NATURAL SELECTION 273 



in a regular way along lines of stress, just as are the cross- 

 pieces between the girders of bridges. If the disposition of 

 the stresses is altered (by an accidental deformation) during 

 the life of the individual, the whole arrangement of the trabe- 

 cules will be altered to meet the new lines of stress. 



In the growth of bone we have not only a striking example 

 of the nature of what we may call ' internal adaptations,' 

 but we are enabled further to define the limitations to all 

 analogy between living and non-living phenomena. If we 

 isolate a single part of an organism, such as an ocellate 

 marking, we may show how such a structure can result from 

 relatively simple chemical processes ; or we may show that the 

 mechanical adjustments of skeletal parts follow the principles 

 of elementary dynamics. But, as soon as we consider the part 

 in relation to the whole, we find a delicate adjustment quite 

 unknown outside living organisms. There is nothing in the 

 analogous laboratory experiments suggesting why the various 

 growth processes stop just at the right point, or why one type 

 of growth occurs at one point and one at another and yet both 

 are so related that a delicately adjusted organism results. 

 Though we can scarcely imagine that the functions of living 

 organisms at any point involve processes different from those 

 known to chemists and physicists, yet the physico-chemical 

 processes might be called the mere bricks of which such 

 organisms are made. It is probable that we are nearer the 

 truth in saying that living organisms have selected certain 

 processes to do their work and elected to follow certain laws, 

 than in adopting the more usual viewpoint that living organ- 

 isms obey physico-chemical laws. The bearing of these facts 

 and speculations on the selection theory may seem somewhat 

 remote, but two points emerge for consideration. First, there 

 are many details of living organisation that are so closely 

 paralleled by processes known to occur outside the organisms 

 that we may believe that the same forces are at work in both 

 cases. This possibility relieves the selectionist of part of his 

 burden, since in such cases it may be unnecessary to treat a 

 structure as the result of the selection of numerous"' small 

 favourable variations : what would have been called the 

 result of evolution may now be called the result of growth, 

 and it has only to be shown that the results of, and not every 

 stage in, such growth are adaptive. Secondly, it has been 



