EXPEEIMENTAL KNOWLEDGE OF THE PKOPEBTIES OF MATTER. 497 



respond with the result obtained from the tracings of the style during 

 the whole period of cooling ; this, however, is not always the case, and 

 in the exceptional cases it is foand that, if the new curve is plotted out 

 from the numbers given by the experiment, the later portion of the cool- 

 ing will be represented by a straight line almost exactly for a consider- 

 able distance; but the line is found in some cases to be for a portion of 

 its length curved, the curved portion corresponding to very great tem- 

 peratures and pressures. The interpretation of these cases is that at the 

 highest pressures, and therefore the highest temperatures, there is dissocia- 

 tion, and that on cooling the dissociated gases gradually recombine till 

 at the temperature and pressure at which they have entirely combined the 

 line becomes straight through the rest of the cooling. Thus the line 

 deduced from the observed rates of cooling by means of the formula is 

 straight except in cases of dissociation. 



A case in point is carbonic acid gas which undergoes dissociation at 

 high temperatures, so that when CO and are mixed without being dried ' 

 and the mixture exploded it is calculated that a temperature of 3100° is 

 reached, and that only about '6 of the mixed gases have combined; on 

 cooling the remaining mixture of CO with O gradually combines until at 

 about 1800° there is no dissociation, and from 1800° to the temperature of 

 the room throughout CO2 is the only gas cooling. HjO vapour is not dis- 

 sociated in the conditions of these experiments below 3350°. 



From Berthelot's and J. Thomsen's long-continued thermo-chemical 

 researches there are obtainable abundant data for the total heat of com- 

 bustion of hydrogen, carbon monoxide, cyanogen, methane, for the heat 

 of formation of hydric chloride from its elements, and of the conversion of 

 cyanogen b^^ oxygen into nitrogen and carbon monoxide, and for other 

 reactions between gases which can be brought about by explosion. 



In the case of an explosive mixture of hydrogen and oxygen it is 

 found that there is no dissociation at the temperature of combustion 

 y360°, the pressure corrected for cooling during combustion being 619 cm. 

 of mercury. By adding to the explosive mixture a volume of hydrogen 

 and exploding after this addition, the temperature of explosion, as found 

 from the observed pressure and rate of cooling, is lower the more the 

 mixture is diluted with hydrogen, remaining explosive. Calling now in 

 any case the total heat of combination Q, the temperature (corrected for 

 oooling) T, and c the mean heat-capacity of 18 grams of water-vapour 

 between. 0° and T°, we have Q=cT ; whence c is found to be 16-5 for the 

 high temperature 3360° of this experiment (allowance having been made 

 for water- vapour present before the explosion). Values of c for lower 

 temperatures are found by adding CO2 to the explosive mixture of 

 hydrogen and oxygen, the explosion giving temperatures at which CO.2 is 

 not appreciably dissociated ; from these values of c the next step is to find 

 a formula of the form c^=a + ht, or with a third term in t-, for the heat- 

 capacities of 18 grams of water- vapour for temperatures up to or beyond 

 3000°. 



Retui'ning now to cases in which oxyhydrogen mixture is mixed with 

 hydrogen and this mixture exploded, Q being the heat produced by the 

 explosion, at constant volume, of the hydi-ogen and oxygen which form 18 

 parts of water, and on being the mass of hydrogen added, c' the mean heat- 

 capacity of this hydrogen from 0° to the temperature T' produced by the 



' See Dixon, Phil. Trans. 1884, Part II.; aud Client, ^en■s, 46, pp. 151, 152. 

 1888. f K 



