Fungus of Lollum. 268 



.as many as three cells. As is well kn(»wn, the cell walls of this 

 layer are thick, and also pitted. The hyphae enter throiifj:h these 

 .pits, and thereby gain access to the cell (Plate XIX. Fig 5). Some- 

 times the opening in the wall of the aleurone cell was smaller in 

 diameter than the penetrating hypha ; when tliis was the case a 

 conspicuous narrowing was Udticed at the point of entrance, but 

 on the far side of the pit tlie hypha again attained its previous size. 

 In addition the Kcutellum showed an extraordinarily large amount 

 of the fungus. Here the intra-cellular course was also very evi- 

 dent (Plate XJX. Fig. 4). Many of the scutellar cells stained 

 vividly; such cells were seen to be fungal containing. The entrance 

 to the, cells was gained through pitted walls, as is the case in the 

 aleurone layer. The remaining cells of the scutellum were mtniial. 

 and the grains did not seem to be any the worse for this excep- 

 tional behaviour on the part of the fungus. In such abnormal 

 grains the liyphal layer was present as usual. Tliere is no doubt 

 "that the hypliae invading the cells are the same as those composing 

 the extra cellular layer. 



These phenomena were not confined to the sample from Ireland, 

 -one of English rye grass from Soutli Africa also contained certain 

 •grains showing an extraordinary distribution and growth of the 

 fungus. As before, both the aleurone layer and the scutellum were 

 permeated by intra-cellular hyphae. In one particular case +he 

 scutellum, which normally is packed with aleurone grains, appeared 

 to consist of a dense sclerotial-like mass of threads. The bulk 

 remained colourless, and they resembled " ghosts," or casts, of 

 former more virile liyphae (Text figure 2). They are represented 

 in the text figure as dotted lines, and they completely filled the 

 whole of the scutellai- tissue, although the cells composing it were 

 not distorted oi- enlarged in any way. Ibis seition cut in the 

 •coronal plane) and the others accompanying it, were later stained 

 with Congo red; it was then easier to decipher these ghost-like 

 contents of the scutellum. Many were cut transversely, but owing 

 to a large an)ount of twisting Kome were seen running lengthwise 

 through the tissue for a short distance. They probably represent 

 fungal hyphae, which were numerous at certain stages in the 

 'development of the grain, canying a, s])ecial footl su]>i)ly to s])ecial 

 -parts, and in giving tliis up to the host-i)lant tliey have undergone 

 a partial dissolution, wliich was not completely carried out in 

 these few exceptional cases l)y tlie time the grain reached mahnity. 



