78 Mr Potter on the Specific Heats of Metals, <$c. 



The great discordance in the results given by the reverse 

 ways convinced me that the tin-plate vessel was an insufficient 

 apparatus ; and also in making use of the second method, or, 

 addino- the cool metal to the heated water, that considerable 

 care and dexterity would be necessary to gain correct results. 

 In lieu of the tin-plate vessel, my later experiments have been 

 made with a compound one, consisting of an inner vessel of 

 thin varnished linen surrounded by a close and friezed woollen 

 cloth, and both these fixed in an earthen-ware one, of a dimen- 

 sion just sufficient to admit them. This compound vessel was 

 capable of holding about 3500 grains of water- In making 

 use of what I shall call the first method, I found this vessel 

 quite sufficient, without need of any other particular precau- 

 tion except that of going as quickly through the experiment as 

 possible ; but, in attempting the second method, I found the 

 results still very unsatisfactory, until I adopted the plan of sur- 

 rounding the compound vessel with water of nearly the tem- 

 perature of that about to be experimented with, and also of 

 preventing the evaporation from the surface by using a cover 

 of varnished linen cloth with the same sort of woollen cloth 

 before-mentioned attached to it. In this manner, and by tak- 

 ing care to raise the cover no more than necessary when the 

 metal was introduced, I got results which accorded as near 

 with those got on the other method as I had any reason to 

 expect. We might anticipate some little difference in the re- 

 sults of the two methods from the alterations in capacity arising 

 from temperature ; but a limit of error must be allowed much 

 greater than any effect arising from this cause, between 32° 

 and 212° Fahrenheit. 



In taking the specific heats on the first method, it is neces- 

 sary, when great correctness is wanted, to have the solid in one 

 piece only, and presenting as small a surface as possible, to 

 prevent to the utmost the loss of heat whilst it is in the course 

 of being transferred from the boiling to the cold water. In 

 this respect, MM. Dulong and Petit committed an oversight 

 in experiments published in a previous paper to the one just 

 alluded to ; and it is on this account that I consider the num- 

 bers found for gold, silver, and copper, in the table below, on 

 the first method, as much too small ; because for these metals 



