166 



hf reniovod very rnpidly nnd rondored iiisolnhU*. nffoftinj; tliorohy the character 

 of the leguminous cmp. 



Then there is another jioint. Kiihn reports a series of experiments where for 

 Iwenty years lie grew nonleguminous phmts on rather light soils, with a gen- 

 erous .ipiilii atiou of potassium salts and phosphates, and there was no diminu- 

 tion in the croi) producing i)ower of that soil. 



Hall, recently, reports an instance showing that on one of their experimental 

 fields where no leguminous plants were grown for a numher of years there was 

 a consider.ahle increase of nitrogen in the soil — several thousand pounds per acre. 



A still more interesting experience of our own. an ohservation made hy Doctor 

 Voorhees and myself, was this: In certain sandy soils, where we applied phos- 

 phoric acid and potash but no nitrogen, we have been gi-owing millet, which is 

 a very exhaustive crop, for three successive seasons, and yet observed no diminu-^ 

 tion in the yield of millet. Taking the sum total of the nitrogen yield in the 

 crops, and of the nitrogen now in the soil, and subtracting that present in the 

 soil at the beginning of the experiment and that subsequently added in the seed 

 and \\ater. we tind a very large gain from some source. 



We do not know what intluence the crop has on the gi'owth of the bacterial 

 flora of the soil. There is no question that different crops affect this flora 

 unequally. We know that from crop to crop the bacterial flora may be changed 

 to an astonishing degree. There are certain classes of nitrogen-fixing bacteria 

 which seem to be found in little nests on particles of limestone in the soil, and 

 we may obtain practically pure cultures of them from those little nodules of 

 limestone. These are but a few scattered facts which may throw some light on 

 the interrelation of soil acidity. liming, crop rotation, and the growing of legumes. 



While we have looked at soil fertility from different standpoints, and while I 

 do not pretend to regard the bacteriological standpoint as the only legitimate 

 standpoint, yet I think, as Professor Thorne has also pointed out, that the l)ac- 

 leriological problems surely deserve more attention than they are receiving. 



H. J. Wheeler, of Rhode Island. What I say to-day you will please accept as 

 tentative and in the line of suggestion rather than as something formal. An 

 ;.ttempt will be made to simply point out some of the peculiar and interesting 

 tilings that arise occasionally as suggestive in relation to soil investigation. 

 Let me refer first to the two-hundredth normal hydrochloric-acid method. This 

 has been tried on some soils, and it apparently shows a considerable amount of 

 phosphoric acid as probably available to plants. The same method of extrac- 

 tion applied in the same lal)oratory to some of our own soils yielded no phos- 

 ])horic acid. yet. strangely (>nough. iipon some of the same plats from which 

 the soil w.is taken we get very good crops of certain varieties of plants. 



Now. while there may perhaps be a likelihood of finding a single method of 

 extraction that will tell us as to the assimilability of the potash in soils, it will 

 seem more sunirising if we find any single solvenjt that will tell us for all 

 soils concerning the condition of the phosphoric acid. To make this more clear, 

 let me state that in our own soil we have about ;?i per cent of humus, and there 

 are very fertile soils in the valley of the Kio Grande that have a comparatively 

 low iier cent of hunuis. In our soil a large proportion of the phosphorus is 

 I'resent in the organic matter. In the soil from some other regions a large 

 proportion of it may lie present in apatite, iron phosphate, or aluminiun i)hos- 

 phate. Is it not too nmch to expect that ;iny one solvent can show the require- 

 ments of these soils in which the ])hosph<)rus is itresent in so many different 

 coinbinationsV This suggests that what we need is a differentiation of the phos- 

 |. horns in ]>articular soils, and perhai)s we shall find that for a certain class of 

 them a certain solvent may be u.sed advantageously, in another class another 

 solvent, and in a third class we nuist perhaps apply two or three treatments 



