EXPERIMENTAL KNOWLEDGE OF THE PEOPEETIES OF MATTER. 487 



for silver, a is -05449, and 6=-0003929; so that ah, which is the 

 coefficient of t, is -0000184. 



And for copper, a=-09205 ; &=-0002308; and the specific heat of 

 copper is 



•09205 + -000021252^; 



which agrees with Bede's 



•0910 + -000023i 



so far as Bede carried his approximation. 



Quite recently Pionchon has made determinations' on the specific 

 heat of quartz up to 1000°, and finds that it increases rapidly np to 400° 

 from -1737 to -805 ; that from 400° to 1000° it remains remarkably con- 

 stant. Now the molecular heat during this interval within which the 

 specific heat of quartz is so nearly constant is 60 X ■305=18'3 ; and this 

 is 3 X 6"1, so that at these temperatures silica in the form of quartz has a 

 molecular heat which corresponds to Cannizzaro's rule. 



In reference to this Kopp quotes, from a memoir which he published 

 in 1879,^ a passage in which he says that it is not unlikely that the 

 molecular \\e?t,ts oi compounds of elements, which at ordinary temperatures 

 have atomic weights not in accordance with Dulong and Petit's law, may 

 possibly when raised to higher temperatures be so increased as to be in 

 accordance with Gamier and Cannizzaro's rule, and that compounds, 

 which are exceptions of Neumann's law as to the molecular heats of 

 compounds of like atomic composition, may in this way come into accord- 

 ance with it ; although it is often impossible to realise the conditions 

 required on account of approaching the fnsing-point of the body, or 

 exceeding the temperature at which the compound begins to decompose. 



Kopp remarks that the molecular heat 18'3 for SiOo not only gives a 

 result 3 x6'l nearly in accordance with Garnier and Cannizzaro's rule, 

 but approaches nearly to the numbers for compounds of the formulas 

 XCI2, XBr2, XI2 ; which are found by his experiments to vary between 

 this number and 19'0 ; and at the high temperatures of Pionchon for 

 quartz this compound ranges itself among the compounds to which 

 Neumann's rule applies. 



The elements which are at ordinary temperatures exceptions to 

 Dulong and Petit's law, so far as is known at present, are elements with 

 small atomic weights, including sulphur, with atomic weight 32. The metal 

 magnesium, with atomic weight 24, has a normal atomic heat; while that 

 of aluminium, atomic weight 27, is low, about 5'7 ; and that of beryllium, 

 atomic weight 9'1, is among the exceptions, the atomic heat being about 

 4; but the specific heat of beryllium varies immensely between 0° and 400°.^ 



Specific Heat — Liquids. 



The specific heat of many liquids was found by Regnault* for various 

 temperatui'es ; these were found always to increase with rise of tempera- 

 ture ; but no general relation could be found connecting specific heats of 

 bodies in the liquid state with molecular weights, temperature, or dilata- 

 tion. In the liquid state the considerable internal, as well as some 

 external, work complicates the expression, the dilatation, in fact, in the 



' a B 106, 1888, p. 1344. 2 Ber. 12, p. 896. 



» See Humpidge, Proc. Bey: Soc. 1885. 

 « Mevi. de VAcad. t. 26, pp. 262-295. 



