7 8 Journal of Agricultural Research voi. xvni. No. 2 



centration the colorimetric methods of Clark and Lubs (8) were em- 

 ployed, or, for estimating soils, direct measurements by means of the 

 hydrogen electrode, in the manner described by Sharp and the author {38), 

 Reactions have been calculated to the customary Pg values. Conduc- 

 tivity measurements were carried out in the usual way at a temperature 

 of 25° C. For standardization of the electrode a N/^o potassium chlorid 

 (KCl) solution was used. 



Since conductivity measurements express the total conducting power 

 of the solution and are necessarily influenced by the nature of the ions, 

 degree of dissociation, etc., they can be regarded as giving only approxi- 

 mate values, showing the general trend and magnitudes of absorption. 

 For the latter purpose they are very convenient, and with the nutrient 

 solutions and concentrations used in this experiment it has been found 

 that the resistance varies with the concentration in a fairly direct ratio. 



In many instances chemical analyses for calcium (Ca), magnesium 

 (Mg), phosphate (PO4), nitrate (NO3), sulphate (SO4), and potassium 

 (K) have been made on the solutions or the plants. Whenever possible 

 standard gravimetric and volumetric methods have been used. Occa- 

 sionally, when very small concentrations were involved, the special 

 technic described by Stewart (42) has been found valuable. 



In the study of the absorption by the plant the culture jars were made 

 up to the original weight with distilled water then thoroughly mixed 

 by passing a current of air through, after which samples were taken for 

 conductivity measurements or composites made for analysis. Before' 

 the jars were made up to weight the cork and plant were removed and 

 temporarily placed in another jar. The roots were allowed to drain as 

 thoroughly as possible, and since they were equally saturated with liquid 

 at all times it is not probable that the general tendency of the results 

 would be appreciably influenced by the loss of solution adhering to the 

 surface of the roots. In several experiments conductivity measurements 

 have been made on each individual jar in order to ascertain the degree of 

 variability displayed by individual plants. 



PRELIMINARY SAND CULTURES, SERIES I 



Several years ago an attempt was made by the use of sand cultures to 

 gain some idea of the effect of concentration of the nutrient solution on 

 the growth of the barley plant. At this time arrangements were not 

 available for changing solutions, and additional quantities of nutrient 

 solution were added as water was lost by transpiration. Thus there 

 was no control of exact concentrations as there was in the later experi- 

 ments. It is thought worth while, however, to give these results a 

 brief consideration, since they clearly indicate a definite relation between 

 the nutrients present and the yield of grain and straw. In the critical 

 discussion of other experiments these earlier data may be very helpful. 



