- 8 3 - 
monia, and precipitated with magnesia mixture. The pre¬ 
cipitate was allowed to stand, though precipitated hot with 
the aid of violent stirring, for twelve hours, and then, after 
washing, dissolved and precipitated, often with the addition 
of a little citric acid if there was any suggestion of the 
presence of ferric phosphate in the phospho-molybdic acid. 
The third portion of the solution was used for the determi¬ 
nation of the alkalies. I added to the boiling solution 
'aric chloride, enough to precipitate the sulphuric acid, 
V xnd ferric chloride to combine all of the phosphoric 
acid and then washed milk of lime to alkaline reaction, 
washing the precipitate free from chlorine. If too much 
milk of lime is added this is rendered much more difficult. 
I prefer to add no more milk of lime than is necessary to 
precipitate the ferric salts and render the solution alkaline, 
filter and wash out the precipitate, add lime water to filtrate 
and evaporate down to a small volume, filtering off the 
magnesic and calcic salts which separate before precipitat¬ 
ing with ammonic carbonate ; by which, together with a 
little oxlate of ammonia, the lime was removed. The 
last portions of the lime are removed as usual. The addi¬ 
tion of baric salts makes this portion of the operation 
more difficult, but if more than traces or only small quan¬ 
tities of sulphuric acid are present, the addition of baric 
chloride is advisable. The potassic-platinic chloride was 
uniformly dissolved in boiling dilute hydrochloric acid and 
crystallized by evaporation on the water-bath. If enough 
hydrochloric acid is added, there will be no trouble experi¬ 
enced by the formation of a crystalline film to prevent 
evaporation ; on the contrary, the salt will crystalize in 
large, well defined crystals, as good as free from water, if 
not entirely so. The potassic-platinic salt was weighed on 
a tared filter after drying in the water-oven for not less than 
two and one-half hours. 
CHLORINE AND SULPHUR. 
The chlorine was determined in two ways ; first, from 
the ash and second from the plant. I was induced to do 
this by two observations, the first of which has already 
been given, i. e., an observed loss of chlorine; and the 
second was the fact that I obtained such unusually high 
percentages of chlorine in the different ashes that I at first 
felt that probably I had made some mistake, particularly so 
as there was not even an approximate agreement between 
the results obtained from what would be considered com¬ 
parable samples. The first method was, to dissolve a 
weighed portion of ash in cold dilute nitric acid with imme- 
