on the new method for the estimation of Nitrogen, §c. 289 
The calculation from the formula C,, H., N, O, gives only 
7°16 per cent. 
The excess of 4°8 per cent. here obtained, estimated by 
weight, amounted to 0024 gramme nitrogen, or in volume 
nearly 25 cubic centimetres; in the above experiments with 
sugar 0°015 gramme of nitrogen was, according to M. Reiset, 
condensed in the soda-lime, and therefore took a part in the 
formation of the ammonia. 
If we consider that the decomposition of organic bodies of 
difficult combustion by the hydrates of the alkalies does not 
take place at a heat below redness, that further, the heating 
of the contents of the tube by the fire placed around it cannot 
be so sudden as to produce in an instant the temperature ne- 
cessary for combustion, but that the heat, even when sudden, 
penetrates the mixture only by degrees, and that the greater 
portion of the inclosed or condensed air is driven out by its 
own expansion, we can scarcely comprehend how M. Reiset 
could entertain the idea that the nitrogen condensed in the 
mixture could take a part in the formation of ammonia. He 
certainly brings forward an experiment apparently supporting 
this view, viz. that by the combustion of 1°500 gramme o 
sugar in a current of air, the combustion being thus quickened, 
only 0:0099 nitrogen was obtained, instead of 0°0149. The 
ammonia did not increase when pure nitrogen was passed 
over the mixture during the combustion. I shall subsequently 
return to this point. ; 
I have repeated and partly varied the experiments of Reiset, 
and have-come to entirely different results. 
1214 sugar-candy of the shops by combustion with the usual 
mixture of soda-lime, which had not been previously ignited, 
gave on evaporation with chloride of platinum and ignition of 
the washed residue, 0°006 metallic platinum = 0:00086 nitro- 
gen = 0:07 per cent. of the sugar burned. 
0°386 pure stearic acid recrystallized from alcohol, gave 
0002 metallic platinum = 0:00028 nitrogen. 
0°430 leguminous starch, prepared in the laboratory of 
Giessen, and purified with sulphuric acid, gave 0:005 metallic 
platinum equivalent to 0:0007 nitrogen. 
A gramme of the above-mentioned starch was submitted to 
dry distillation. The product of distillation was mixed with 
hydrochloric acid, evaporated, the residue dissolved in water, 
mixed with chloride of platinum, and again evaporated. After 
treatment with alcohol and ether, a portion of ammonio- 
chloride of platinum remained, which ignited left 0°004 me- 
tallic platinum. The ammonia obtained by the combustion 
with soda-lime was thus, in part at least, contained in the 
starch, and was no product of the operation. 
Phil. Mag. S. 3. Vol. 22. No. 145. April 1843. U 
