380 GROWTH OF PLANTS 



and chlorine compounds are included, than when the correlation includes 

 also chi'omium, uranium, and copper compounds. 



In most cases the correlation between laboratory slide-germination and 

 tomato foliage disease methods was good for compounds of copper, chlorine, 

 nitrogen, and nitrogen plus sulfur or oxygen where tautomerization is 

 impossible. However, some uranium and chromium compounds were more 

 effective in the laboratory than in the greenhouse, perhaps because they 

 are fungistatic rather than fungicidal. Also, some nitrogen plus sulfur 

 or oxygen compounds in which tautomerization could exist were markedly 

 more effective in controlling the tomato foliage diseases than would have 

 been predicted from the laboratory tests. This difference held even when 

 the same fungus was used in both methods. 



Wheat smut results were more highly correlated with tomato foliage 

 disease results than with slide-germination results, though the difference 

 was largely attributable to the same groups that gave the poor correlation 

 between slide-germination and tomato foliage disease. Thus, with these 

 exceptions, the three methods are reasonably well correlated. 



Despite these correlations, in the search for new fungicides it is advisable 

 to use all the available methods for screening and to select for further testing 

 those compounds which show promise by any one of the methods. 



New Fungicides 

 ' Until very recently the only agricultural fungicides of economic impor- 

 tance were compounds containing copper, sulfur, mercury, or formaldehyde. 

 These compounds do not by any means fulfill all requirements and there is 

 a pressing need for new and better fungicides. 



Periodic table of the elements. In the search for new fungicides a 

 survey was made of the toxicity of representative inorganic compounds 

 relative to the position of the elements in the periodic table.-^ The labora- 

 tory slide-germination or fungistatic technique was employed using four 

 different fungi. 



In general, toxicity within a group increased with increasing atomic 

 weight. Compounds of the more positive elements showed practically the 

 same toxicity regardless of the particular compound used, but in the case 

 of the more negative elements the toxicity varied greatly with the particular 

 type of compound used. The volatile hydrides, so far as tested, were all 

 highly toxic, but the most highly oxidized forms showed little or no toxicity. 

 There was a considerable tendency for an element which is toxic to one 

 fungus to be toxic to others also, but the correlation is by no means perfect 

 and many exceptions may be observed. The results with one fungus, 

 Botrytis sp. (cinerea type), are sho^\^l m Fig. 151. Compounds of silver, 

 osmium, and mercury were most toxic. In addition to copper, toxic ele- 

 ments which are relatively non-hazardous to humans and which may find 

 more extensive application as fungicides are silver, cerium, cadmium, and 

 chromium. 



