Fungus of Lolium, 201 



'.quote directly: "It is not impossil)le perhaps that infection 

 may, in exceptional cases, take place from this side of tlie 

 scutellum (dorsal); but, if so, it occurs very seldom. I have 

 seen no evidence either in the mature grain or in the developing 

 ovary to indicate that such an infection is ever accomplished." 



My observations permit of a different answer to this question. 

 Hyphae occur at the junction of the scutellum and endosperm, not 

 only near the ventral surface (Freeman's infection layer), but 

 wherever these tissues are in contact I was una])le to demonstrate 

 the existence of an isolated patch as described by Freeman, Fur- 

 thermore, it is impossible to agree with the statement that the hyphal 

 layer does not reach the •end of the aleurone layer on the dorsal 

 side of the grain. As is shown in Plate XVIII. Figs. 2-5, hypliae 

 can and do occur right round the periphery of the embryonic area. 



These facts in themselves are interesting, but they do not answer 

 our question. At a later stage, in this paper, it will be shown that 

 infection of the embryo takes place at a very early stage in develop- 

 ment, and that the distribution of hyphae in the mature grain has 

 no bearing on this point, but is a result of the special function 

 carried out by this partner in the development of the grain. 



It is only fair to emphasise the fact that Freeman dealt only with 

 Lolium temultntmn when working out his idea of an infection 

 layer, and that this criticism is based mainly on work done on 

 .Lolium perenne. However, if the facts demonstrated in the embryo- 

 logical section (pp. 267-281) are true, they apply equally well to 

 both forms, and it becomes abundantly clear that the distribution 

 in the adult grain is not associated especially with the infection of 

 the embryo as Freeman suggevsts. 



Previous workers have described the hyphal layer itself in 

 detail. Australian grown grains of either grass seem to shew a 

 very rich growth of hyphal tissue. Some grains of Darnel grown 

 in the University grounds, Melbourne, had an average layer of 

 31.6 u. Grains of English rye in many cases showed a layer quite 

 •as broad as that showm by an average Darnel, but in both the width 

 or extent of the layer is extremely variable, depending largely on 

 the activity of the fungus during the period l)etween fertilisation 

 and formation of the seed. Aniline gentian violet, followed by 

 Gram's iodine water, was used solely for staining the adult grains. 

 The hyphal layer does not stain uniforndy, liowever, with this stain, 

 some portions of the hyphal threads reacting to the violet colour, 

 other parts remaining colourless. This variation in the stainiug 



14a 



