184 THE AMERICAN NATURALIST [Vol. LII 



directly suggestive of Mendelian segregation? A defi- 

 nite answer to these questions I am likewise obliged to 

 defer for the present. 



Truly representative collections have been made by me 

 thus far at only four stations within the State of Cali- 

 fornia, though various other points have been visited 

 and considerable numbers of the mice have been trapped 

 there. My four principal collecting stations are located 

 near Eureka, Berkeley, La Jolla and Victorville. 1 Me- 

 teorological records were kept for about two years at 

 each of these points. These records have not thus far 

 been carefully analyzed, however, so that their publica- 

 tion must be postponed. A preliminary comparison of 

 climatic conditions at these four points has already been 

 made (Sumner, 1915a). It will suffice, for present pur- 

 poses, to state that, as regards both atmospheric humid- 

 ity and rainfall, these stations rank (from highest to low- 

 est) in the order given above, i. e., Eureka, Berkeley, La 

 Jolla and Victorville ; while as regards mean annual tem- 

 perature the reverse order holds. 



The distribution of the three subspecies of Peromyscus 

 mcmiculatus, recognized by Osgood as occurring within 

 the limits of California, is represented in Fig. 3. It will 

 be seen that one of my stations (Eureka) lies within the 

 range of rubidus, another (Victorville) within the range 

 of sonoriensis, while the other two (Berkeley and La 

 Jolla) lie within the range attributed to gambeli. 



In the ensuing pages, I am not in the least concerned 

 with characterizing and defining those taxonomic groups 

 which have been called Peromyscus maniculatus gambeli, 

 rubidus and sonoriensis. I shall merely discuss the dif- 

 ferences between (and within) four representative collec- 

 tions taken by me in widely separated and climatically 

 different regions of the state. The question as to what 

 ' 1 subspecies ' ' a given mouse 4 'belongs to" is for my pur- 

 poses a distinctly minor consideration. 



ten, but the data derived from these can not be included here f 



