238 COSMIC PJlILOSOrUY. [pt- *• 



between the variation of the cause and that of the effect; 

 and secondly, we need to know that the intermixture of 

 circumstances will not, after a certain point, alter the order of 

 the variations. In the case of the pendulum, just cited, we 

 know both of these points. We know that the only factors 

 in the case are the momentum of the pendulum, acting in 

 concert with gravity, the friction at the point of support, and 

 the friction and resistance of the atmosphere ; and as we 

 progressively diminish these latter retarding factors, we can 

 calculate the exact ratio at which the retardation diminishes. 

 We are therefore perfectly justified in concluding that if 

 the friction and resistance could be utterly abolished, the 

 momentum of the pendulum, acting in concert with gravity, 

 would carry it backward and forward for ever. But because 

 the abstraction of heat causes the molecules of a body to 

 approach each other, it is not safe to infer that, if all the 

 heat were abstracted, the molecules would be in complete 

 contact. This is a more or less plausible guess, not a true 

 induction. "For since we neither know how much heat 

 there is in any body, nor what is the real distance between 

 any two of its particles, we cannot judge whether the con- 

 traction of the distance does or does not follow the diminu- 

 tion of the quantity of heat according to such a numerical 

 relation that the two quantities would vanish simulta- 

 neously." x In similar wise, from the fact that in alcoholic 

 intoxication the severity of the narcotic symptoms varies 

 according to the size of the dose, it is not legitimate to infer 

 that a very small dose will cause slight narcotic symptoms 

 or even a tendency to the production of such symptoms. 

 For we can neither ascertain the quantitative ratio between 

 the variation in the dose and the variation in the narcosis, 

 nor in the case of such a complex aggregate as the human 

 organism can we assert the absence of interfering conditions 

 which, after a certain point, will entirely change the order oi 

 1 Mill, System of Logic, 6th edition, vol. i. p. 447. 



