DEVELOPMENT BY NATURAL SELECTION. 125 



again from the stage at which this structure is in defect , 

 repeating in outline the slow stages of ancestral history. 

 Progress is easy and rapid, because the proper environ- 

 ment and proper stimuli are provided in advance ; 

 just as a man who is given proper tools and material, 

 and knows what is required, rapidly succeeds in doing 

 work for which it took centuries to establish the neces- 

 sary conditions. 



But if not mechanism, what is it in a living thing, or 

 in the atoms composing it, that makes it react in this 

 particular way to environment ? Saturated as we are 

 with mechanistic conceptions of reality, this question 

 inevitably arises. The action seems to depend some- 

 how on the properties of the atoms. An atom for ordi- 

 nary chemistry is a material unit of a certain mass, 

 and which is found to possess one, two, or more definite 

 " affinities " for other atoms. But when we follow the 

 atom into the living body we find that its properties 

 cannot be thus summed up simply : for the specific 

 peculiarities connected with different saturations of the 

 affinities are endless ; and evidently we have also to do 

 with accessory affinities manifesting themselves in the 

 formation of hydrates, crystals, liquids, solutions, and 

 various other complexes. In all this complexity of 

 properties we lose sight of the old simple atom. The 

 . conception of it as a " thing " with definable properties 

 has failed us. To define the " thing " which we call an 

 atom, and which at first seemed so simple, we should 

 have to define the whole universe of "things." The 

 conception of a " thing," or material unit, is thus useless 

 in the interpretation of distinctively biological facts. 



To the question why living organisms behave as they 



