22 METHODS OF INVESTIGATING 



solutions devoid of chlorine. It appears clear that by this 

 procedure seed was obtained containing a negligibly small 

 quantity of this element. 



The problem of obtaining water and salts free from the trace 

 elements for work on the nutrition of fungi has been dealt with 

 by a number of workers, notably Steinberg (1919, 1935 6). It 

 may be stated at once that pure salts sold as analytical reagents 

 may contain a sufficiency of trace elements present as impurities 

 to allow the growth of plants, nor does recrystallization neces- 

 sarily afford an adequate means for the removal of these 

 contaminants. 



Steinberg's original precedure for obtaining a nutrient solu- 

 tion free from so-called heavy metal contaminants consisted in 

 heating the complete nutrient solution with pure calcium 

 carbonate under pressure. The nutrient solution used was one 

 due to Pfeffer and was made up as follows : 



Sucrose 50 g. 



Ammonium nitrate ] g. 



Potassium dihydrogen phosphate 5 g. 



Magnesium sulphate 2-5 g. 



Ferrous sulphate Trace 



Since Steinberg's treatment of the solution leads to the 

 removal of iron it is obvious that the trace of ferrous sulphate 

 may be omitted. 



To a litre of this solution 15 g. of pure calcium carbonate were 

 added and the mixture heated in an autoclave for 20 min. under 

 a pressure so as to give a temperature of 120-5° C. The mixture 

 was then allowed to stand overnight and the clear solution 

 then decanted from the sediment. Subsequently, Steinberg 

 recommended filtering the solution from the sediment im- 

 mediately after autoclaving. 



The principle involved in this method of purification consists 

 in increasing the alkalinity of the solution so that the traces of 

 heavy metals are precipitated as carbonates, hydroxides and 

 phosphates which are adsorbed on calcium salts precipitated in 

 some bulk. In this way traces of iron, manganese, zinc and 

 copper are removed from the solution. 



Various modifications of this procedure have been proposed. 

 Steinberg himself points out that the substitution of basic 

 magnesium carbonate for calcium carbonate has certain advan- 



