460 Mr. Griffin on the Constitution of 



from 50|° to 51^°. A solution of bicarbonate of potash from 

 50±° to 5 If °. Oil of vitriol from 51£° to 52^°. 



These experiments show how difficult it is to determine 

 the amount of heat that is disengaged when liquors are mixed, 

 since the agitation necessary to effect the mixture can of 

 itself produce part of the rise of temperature which is attri- 

 buted wholly to the chemical action of the substances mixed 

 together. 



I shall conclude this notice of ammonia by referring to a 

 fact of considerable importance which is disclosed in Table 

 XVI. This is, that it is impossible to state accurately the 

 specific gravity of solutions of ammonia with less than six 

 places of decimals, since the effect of every chemical degree 

 on the specific gravity is '000125. Thus while a solution of 

 800° is accurately expressed by '9 or '900000, a solution of 

 799° requires "900125, and a solution of 801° requires 

 •899S75. This peculiarity ought to be attended to in future 

 researches on aqueous solutions. 



This observation leads me to make another which may be 

 interesting to practical men, who are accustomed to estimate 

 the value of ammonia and similar liquors after the indications 

 afforded by hydrometers that have scales with large divisions 

 — TwaddelFs, Baume's, and the like. Now 40 chemical 

 degrees of ammonia, or §■ of a test atom, affect the density of 

 the liquor only so much as answers to 1 degree of the scale 

 of TwaddelFs hydrometer. Such an instrument therefore, if 

 perfect in construction, and used with unerring skill, is unfit 

 to tell the strength of solutions of slight density, such as those 

 of ammonia and acetic acid : methods of chemical testing are 

 greatly preferable. 



Since the preceding section on ammonia was written, I have 

 seen an abstract of Dr. Faraday's recent researches concern- 

 ing condensed gases. In that it appears that his present 

 estimate of the specific gravity of condensed liquid ammonia 

 at 60° F. is 0*731. His former estimate was 0*76. The 

 specific gravity which theoretical considerations led me to 

 adopt for anhydrous liquid ammonia at 62° F. is 0*708333. 

 This differs from Dr. Faraday's new experimental determina- 

 tion by only 0*022667. The number of test atoms of ammonia 

 contained in a decigallon is, according to Dr. Faraday's spe- 

 cific gravity, 24*08, according to mine 23*33. The difference 

 is only f of a test atom per decigallon. The atomic measure 



