236 Dr. J. H. G-laclstone on the Want of Uniformity in 



examined. The alloy Cu 38*38 per cent, appears, however, 

 to be pretty uniform. 



2. Supposing, however, that we had the true constitution 

 of the specimen and the exact products of the chemical action 

 both in solution and as gases, there would still remain an un- 

 certainty about the number of calculated calories. The original 

 determinations of Thomsen and the recent ones published by 

 Berthelot are very discrepant ; and the various corrections to 

 be made, especially in regard to the gases, are too uncertain 

 to admit of any trustworthy comparison within the narrow 

 numerical limits with which we have to deal. 



There is, however, another method of comparison which 

 appears to be more promising. Although Thomsen and 

 Berthelot differ as to the calories produced by any particular 

 reaction, there is one point on which they practically agree : 

 that is, the excess of calories in a zinc reaction over those in 

 the corresponding copper reaction. Thomsen makes it for 

 one gramme 750 calories, Berthelot 756 calories, which is 

 a practical agreement (say 753 calories). Starting from 

 604 calories, the value, according to Gait, when pure copper 

 is dissolved in nitric acid of sp. gr. 1*360, we should have 

 1357 calories when pure zinc is dissolved, provided the 

 chemical action were the same in each cane ; and all the calori- 

 metrical results from the different specimens of alloy would 

 theoretically lie, not upon the line drawn in Dr. Gait's 

 diagram, but upon the straight line drawn between 604 

 and 1357. On referring to the diagram, we see that the 

 little crosses fairly coincide with this line from pure copper 

 to about the Cu 70 alloy, but beyond that there is less heat 

 produced than the mean indicated by the straight line, the 

 maximum deviation being at about Cu 37. Beyond that point 

 the heat gradually augments, exceeding the mean at about 

 Cu 30, with a somewhat irregular progress to 1432 calories, 

 the amount experimentally obtained by Gait from pure zinc, 

 — an excess of as much as 75 calories above the theoretical. 

 The diagram of Baker's experiments shows the same general 

 result, with the difference previously referred to. 



Reference to Table I. will show that the products of 

 decomposition of the nitric acid between pure copper and the 

 Cu 38*38 per cent, alloy grve nearly the same amount of 

 nitrous acid in each case (representing about 80 per cent, 

 of the possible amount), and practically no ammonia. The 

 remaining 20 per cent., or thereabouts, consists mainly of 

 nitric oxide. As the products in each of the five cases seem 

 to be the same, and nearly in the same proportions, we should 

 expect that the amounts of heat evolved would form a pretty 



