Fivngtis of Lolium. 281 



hyphae. They may occur near the vascular tissue, and also 

 towards the periphery of the stem. 



When the inflorescence is formed, they ai-e especially abundant 

 at the base of the carpels. The cells so affected do not increase in 

 size, and are only to be distinguished from a normal unaffected 

 cell by their different staining properties. It is not till the ovule 

 is well advanced that any great increase in the fungal partner takes 

 place, when the phenomena already detailed follow in their natural 

 sequence. 



Ctdtivatiou of tlie Fungus in Artificial Media. 



All attempts by previous workers to obtain a pure culture of The 

 fungus have been unsuccessful. Their work lias been limited 

 mainly to the nucellar hyphae. So far I have been no more suc- 

 cessful than Xestler and Freeman in endeavouring to get the 

 fungus to grow outside its host. As further work is being done 

 in this direction, it has been thought advisable to give a short 

 account of the methods employed^ and the results so far gained. 



Since hyphae isolated from the hyphal layer of the grain had not 

 yielded any result, and as they represent the dormant stage of the 

 fungus, I thought greater success might be attained if the cultures 

 were made from a more active stage in its life-history. Accord- 

 ingly, the ovary was thought to be a suitable stai'ting point, and 

 stages ranging from A-C in the development of the grass have been 

 used for infecting the culture media. 



For the most part the culture medium has been made up in the 

 following way : — 



A decoction of Lolium iierennt in water was autoclaved, then 

 filtered and cleared with egg albumen. The liquid so obtained was 

 made into a 1% agar solution, and autoclaved. It was subsequently 

 filtered, titrated, tubed and sterilised. 



Other media have been tried, e.g., honey agar, starch agar, etc., 

 but with no better results. 



The ovaries were treated in various ways, before using them for 

 infecting the plates. 



(1) Some were washed for one minute in equal parts of a 1% 



mercuric chloride solution and 45% alcohol, followed 

 by a thorough washing in sterile distilled water. 



(2) Others were washed in ether for varying lengtlis of time, 



from five minutes to one minute. 



(3) Others, again, were shaken for some time in sterile dis- 



tilled water. 



