XIV THE ORIGIN OF SPECIES AND GENERA 295 



length of any toe varies independently of the length of 

 the tarsus ; a long head sometimes goes with a short, 

 sometimes with a long wing ; while the width of the bill 

 seems to vary independently of its length or of any of the 

 other parts of the body. All these variations, too, are 

 very considerable in amount. Thus among twenty male 

 Baltimore Orioles the total length varied from 7 to 8 

 inches ; the wing from 3-45 to 3-85 inches ; the tail from 

 2-70 to 3"10 inches ; the primaries extended beyond the 

 secondaries from 0-56 to O'BO inch; the tail extended 

 beyond the upper coverts from 1*37 to 1"87 inches; 

 the tarsus varied from 0*83 to 102 inch ; the hind toe 

 varied from 0-62 to 0*75 inch, and the middle toe from 

 0*82 to 1*00 inch ; the head varied in length from 1'50 to 

 1-62 inches; the beak in length from 0-74 to 0-84 inch, 

 and in width from 0-32 to 0*38 inch. And if these differ- 

 ences and these combinations, indicating many diverging 

 proportions between two or more characters, are found 

 among only twenty specimens, we may certainly expect 

 much greater differences in every character, and these 

 differences combined in an endless variety of ways, among 

 the millions of individuals which constitute every common 

 species. Not only, therefore, is it clear that there is, 

 among birds at all events, ample individual variation for 

 natural selection to work upon, but, what is even more 

 important, that coincident variations in every conceivable 

 comhination are also available. 



Among mammalia we have fewer materials for com- 

 parison, but there is good reason to believe that they are 

 quite as variable as birds, if not even more so. Among 

 twenty males of the Grey Squirrel, whose dimensions are 

 given by Mr. Allen, we find the length of the tail to vary 

 as 3 to 4, of the fore foot as 9 to 11, and of the hind foot 

 as 6 to 7. The Virginian Opossum also varies greatly in 

 colour, and in the size and proportions of all the parts, 

 including the skull, the variation amounting to nearly 

 twenty per cent.^ 



^ A number of diagrams illustrating these and many other facts of 

 variation, are given in my Dariciuwm (Chap. III. ), where the subject of 

 the nature of variation is more fully discussed. 



