300 
ANNALES DE L’INSTITUT PASTEUR 
was testée! by inoculations on the shaven skin of the rabbit 
according to the method of Galmette and Guérin [3]. We 
found that the chloroform method, when used for a sufficient 
length of lime to purify, frequently greatly weakened the 
strength of the viras. 
With ether, our results were not as good. We were unable 
to obtain a pure virus that was not almost completely inac- 
tivated. 
We bave devised the following means for purifying the 
virus without greatly weakening its activity. 
The glycerinated, carbolized calf viras, as prepared by the 
Board of Health of New York City (1), is pat in stérile collodion 
sacs and dialyzed (method of Poor and Steinhardt) [4]. The 
dialysis can be carried ont in either distilled waler or physio- 
logical sait solution. If the virus lias been subjected for some 
weeks to the action of the glycerin and carbolic acid at ice-box 
température, most of the contaminating organisms hâve been 
killed. Our method for cultural experiments lias then been, 
to put severalcubic centimeters ofthe dialyzed virus on poured 
agar plates and incubate 1 to 2 days, lhen from between the 
bacterial colonies the uncontaminated virus is eut out, and 
these pièces are used for inoculating other media. 
We hâve found that if fresh pulp is prepared in the usual 
way, and left in the ice-box a week or two, dialyzed and to this 
dialyzed virus, glycerin and carbolic acid (2) are again added, 
placed in the ice-box again for a short time, and re-dialyzed, 
a pure active virus is usually obtained, but it is considerably 
diluted (about 1 to 40). It, however, gives a conlluent éruption 
when inoculated on the back of a rabbit. The exact number 
of days of contact with the disinfectants necessary to purify 
the virus differs with the nature of lhe contaminating orga¬ 
nisms and must be determined in each instance. From a few 
preliminary experiments, it seems thaï the glycerin is not 
essential. In wliich case, by only using the carbolic acid, a 
(1) The method used is one part of calf pulp emulsifîed very finely with 
four parts of a solution of glycerin 50 p. 100; carbolic acid, 1 p. 100, and 
distilled water, 49 p. 100. The émulsion is then passed through a very fine 
sieve several times. 
(2) It is usually better to add carbolic acid only in the proportion of 
1/2 p. 100 in this second sterilization. 
