40 
sodium hydrate. The reaction is indicated by the first appearance 
of a distinct rose-red color, best seen in a white porcelain dish before 
a well-lighted window. 
Example: 2.6 c. c. of the ^ l^aOH solution is found, necessary to 
neutralize 5 c. c. of the bouillon in order to obtain the desired acidity 
of 0.5 per cent, add 2.1 c. c. (2.6 — 0.5 = 2. 1), of the normal soda solu- 
tion (40 to 1,000) to each 100 c. c. If an acidity of 1.5 per cent is 
desired, add 1.1 per 100 c. c. (2.6 — 1.5 = 1. 1). 
Each lot of bouillon is numbered and a complete record kept of 
eyery stag-e of the operation upon the blank shown in fig. 6." 
The bouillon is distributed in modified Fernbach fiasks. presenting 
a large surface to the air for pellicle growth. Seyen hundred and fifty 
c. c. of the bouillon is placed in each fiask. 
The freshly prepared media is inoculated upon the surface from a 
twenty-four-hour-old culture by means of a platinum spoon. 
The fiasks are incubated for seyen days at 37.5° C. 
They are then remoyed and examined for purity by means of coyer 
slips. All atypical growths are discarded. 
The final reaction of the culture when taken from the incubator is 
taken for each fiask. This usually yaries from 0.6 to 0.8 per cent. 
Very acid results, such as 1.5 or oyer, would, according to Smith's 
work, indicate the absence of a strong poison and may be discarded 
without further testing upon guinea pigs. 
The bouillon is now filtered through a pear-shaped, unglazed porce- 
lain filter or a Berkefeld candle by means of a yacuum. The arrange- 
ment for filtering the yirus is shown on page 38. This diagram also 
illustrates the character of the fiask used to store the toxine in bulk. 
It is yery conyenient to draw ofi' small amounts from time to time by 
means of the siphon and Maasen nozzle without danger of contamina- 
tion. The especially strong and otherwise suitable toxines are bottled 
from these fiasks without the addition of an}^ preseryatiye. 
BottUng and lyi'e^eri'ing . — The toxine is kept in bulk in the 2 liter 
bottle shaped flasks described aboye, and later diyided into small 
ground-glass stoppered bottles holding 5 and 10 c. c. each. 
The small bottles are filled to the neck with the toxine and the 
stopper inserted so that the air is all displaced and the fluid completely 
fills the yial. 
A minute quantity of sterilized liquid petrolatum is touched to the 
ground glass of the stopper so as to preyen t the two surfaces of glass 
am indebted to Dr. Herbert D. Pease, director of the antitoxin laboratory, 
New York State department of health, for suggestions on the method of keeping 
records on the card system shown in figs. 4 and 6. Doctor Pease ver^" kindly let 
me have blanks used by liim, which were modified to suit our purposes. 
