( 786 ) 
has now, besides its sensitizing power, retained another quality: it 
has remained vaccinating, i.e. that sensitized guinea-pigs, treated 
with it, are already within a few hours immune against an other- 
wise mortaliy large dose of horse-serum not treated beforehand. Also 
fresh serum itself, in the usual toxical dose (4—5 cM’) has vaccinating 
action, but this is a two-edged sword, because in the majority of 
cases the animals are killed thereby. BESREDKA, *) however, was able 
to prove, that already with a very small dose of serum (e.g. 
0.05 cM®*), injected into the abdomen, a sensitized animal could be 
vaccinated without showing symptoms of disease. Now this fact 
— it seems to me — may be said to be ona par with the vaccinating 
action that I found of larger doses of serum rendered atoxical, and 
at the same time it suggests the following explanation: for the 
dissensitizing (vaccinating) of an animal rendered sensitized a definite, 
very small quantity is wanted of a substance found in fresh horse- 
serum. The latter, administered at once and in larger quantity, acts 
moreover toxically (sudden dissensitization, shock); a small dose of 
serum, however, \BESREDKA), or — as we have seen — a larger 
quantity of serum from which the greater part of this substance has 
been eliminated, contain still enough of it to dissensitize gradually 
and without toxical by-actions. In passing it may be remarked here 
that for a probable practical application in the re-injection of thera- 
peutic sera only such a method is practicable, in which is obtained 
an atoxical and dissensitizing serum, which at the same time has 
retained a sufficient quantity of antitoxine-units 
Further I have asked myself whether it would be possible, in a 
simple way to get a closer determination of the generally chemical 
nature of the antigen of anaphylaxis (by this I mean the toxical and 
vaccinating matter which I consider identical with the sensitizing 
principle). Does this material belong to the proteins or to the lipoids? 
1 have tried to answer this question by means of a method which 
by Borprr and the present writer in a still unfinished investigation 
about the causes of specificity is applied in the splitting of antigens 
into a soluble part and another part insoluble in absolute methyl- 
alcohol free from aceton. Now we take 5 cM’ of horse-serum (conse- 
quently a toxical dose for sensitized guinea pigs), it is dried, rubbed 
into powder and repeatedly extracted with methyl-aleohol. The 
alcoholic extract is then, by its being taken up again into a small 
quantity of aleohol — in which the salts do not dissolve — almost 
l) G. R. Soc. de Biol, 23 Jan. 1909, 
fc 
