Prof. Frankland on the Origin of Muscular Power. 185 



same period of time as the gaseous products of combustion, 

 depressed the temperature of the water by only o, 02 C. 



By placing a delicate thermometer in the escaping gases, and 

 another in the water, no appreciable difference of temperature 

 could be observed. Both these corrections may therefore be 

 safely neglected. 



The two remaining corrections can be best considered together, 

 since a single careful determination eliminates both. When a 

 combustible substance is burnt in gaseous oxygen, the con- 

 ditions are essentially different from those which obtain when 

 the same substance is consumed at the expense of the com- 

 bined or solid oxygen of potassic chlorate. In the first case 

 the products of combustion, when cooled to the temperature of 

 the water in the calorimeter, occupy less space than the sub- 

 stances concerned in the combustion, and therefore no part of 

 the energy developed is expended in external work — that is, in 

 overcoming the pressure of the atmosphere. In the second case 

 both the combustible and the supporter of combustion are in 

 the solid condition, whilst a considerable proportion of the pro- 

 ducts of combustion are gases. The generation of the latter 

 cannot take place without the performance of external work ; for 

 every cubic inch produced must obviously, in overcoming atmo- 

 spheric pressure, perform an amount of work equivalent in 

 round numbers to the lifting of a weight of 15 lbs. to the height 

 of one inch. In performing this work the gases are cooled, and 

 consequently less heat is communicated to the water of the calo- 

 rimeter. Nevertheless the loss of heat due to this cause is but 

 small. Under the actual conditions of the experiments detailed 

 below, its amount would only have increased the temperature of 

 the water in the calorimeter by o, 07 C. Even this slight error 

 is entirely eliminated by the final correction which we have now 

 to consider. 



It is well known that the decomposition of potassic chlorate 

 into potassic chloride and free oxygen is attended with the 

 evolution of heat : if a few grains of manganic oxide, or, better, 

 of ferric oxide, be dropped into an ounce or two of fused po- 

 tassic chlorate, which is slowly disengaging oxygen, the evo- 

 lution of gas immediately proceeds with great violence, and 

 the mixture becomes visibly red-hot, although the external ap- 

 plication of heat be discontinued from the moment when the 

 metallic oxide is added. The latter remains unaltered at the 

 close of the operation. It is thus obvious that potassic chlorate, 

 on being decomposed, furnishes considerably more heat than that 

 which is necessary to gasify the oxygen which it evolves. It 

 was therefore necessary to determine the amount of heat thus 

 evolved by the quantity of potassic chlorate (9*75 grins.) mixed 



