New YorK AGRICULTURAL EXPERIMENT STATION. 65 
This card consists of three essential parts: (1) A group number 
(see p. 64) as suggested by Gage, which records the more im- 
portant facts regarding an organism and provides a means of ar- 
ranging the records so that duplicates can be found readily; (2) a 
brief characterization (see page 116) which may be filled in by 
means of + or — as suggested by Fuller and Johnson and which 
serves as a further means of comparing germs with the same group 
number; and (3) a provision for tersely recording the detailed 
features of cultures upon the common media as suggested by 
ester 2 
The significance of each figure in the group number as endorsed 
by the Society of American Bacteriologists at its annual meeting in 
December, 1907, is shown by the key given on the opposite page. 
Our records of the cheese flora were begun with Conn’s modifica- 
tion of Fuller and Johnson’s table but were largely entered upon a 
card prepared by Conn as a modification of Gage’s card. This card 
furnished practically all of the data called for by the Society card of 
1906 aside from the detailed features of the culture growths. As 
soon as the Society cards were available in 1906 the work was con- 
tinued with them and the records transferred from the other cards. 
The Society card of 1907 appeared too recently to be utilized in this 
study. However it differed from the former card mainly in calling 
for additional items and in an extension of the group number as in- 
dicated above. While the lack of some of the data for this Society 
card is regrettable sufficient are at hand to characterize the cultures 
to an unusual extent. 
As has been explained, pure cultures were obtained from two or 
more colonies representing each type present on the plates at each 
analysis. The purity of these cultures was first tested and their 
group number determined except that relating to lactose and saccha- 
rose. At this point cultures which gave every appearance of being 
duplicates were rejected and further studies confined to a single 
_representative of each type. These selected cultures were tested 
with regard to practically all of the points called for by the classifi- 
cation card. 
In the presentation of our results it has seemed best to follow the 
Society card in designating the types of germs present. Accord- 
ingly, in the discussion of the individual. cheeses the germs will be 
referred to by their group number and additional data regarding them 
> Chester, F. D. A manual of determinative bacteriology. New York: 
Macmillan. gol. 
