52 THE NATURE AND [SECT. n. 



These experiments are valuable, because they afford a proof that the additional 

 heat required for steam is either accurately or nearly a constant quantity. 1 



78. And they also show that the bulk or volume of steam is inversely as the 

 pressure, when the temperature is not altered. For as 80 : 40 : : 1208 : 604, which 

 added to the expansion would be 635, nearly; and 120 : 40 : : 1208 : 402, and 

 adding the expansion it is 427, nearly ; and conversely the density is directly 

 as the pressure ; the experiments being quite as near as could be expected in so 

 extremely delicate an operation. 



79. Count Rumford obtained a higher result ; and from his known skill in 

 such inquiries, much confidence may be placed in his experiments. The heat 

 was measured by means of the temperature communicated to a copper vessel 

 filled with water, which he called his calorimeter. Within this calorimeter a thin 

 serpentine pipe of copper contained the steam to be condensed ; hence the fluids 

 did not mix together, and loss by the escape of vapour was prevented. 



The water which the calorimeter contained was of a lower temperature than that 

 of the room by 5 or 6 ; and when the thermometer of the calorimeter announced 

 an augmentation of temperature of 10 or 12, an end was put to the experiment. 



The water produced by the condensation of the vapour in the serpentine was 

 carefully weighed, and from its quantity, as well as from the heat communicated 

 to the calorimeter, the heat developed by the vapour in its condensation was deter- 

 mined. 



As a small part of the heat communicated to the calorimeter was produced 

 from the cooling of the water, condensed in the serpentine pipe after the vapour 

 had been changed into water, an account was kept of this heat. It was supposed 

 that the water at the moment of condensation was at the temperature of 212, 

 being that of boiling water ; and it was determined by calculation, what part of 

 the heat communicated to the calorimeter must have been owing to the boiling- 

 water. 



In making this calculation, Count Rumford remarks, no " account was taken of 

 the difference in the capacity of water for heat, which depends on its temperature : 

 this is but imperfectly known ; and besides, the correction which would have been 

 the result could not but have been very small." 



The following are the details and results of two experiments made on the 21st 

 of January, 1812. The duration of each of the two experiments was from ten to 

 eleven minutes. The water had been boiled for some time to drive out the air 

 which it contained, before the steam was directed into the serpentine pipe of the 

 calorimeter. 



1 M. Despretz, Ann. de Chim. et de Phys. xxiv. 329. makes it 955 c -8. 



