194 Scientific Sophisms, 



chemical properties : to these last heterogeneity 

 is essential. To deduce chemical phenomena 

 from mechanical conditions, if it be not an 

 impossible conception, may possibly be a " fig- 

 ment of the intellect," but it is a figment with- 

 out any pretence to " verification." 



" Even in the last resort, if we succeed in getting all 

 our atoms alike, we do not rid ourselves of an unex- 

 plained heterogeneity ; it is simply transferred from their 

 nature as units to their rules of combination. Whether 

 the qualitative difference between hydrogen and each of 

 the other elements is conditional upon a distinction of 

 kind in the atoms, or on definite varieties in their mode 

 of numerical or geometrical union, these conditions are 

 not provided for by the mere existence of homogeneous 

 atoms ; and nothing that you can do with these atoms, 

 within the limits of their definition, will get the required 

 heterogeneity out of them. Make them up into molecules 

 by what grouping or architecture you will ; still the 

 difference between hydrogen and iron is not that be- 

 tween one and three, or any other number ; or between 

 shaped solids built off in one direction and similar ones 

 built off in another, which may turn out like a right and a 

 left glove. If hydrogen were the sole ' primordial/ and 

 were transmutable, by select shuffling of its atoms, into 

 every one of its present sixty-two associates, both the 

 tendency to these special combinations, and the effects 

 of chem would be as little deducible from the homogene- 

 ous datum as, on the received view, are the chemical 

 phenomena from mechanical conditions. I still think, 

 therefore, that M you assume atoms at all, you may as 

 well take the \\hole sixty-three sorts in a lot. And this 

 startling multiplication of the original monistic assumn- 



