46 THE LIFE OF SCIENCE 



remarks. Any amount of them can be found in the history of 

 medicine; we need but recall how greatly the whole of medical 

 evolution has been influenced by the Hippocratic teaching, our 

 modern ideas on humorism and naturism; or, again, the organo- 

 therapeutic theories. Not only are the old ideas restored to vogue, 

 but it sometimes seems that a kind of rhythm brings them back 

 to light periodically. Georges Bohn has shown the periodical 

 return, in the domain of comparative psychology, on one hand, 

 of the animistic and anthropomorphic conceptions, and on the 

 other hand, of the positivist conceptions. As a rule, the further 

 science is removed from the mathematical form, the more likely 

 these vicissitudes are. One can also say that when science is more 

 accurate, that is to say, when the domain of uncertainty and 

 hypothesis becomes narrower, the oscillations of the mind between 

 divergent points of view are so much the less numerous, — but they 

 do not cease entirely. Thus E. Belot reintroduced into cosmology, 

 in a very seductive shape, the vortex theory that one would have 

 thought had been entirely banished by Newton's criticisms. 

 Similarly weighty reasons exist for reinstating into optics the 

 emission theory, which seemed to have been forever exploded 

 by the discoveries of Huygens, Young and Fresnel. 



But the best examples of such return to ancient knowledge are 

 given to us by the history of technology. The history of chemical 

 industries is very significant from this point of view : this is due to 

 the fact that here economic conditions play a considerable part. In 

 order that an invention may be realized it does not suffice that it be 

 theoretically possible; it must pay. Now thousands of circum- 

 stances continually modify the material factors which the engineer 

 is struggling with; many are of such a nature that nobody could 

 foresee them, or (what amounts to the same thing) , that it would 

 cost too much to insure oneself against all of them. If new products 

 appear on the market, or if the prices of some of the raw materials 

 happen to vary considerably, or if new discoveries are made, or 

 if new residues are to be employed, old methods that were once too 

 expensive may become economical, or vice versa. Hence the 



