﻿Intelligence and Miscellaneous Articles. 159 



foreign gas, but humid, the real temperature of combustion wilJ be 

 less for the latter. 



It is clear that these formulae apply to oxide of carbon with mere 

 numerical modifications. They likewise suit, whatever be the pres- 

 sure, provided this is constant in each case. 



Without trying prematurely to indicate the arrangements by 

 which I hope to realize some, at least, of the cases under consider- 

 ation, I pass to the examination of a recent memoir by M. Bunsen 

 on the present question. 



M. Bunsen detonates gaseous mixtures in a valved eudiometer, 

 and calculates the temperature from the pressure which is developed 

 at the moment of the explosion. This case differs in two points 

 from those which we have considered : in the first place, we must 

 introduce into the formulae specific heats at constant volumes, and 

 not at constant pressures ; and then the pressure does not remain 

 constant, but it increases up to the moment of the maximum of 

 temperature. These two circumstances tend to augment the real 

 temperature of combustion, at least if we assume that pressure tends 

 to diminish the tension of dissociation. 



It will thus be understood how M. Bunsen finds for the gas of the 

 pile 2800°, and M. Deville, operating under the ordinary pressure, 

 only 2500°. 



I next discuss M. Bunsen's theoretical conclusion, according 

 to which dissociation is not a continuous phenomenon, but one 

 varying suddenly at certain temperatures, in the interval of which 

 it remains constant, varying moreover in such a manner that there 

 is always a simple proportion between the dissociated part and that 

 which is not dissociated. I show that this conclusion has but little 

 probabilit)' - a priori ; for this instantaneous production of a phenome- 

 non which induces an evolution or an absorption of heat, when the 

 exchange of heat with surrounding bodies is necessarily continuous, 

 must cause sudden variations of temperature such as we do not 

 observe. 



M. Bunsen shuns the difficulty by assuming that, during the pas- 

 sage from one simple proportion to another, the temperature remains 

 constant. But we ought to observe this period of constancy. Thus 

 in the flame of a blowpipe with mixed gases, of which the various 

 parts offer precisely the successive conditions of a mixture in com- 

 bustion which is cooling, we ought to observe a very appreciable 

 space during which the temperature would be constant. JN"ow the 

 accurate experiments of M. Deville upon the blowpipe with oxide of 

 carbon indicate a continuous decrease of both the temperature and 

 the dissociation. 



Moreover, during this period of invariable temperature, which 

 should represent a very noticeable fraction of the total duration of 

 the refrigeration, there would be a gradual, and not a sudden, passage 

 from one simple proportion to another. The simple proportion 

 would therefore not be an absolute natural law. 



Finally, even the discussion of the experiments leaves, at least, 

 much doubt as to the legitimacy of the conclusion.— Comptesllendus, 

 December 28, 1868, vol. lxvii. pp. 1348-1352. 



