— 8o — 
der, placed over the already charred mass became coated 
with a white film whose solution in distilled water reacted 
strongly for chlorine. I he temperature of the mass was at 
this time very low, and the escape of the potassic and sodic 
chlorides from the mass, whose surface was covered with a 
layer of already cooled ash, even if the temperature at the 
glowing points was high enough to volatilize them, is diffi¬ 
cult to believe. But one proof has already been adduced, 
thatwiththis highly nitrogenous plant, chlorine did escape, 
whatever mav be the facts relative to the sodium and. po¬ 
tassium. The loss of chlorine was also noted by direct 
observations in the incineration of a sample of seed, but 
the conditions were not similar and the film of chloride, 
collected on the cool platinum foil used in this instance, 
may have been the chloride of one of the fixed alkalies, or 
perhaps of both. No less care had to be exercise to avoid 
the formation of the alkaline sulphides either directly or by 
reduction of the sulphates, principally, if not exclusively, 
by the latter process. Close observation shows that, if the 
combustion is slow enough to avoid a high temperature in 
any part of the mass, there will be no sulphides formed, but 
otherwise yellow points may be detected. in the mass by 
careful examination before it has been stirred too much. 
In burning large quantities it is almost impossible to avoid 
its formation because of our inability to control the rate of 
combustion in all parts of the mass. 
The time consumed in preparing the samples of ash 
was very great,and, as we shall subsequently see,did not pro¬ 
duce results commensurate with the care bestowed upon it. 
The only point in which satisfactory results were obtained 
was in producing an ash practically free from carbon and 
one in which we had no free bases, either lime or alkalies. 
The evaporation of the ammonic carbonate solution to dry¬ 
ness and heating to 200 degrees C., with the addition of the 
insoluble, was probably without any . other effect than to 
assure the conversion of any caustic lime into the carbona-te. 
That it may have been the cause of the loss of any chlorine 
ors ulphuric acid, particularly the latter,is verydoubtful. This 
question suggested itself,and calcic sulplate .was repeatedly 
heated in this manner without change in weight. 
THE METHODS OF ANALYSES. 
. 
As the methods adopted in this work are not identical 
with those commended by the Association of Official Agri¬ 
cultural Chemists, we deem it just to give the methods used, 
which, in our hands, are convenient and very satisfactory. 
The general method was to dissolve 4-5 grams of the 
