76 
experimental way we filtered malarial blood in order to determine 
whether there might be forms of the malarial parasite which are even 
smaller than this spore. 
We know from the work of Novy that a trypanosome ( Trypano¬ 
soma Lewisi ), which is a colossal organism when compared with a 
malarial spore, has forms which are so minute that they pass a 
Berkefeld filter, for he has succeeded not only in artificially culti¬ 
vating the adult trypanosome parasite, but in infecting animals with 
the filtrate from these cultures. We also know from the recent work 
of Schaudinn ° that some of the animal parasites ( Spirochaeta ), 
multiply by reducing division; that is, each time cleavage takes 
place the organism is reduced in size^, and this process continues until 
the divided forms become too small to be seen as individuals and can 
be made out only as clusters. 
We therefore reasoned that if the malarial parasite has an ultra- 
microscopic form minute enough to pass the pores of a filter, it 
would encourage us very much to look for a visible form of the 
yellow fever organism in the blood and tissues of man and the mos¬ 
quito by the aid of technique that had not previously been employed. 
Our filtration experiments with malarial blood resulted negatively 
so far as demonstrating the presence of a minute or ultramicroscopic 
form of this parasite was concerned, but there developed unexpectedly 
what appears to be a demonstration of the malarial toxin. We pro¬ 
duced a definite paroxysm by the inoculation of blood serum freed 
of the malarial parasites by filtration; and it is reasonable to 
suppose that the same substance circulating in the blood, which caused 
the chill, fever, and sweat in one man, caused a precisely similar chain 
of symptoms in the other two into whom this serum was transferred. 
We found that if the blood is drawn after the height of the parox¬ 
ysm and while the fever is declining this poison is not manifest; but 
if the blood is taken during the chill and while the temperature is 
rising, it is present. 
If this poison is the toxin causing a malarial paroxysm it is remark¬ 
able that it should be present in the blood serum in such a considerable 
quantity and disappear so very rapidly. Still, the clinical symptoms 
of the disease would indicate the sudden production of a large quan¬ 
tity of toxin and its rapid elimination, neutralization, or destruction. 
So far as we know, this is the first time that a poison has been demon¬ 
strated which is capable of reproducing the symptoms of a disease 
due to an animal parasite of microscopic size. 
It would be folly from a few observations to claim that we have dis¬ 
covered the malarial toxin. The only conclusion justified is that we 
have demonstrated the existence of some poison in the blood which is 
«Loc. cit. 
