the locality at which it was trapped. No such 

 confusion would be possible, however, between the more 

 divergent races of our series, e. g., those from Eureka 

 and the desert. I have never seen a nibidus which could 

 not, by color alone, be readily distinguished from sonori- 

 ensis and vice versa. 



Any attempt to give verbal equivalents for these color 

 differences is highly unsatisfactory. In a later report I 

 expect to undertake the analysis of these shades by means 

 of a color wheel. For the present I will content myself 

 with a very brief statement. 3 The dorsal darker stripe 

 of the Eureka mice is of a shade lying somewhere be- 

 tween Ridgway's ' ' sepia" and black, the paler lateral 

 region lying between "Saccardo's umber" and ' 'sepia." 4 



3 The ensuing remarks apply only to the mature pelage. These mice pass 

 through three distinct pelage phases: (1) the juvenal, which, in all races, is 

 neutral gray in hue, and considerably darker than the adult shade; (2) the 

 post- juvenal or adolescent, commonly paler and yellower than the last; (3) 



approximate the shades i 



