76 Intelligence and Miscellaneous Articles. 



to be ; in fact his whole purpose is to arrive at a complete analysis 

 of the contents of the mind. The constant question in metaphysic 

 is, What do we mean by such words as Time, Space, Cause, Neces- 

 sity, Power, and other facts which are taken as ultimates by Science? 

 It does not aim at discovering the cause of these facts, but simply 

 at studying their relations. The introduction of this method is 

 mainly due to Kant ; and it is on this account that his claim to so 

 high a place in the history of philosophy rests, although his method 

 still retained a large admixture of the old conception. This new 

 method of philosophizing without the assumption of entities, which 

 we may call the New Metaphysic, will be understood by reference 

 to such works as Mr. Shadworth Hodgson's ' Philosophy of Reflec- 

 tion,' or M. Penouvier's Essais de Critique Generate. "What may be 

 the merits of the systems propounded in these books, taken as a 

 whole, I do not here discuss ; but of the former of them, at least, it 

 may be said that there is no conclusion touching scientific questions 

 which Science, as such, would hesitate to indorse. 



Tours truly, 



Wtndham P. Dxthbtab - . 



ON THE DEPRESSION OF THE ZERO-POINT IN MERCURIAL 

 THERMOMETERS. BY J.-M. CRAFTS. 



It is well known that a thermometer which has remained for a 

 long time at the ordinary temperature presents when it is heated a 

 depression of its zero-point. The amount of the depression depends 

 on the temperature to which the thermometer is raised ; and 

 M. Pernet has demonstrated that the depressions are proportional 

 to the squares of the temperatures when it is heated from 0° to 

 about 100°. For higher temperatures Mr. Mills has found a Aery 

 different course. He does not indicate all the details which 

 would be necessary to enable one to usefully criticise his data; 

 and I will confine myself to saying that the numbers he gives 

 resemble those obtained with thermometers which have not under- 

 gone a suitable preparation, and that the method of experimenting 

 which I desire to present to the appreciation of the Academy 

 leads to very different results. 



Before measuring the depressions produced by a reheating, it is 

 necessary to make sure of two things : — (1) that the starting- 

 point is the zero raised to the maximum ; (2) that the phenomenon 

 one wishes to observe is not masked by the permanent elevation of 

 the zero, which may take place at the same time as the depression. 

 I have ascertained that this last effect is invariably produced when 

 one heats even for a few minutes a thermometer which has not 

 been previously heated for a long time to a temperature near that 

 of the experiment. This movement can be effectually suppressed by 

 preliminary processes of heating which w ill be described further on ; 

 and each series of depressions and reelevationsmust be repeated, in 

 order to make sure that no permanent displacement of the zero has 

 taken place. Let us suppose this preparation completed: the ther- 





