136 BOARD OF AGRICULTURE. [Pub. Dov. 



logic, I would «ay that their syllogism is practically the 

 followino^ : — 



1. Plants feed entirely upon the materials dissolved by 

 soil water, and the plant roots do not function in making the 

 solutions u})on which the plant feeds. 



2. The fertility of soils — that is, their crop-producing 

 power — has "no apparent relation to the dissolved salts 

 of the soils " (in water) as determined by methods devised 

 by the authors. 



3. Therefore, practically all soils will remain indefinitely 

 fertile so long as right physical conditions and a sufficient 

 supply of water are maintained. This is a remarkable syl- 

 logism, — remarkable in its premises as well as in its con- 

 clusion. The first weakness lies in the premises. 



It yet remains to be proved that the present belief of 

 plant physiologists in the power of plants to make their own 

 solutions of plant food is erroneous. Certainly we have no 

 demonstration that the soil water is the sole source of a 

 plant's food, and there are many facts which inferentially, 

 at least, indicate that our present views are correct. I have 

 personally observed the fact that plants belonging to the 

 Cruciferous family are able to secure their needed supply of 

 phosphoric acid from the crudest and most insoluble materi- 

 als, — a power which seems to be very largely wanting in 

 Graminaceous plants. Observations of this kind have been 

 repeatedly made in forcing-house experiments of the most 

 exact character, and which are ratified ])y the experiences of 

 field culture in the application of crude phosphates to tur- 

 nips. How shall we explain these observations, unless we 

 decide that one class of plants has a feeding power unlike 

 that of another class? It is idle, with our present knowl- 

 edge, to declare that the ability of one plant to secure its 

 food where another one would starve is due Avholly to a 

 more sympathetic environment for the one plant than for 

 the other ; it is much more rational to conclude that one 

 plant has a feeding power unlike the other. If, then, the 

 plant itself is a factor, we step outside of the jiiere question 

 of soil solutions. 



In the second place, docs the method of testing soils, 



