758 EXPERIMENT STATION RECORD. 



ex|H'riiiient)^ rcqiiiro as luut'li skill us laboratory methods, and are tof) complicated 

 for llu' average fanner to nndertake. Fertilizer exjjerinients are the logical supple- 

 ment of i)hysiolof:i(al analysis, since their o])ject is to determine which of the various 

 commercial fertilizing materials available will most etliciently and economically supply 

 the fertilizer requirements indicated l)y the physiological analysis. Such exjjeri- 

 ments may be made with advantage by farmers. 



jNIodern investigators clearly recognize the fact that the simple methods advocated 

 by Ville and others are entirely unreliable, and that it is not possible to deduce from 

 the results of such experiments by a simple calculation the exact quantities of fertil- 

 izer to l)e restored in practice. Reliable results can be obtained, and these results 

 can be properly interpreted only when the nuuienjus possible sources of error, })oth 

 of execution and of interpretation, are taken into account and precautions taken to 

 eliminate them as far as possible. The author enumerates the following points: 



(1) In the first place the law of minimum on which the method so largely rests is 

 by no means absolute as applied to such experiments — i. e., the fertilizing constitu- 

 ents are more or less replaceable or interchangeable, and the mineral composition 

 of the plant varies to a considerable extent with the excess or deficiency of certain 

 elements of ])lant food. Then again certain fertilizing constituents may improve onlj'^ 

 the quality and not affect the quantity of the product. 



(2) A disturbing factor is also introduced by the fact that even in the simplest 

 fertilizing material the constituent under investigation is necessarily combined with 

 other constituents, and it is often difficult to tell how far the particular effect observed 

 is due to the form of combination or to the associated constituents. 



(8) Moreover, certain fertilizers do not serve simply as sources of plant food, but 

 also modify the physical conditions in the soil, act as solvent and diffusing agents, 

 correct acidity and other unfavorable conditions, aid nitrification, etc. 



(4) The soil renders certain of the fertilizing constituents applied unavailable. If, 

 for example, by analysis by the plant of a soil rich in organic matter a lack of 

 phosphoric acid is found, one must not conclude a priori that restitution of phos- 

 phates is necessary. This soil may be rich in unavailable phosphoric acid, and in 

 that case it would be of greater advantage to destroy the excess of humus by liming, 

 harrowing, etc. 



(5) Account nmst also be taken of the fact that some fertilizers are more soluble 

 and diffusible than others and are less readily absorbed by soils. 



(6) With an abundant supply of water in the soil and free transpiration from the 

 plant a smaller amomit of plant food in the soil may suffice for the needs of the 

 plant than if the reverse were the condition. In other words, fertilizers are most 

 effective when the water conditions are favorable. 



(7) The behavior of the plant during growth as well as the final weights must be 

 recorded, because the two are sometimes contradictory. 



(8) In experiments with mixed herbage there is a special source of error due to 

 the fact that a form of fertilizer which favors one plant of the mixture may be 

 injurious to another. 



(9) There is always the injury due to birds, insects, and fungus diseases, etc., to 

 be guarded against and taken into account. 



The results of physiological analysis of a soil vary with the species, variety, and 

 sometimes the individuality of the plant, the quality of the seed, the physical con- 

 ditions of the soil and air, and the form in which the fertilizers are applied. The 

 results of experiments in pots and boxes are not strictly aj^plicable to field conditions, 

 nor those of one year to another. From a physiological analysis made by means of 

 oats we may not legitimately conclude that sugar beets, potatoes, clover, etc., would 

 behave in the same manner in the same soil and under the same conditions of en\a- 

 ronment. Such an analysis would have a considerable relative and indirect value 



