6 BULLETIN 259, U. S. DEPARTMENT OF AGRICULTURE. 
send samples, many of this set of experiments were discontinued 
at just about the time when the second stage probably had been 
reached, but the phenomenon will be better illustrated in the series 
of experiments in 1914. 
In explaining the occurrence of the second stage in the history of m 
the vats which show it, it is not enough to say that the considerable 
degree of usage to which the vats were subjected and the consequent 
copious renewal with fresh bath allowed no time for the arsenic to 
become largely oxidized before it had been carried out by the cattle, 
for in several cases it is very plain that the lines tend to converge 
under a figure of use that is no greater but even decidedly less than 
the figure of use under which they had previously diverged, a fact 
explicable only by the existence in the former case of some agent 
which positively prevents oxidation. During the examination of 
samples from vat A, it happened several times that the field test made 
at the vat by the field agent of the bureau would be reported as 
showing a figure for actual arsenious oxid decidedly lower than the 
figure obtained through laboratory analysis, while in some such cases 
laboratory analysis showed no difference at all between the results 
for actual and total arsenious oxid; that is, the bath appeared to 
be entirely unoxidized. While the laboratory was puzzling over 
this annoying phenomenon and contemplating active steps to inves- 
tigate the cause, the previously noted article by Laws came to its 
attention. If Laws's conjecture of the action of reducing bacteria is 
correct, then it was possible that in the case of these doubtful sam- 
ples the analyses of the field agent and the laboratory were in fact 
both correct; that is, the composition of the sample had changed en 
route. Therefore, laboratory experiments were undertaken to deter- 
mine the truth of Laws's conjecture, which was not supported by any 
direct experimental evidence. At the same time it seemed desirable 
also to run some experiments on oxidation. 
LABORATORY EXPERIMENTS. 
The plan of investigation was simple. Strong stock solutions of 
sodium arsenite and sodium arsenate were prepared by boiling the 
corresponding oxids of arsenic with sodium carbonate in amounts 
proportionate to the regular bath formula, Nutrient broth was 
prepared according to the formula, 5 grains solid meat extract, 10 
grams Witte's peptone., and 24 grams sodium chlorid per liter, but 
sterilization appeared unnecessary for the work in view. For the 
tests the stock arsenic solutions were diluted, mixed with broth 
(about 30 per cent of the final volume), antiseptic added if desired, 
and the mixture inoculated with a portion of used dipping-bath 
from the field. No tar was added. The freshlv made mixtures 
