376 On the Heat developed in Combustion 



the aperture of the worm, it is continually cooled by 

 the ambient air that surrounds it on all sides. It would 

 be possible, no doubt, by calculations founded on a 

 knowledge of the law of refrigeration of the receiver, 

 which might be found by separate experiments, to ascer- 

 tain the quantity of the effect produced by the refrigera- 

 tion in question ; and this even with a certain degree of 

 precision: but -it would have been impossible by this 

 method, or by any other known, to calculate the effects 

 of another cause of error, less obvious perhaps, but 

 certainly more weighty, than that of the refrigeration 

 of the external surface of the receiver. 



The nitrogen which is mixed with the oxygen of the 

 atmospheric air is necessarily carried into the worm 

 with the proper products of the combustion ; and with- 

 out a precaution, which it occurred to me to employ to 

 prevent the effects of this cause of error by making 

 a compensation for them, all the experiments would 

 have been of no value. 



Fortunately the method I employed to obviate the 

 effects of this cause of error was sufficient to prevent at 

 the same time those that might have arisen from the 

 cooling of the outer surface of the receiver. 



As the receiver is cooled, whether by the atmospheric 

 air in contact with its external surface or by the nitro- 

 gen and other gases traversing the worm with the 

 products of combustion, only so far as the worm is 

 hotter than the surrounding air, while, on the contrary, 

 it is heated by these elastic fluids whenever it is at a 

 lower temperature than they are, by arranging matters 

 so that the temperature of the water in the receiver shall 

 be a certain number of degrees, 5 for instance, below 

 the temperature of the air at the beginning of the ex- 



