EM AN CI PA TION FROM SCIENTIFIC MA TERIALISM. 433 



free to study objectively the various properties of these 

 forms of energy. We can, further, by rational consideration 

 and arrangement of these properties, establish a system 

 which shall represent explicitly not only the similarities, 

 but also the differences in these forms of energy, and which 

 will therefore lead us much further scientifically than would 

 be possible when slurring over their differences through the 

 hypothetical assumption of their " inner" identity. A good 

 illustration of the meaning I intend to convey may be 

 found in the kinetic hypothesis of gases which at present 

 enjoys almost universal recognition. According to this 

 hypothesis the pressure of a gas arises from the blows 

 delivered by the moving particles. Now pressure is a 

 quantity which possesses no special direction : a gas exerts 

 pressure in all directions equally ; but a blow is caused by 

 a moving object, and this motion has a definite direction. 

 Consequently one of these quantities cannot be referred 

 directly to the other. The kinetic hypothesis gets round 

 this difficulty by the assumption that the blows occur 

 uniformly in all directions, whereby the vector-property 

 really possessed by the blow is artificially done away with. 

 In this case the artificial adaptation of the properties 

 of the different energies is successful, but in other 

 cases it is not completely possible. The factors of electri- 

 cal energy, potential and electrical quantity, form a case in 

 point — quantities which I propose to call polar, i.e., they 

 are characterised not only by a numerical value but also by 

 a sign in such a way that the sum of two equal values 

 with opposite signs is equal to zero and not to their double 

 value. In mechanics the purely polar quantities are 

 unknown, and this is why all attempts to set up even a 

 partially workable mechanical hypothesis for electrical 

 phenomena must essentially fail. Should we succeed in 

 contriving a mechanical magnitude with polar properties (as 

 is perhaps not impossible, and certainly worthy of careful 

 consideration) we should have the means wherewith to 

 picture to ourselves mechanically at least a few phases of 

 electrical phenomena. We may say with certainty that 

 it will be a question of a few only, and that the imper- 



