( 783 ) 
been able to fix these differences in a series of exact quantitative 
complement-titrations. From these it appeared among others that a 
few minutes after intravenous injection of 0.2 cm* of horse-serum 
the complement-quantity of the testanimal-serum may have decreased 
nearly as strongly as at intraperitoneal injection of 3 cm* after 
half an hour. 
My investigations and those of Friedberger, accordingly, do not 
contradict each other; they complete each other. Therefore I cannot 
see in Friedberger’s results anything but an essential corroboration 
of my observations. The same thing holds good for the haemolysis, 
which in the anaphylactic shock shows itself in the test-animals. 
Friedberger corroborates also this fact, just as Pokls *) does. My 
contention, therefore, that Besredka with his exclusivistic opinion 
that it is only the elements of the central nervous system which 
are to be brought to hypersensibility, is wrong, finds satisfactory 
support in this haemolysis which has been proved in more than 
one hand. 
Now as to the alexine-fixation of the anaphylactic serum with its 
antigen in vitro, I think I can maintain my negative results over- 
against Nicolle and Abt. In many series with mixtures of falling 
quantities of horse-serum with rising quantities of anaphylactic 
gnineapig-serum I could not observe anywhere a specific retardment 
of the haemolysis. Not even though I stuck accurately to the quan¬ 
titative proportions, as Nicolle and Abt have mentioned. In the 
meantime, thanks to the necessary controlling experiments, I came 
to the following conclusions. Even normal guinea-pig serum (inactivated) 
often retard the haemolysis, quite independent of the presence of 
horse-serum; nay, we sometimes meet a normal serum which has 
a stronger fixing power than an anaphylactic serum which has also 
been examined 2 ). Therefore I abide by my former contention that 
there is here no question of a specific complement-fixation in the 
test-tube. This incongruity of the alexine-fixation in vitro and in 
vivo, to which I drew attention already a year ago, was the other 
day corroborated with certainty by Michaelis in the meeting of the 
“Physiologische Gesellschaft” at Berlin (21 January 1910). Also 
Friedberger seems after all to share this opinion (Zeitschr. f. Imro. 
forsch Bd IV H 5). It now seems to me that the labile state of 
physiological equilibrium, in which the hypersensitive organism finds 
itself, is biologically characterized by the incongruity referred to 
just now. 
iJ^Handelingen v. h. Nederl. Natuur- en Geneesk. Congr. te Btrecbt, April 1909. 
») Ampler details I hope to publish elsewhere. 
