the specific heats of certain solids. 269 



100° and SOO"^. Yet in a subsequent table of the memoir, Petit 

 and Dulong have given the indications of thermometers formed of 

 the different metals, on the basis of their specific heats, compared with 

 those of an air thermometer at 300°, and they have put down that 

 of platina 317.9, which it obviously coidd not be, if its specific heat 

 were invariable, but supposing that heat to increase from .0335 at 

 100° Cent, to .0355 at 300°, the indication ascribed to it would be 

 correct. In a recent edition of Turner's Chemistry, by Dr. Bache 

 of this city, this error has been corrected. 



But, to return to the subject of iron, we find, in the various works 

 of philosophers, a remarkable discrepancy between their statements 

 of the specific heats of this metal. The following are among the 

 results obtained by the different individuals whose names are an- 

 nexed. 



On iron of sp. gr. 7.87G, the specific heat -was found, .1200 by Wilcke. 



" soft bar iron, sp. gr. 7.724, . . . .liyo " Gadolin. 



" sheet iron, ..... .1000 " Lavoi.sier. 



" iron, of M'hat quality not specified, - - .1"250 " Kirwan. 



" do. do. do. - - - .12G9 " Crawford. 



" do. do. do. - - - .1450 " Irvine. 



" do. do. do. - - - .1300 " Dalton. 



" oast iron, abounding in plumbago, - - .1240" Gadolin. 



" white cast iron, .... .1320 " do. 



" iron, (kind not specified,) between 32° and 212° F. .1098 " Petit and Dnlong. 



" do. do. do. do. 32 " 392 .1150 " do. do. 



" do. do. do. do. 32 " 572 .1218 " do. do. 



" do- do. do. do. 32 " Gli2 .1255 " do. do. 



The mean of these thirteen numbers is 0.12377. The wide dis- 

 crepancies are probably owing to the circumstances under which the 

 authors respectively operated, and to physical differences in the 

 metal. Nor is the disagreement confined to these results ; for while 

 Crawford and Irvine contend that the specific heats of bodies re- 

 main constant, at all temperatures, Dalton, Dulong and Petit main- 

 tain that they increase with the increase of temperature. But it 

 seems difficult to reconcile this supposition with another result of 

 Petit and Dulong, viz. that the specific heat of all bodies is inverse- 

 ly as their atomic weight, unless we could suppose what is manifestly 

 absurd, that the atomic weight varies with the temperature, or that 

 in different bodies the rate of increase in specific heat varies always 

 inversely as the atomic weight. Thus, if H were supposed the spe- 

 cific heat of any body and A its atomic weight, and if ^H were the 

 increment of specific heat for a given rise of temperature, then not 



