ANALYSIS OF CONTINENTAL JOURNALS. 
763 
therefore, completely fails to clearly demonstrate what those 
are, among the elements of the blood, which possess the 
infecting aptitude. 
This is, definitely, what this attempt to separate the active 
elements of anthrax by filtration of the blood through the 
placenta of pregnant animals is reduced to. 
I have the same objections to make with regard to ano¬ 
ther tentative of Davaine’s to isolate the anthrax bacteria. 
In this experiment the diseased blood is mixed with a large 
quantity of water, and allowed to stand for some time in a 
test-tube; the bacteria fall to the bottom, and the fluid above 
them is removed. Inoculation with the deposit at the bottom 
of the vessel produces anthrax, but that with the super¬ 
natant liquid does not. It will be remarked that this experi¬ 
ment can have no other signification than the preceding; for 
if the inoculable deposit contained anthrax bacteria, it also 
contained the other solid elements of the blood, except the 
haemoglobin which the water readily dissolves. But again 
you cannot,fail to observe that there is little more to be done, 
perhaps, in this very remarkable experiment than to cause 
the total destruction of all the normal elements of the blood, 
leaving the bacteria intact. The direct demonstration of the 
action of the^e bodies would then be placed beyond all dis¬ 
pute. We shall see what the future will do for us in this 
respect. . 
From these examples you will perceive how easy it is to be 
led to accept too readily the scientific proof of the determina¬ 
tions belonging to the order of facts under discussion. I have 
designedly borrowed these instances from a work which is 
worthy of esteem among those which treat of subjects ana¬ 
logous to own ; and in doing so I shall have forewarned you 
against the danger to which we are exposed if we neglect to 
be careful in the choice and criticism of the proofs we have 
to make in order to arrive at the determination of the virulent 
agents. 
Every means of demonstration should be had recourse to. 
We shall be aided by observation, but it is especially to the 
experimental method that we must address ourselves. No 
determination should be received if it cannot be based on a 
direct and rigorous demonstration. 
Our first care shall be to determine in what physical con¬ 
dition the active elements of virulent humours are to be found ; 
as it is only after having assured ourselves on this primary 
point, the most important of all, that we may inquire if it is 
possible to arrive at a more special determination of the viru¬ 
lent agent. 
