186 REPORT OF DEPARTMENT OF BACTERIOLOGY OF THE 
suffice here to say that their comparative studies as there re- 
corded lead them to conclude (p. 138) that these forty strains — 
probably constitute only one somewhat variable species. The 
following comparisons as to enzym production were, however, 
completed before their decision was reached and we believe 
they were worth recording partly as contributing further evi- 
dence as to the general likeness of these strains and partly 
as emphasizing the minor variations which we believe must 
always be expected to occur with different bacterial strains, 
even of the same (so-called) species. 
The comparisons as to enzym production have been made 
by testing both the living cultures and the alcoholic precipi- 
tates, following the general methods outlined earlier in this 
article. Inasmuch as the alcoholic precipitate yields the enzym 
in a more concentrated form, and one in which it can be pre- 
served indefinitely for comparative trials, this method has been 
chiefly relied upon. In all cases much care has been exercised 
to insure uniformity in the several series as to medium and 
cultural conditions. The trials of cytolytic activity were made 
on razor sections of vegetables, carefully selected for uni-— 
formity, and the trials repeated until convincing evidence was 
. obtained as to relative activity.. Inasmuch as it was not prac- 
ticable, for obvious reasons, to make close comparison of more 
than a limited number of these at a time, they were handled 
in series of six or eight in a group. Since there was some 
variation in the vegetable upon which the test was made in 
different cases it is to be noted that B. carotovorus of the 
original strain, or some other organism whose activity. was 
well known, was included in each series. In all cases except 
Pember’s organism R. cytolytic action occurred. In order to 
have a basis for comparison the organisms of each series were 
grouped in order of cytolytic activity into three classes, as 
follows: 
