F, J, F. SHAW 43 
concentrated sulphuric acid and titrated with a solution of potassium permanga- 
nate (1 grm. in 1,000 c.c.). The liquids from the control flasks required respect- 
ively an average of 0°85, 0°75 and 0°85 c.c. of the potassium permanganate 
solution before a permanent colouration was obtained. For the entire quid 
content of each control flask therefore approximately 3:2 ¢.c. of potassimum 
permanganate solution were required. In the case of the three infected flasks 
25 ¢.c, of the liquid required an average 2°3, 2°2 and 19 ¢.c. of potassium 
permanganate before a permanent coloration was obtained. For the entire 
liquid contents of an infected flask, therefore, an average of 84 c¢.c. of 
potassium permanganate solution was needed. Thus the liquid from the 
infected flasks had more organic matter in solution than that from the uninfected 
flasks, and this could only have come from the solution of the cellulose by the 
action of the fungus. | : 
The power of setting up a rot in cellulose tissues does not, however, explain 
the manner in which the fungus gains entry into the stem of the host. This can 
only result from ingress either at some break in the superficial tissues or from 
direct penetration of the cuticle. An infection in which a minute piece of agar 
culture was placed on the surface of the stem resulted in direct penetration 
of the epidermis and cuticle (Pl. IV, fig. 2) within 12 hours (Experiment IX). 
There exists, however, in this case the possibility that the presence of a small 
piece of agar in contact with the stem may cause a local softening of the cuticle, 
rendering the passage of the fungus more easy than under natural conditions, 
The following experiment was, therefore, carried out with the object of deciding 
whether the fungus could penetrate the uninjured cuticle. Three glass rods 
terminating in a small funnel-shaped expansion were placed upright in the 
soil next to three jute plants so that the edge of each funnel was within 1 m.m. 
of astem. A small piece of an agar culture of the fungus was then placed in 
the bottom of each funnel and the whole enclosed in a lamp chimney plugged 
with cotton wool. In the moist atmosphere the hyphe grew out over the edge 
of the funnel and made contact with the jute stem. A dark stain appeared on the 
Stem in the region of infection, but the plants showed no sign of wilting, and 
after a week the infections were opened and portions of the stem from the region 
of apparent infection were fixed for microscopic examination. Sections showed 
clearly that penetration of the stem had taken place, and scattered hyphee, causing 
a local disintegration of the tissues, could be seen in the cortex (PI. MM yiie..2). 
Unfortunately in this case cork formation had just commenced, and no cases could 
be seen in which hyphe were directly penetrating the cork layers from the outside. 
The only place at which entry was obvious being in the region of a lenticel. The 
question, therefore, whether the hyphze of D. Corchori can ord inarily penetrate the 
