254 THE CALCARIMETER. 



too often by paying attention to the strength of the reaction 

 of the fluid upon the litmus-paper. 



In most marls which have served as the subjects of my ex- 

 periments more or less alumina is to be found, a part of which 

 is dissolved by the acid, of which part a very good use can be 

 made. While adding the ammonia the alumina immediately 

 around where the ammonia falls is thrown out of solution ; and 

 if we stir the liquid, the alumina will be redissolved so long as 

 there is any free acid ; so that when the flocks of alumina are 

 no longer taken up we are furnished with an assurance that 

 the process is nearly completed. The acid that the alumina 

 and iron take up is acted upon by the ammonia with almost 

 the same readiness as if free, so that no cause of error is to be 

 apprehended from that source. 



It may sometimes happen from oversight that too much 

 ammonia is added. Notwithstanding this the analysis need 

 not be lost. Still holding the instrument in the left hand over 

 the cup, having of course arrested the flow of the fluid, we 

 pour some of the acid solution into a wine-glass, introduce 

 the small end of the acid instrument into it, and allow it to 

 rise on the inside to either of the small marks, and add this 

 acid to the liquid, and go on as before with the experiment, 

 and at the conclusion read off what is indicated, and to it add 

 10 or 20 according as we may have added the acid measured 

 by the first or second mark. 



After what has been said a few words will suffice to explain 

 how the instrument operates. 



It takes 50° of acid to dissolve fifty grains of carbonate of 

 lime, or 1° to dissolve one grain; and it takes 2° of the am- 

 monia solution to neutralize one of the acid; and therefore in 

 treating a substance consisting in part of carbonate of lime, 

 for every grain that is present one degree of the acid is taken 

 up, so that when we come to add the ammonia we know how 

 much of the acid is taken up by the quantity of ammonia left 

 behind, thereby knowing the number of grains of carbonate 

 of lime, which we multiply by two (as fifty grains of the sub- 

 stance was used) to arrive at the percentage. This multi- 

 plication is not actually performed, as the instrument is so 

 graduated as to dispense with it. 



Were it at all necessary to give any evidence of its easy 



