196 THE AMERICAN NATURALIST [Vol. LII 



to determine whether or not these various pigment char- 

 acters are correlated. If they are, the correlation is cer- 

 tainly not a close one, as frequent observations have 

 shown. For example, the grade of foot-pigmentation was 

 determined for the paler and darker halves of a series of 

 sonoriensis and also for a series of gambeli. In both 

 cases the average grade for the foot was slightly greater 

 for the darker half than for the paler; but the difference 

 was so small that I am not sure of its significance. Again, 

 in the lot of 38 gambeli comprised in Table II the darkest 

 individual (dorsally) and the one with the darkest feet 

 were both devoid of visible pigment on the scrotum. 

 Similar entries are frequent among my notes. 



2. Structural Differences 



The structural features which I have subjected to quan- 

 titative determination are (1) weight, (2) body length, 

 (3) tail length, (4) foot length, (5) ear length, (6) number 

 of tail vertebra' ; together with several other skeletal char- 

 acters which I shall not discuss in the present paper. 

 The methods employed throughout these studies will be 

 described more fully in a later report. A brief statement 

 will suffice for the present. Body length, as here em- 

 ployed, is the total length, minus the length of the tail. 

 In taking the total length, a special contrivance is em- 

 ployed, the body being stretched slightly and to a uniform 

 extent. A constant procedure is likewise employed in 

 measuring the tail length. The figure recorded for the 

 latter represents the distance from the first free caudal 

 vertebra to the tip of the tail, under a uniform degree of 

 tension. The ear length here used is that from the sum- 

 mit of the ' ' notch" to the tip of the ear. Foot length is 

 the distance from the heel to the tip of the claw of the 

 longest toe, the foot being pinned, sole downward, to a 

 blackened board. 



The statistical methods employed in analyzing these 

 data have been rather fully discussed in a former paper 

 (1915), to which the reader is referred. 



