224 TWENTY-FIRST RErORT. 



plots on whioh applications of single salts and double salts were made. Here 

 and there scattered through the series were check or control plots. The soil 

 was a sandy loam and the crop oats. At the end of the experiment the best 

 combination of fertilizer salts was determined by the yield of grain and straw. 

 Although it is hazardous to judge from one season's work, the results indicate 

 that the best combination is that in which there is a large amount of acid 

 phosphate. This is true only for this particular .soil. The plot that was 

 ti'eated with acid phosphate alone, however, did not yield as well, showing 

 that a small amount of the other two fertilizers was needed. A subsequent 

 experiment should consider the ratios in this region of the triangle in more 

 detail and on larger plots. With another crop on another type of soil, some 

 other ratio might be the best. These matters have yet to be determined. 

 Three experiments are now running : One on the college farm with oats on a 

 poor sandy soil ; one at Clio, Michigan, on the farm of L. J. Reed with corn 

 on sandy loam ; and the third on the farm of Ezra Levin with potatoes on a 

 fine sand, containing only a little organic matter. It is the hope of this article 

 that others will be interested to work out their soil and crop needs by the 

 ratio method. If so inclined one could collect a large amount of a certain soil 

 type in the fall, and during the winter set up some such series as outlined 

 above. In this way the preliminary work could be done away with and a 

 small series run on a large scale in the field in the spring. The farmer, not 

 inclined to study the soil because of inadequate apparatus and supplies, could 

 send the soil samples to a laboratory, where such a study as outlined above 

 could be made and from the data collected a method of treatment devised and 

 forwarded to him. 



It is believed that three or four weeks' growth of plants in pot cultures 

 will give data which can be used to predict the probable yield at maturity. 

 In fact, there is every likelihood that a mathematical scheme can be worked 

 out to show what the final yield will be after merely a few weeks of growth. 

 Thus in the winter time one can wisely plan what sort of combination will 

 be best to use for a particular crop on a particular soil. Truck growers who 

 use large greenhouses have excellent facilities for working out such a method 

 which will aid them in planning the fertilizer treatment of their soils in the 

 field. 



In conclusion it may justly be said that chemical analyses of soils and 

 crops have taught us much in some lines, but little or nothing as to the fer- 

 tilizer requirements. The crop test method is the only reliable one left, and 

 if a wide range of combinations are selected and logically worked out, some- 

 thing definite and tangible can be obtained. There are great possibilities for 

 improving our practice in the utilization of fertilizers in plant production. 

 We need exact knowledge as to the best salts to use together and the best 

 ratios or combinations. 



Michigan Agricultural College. 



