450 
Tliis initial power of milk to restrain or destroy bacteria is exceed- 
ingly feeble and variable. The germicidal properties of milk have 
been much misunderstood, especially by dairymen, some of whom 
argue that advantage may be taken of this property for the preser- 
vation of milk without the use of ice. 
When we stop to consider that bacteria frequently enter the im- 
perfectly closed orifice of the teat and grow well in the milk contained 
in the milk cisterns, and that they often invade the finer tubules of 
the gland structure itself where the milk is being freshly secreted, 
we must be convinced that the “germicidal” power of milk must be 
exceedingly feeble, if it exist at all. 
This property varies with the milk of different animals, and also 
with the milk of the same animal at different times. 
There is evidence to show that the restraining action of fresh raw 
milk upon the growth of bacteria varies with the bacterial species, and 
when we inquire into the causes of the phenomenon we find that this 
is what we might expect. 
When micro-organisms are transferred to a strange medium they 
sometimes hesitate, until they become sufficiently accustomed to 
the new surroundings, before they begin to grow and multiply. 
Our experiments show that this is by no means always the case and 
can not account for the facts now under consideration. 
We know that the serum -of milk may contain “antibodies” in 
appreciable and variable quantity similar to those found in the 
blood. For instance, diphtheria, tetanus, and other antitoxins 
have been demonstrated in the milk of immunized animals. We 
might also expect slight and variable quantities of the agglutinating, 
bactericidal, and bacteriolytic substances present in blood serum 
to pass into the milk serum. Agglutinins in milk would apparently 
diminish the number of bacteria contained therein when estimated 
by the number of colonies that develop on agar plates. Such a 
decrease would be apparent, not real. Microscopic examination of 
the bacteria in milk made at once after milking, and again in eight 
hours, demonstrates that such agglutination actually takes place. 
This is confirmed by our other technique. 
We know that milk always contains large numbers of leucocytes — 
many of them of the polymorphonuclear variety. These are known 
to be active phagocytes, and we might assume that they are active in 
milk for a short time after it is drawn. In fact we have found that 
. some of the leucocytes actually contain more bacteria after eight 
hours than when freshly drawn. 
If phagocytosis plays a part in the diminution in the number of 
bacteria in milk, we must assume that the milk serum must also 
