362 Bancroft.—Researches on the 
in the one a sterilized needle was used, in the other they were mixed with 
sterilized water in a spraying-apparatus which had been previously washed 
out with absolute alcohol and allowed to dry. Six young Cabbage plants 
were selected as being healthy and intact and were arranged in two lots of 
three each. In one experiment the leaves of three plants were sprayed 
with the conidia ; in the other experiment the conidia were transferred 
to definite spots on the leaves of the other three plants and the spots were 
moistened with distilled water. Each lot of three plants was covered with 
a bell-jar of which the inner surface had been cleaned with absolute alcohol. 
In from four to five days the leaves thus treated were found to have per¬ 
forations in them ; the leaves of those plants which had been sprayed 
showed numerous perforations, while in the others the perforations were 
limited to the spots on which the conidia had been placed. The per¬ 
forations increased in size and soon became large holes. A microscopic 
examination showed that the conidiophores of the fungus were arising from 
the tissues of the leaf surrounding the holes and were producing conidia 
abundantly. The symptoms developed by these plants which had been 
artificially infected were similar to those observed naturally on the infected 
plants. All of the six plants yielded the same results, while a control in 
the form of a young Cabbage plant kept under a sterilized bell-jar and 
to which no conidia had been transferred showed no signs of infection. 
Another series of experiments was made for the purpose of determining 
with what degree of facility the fungus could be transferred from one 
host to another. For this purpose excised leaves were used; the leaves 
were placed on filter papers in Petri dishes and the filter papers were 
moistened with distilled water, the Petri dishes and filter papers having 
been previously sterilized by heating in an air-oven at ioo°C. for five 
hours each day on two successive days. Conidia were taken from an infected 
leaf of one species and were placed on definite spots of a leaf of another 
species in the Petri dish ; the spots were moistened with distilled water. The 
experiments showed that conidia from one species would grow on another 
with apparent facility. The fungus on Cabbage could be transferred to 
Cat alp a, that on Melon to Cabbage and vice versa, that on Catalpa to 
Mallow and vice versa, and that on Catalpa to Melon. It is concluded that 
the form Hormodejidron has not adapted itself closely to any one host, 
at least in the case of the species experimented with. Further, it may be 
mentioned that the disease caused by this form would be particularly 
difficult to exterminate, since the fungus attacks many species of plants and 
can apparently pass from one species to another, and this apart from its 
connexion with Cladosporium herbarnm, which is so common everywhere 
on the dead leaves of plants. 
