Lecture 1 — 19 — Survey 



best proportions of nutrient salts in a culture medium. 

 It was thought that this knowledge would have a prac- 

 tical application in guiding fertilizer practice. A con- 

 siderable number of investigations were carried out. 

 Several difficulties of technique and interpretation were 

 involved, and it is well to examine one or two of these 

 in the light of subsequent trends. 



In the first place, plant physiologists at that period 

 had given little attention to biological variability in 

 plant culture experiments. An array of solutions of 

 compositions varying by steps would be tested, and 

 average values plotted on a triangular diagram, each 

 axis representing one of the component salts of the 

 nutrient solution. The attempt was made to assign 

 small regions within the triangle as representing the 

 best proportion of salts. One of my colleagues, A. R. 

 Davis (1921), made an early experiment with a large 

 number of replicate cultures of young wheat plants 

 for each of several solutions and showed that, statisti- 

 cally evaluating the data, one could not single out 

 within any narrow range a particular proportion of 

 nutrient salts as representing a "best" solution. The 

 results of the soil solution investigation that I have 

 described were consistent with this conclusion. Sev- 

 eral soils produced almost the same crop yields and 

 yet the composition of the soil solutions differed 

 markedly in ionic relations. Now the pendulum has 

 swung in the opposite direction with regard to sta- 

 tistical evaluation of results in studies like these and 

 the orthodox thing to do is to present an analysis by 

 some statistical method. The mathematical analyses 

 proposed by R. A. FiSHER are most frequently em- 

 ployed. 



Leaving aside the matter of statistical validity, 

 there remained another difficulty in the water culture 

 experiments. The solution placed in the culture vessel 

 did not long remain the same, because of the absorp- 

 tion of salt by the plant. The solution was in fact 

 continuously undergoing alteration. It was this ques- 



