100 h'ooi) Koi; I'la.nis. 



wahT of llii' x'il wliicli holds in solution inhicral 

 plaiit-t'ood. 



Siller nil I'oiicii is Mipplicd in ni;in_\- diiTercnt forms, 

 it at oiR't' bccomi's a (|U('s1ion as to which ol' tlu'se is most 

 officient in ci-o]! i)roduction. Far too little attention has 

 hoQii o-iven to tliis important (pu'stion. Too often a cer- 

 tain material has been ehosen because there was anion"' 

 farmers a s>enei"al im])ression that thus particuhir 

 material was better than some other, when, as a matter 

 of fact, there was' no scieiditic basis for such conclusion. 

 As an exami)le, nitrogen from organic sources has been 

 preferred 1)> many because it was Ix'lieved that organic 

 nndter thus supi)lied woukl be of great vahie in improv- 

 ing the physical condition of the soil, but in making this 

 choice farmers overlooked the possibility of using a more 

 readily-available material which would increase the crop 

 residues sufficiently to more than make up for the small 

 amount of organic matter contained in the few hundred 

 ])ounds of divied blood, iish or tankage. Also, there is a 

 widespread impression that the loss of nitrogen is greater 

 Avhen nitrates are used, than wdien organic nitrogen is 

 used. But experiments both in this country and abroad 

 show beyond a doubt that the crop yields and the per- 

 centage of nitrogen recovered in the crop were greater 

 (and hence the loss must have been less) when nitrates 

 were used than Avhen organic sources of nitrogen ^vereused. 



The question of availability of nitrogenous fertilizers 

 began to receive serious consideration at several of the 

 leading European experiment stations some 30 years 

 ago and much valuable information has been accumu- 

 lated by these stations. 



About 10 years later the subject began to receive atten- 

 tion in this eouidry and it is a satisfaction to lind that 

 the results obtained here are fairly in accord with the 

 lindings of the Kurojx'an investigators. 



P^'airly complete i-eviews of this early work liave been 

 given in recent pul)lications (2,8) and no attempt will h' 

 made here to cover this field. 



The completion in 1!)17 of 20 years' work in which a 



