338 BOTANICAL GAZETTE [NOVEMBER 
chemotropism) ; while still others produced no effect. Sugars were 
especially attractive to the molds. 
The most important methods employed by Mivosui were as 
follows: (x) the substances to be tested were injected in solution 
into leaves, which were then sown with spores; or (2) the substances 
were incorporated in a gelatin layer, on which was laid a perforated 
mica plate or membrane, the spores being sown above, either on the 
free surface of the plate or membrane, or in a superimposed layer 
of medium lacking the substance to be tested; or (3) fine glass 
capillary tubes holding the substance to be tested were introduced 
into a preparation containing germ tubes of the fungus growing in 
water or in a very dilute nutrient solution. 
Where the germ tubes showed a marked turning toward the 
stomata, perforations, or open ends of the capillary tubes, as the 
case might be, and thus toward the diffusion centers of the chemical 
substances, the turnings were interpreted as instances of positive 
chemotropism. Moreover, there appeared to-be an optimum con- 
centration, beyond which such substances exerted a less and less 
attractive and at length a repellent effect. On the other hand, 
below the optimum, with gradually decreasing concentrations, the 
attractive effect also gradually decreased, ultimately to zero. 
Acids, alkalis, and some salts always produced a repellent effect, 
however, even in weak solutions. 
Perhaps Mryosut’s strongest point in support of his view was 
his statement that when no special chemical substances were 
offered, as, for example, when leaves were injected with pure water 
only, no turning resulted. 
When we survey the investigations of FuLron (5) and CLARK 
(3), we find that, taken in combination, they completely contra- 
dict these results. CLARK injected leaves with various concen- 
trations of copper and cobalt salts, etc., and found that in this 
case the germ tubes near the stomata curved toward and grew 
directly into them; but he obtained a similar result with leaves 
injected with pure water only. In experiments in which he used 
perforated mica plates, he found that the germ tubes near the per- 
foration always “grew toward the opening if it communicated with a 
layer in which no spores had been placed.’ Although he did not 
