THE METABOLISM OF THE BODY. 449 
when blood containing benzoic acid is mixed with fresh minced 
kidney it is transformed to hippuric acid. Hay contains a ben- 
zoic compound, so that it is not difficult to find a starting-point 
for the hippuric acid of the herbivora. In these instances it is 
assumed that glycin is added in the kidneys; but, as a matter 
of fact, this substance has not as yet been found anywhere in 
the body, though it is possible to conceive that, like peptone, 
it might be formed and disappear (be used) as fast as gen- 
erated. ; 
The above is one of the clearest cases favoring the view that 
the chemical processes of the body do really very much resem- 
ble those of the laboratory. But, considering the. difficulty as 
to glycin, and that the liver also can form hippuric acid under 
similar circumstances (those mentioned above), and that there 
are several laboratory methods for the synthesis of hippuric 
acid, it behooves us to be cautious even in this case, the chain 
-of facts being by no means complete. 
Of the origin of the allied bodies—xanthin, etc.—or their 
fate and purpose, we know very little. Their resemblance 
chemically to certain alkaloids in tea, coffee, etc., is suggestive. 
Are they natural stimulants ? 
THE STUDY OF THE METABOLIC PROCESSES BY OTHER 
METHODS. 
It will be abundantly evident that our attempts to follow 
the changes which the food undergoes from the time of its 
introduction into the blood until it is removed in altered form 
from the body has not been as yet attended with great success. 
It is possible to establish relations between the ingesta and the 
egesta, or the income and output which have a certain value. 
It is important, however, to remember that, when quantitive 
estimations have to be made, a small error in the data becomes 
a large error in the final estimate; one untrue assumption 
may vitiate completely all the conclusions. 
In discussing the subject we shall introduce a number of 
tables, but it will be remembered that the results obtained by 
one investigator differ from those obtained by another; and 
that in all of them there are some deviations from strict ac- 
curacy, so that the results must be regarded as only approxi- 
mately correct. It is, however, we think, better to examine 
such statistical tables of analyses, etc., than to rely on the 
mere verbal statement of certain results, as it leaves more 
29 
