250 E. J. BTTTLER. 



successive leaf-sheaths, it is easy to find spots in the innermost lay- 

 ers where the parasite has only penetrated a portion of the sheath 

 thickness, the spot being visible on the outside but not on the inside 

 of the sheath. Sections through these show the hypha? passing 

 through the stomata. Some of these appear to be hyphse of entry, 

 though the fact is extremely difficult to prove. Others are hyphae 

 of exit, as is shown by branching taking place just outside the 

 stoma. Hence it is evident that the parasite, after invading a 

 sheath from the outside, may send back branches through the 

 stomata to the outer surface, and it is this fact which makes it 

 difficult to be certain that entry occurs along the same road. In a 

 few cases the mode of entry or exit shown in plate V, fig. llr/, where a 

 hypha engages in the thickness of the cuticle and passes along the 

 vertical walls of the epidermal cells, was seen, but whether such 

 hyphse were going in or coming out could not be decided. The rarity 

 of these cases, compared with those in which hyphee pass through 

 the stomata, makes it very probable that the latter is the ordinary 

 mode both of entry and of exit. 



When the first infection is from zoospores no evidence is avail- 

 able as to the route followed. As already mentioned, the germ-tube 

 can bore through cellulose walls, but this does not imply that it can 

 penetrate the cuticle. The fact that the internal mycelium is purely 

 intercellular does not preclude the possibility of entry directly across 

 the epidermal cells, for in Phytophthora Colocasiee Racib., a com- 

 mon parasite in India, the germ-tubes penetrate the leaf frequently 

 in this manner, though below the epidermis they are strictly inter- 

 cellular. The only reason for believing that zoosporal infection 

 in Pythium palmivorum occurs through the stomatic openings 

 is that, in many sections examined, no case was seen of hyphse pass- 

 ing across the epidermal cells, though some of the spots sectioned 

 had resulted from inoculation with zoospores. Still some cases 

 were seen in which hyphse occupied the cavities of the epidermal 

 cells as shown in plate V, fig. 116, and this seems to show that these 

 cells offer less resistance to the entry of the parasite than those of 

 the parenchyma. 



