RICIIAIIDS AND LAMB. — SPECIFIC HEAT OF LIQUIDS. G79 



at 24°, were respectively —0.002, — O.OOo, — O.OOG, -0.006; or on the 

 average —0.005 = t/r^,. At 1G° this had previously been found to be 



r. ^. ,, rr., r^ r. (" 0.006° + 0.005°) 7000 



-0.006 =:: i'j,. Then Ai - K = ^ ^ = - 0.9. 



o 



The total heat capacity having been in the first place 0.947G (170 + 

 196.7 X 18) + 196.7 x 18 = 7056.8 (that is, the sum of the separate 

 heat capacities of the argentic nitrate solution and tiie water) the heat 

 capacity of the mixture will be 7056.8 — 0.9 = 7055.9, and the specific 

 iicat will be this (luautity divided by the heat capacity of an equal weight 

 of water, or (170 + 196.7 X 18) + 196.7 X 18 = 7251.2.* The spe- 

 cific heat of the dilute solution is thus 0.9731. 



Further determinations of many substances by all these methods will 

 soon be made in this Laboratory. The determinations above given are 

 only preliminary, as has been said ; and they are recorded here I'ather to 

 indicate the working of the methods under consideration than to fix the 

 values in question. Much more accurate work could undoubtedly be 

 done with the method if more pains was taken with the thermometry, 

 without question the least certain of the measurements involved. 



We take pleasure in acknowledging our indebtedness to the Rumford 

 Fund of the American Academy of Arts and Sciences for generous 

 pecuniary help in the construction of the somewhat complicated 

 apparatus. 



Summary. 



1. A new method of determining the specific heat of solutions is 

 described. 



2. A new device for determining accurately the heat of dilution of a 

 solution is described. 



3. Both these methods involve only the reading of stationary ther- 

 mometers, and therefore avoid error from thermometric lag. 



4. In both methods error from uncompensated cooling is avoided as 

 mucli as possible. 



5. Each of these methods may be used for certain types of chemical 

 reaction, as well as for the physico-chemical measurements above 

 mentioned. 



6. Tiie method proposed by Berthelot for determining from the change 



* It will be easily seen that the rounded-off values of the atomic weights suffice 

 in this calculation, if a result no nearer than O.OI per cent is needed. 



