482 EEPOET— 1888. 



of diamond between 0° and 200°, he finds the mean heat-capacity of 

 diamond at twelve nearly equal intervals between 0° and 200", and thus 

 finds the constants in an equation of the form. C'o= a + bt — ct^ to 

 represent the mean heat-capacity between 0° and t° ; whence the specific 

 heat at t is a + 2bt — 3ct~. Similarly he finds for graphite an equation of 

 the form, y^=za-\-lit for specific heat at temperatures between 0° and 100°. 



Dewar by an experiment in which varieties of carbon were succes- 

 sively heated by an oxyhydrogen flame on a bed of quicklime, using a 

 water-calorimeter, found a mean specific heat "42 up to a temperature of 

 about 2100° ; which would give an atomic heat of about 5 at some tem- 

 peratures very much higher than 200°. This shows that a temperature 

 can be reached at which the specific heat of carbon gives an atomic heat 

 not much less than that required by Dulong and Petit's law. 



Previously Dewar had found^ a mean value of the specific heats 

 of varieties of carbon between 20° and 940°, the temperature of boiling 

 zinc ; this mean specific heat was 3"2. 



Weber in 1874^ continued his researches on the specific heats of 

 varieties of carbon, and found results from a large number of experiments, 

 from which he was able to deduce specific heats of diamond at a variety of 

 temperatures up to nearly 1000°, the determinations of the specific heats 

 of other forms of carbon being made at temperatures under 250°, at 

 which their specific heats began to approach that of diamond closely. 

 Weber's determinations showed that in reference to specific heat the 

 forms of carbon other than diamond behave alike, but differently from 

 diamond at the lower temperatures. 



In accordance with what is known with regard to the relation of 

 specific heat to temperature generally, there is no point at which it ceases 

 to increase with rise of temperature ; and although the true specific heat 

 of a solid body will not be given at any temperature at which it begins 

 to soften or to approach its liquid state, in the case of carbon there is no 

 such danger ; for the temperature at which it tends to soften or become 

 liquid is so high that we cannot suppose that we are in the neighbourhood 

 of it at 1000°. An inspection of the curve in Weber's paper of 1874, taken 

 in conjunction with the curves of variation of specific heats of metals 

 alluded to later on, will suggest that it is possible, by taking temperatui-es 

 higher and higher by hundreds of degrees, to arrive at results in excess 

 of the specific heat demanded by Dulong and Petit's law. The tempera- 

 tures at which Dulong and Petit's law holds good for metals are not 

 temperatures at which the specific heat is constant, but at which it varies 

 steadily and slowly ; and if this is the case, as it is for metals generally 

 at ordinary temperatures, the specific heats are such ac ai'e in accordance 

 with Dulong and Petit's law ; this is a general result of experiment, but 

 in the case of metals, which still at higher temperatures have specific 

 heats increasing steadily above 100° as they do between 0° and 100°, 

 experience does cot warrant us in using the specific heats at higher 

 temperatui'es (see Pionchon's results further on) . The case of carbon has 

 been shown by Weber to be entirely different from that of the metals, for 

 at ordinary temperatures the specific heat increases with rise of tempera- 

 ture very much more rapidly; but after the temperature has been raised 

 suSiciently high the specific heat at higher temperatures increases only 

 naoderately with rise of temperature, and carbon reaches the steady state 



' PkU. Mcuf. 4, 44, p. 461. 



2 Jahresh.'f. Chem. 1874, p. G4; and Ann. Chim. 5, 1876, p. 7. 



