82 
The institute for testino' the serum does not determine in its official 
tests the exact antitoxic value of sera sent it, but determines by 
means of the L-j- dose of the test poison whether the serum contains 
at least the value claimed for it by the manufacturer. The testing 
for the sterility of the sera is done in accordance with well known 
bacteriologic rules, by making aerobic and anaerobic agar and bouillon 
cultures. 
In order to determine the amount of disinfectants added to the serum 
white mice are used. They are inoculated subcutaneously with 0.5 
c. c. of the serum. Solutions of carbolic acid containing more than 
0.5 per cent unerringh^ cause, in this amount, the death of the animal. 
Finally, guinea pigs are inoculated with 10 c. c. of the serum, and 
they must remain well. 
If the serum has complied with all the official tests, it is considered 
reliable, and, after due notification of the fact, it is immediately 
bottled in the presence of the local official. It is the dut}" of this officer 
to seal each bottle with his lead seal. His control is a guarantv of the 
procedure of this part of the process, and he is responsible for the 
proper labeling of the bottles that are sold on the market. 
A further control is carried out in this way: Of every test number 
in the testing institute six and twenty-four months afteiuvards the 
serum is again tested to determine whether it contains the stated value 
and has not lost strength. Also, a number of hospitals are commis- 
sioned, from time to time, as they use the bottles of serum, to test 
them for sterility. Sera that lose streitgth or show bacterial contami- 
nation are withdrawn from the market. 
From experience gained in his work, Ehrlich suggested the following 
amendments to the instructions for the testing of diphtheria antitoxin, 
which were officially promidgated by the minister of education, etc., 
on March 29, 1897: 
1. A powdered serum protected against moisture and oxygen serves 
as the measure bv which the strengths of other serums are determined. 
This powdered serum is preserved in exactl}^ weighed quantities in 
special vacuum tubes. At this time in the Institut the tubes contain 
2 grams each of the diy powdered antitoxin, ^rhich have the value of 
1,700 times normal strength (1,700 facher Starke). 
2. The serum is dissolved in a mixture of equal parts of a 10 per 
cent solution of sodium chlorid and glycerin, this mixture favoring 
the stability of the antitoxin. Eveiw three months a tube is opened 
and a new" solution made. At present the contents of the tubes pre- 
served in the Institut are dissolved in 200 c. c. of the above-mentioned 
mixture, wffiich gives a test serum containing 17 units in each cubic 
centimeter. 
3. The present test dose of toxin is determined by means of an 
