PREVENTION OF RUSTS. 69 



morphology of grains has little or no effect upon the resistance, physi- 

 ological effects, such as described, undoubtedly will influence the 

 general resistance or predisposition of plants to disease in some degree, 

 the extent of which has not yet been determined. Experiments in soil 

 treatments for disease prevention have, however, been made from 

 time to time, a few of which will be cited. 



In 1891-92 Galloway (53, p. 208) at Garrett Park, Md., treated 

 the soil with various chemicals, among which were flowers of sulphur, 

 air-slaked lime, ferrous sulphate, Bordeaux mixture, potassium 

 sulphid, ammoniacal copper carbonate, and potassium bichromate 

 in various quantities and proportions. No practical results were 

 apparent, and he concluded that "in no case did these chemicals 

 have any appreciable effect on the prevalence of rust." On the 

 other hand, Petermann (83, p. 15) claims that wheat on land fer- 

 tilized with superphosphate rusted badly, while wheat under similar 

 conditions, but manured witji Martin slag (a commercial fertilizer), 

 remained almost rust free. He was inclined to believe that the 

 silicic acid present in the fertilizer was an effective agent in preventing 

 rust. 



Further experiments on the effect of fertilizers on crops, both in 

 the United States and in Europe, have been exceedingly numerous in 

 the last few years, but very little careful attention seems to have been 

 given to their effect on cereal diseases. General observations have 

 been made,, however, and it is now well established that where there 

 is an excess of nitrogen in the soil, other things being equal, grains 

 are more severely attacked by rust than crops on soil containing less 

 nitrogen (28, p. 659; 60, p. 245; 76, pp. 72, 73; 95, pp. 263-270). 

 Where barnyard manure has been applied heavily the result is simi- 

 lar, and where grains are grown after a crop of clover, beans, or 

 vetch, rusts may be expected. In fact, it may be generally stated 

 that where soils are rich in nitrogen, producing rank and succulent 

 plant growth, rust attacks will, as a rule, be most severe on account 

 of increased succulence of the plants, increased rankness of growth, 

 delay in drying out after showers and dews, and slight delay in the 

 ripening period. On the other hand, phosphate of lime tends to 

 shorten the ripening period and thus acts as a rust preventive to 

 some extent. Careful observations and experiments along this line 

 in the future should give both interesting and valuable results. Care 

 should be taken, however, to differentiate the results in experiments 

 on fertilizers with relation to rust resistance of cereals. In general, 

 a rust attack is most virulent on a healthy plant. This is particu- 

 larly true of succulent plants in thick stands. As delay in ripening 

 and other effects may also be produced by fertilizers, their relation- 

 ship to the rust must be carefully kept in mind. The effect of such 



216 



