106 REporRT OF DEPARTMENT OF BACTERIOLOGY OF THE 
given can be duplicated in any laboratory where care is given 
to approximating standard conditions of work. 
One of the important items which should never be disre- 
garded in comparative work of this kind is the vigor of cul- 
ture. Especially when they have been long subjected to arti- 
ficial conditions in ‘a laboratory it is necessary to put the 
cultures through a revivifying process, a series consisting of 
transfers of young cultures through bouillon, gelatin and agar 
having been used with these germs. Comparative cultures 
were rarely made in duplicate but almost uniformly in tripli- 
cate. When cultures long in stock were tested without pre- 
viously revivifying, discordant results from the three simul- 
taneous cultures were occasionally obtained, while such com- 
parisons made after the stock culture had been revivified 
practically always gave more accordant results. 
In separating these cultures the main difference is in their 
ability to ferment the different sugars. Accordingly it was 
absolutely necessary for accurate results that the broth which 
was used as the basis of these tests should be free from muscle 
sugar. The absence of this sugar was determined by testing 
each lot of broth with B. coli, and when found to be present 
the sugar was destroyed by the growth of this organism. By 
using meat extract it was found possible to prepare bouillon 
free from this sugar, and all of tke broth used at the New 
York laboratory was in this way made free of muscle sugar. 
METHODS OF CLASSIFICATION, 
Those familiar with such work will appreciate the difficulty 
encountered in comparing determinations upon so many or- 
ganisms scattered over a term of years and made in different 
laboratories. It is rare that two sub-cultures from the same 
original will give identical quantitative and qualitative re- 
sults on a variety of media when tested side by side and when 
considering the large number of cultures tested at different 
times and places the slight variations are bewildering. Some 
system of classification is absolutely essential to progress. 
