SELECTION: ORGANIC AND SOCIAL 205 



they seem sometimes to have this value. But 

 this camiot be the whole truth. Apart from the 

 just objections that thorns are often prevalent in 

 countries where there are few grazing animals and 

 that they do not appear in the early stages when 

 they are most needed, experiment has shown that 

 many thorns arise in response to poor nutrition.' 

 Thorns are the natural outcrops of a kind of con- 

 stitution suited to dry countries. 



This idea was familiar enough to Darwin, as we 

 see from the emphasis which he laid on instances 

 of " correlated variabiHty." . In this connection, 

 Sir Kay Lankester observed, at the Cambridge 

 Centenary Celebrations : " In my opinion he has 

 thus furnished the key to the explanation of what 

 are called useless specific characters and of in- 

 cipient organs. That key consists in the fact that 

 a general physiological property, or character of 

 utihty, is often selected and perpetuated which 

 carries with it distinct, even remote, correlated 

 growths and pecuharities obvious to our eyes, yet 



1 See Coulter, op. cit. (1909), and Geddes, "Proo. Brit. Assoc." 

 (1889.) Prof. Coulter points to significant facts like the following : 

 the nettle can get on quite well without its stinging hairs ; many- 

 seeds, especially in arid regions, develop a testa so hard that it inter- '^ 

 feres with the breaking through of the embryo — which looks like 

 " over-adaptation " ; further investigation has played havoc with 

 the pretty story of the extra-floral nectaries attracting a body-guard 

 of harmless ants. It is probable that in these and a hundred other 

 cases our task is rather that of discovering the phjrsiological and 

 embryological significance of the structures in question, than that 

 of searching diligently for a utilitarian justification which does 

 not exist. A familiar example may be found in our finger-prints, 

 which illustrate discontinuity in evolution — the apparently abrupt 

 origin of new patterns ; but, as we have no warrant for supposing 

 that natural selection operates in any way in this case, we must 

 suppose that these patterns are the expressions of internal growth- 

 conditions. 



