
44 STORRS AGRICULTURAL EXPERIMENT STATION. 
this case. Tables 3 and 4 and Tables 5 and 6 represent the 
same milk, and this difference in results shows the difficulty of 
obtaining uniform analyses. In both there is a small differ- 
ence in favor of straining. But the difference is not very large, 
particularly in the second case, and cannot be regarded as very 
significant. The conclusion to be drawn from these tables, 
taken with Table 2, is that straining milk through cheese cloth 
does have a slight ettoet of removing the bacteria. 
It is something of a surprise that no larger benefit is shown, 
for, as has been indicated by previous experiments,(the amount 
of dirt which is removed by the straining is about 4o per cent) 
It would have been expected, therefore, that the removal of so 
much dirt would remove a larger proportion of the bacteria 
than the experiments seem to indicate. Furthermore, it is 
seen that in some of the samples the strained milk contained 
more bacteria than did the same milk before straining. This 
is shown, for example, upon December 28, January 14, and 
January 18. The differences in these cases were not very 
great, and possibly not greater than the limits of error would 
permit, but certainly in these cases the strained milk showed 
more bacteria in the plates than the unstrained milk did. This 
result, however, 1s not very surprising, since similar results 
have been obtained by others. It has been found by Weil 
(Milchztg 1901, p. 739) that filtering milk through filters fre- 
quently increases the apparent number of bacteria present. 
This author, however, concluded that the cause of the apparent 
increase was that the filter was not sterilized and contained 
bacteria which were washed through by the filtering.’ This does 
not apply to our experiments, inasmuch as the filter, being 
simply cheese cloth, was thoroughly sterilized before each ex- 
periment. ‘The probable explanation of the result is the condi- 
tion of the bacteria existing in the milk. A microscopical 
examination shows that the bacteria of milk have a tendency 
to group in little clumps. Each of these clumps would, asa 
rule, result in showing a single colony upon the gelatin plate, 
with no indication of the number of bacteria present in the 
group. When the milk is forced through the strainer under 
ate 
the influence of the stream of milk from the pail, these clumps 3 
are more, or less broken up. ‘The result is a larger count of — 
colonies, although, of course, there is no increase in bacteria. — 
. | 
<.: 
*: 
. 
. 
: 

