Fungus of Loliwni. 283 



The ovax'v used in this plate was obtained from a plant growing 

 in the Mellouine University grounds. It ■was apparently the pro- 

 <duct of a second flowering resulting from heavy early autumn 

 rains. Since all attempts t) obtain this form again have l>een 

 -unsuccessful, it might be argued that it cannot be the fungus found 

 associated with Loliuiu pereune. lliis may be so, but it is just 

 probable that since the ovary represented a second flowering the 

 fungus may be growing more actively than the developing grain, 

 the grass being naturally weakened by its previous flowering, so 

 that the fungus may have been in a suitable condition to grow 

 ■on the artificial medium provided. 



Concerning the Function of the Funr/us. 



It has been suggested that the fungus associated with Darnel 

 grass possesses the power of nitrogen-fixation. Hiltner (5) was the 

 first to formulate this idea, and after testing it by experiment, he 

 concluded that Loliuni frni>ile?i,tum grew as well in nitrogen-free 

 sand, as in sand to which nitrogen, in the form of potassium 

 nitrate, had been added as a fertiliser. As a control he grew 

 Loliuui italicum under similar conditions. This species, at the 

 time of Hiltner's work (1899), was regarded as being fungus 

 free. Later, Freeman (1903) found in a sample of 59 grains two 

 -contained the fungus and 57 were devoid of it. This, although it 

 is a low percentage of infected grains, could introduce a serious 

 error into such work when using this species as a control. 



The experimental methods employed by Hiltner are also open to 

 •criticism. He planted grains of both species in pots, which were 

 •completely nitrogen-free, but he watered one set of two with tap- 

 water, whicli contained 0.84 nig. of nitrogen per litre. To the 

 •other set of two he gave in addition 50 mg. of nitrogen in the 

 form of potassium nitrate. These pots were appai-ently left exposed 

 to the air, and so were subject to many sources of external nitrogen 

 contaminatitm, the most formidable perhaps being nitrogen-fixing 

 bacteria. 



An experiment can-ied out in this manner could not aim at deter- 

 mining whether the fungus is capable of fixing free atmospheric 

 nitrogen in the coinpelett absence of comhiiied nitror/en. How- 

 ever, as several investigators have shown. Berthelot (17), Purie- 

 witsch (18), and Latham (19), that certain fungi can fix free 

 nitrogen if supplied with a small amount of this element in a com- 

 bined form, the results given by Hiltner might have some bearing 

 • on the latter point. 



