PARASITISM IN FUNGI 135 



.plant. They develop readily on the dung of animals. 

 Certain species were also grown in nutrient solutions. 



It has long been known that the parasite gains an 

 entrance into the host by different methods. When the 

 spore or conidium of a parasitic fungus alights on the 

 surface of a leaf that it can infect, germination takes place 

 if conditions are favourable, a certain amount of moisture 

 being usually necessary. In some species the germ-tube 

 produced by the germinating spore always enters through a 

 stoma ; in other instances the germ-tube pierces the walls 

 of the epidermal cells, and thus gains an entrance. It 

 has also long been known that a parasitic fungus as a rule 

 can only develop on one particular kind of host-plant, or on 

 a few closely allied plants. The spores and conidia of para- 

 sites, carried by wind or other agents, alight on living leaves 

 of all kinds, and readily germinate, but the germ-tube never 

 enters a plant to the extent of developing a mycelium and 

 producing its fruit, unless it happens to be the kind of 

 plant on which the fungus is habitually parasitic. This 

 apparent selection of hosts by fungus parasites I have 

 endeavoured to prove to be due to chemotaxis or chemo- 

 tropism, which may be defined as the direction of move- 

 ment due to chemical stimuli. This again means that a 

 germ-tube will only enter the tissue of a living plant, and 

 run its course of development, when there is some special 

 substance or substances present in the cells of the plant 

 which exercise an attractive influence on the germ-tube. 

 Miyoshi, a Japanese botanist, first suggested this idea. 

 He proved that by placing the epidermis of an onion bulb- 

 scale or a thin film of mica through which very minute 

 holes had been pierced, on the surface of gelatine contain- 

 ing a positively chemotactic substance, as dextrin, plum 



