TRANSACTIONS OF SECTION K, 857 



But there are other cases which prevent our casting aside as impossible the 

 view that Eriksson has put forward. 



I showed some years ago that the mycelium of the Botrytis of the lily disease 

 can lie dormant for some time in the cell-walls, and I have ohservations showing 

 that other forms of Botrytis which attack roses and chrysanthemums only gain 

 the upper hand when the cold autumn nights so chill the attacked cells that they 

 succumb ; the mycelium was there long before, but so long as the cells were active 

 no progress could be made, and only when the plasmolysed chilled cells exude their 

 sap can the mycelium advance. 



Many cases of similarly dormant mycelia appear to exist in those cortex and 

 cambium diseases which result in the production of cankers — e.g., Nectria ditissima 

 and Peziza Willkommii, and Tubeuf's experiments with Gymnosporangium are even 

 more suggestive. Tubeuf found that if G. claoariceforme is sown on hawthorn 

 seedlings the fungus forms yellow spots and induces marked hypertrophy, and normal 

 spermogonia and secidia — Roestelia lacerata — are developed ; but if Pyrus Aucitparia 

 is used as the host, no yellow spots or hypertrophy result, though a mycelium is 

 formed and will even produce a few starved spermogonia. On allied species of P«/7-?<s 

 the fungus may even succeed in forming a few poorly developed secidia. But on the 

 quince the fungus only just succeeds in establishing an infecting mycelium, and 

 soon dies ; and Wagner describes similar events with fungi on Stellaria. 



These cases point to a struggle between the protoplasm of the cells of the 

 diflerent hosts, and of the fungus respectively : sometimes one wins, sometimes 

 the other. The following cases are also suggestive. De Bary found that the 

 germinal hyphfe of Peronospora pyymcea, which is parasitic on Anemone, will 

 penetrate the tissues of Ranuncidics JFicaria, but cannot maintain its hold, and the 

 mycelium soon succumbs and dies. 



Still more remarkable and to the point is the following case. Soppitt and 

 Plowright in England, and Klebahn and others on the Continent, have gradually 

 unravelled a curious case of hetercecism and specialised parasitism among certain 

 Puccinias found on Smilax, Convallarin, Paris, and Diyraphis. The story is too 

 long to recount in detail, hut the Puccinia-spores from Phalaris were found by 

 Klebahn to refuse to infect Polygonatum leaves successfully, though they readily 

 infect the allied Convallaria. Close investigation showed, however, that although 

 the sporidia failed to develop a mycelium in the Polygonatum leaves, they really 

 penetrate the cells, and the delicate germ-tube is killed off' by the protoplasm, a 

 red spot marking the place of entrance. 



The perennial mycelia of Witches' Brooms, secidia in Euphorbia, Taphrina, 

 and many other perennial mycelia are also cases in point. 



It is not my purpose to hold a brief for Eriksson's hypothesis, but I may point 

 out that it is in no way contradictory to the facts already known since De Bary's 

 time. Its most serious aspect is with regard to possible treatment, and it is obvi- 

 ously essential that we should have it tested to the utmost, for it must be remem- 

 bered that no method of spraying or dusting has been, or apparently can be, devised 

 for cereals ; hence the questions as to the existence of really resistent forms, and 

 whether dormant mycelia lurking in their tissues have deceived us in these cases 

 also, require sifting to the bottom. Experience, so far, points to the selection of 

 pedigree wheats and careful cultivation as the first necessities ; how far the ques- 

 tion of spring versus winter wheat aids us is still matter for further experiment ; 

 early and late ripening are also concerned. Climate we cannot hope to control, 

 but it remains to be seen — when the facts are known — how far it can be ' dodged.' 



Clearly what is needed, then, is experiments with varieties of wheat under all 

 conditions, and we may congratulate the Australian, Swedish, and United States 

 experimental stations on their preliminary efforts in this direction. 



I have only been able to give a mere sketch of this rapidly growing subject, 

 but I think you will agree that we are justified in saying that an epidemic of para- 

 sitic fungi depends on the interaction of many factors, congenital variations of 

 the host-plant and topical variations of its cell-contents being probably among the 

 most important ; and since we cannot hope to control the variations of the parasite, 

 or the meteorological conditions, it behoves agriculturists to pay more systematic 



