28o Transactions British Mycological Society. 



briefly with the question of " subinf actions. " Salmon {4) (pp. 270- 

 271) in his work on Erysiphe graminis and the Bromes found 

 that in some inoculations the only result was a few minute 

 flecks of mycelium and a few scattered conidiophores ; in some 

 cases these minute infections disappeared within a few days 

 and in others small flecks of mycelium persisted. He came to 

 the conclusion that a faint infection does actually occur and 

 that the flecks of mycelium and few conidiophores are not 

 merely the production of a conidium germinating and living 

 independently for a short time. He called these slight infections 

 "subinfections." 



In the work described here the phenomenon of " subinfections " 

 was encountered early. As will be seen from Table I above, 

 inoculation of conidia from Swede on to cultivated species of 

 Brassica Meracea invariably produced "subinfections" only. 

 However in these particular cases of "subinfections" there is 

 another detail to be taken into account and that is the distinct 

 discolouration of the epidermis of the host which takes place. 

 Superficially the epidermis at a point where a "subinfection" is 

 present shows a series of minute black spots giving the appear- 

 ance of local death of the cells. In section it was found that in 

 some cases only an epidermal cell or a small group of cells were 

 affected, the walls being very dark brown and the cells being 

 filled with a dark brown disorganised content so dense that it 

 was impossible to determine if normal haustoria had been 

 formed or not; in other cases the discolouration radiated into 

 the subepidermal cells changing the cell walls to a dark brown 

 colour and sometimes the cell content also. 



The discolouration of the epidermis by species of Erysiphaceae 

 has been noted by several investigators and was discussed by 

 Grant Smith (i), but the discolouration noted by him did not 

 seem to be of such an extreme nature as that under considera- 

 tion now, as Grant Smith says it was not noticeable in sections 

 when cut and stained and seemingly normal haustoria were 

 visible. In fact the discolouration was so distinct in the present 

 instance that by its means the phenomenon of "subinfection" 

 was later discovered in the field. Salmon ((4) p. 270) says, "It 

 may be that, in some cases under certain favourable conditions 

 the fungus could exist permanently and increase on such a host 

 plant." That this is the case at any rate with "subinfections" 

 on cultivated Brassica oleracea is confirmed by the following 

 field observations. 



Oct. 1913. After long search a Rape plant severely infected 

 with the oidium of Erysiphe Polygoni was discovered in the 

 centre of a large field of Kohl-rabi. An investigation of the 

 Kohl-rabi plants showed that all those plants within a few yards 



