443 



quantities of glucose in Erlenmeyer's conical flasks 

 of 450 ciP. 



Xot one of thèse cultures which I kept from July 1911 

 till May 1912 showedany signof spore-or fruitbody formation. 



c. Artificial inoculations. 



Infection is accomplished very easily by binding round 

 a branch leaves already attacked; under fa vourable condi- 

 tions c. q. a mnist atmosphère the fungus develops rapidly. 



On June 30th in this manner a branch of Liberian coflFee 

 was infected; above the place of inoculation it bore 8 

 leaves. On July 21st four leaves were already attacked so 

 bad that they showed brov^n patches ; on July 28th thèse 

 patches had become larger and the fifth leaf was attacked ; 

 hère the filament was still white, and no discoloured spot 

 could be seen; but the following day, July 29th, the leaf 

 had a large brown spot; this proves that death by 

 suffocation, (the stomata becoming clogged) occurs very 

 quickly, so it can be understood, that this stage is not 

 easily found. 



' On August 6th ail the leaves showed a bad infection 

 while the day before no brown colour was seen on the 

 eighth and last leaf. 



Infection from a pure culture cannot be managed so 

 easily; I obtained some good results by keeping freshly 

 eut branches under a glass bell very moist, after putting 

 little pièces of agar with the mycélium on the leaves. Of 

 course the leaves had been thoroughly rinsed with steri- 

 lised water in order to remove a good many attacking 

 spores. One can follow the infection under the microscope 

 so that no confusion is possible. After some time the 

 hyphae spread over the leaf; when I bound those so 

 attacked round coffee branches on the tree, the infection 

 soon gave positive results. 



