Fungus of Lolium. 285' 



Although many mycorrhizal fungi are thought tn aid their host 

 23lant in this way, consideralile luicertainty exists cimcerning the 

 determination of the species producing mycorrhiza and their actual 

 function. 



These facts suggest that it is not improbable that the fungus 

 associated with Darnel or English rye grass might act as a nitro- 

 gen-fixer, so an experiment was devised to try .and establish a defi- 

 nite answer to this suggestion as regards Lolium perenne. 



Materiah and Apparatus. 



(1) Method of rreparing Sand. 



Sand cultures were chosen, as they supply a rather more natural 

 condition for the plant roots than water cultures, and sand has 

 the additional advantage of being practically insoluble, and it 

 does not interact with the nutritive compounds used in the water- 

 ing solutions. In order to obtain it free from all traces of nitrogen, 

 it was subjected to the treatment recommended by Schramm (22), 

 A good sample of fine quartz sand was chosen. This was thoroughly 

 washed for about two hours in running tap-water. It was next 

 boiled in strong hydrochloric acid for about one hour, and then 

 washed with distilled water until chlorides could no longer be 

 detected on the addition of silver nitrate. The sand was then 

 heated to a red heat in a furnace for eight hours. This effectively 

 removes any organic material which may be j^resent. The ash 

 formed in this way and any remaining traces of nitrogen were 

 removed by a second boiling in pure strong hydrochloric acid. A 

 second washing with distilled water ensued, and was carried on 

 until the sand was free from chlorides. Finally it was washed 

 a dozen times with nitrogen free water, and then dried in a 

 drying oven. After this treatment, on testing for ammonia, 

 nitrites and nitrates, onlv nesative results were obtained. ^ 



.5. Nessler's reagent wns used in testing for ammonia. The Lunge test 

 (Dlplienvlamine) was used in testing for nitric acid. A modification of the 

 Peter-Griess method was used in testing for nitrous acid. This test is^ 

 extremely delicate, according to Anderson (25). One-thousandth ot a 

 milligram can be detected with certainty. The Griess-Ilosvay method is as 

 follows : — ■ 



(1) Dissolve 0.5 gm. of sulphanilic acid in 150 cc. of 2-normal acetic 

 acid. 



(2) Boil 0.2 gm. of a naphythylamine in 20 ccs. of water. Pour 

 off the colourless solution from the violet residue, and add to the 

 solution 150 cc. of 2-normal acetic acid. Mix the two solutions^ 

 (This mixture must be kept in a dark place.) Take 50 cc. ot 

 the material to be tested with 2 cc. of above reagent, and allow 

 it to stand 5 or 10 minutes; it will be coloured red if a trace of 

 nitrous acid is present. 



In using this test, the flask should be plugged with cotton wool, to prevent 

 dust from entering the solution and disturbing the result. 



