144 Mr. Emmet t on the [Aug. 



follows that at aiiy temperature the specific heat of a mass of 

 matter is made up of the sum of all its several capacities from 

 the point of absolute cold, or the true zero, to the temperature 

 at which it is examined. Let, therefore, this range of tempera- 

 ture be divided into an indefinitely great number of equal inter- 

 vals ; or let the interval between the point of greatest density of 

 the solid and its density at any given temperature be similarly 

 divided ; and let the quantities of heat absorbed or evolved 

 while the body passes through these intervals be c, c% c" , &c. 

 and let s represent the specific heat ; then c -^ c' -\- c" -|- c'" + 

 &c. = s. Now if c = c' = c" = c'" = &.C. in what degree 

 soever s is diminished, the heat is evolved (the quantity evolved 

 at each interval being denoted by c), still no change takes place 

 in the capacity. If c" > c' > c, then by reduction of tempera- 

 ture, at each interval a quantity of heat = c", or c', or c, &c. 

 will be evolved, and the diminution of capacity at each succes- 

 sive interval will be c" — c', c' — c, See. ; but if c > c' c", 

 &c. then by diminishing s, the capacity will increase, the heat is 

 evolved. It may now with propriety be asked, what experiments 

 have been made upon the capacities of the metals for heat that 

 decisively prove that the capacity of the same metal changes in 

 the least sensible degree under the different circumstances of 

 temperature or pressure while it retains its solid state? Or can 

 we expect a sensible difference to be perceived ? The best expe- 

 riments hitherto made differ too much from each other to show, 

 with any degree of certainty, such small differences, which are 

 less than the probable and almost certain errors of the experi- 

 ment ; so that we have no proof from experiment of any such 

 change ; nor has any principle yet been developed which shows 

 whether the capacities of the same metal under different circum- 

 stances may not be all equal, or even greater, at a low than a 

 high temperature, or after tlian before compression. That the 

 quantity of heat evolved in this experiment was great cannot be 

 disputed, yet it was by no means sufficient to warrant the con- 

 clusions that have been drawn. The ignition of soft iron by 

 hammering demonstrates that caloric exists in the metals in a 

 state very little, if at all, inferior to that of ignition, and though 

 an apparently great quantity may be thus evolved, it may, and 

 most probably does, bear a very small proportion to the whole 

 quantity contained in the metal, as will subsequently be demon- 

 strated ; besides, we do not know how small a quantity of caloric 

 is sufficient to boil the quantity of water employed in this expe- 

 riment ; there is every reason to suppose it much less than is 

 generally imagined ; and in these experiments, a veiy large 

 mass of metal was submitted to an excessive pressure, and of 

 the mass, fresh strata were continually exposed to the compres- 

 sion by the wearing off of the brass : hence a definite quantity 

 of heat was separated from each stratum in succession. Now if 

 we admit the existence of caloric in a state of great density in 



