AND ITS BEARINGS ON PATHOLOGY 359 
perature which is attained proves adequate for the destruction of organisms 
in the liquid. All we have to do then, supposing the interior of the flask per- 
fectly pure, is to introduce the liquid into the lower part of the flask with a view 
to immersion in a saucepan of boiling water: but then the most scrupulous 
care must be taken that no portion of the liquid so introduced shall come into 
contact with the upper parts of the interior of the flask; for any particles 
deposited there would fail to be acted on by the full heat of the water in the 
saucepan. Now the mode in which I filled the flasks in my earlier experiments 
was this. Having provided myself with a rag soaked with a strong solution of 
carbolic acid (1 to 20), I washed with it the exterior of a long funnel, and, wrap- 
ping the rag round the lower end of the funnel in sufficient mass to cover the 
mouth of the flask, I substituted the mass of antiseptic material for the cotton 
cap with which the mouth was previously covered, pushed down the funnel 
through the rag, poured in the liquid and withdrew the funnel, taking good care 
that the drop at its extremity did not touch the interior of the flask. A freshly 
prepared carbolized cotton cap was then applied to the mouth of the flask at 
the moment of withdrawal of the antiseptic rag, and the flask was immersed 
in the saucepan.' 
If I proceeded in this way with Pasteur’s solution or with turnip infusion, 
I always had success; but when I did the same thing with milk, time after 
time, to my great disappointment, I failed altogether. What was the explana- 
tion of the failure ? Some persons might have said, ‘Oh! the explanation is 
very easy to find. There are in milk complex organic molecules which, though 
as yet mere chemical substances, are, we imagine, ready to develop into living 
beings, and it is this complex constitution of the milk that makes you fail ; 
whereas, your Pasteur’s solution is a comparatively simple material, and turnip 
infusion may, for aught you know, have its molecules more simply constituted 
than milk.’ I felt sure that this was not the true explanation, but that there 
must have been some defect in my method of procedure. It may perhaps have 
occurred to some of you what that defect was ; it was this, that if we pour any 
liquid through a funnel, we invariably have air pass along with it. Air-bubbles 
consequently formed upon the surface of the liquid, and those bubbles bursting 
dispersed their dust in the air within the flask ; so that it might well happen 
* Means must be taken to prevent the flask from being floated up by the water in which it is 
immersed. This I have done by tying the flask into a vessel, such as a soap dish, containing shot. The 
flask so weighted is placed in the empty saucepan, warm water is poured in to a level considerably above 
that of the liquid in the flask, and brought to boil as quickly as possible and kept simmering for an hour. 
A piece of thin macintosh cloth, cut so as to adapt itself to the flask and cover the top of the saucepan, 
has the double advantage of preventing the water from boiling quickly away and avoiding moistening 
of the cotton about the flask by drops of condensed vapour. 
