298 WRITINGS OF JOSEPH HENRY. [1851 



therefore superadd to the hypothetical points of Boseovich 

 these other conditions; but in so doing we arrive at a consti- 

 tution of matter precisely similar to that adopted b}'^ Newton, 

 namely, a system of indivisible and indestructible atoms 

 endowed with the essential properties of matter in masses. 

 Indeed, this is the only hypothesis which we can adopt in 

 strict accordance with analogy, reasoning from the known 

 to the unknown. 



Besides the phenomena of the action of invisible atoms of 

 gases on each other we have a large classknown under the gen- 

 eral name of the phenomena of the "imponderables." This 

 name has been given because it is supposed that it is neces- 

 sary to refer them to hypothetical fluids not subjected to the 

 ordinary laws of force and motion. The term imponderable 

 however expresses a quality with reference to the constitu- 

 tion of such fluids not warranted by the facts. A mass of 

 air poised in air has no weight, and in this case may be 

 considered imponderable. In the same way, if we suppose 

 an elastic medium to pervade all space, any portion of this 

 will be imponderable, even were our balances sufficiently 

 delicate to detect its absolute weight. The existence of an 

 elastic medium pervading all space is assumed in order that 

 the phenomena of light, heat, electricity, and magnetism 

 may be brought within the category of the laws of force and 

 motion, and that we may be able to apply the principles of 

 analytical mechanics in the way of deducing consequences 

 to be afterwards tested by an actual appeal to experiment. 

 Without assumptions of this kind it is impossible to arrive 

 at the general expressions which constitute science in the 

 proper sense of the term. 



It is not necessary that an hypothesis be absolutely true in 

 order that it may be adopted as an expression for a generali- 

 zation for the purpose of explaining and predicting new phe- 

 nomena; it is only necessary that it should be well condi- 

 tioned in accordance with known mechanical principles. 

 We have a remarkable instance of this in the Newtonian 

 theory of emission of light. According to this, light is first 

 considered as consisting of atoms of matter moving with 



