24 BULLETIN 600, U. S$. DEPARTMENT OF AGRICULTURE. 
MANGANESE. 
It was first hoped to make the manganese determination in the 
same subsample as that used for the determinations of barium, 
chromium, vanadium, and molybdenum. After the ash is fused with 
sodium carbonate it seems possible to precipitate the manganese with 
alcohol or hydrogen peroxide and determine the manganese in the 
insoluble portion, but while this method gives good results in many 
cases occasionally large errors were found. Using the Weller method 
the color seems to develop quicker and more uniformly when all the 
constituents of the plant ash are present, and the following pro- 
cedure was adopted: A 2-gram sample of the well ground and mixed 
plant was burned, dissolved as much as possible in sulphuric acid, 
and a little sulphurous acid added to dissolve any manganese peroxide 
that may have been formed. If the concentration of acid is kept 
within the proper limits this method leaves nothing to be desired. 
BARIUM, CHROMIUM, VANADIUM, AND MOLYBDENUM. 
For this determination 20 grams of the ground plant were ashed, 
fused with 10 grams sodium carbonate and 0.25 gram sodium 
nitrate. The methods outlined by Hillebrand?+ were applied. Except 
when grasses (high in silica) and pine needles (high in aluminum) 
are to be analyzed it is necessary to evaporate the solution only to 
near dryness, for the normally trifling amount of silica and aluminum 
present will do no harm. It has been found helpful to titrate the 
dissolved and filtered melt nearly to the neutral point, using methyl 
orange as an indicator, prior to the precipitation of the mercurous 
salts. The results for barium and chromium were satisfactory and 
are believed to be reliable. The vanadium tests, however, were not 
so sharp cut. The color reaction with hydrogen peroxide in nitric- 
acid solution was used. No samples of plant ash contained enough to 
titrate with permanganate. It required 0.0002 gram V.O; to give 
_ the characteristic vanadium test when added to 20 grams of the dry 
plant before incineration. The chromium and molybdenum tests 
were slightly more delicate. Spectroscopic tests for these elements 
were not used. Where vanadium, chromium, and molybdenum are 
reported as not found, there is less than 0.0002 gram of any one of 
those elements in 20 grams of the dry plant, or less than 0.001 per 
cent. 
| STRONTIUM. 
The determination of strontium in plant ash is not satisfactory. A 
large amount of calcium is always present, a fact which necessitates 
a large volume in which the precipitation is to be made. It seems 
probable that a considerable portion of the small amount of strontium 
1 Bul. 422, U. S. Geol. Survey, pp. 138 and 150 (1910). 
