310 On the Rapid Production of Stemn. 



stated at 8 atmospheres ; so that by adding the 52f derived from 

 the over-heated metal we should have no less than 60^ atmospheres 

 or 906 lbs. to the square inch for the resulting elasticity. This is 

 upon the assumption that steam obeys the same law in regard to its 

 relative bulk and elasticity, as that which governs atmospheric air- 

 But if it do not follow that law, there is no probability whatever that 

 the pressure would be less than in the direct ratio of the density. 



Before proceeding to the detail of experiments on other metals, I 

 think it proper to present the following series of results, in which my 

 main object was to ascertain, accurately, the rapidity of cooling of 

 iron from incandescence down to 212°, taking into consideration the. 

 temperature of the water, both at the beginning and end of the ex- 

 periment, its weight in some cases, and the relation, in all cases, be- 

 tween the weight of metal and the amount of its generating surface. 

 These experiments were performed in an apparatus similar to that 

 described in my former communication, but furnished with an at- 

 tached thermometer to mark with accuracy the tfe^mperatures attained. 

 The result, as will be seen, is, that the times approximate to an in- 

 verse proportion to the generating surface. This proportion will not 

 be found to obtain, where part of the heat was employed in raising 

 temperature, and a part in generating steam. The time demanded 

 for cooling a given mass of metal from redness to 212°, by the latter 

 process, must be greyer than by the former, both because the tem- 

 perature of the liquid, which is to receive heat, is greater, and the 

 difference betwpen it and the metal less, and because the surface of 

 the iron is momentarily denuded of water and prevented from acting 

 by a constant and uniform communication. The temperature, in a 

 few instances, was calculated by multiplying the weight of water by 

 the number of degrees through which it was heated, and dividing the 

 product by the weight of metal multiplied into its specific heat. To 

 the quotient was, of course, added 212°, the temperature at which 

 the metal was withdrawn after every trial. 



