THE AIM AND ACHIEVEMENTS OF SCIENTIFIC METHOD. 85 



contact with the body it is taken for granted that the different 

 positions of the surface of the mercury are each correlated 

 with one objective condition of the body. Thus if the thermo- 

 meter gives the same reading in the wind as it does behind a 

 screen, then the air, although it feels colder in the open, must 

 really be in the same objective condition, have, as Boyle 

 vividly expresses it, the same temper, in both places.* If now 

 we " graduate " the stem of the thermometer upon the foot-rule 

 method, we shall have a series of numbers correlated with the 

 various " tempers " or temperatures of the body. In this case 

 the statement that one difference of temperature is double 

 .another has obviously still more of the conventional character 

 than we noted in the case of distances,! for we have no 

 method of deciding that the difference between temperatures A 

 and B is equal to the difference in the case of B and C com- 

 parable with the use of the foot-rule in spatial measurement or 



of the pendulum in time determination. 



* 



33. 



The relation of this " objective condition " to what we have 

 hitherto described as " elements of the Objective " obviously 

 demands discussion. It will at once be noted that there is no 

 reason to suppose that the " real " hotness of the body is ever 

 presented to any of the observers ; at any rate, it is certain 

 that it is never presentecl with any certificate of its supremacy to 

 the presentations of hotness which other observers experience. 

 We have found ourselves, in fact, compelled to regard all such 

 presentations as'standing on the same level of Objectivity^ and 

 have been driven in consequence into a certain amount of 

 metaphysical system-building. It is clearly opposed to the 



* This problem is discussed by Boyle in his Experimental History of 

 Cqld, 1665, First Discourse ; also p. 513. 



"'t Kelvin's "absolute thermometric scale" seeks to avoid this 

 conventionality. A brief account of the theory of this scale of tem- 

 peratures is given in Ch. IV, infra, 52. 



J Supra, p. 15. 



