14 THE LIMITATIONS OF SCIENCE 



can be perceived by the senses. If the consequences 

 of such a hypothesis are afterwards found to be in 

 agreement with the results of observation and experi- 

 ment, then the laws, known to be true for the class of 

 real objects or phenomena, may be applied to the hypo- 

 thetical class. The objects or phenomena considered 

 by this method are thus merely matters of conjecture, 

 and their nature may be modified at any time so as to 

 make the propositions derived from them conform to an 

 expression of experimental fact. Such, for example, has 

 been the method followed in the wave theory of light. 

 To explain the observed action of light, the existence 

 of hypothetical bodies, called atoms, and the luminifer- 

 ous ether, is assumed, and properties are assigned to 

 them similar to those of sensible matter. As new 

 phenomena are discovered the attributes of the atom 

 and the ether are modified to fit the requirements. 

 This theory can be considered only as a convenient 

 means of expressing natural laws, and is always sub- 

 ject to change, as it does not depend on the objective 

 realities fundamental to an abstractive method. 



Just because the theory of the mechanical motions 

 and motive forces of sensible bodies is the only com- 

 plete physical theory, and because it does not require 

 the use of a hypothetical method in its development, 

 we have been led to give the hypotheses, advanced as 

 theories in the other branches of physics, a mechanical 

 form. The classes of phenomena considered in all 



