136 PROBLEMS OF EVOLUTION 



cannot help asking, What is the mean ? Is there a separate 

 mean, to take the last instance, for each ear ? If so, the normal 

 correlation between the two is left out of sight. But if there is 

 one and the same mean for the two ears, then the dwindling of 

 one is not a simple deviation from it. It is impossible to frame 

 a formula to cover all sorts and kinds of variations. The shift- 

 ing forms which living organisms assume suggest crystallization 

 Positions more than anything else. There are certain positions of stability, 

 llty to use Mr Galton's phrase, and some members of a species 

 settle in one of them, others in another. The right-handed 

 helix is the rule among some species of snail, but occasionally 

 the left-handed form makes its appearance ; there is a sudden 

 shift, not a gradual accumulation of small variations on one side 

 or other of a mean. There is good evidence of peach stones 

 producing nectarine trees, and nectarine stones peach trees ; 

 there are two forms in fact, and the species shifts from one to 

 the other, but there is no mean between the two. 1 Five hundred 

 and eighty-three male earwigs in which the forceps was measured 

 divided up into two groups characterised respectively by long 

 and short forceps ; an intermediate length 'was found in very 

 few. 2 When the cephalic horn of three hundred and forty-two 

 Lamellicorn beetles was measured, a similar grouping showed 

 itself. The species shifts between two forms ; the grouping is 

 not that of the law of chance. Sometimes, again, there is a 

 sudden multiplication of parts ; two spurs appear in a species of 

 bird that hitherto has had only one. 



But an objection may, perhaps, be urged against all this. 

 " Such crystallization, such kaleidoscopic shifts can only multiply 

 or rearrange organs that were already there, it cannot produce 

 any new organs." Strictly speaking, there is nothing new at any 

 stage in evolution; you cannot find an absolutely fresh beginning. 

 To take an instance, the outer cells harden and there is an 

 epidermis ; further specialisation follows. Scales are elabora- 

 tions of the epidermis, outgrowths from it ; feathers are 

 elaborations of scales. Again, hair and feathers are equally 



1 Darwin: Animals and Plants under Domestication, vol. i. p. 340. 1868 edition. 



2 See Mr Bateson's Materials for the Study of Variation, p. 41. 



