2 18 Prof. Magnus on the Condensation of Vapours. 



millims. in the level of the liquid was occasioned. It results 

 from this, that the double shades of Melloni, which are so much 

 used, afford a sufficient protection only at low temperatures, such 

 as occur in the common thermo-electrical experiments; for if 

 in such experiments the source of heat has sometimes a tolerably 

 high temperature, still it operates from so great a distance that 

 the heat which passes from it to the shade is always minute. 



In the analytical theory of heat, the alteration of temperature 

 of a point situated at a distance from a constant source of heat 

 is regarded as a function of the distance and of the time elapsed 

 since the presentation of the source of heat. Also the means 

 are given for estimating this heating numerically ; these 

 estimations are, however, only relative to the heating which 

 would take place in another substance under similar circum- 

 stances. The absolute measurement of the time which is requi- 

 site for the attainment of a certain temperature in a point does 

 not enter into the account. After the publication of his Theorie 

 de la Chaleur, Eourier proposed to measure the conductivity of 

 thin plates by placing them on a bed of constant temperature, 

 and laying his contact thermometer (consisting of a mass of 

 mercury in which there is a very delicate thermometer), after it 

 had been heated, upon this plate, and noting the times which 

 must elapse in order that the thermometer, by the employment 

 of different plates, may become cooled to a like difference of 

 temperature. 



Here the times, which are taken up by each cooling, are indeed 

 directly observed ; but Fourier gives no information about the 

 duration of these times in the few experiments which he has made, 

 and, except himself, so far as I know, no one has employed his 

 method. 



Moreover the theory of the propagation of heat will gain 

 nothing from the above-described experiments. Nor are they 

 adapted for the measurement of the times which elapse before 

 different plates undergo a given alteration of temperature ; for, 

 to this end, the degree of sensitiveness of the thermoscope should 

 be quite unalterable ; and that is hardly attainable. Moreover, 

 various other difficulties stand in the way of such admeasure- 

 ments. But notwithstanding that one cannot make accu- 

 rate measurements, it remains a very remarkable fact that the 

 minute amount of heat radiated by a flame at a distance of 60 

 millims. from a copper plate of 1*5 millim. in thickness should 

 pass through that plate and through seven other similar plates 

 standing 15 millims apart, and should be appreciable after the 

 lapse of one minute. 



Collecting together the results of this investigation, it has 

 been established that the most various organic and inorganic 



