APRIL 12, 1912] 
gations lasting not less than four months. 
There were thus 16 X 120—1,920 test days. 
On 40 of the 120 days, 7. e., on 640 days, exact 
urinary analyses were made. The benzoic 
acid was given in the following manner: In 
a fore period of 10 days no benzoic acid was 
given, then for two months 0.3 gram was 
given daily, then none for 5 days, 0.6 gram 
for 7 days, 1 gram for 7 days, 2 grams for 7 
days, 4, 6 and 10 grams, respectively, for 7 
days, and finally for 10 days no benzoic acid 
was given (the individual experimenters 
worked a little differently); the benzoic acid 
was mixed with the food and given as sodium 
benzoate. It must also be pointed out that 
the test persons were not exclusively of espe- 
cially robust health; in fact, some of rather 
moderate constitution were taken. 
In the first pages of the work is given, 
briefly, clearly and concisely, what interests 
us most, the conclusions of the three investi- 
gators. 
Sodium benzoate in small doses, less than 0.5 
gram mixed with the food, is without toxic effect 
and does not disturb health. Sodium benzoate in 
large doses, up to 4 grams mixed with the food, 
showed, upon investigation, no harmful effect in 
the general sense of the word. In some directions 
were observed small changes in certain physiolog- 
ical processes, the exact significance of which 
changes is not known. 
According to the investigations, the mix- 
ing of sodium benzoate in smaller or larger 
doses has no unfavorable or deteriorating in- 
fluence on the value of such food. In this 
series of experiments, also, the test persons 
were not in exact nitrogen equilibrium. They 
were allowed to enjoy at will, within certain 
limits, quite complicated, extremely varied 
and dainty foods, the food being merely 
weighed and analyzed. The daily consump- 
tion of food, which, in Germany in such ex- 
periments, we keep as nearly as possible 
uniform in nature and equal in amount, 
varied not inconsiderably, even though it must 
be confessed that the daily consumption of 
food by this method was relatively uniform. 
We can, therefore, without more evidence, 
just as easily draw favorable conclusions for 
SCIENCE 
581 
benzoic acid from the slight increases in 
weight which almost all of these well-fed test 
persons showed as unfavorable ones from the 
opposite facts in the Wiley experiments. It 
only shows that benzoic acid is not grossly 
harmful. Exact metabolism observations in 
the absolute sense are, also, impossible; only 
relative values can be discussed exactly, and 
here, too, the so widely varying nutrition is a 
somewhat disturbing factor. 
The utilization of dry substance, fat and 
albumin, as we are accustomed to observe it 
in Germany, is not ascertained here. But I 
have convinced myself, from a series of values 
which I have calculated, that no influences 
on the utilization are to be observed. The cor- 
puscular elements in the urine were not, con- 
trary to the observations of Wiley, increased. 
No general symptoms were noted, either with 
the smaller or the larger doses. Of the num- 
berless details of the three investigations we 
ean safely say that, like Wiley’s, they show 
that in almost all special cases nothing essen- 
tial is changed by the administration of ben- 
zoic acid; I may therefore confine myself to 
picking out a few points. Chittenden ex- 
presses, at the end, his extraordinarily favor- 
able opinion that sodium benzoate, up to a 
dose of 4 grams daily, is no more disturbing 
or harmful to the human organism than the 
same quantities of salt. He makes no reser- 
vation in any direction, draws no suspicious 
conclusions from his observations, not even 
from the establishment, by Wiley’s results, 
that the excretion of benzoic acid as hippuric 
acid in the urine is somewhat retarded. 
Long obtains very favorable results; he did, 
indeed, observe in his uneducated test persons 
(institute help, ete.) a slight disturbance at 
times (headache, vomiting, disturbance of 
sight, excitement) but always found a plaus- 
ible explanation for it and never feels forced 
to ascribe it to benzoic acid. Herter noticed, 
especially with large doses, a slight increase 
of indican in the urine, but no change in the 
alkylsulphurie acid, a decrease of the fecal 
bacteria which evolve gases and an increase of 
the cocci as compared with other bacteria in 
