404 Brown.— Studies in the Physiology of Parasitism. III. 
A. On the question of the presence of a crystalloidal toxin in the 
infection drop. 
The following observations have been made : 
1. In many cases the first signs of discoloration of the tissue under- 
lying the infection drop appear along the margins of the latter. This effect 
is best seen when the drops do not spread over a large surface of the leaf. 
Microscopic observation shows this effect to be correlated with more rapid 
germination of the spores along the margin. The fact that such a variation 
in the time of attack can be obtained within the limits of a drop so small as 
o 03 of a c.c. cannot be explained on the basis of the de Bary hypothesis. 
3. If a drop containing spores be placed on the leaf, the spores 
allowed to settle into contact with the epidermis (a process which requires 
about one hour), and the drop be now slightly displaced so as to include a 
new portion of the leaf surface, the discoloration as it first appears, and for 
a considerable time afterwards, is seen to outline strictly the area originally 
occupied by the drop ; in other words, the discoloration effect is strictly 
localized to the immediate neighbourhood of the germinating spores. This 
result is again directly contrary to the hypothesis of de Bary. 
3. With the same number of spores, the size of the drop can be varied 
within wide limits without appreciably affecting the time of appearance of 
discoloration ; and similarly with drops of the same medium and of the same 
size the number of spores can be varied within wide limits without produc- 
ing any marked variation in the time of appearance of discoloration. Thus 
sowings on rose petals in which the concentrations of spore suspension were 
in the ratio 10 : 1 gave 10-13 hours and 12-14 hours respectively as the 
times at which discoloration could be first detected by the naked eye. 
If the discoloration effect were due to the accumulation of a toxic substance 
in the drop, the effective concentration should be reached much more rapidly 
in the case of the denser sowing (the concentration of nutrient in the drop 
being sufficient to allow of vigorous germination), and the time of appear- 
ance of discoloration should therefore be correspondingly accelerated. 
Nevertheless we find that the shortening of the period between sowing and 
the appearance of discoloration with increasing concentration of spores 
is very slight. Bearing in mind the fact that discoloration in its earliest 
stages is not apparent to the naked eye, and further that when once started, 
the degree of discoloration would be proportional roughly to the number of 
spores present, we can readily conceive that the slight time difference above 
noted is explicable in this manner. 
4. When infection drops, in which discoloration was just becoming 
visible, were collected, cleared of spores by centrifuging, and tested on 
the surface of the most sensitive petals, no toxic action whatever could 
be demonstrated. 
