I(i8 RED ROT OF SUGARCANE. 



peiimoiit was repeated on 4 shoots in the hiboratory, kept under 

 bell jars to prevent the inoculated spot from drying rapidly. These 

 also failed. In a third series, similar to the last, 6 inoculations were 

 made without result. A month later, however, out of 12 similar 

 inoculations, 5 succeeded. When the midrib was wounded, much 

 better results were obtained. In the first trial 5 leaves were inocu- 

 lated in plants growing in tubs in the laboratory, the upper epider- 

 mis being first removed by scraping with a sterile knife and the spot, 

 after inoculation, being covered with damp sterile cotton wool to 

 keep moist. All succeeded well, the characteristic red discoloration 

 being well developed by the 9th day. In another series, 4 inoculations 

 were made after injuring the epidermis of the midrib by touching 

 it with a hot knife blade for 2 or 3 seconds. All took the infection 

 severely. Four more were inoculated on another occasion, after 

 scraping off the epidermis, and again all took. 



Experiments were next made to test the parasitism of the stem 

 form of Colletotrichum, falcatum on the leaves. In the first experi- 

 ment, spores from a pure culture were sown in drops of water on the 

 upper surface of the uninjured midrib of canes growing in a tub 

 in the laboratory. Of 15 inoculations, none succeeded. The ex- 

 periment was again tried and of 7 inoculations, all succeeded, in 

 a third series, out of 13 inoculations. 5 succeeded. When the midrib 

 was wounded before inoculation the results were as follows. In 

 the first trial 5 inoculations were mafic after scra])ing off the ei)i- 

 dermis. The inoculated spot was covered with a pad of (himp 

 sterile cotton wool. None succeeded, the fungus growing by choice 

 into the cotton wool. A similar experiment at a later date was 

 made on 3 shoots standing in water, no cotton wool heini; used l»ui 

 tli(» siioots being covered by bell jais. All succeeded. In a lliird 

 experiment, the ej)idermis was injui'cd i)V loudiinu- willi .i hoi knife 

 blade and all of 4 inoculations succeeded. 



Though both leaf and stem forms, are capable of penetrating 

 uninjured leaves, infection occurs much more readily when the 

 leaf is wounded. The microscopic details of penetration will be 

 described below. In nature, it has been observed tliat CoUetotri' 



