4 BULLETIN 76, U. S. DEPARTMENT OF AGRICULTURE. 
the strength of arsenical solutions prepared after the ''standard 
formula." 
It is necessary therefore to describe (1) methods for the determina- 
tion of actual arsenious oxid, and (2) methods for the deter min ation 
of " total arsenic/' that is, methods which shall include arsenic 
present in the oxidized form as well as that existing as actual arse- 
nious oxid. It is necessary also to describe different variations or 
modifications of processes for these two determinations, respectively 
appropriate for use by (1) trained chemists with abundant laboratory 
facilities, (2) persons possessing but slight chemical training and 
equipment, and (3) persons in the field possessing no chemical 
knowledge or training whatever, who obtain their results by manipula- 
tion of an " outfit" prepared by a trained chemist. 
LABORATORY METHODS. 
It is to be distinctly understood that the methods here described 
are not in all cases adapted or intended to reach a high degree of 
accuracy from a chemist's viewpoint, though as a matter of fact 
most of them are very accurate under especially favorable conditions. 
A certain variation must necessarily be allowed in the composition 
of baths prepared in the field. Again, experience has shown that the 
percentage of arsenic in such baths may vary within certain limits 
without perceptible effect upon either then- effectiveness or then- 
safety. Therefore the purpose of assay is not primarily to determine 
the exact percentage of arsenic — though the nearer this result is ap- 
proached the better — but to insure that the composition of the baths 
used shall fall within certain limits. Extreme accuracy, which is not 
necessary, must be sacrificed to simplicity and convenience, which 
are necesssary to permit the execution of very frequent tests. 
METHODS FOR ACTUAL ARSENIOUS OXID. 
Whatever method may be employed for the determination of actual 
arsenious oxid, there are certain precautions connected with the tak- 
ing and the storage of samples which can not be omitted if the results 
are to be of any value whatever. As already noted, the arsenic in 
used arsenical baths tends to change its degree of oxidation, 
mainly through the action of microorganisms. Other things being 
equal, the rapidity with winch this change takes place is greatly 
influenced by temperature, so much so that a sample contained in an 
ordinary cork-stoppered bottle and exposed to summer heat may show 
more change after one day than it would show after a week in the 
original vat in its comparatively cool location underground. The rate 
of the change can accordingly be retarded by keeping the sample cold, 
as by storage in an ice box after receipt; but if the sample has trav- 
