Growth Phases. These are the same as for water-cultures 

 with wheat. For the reproductive phase the plants will require 

 mechanical support. (See McCall, Jour. Amer. Soc. Agron. 10: 

 127-134. 1918.) 



Location, Exposure, Climatic Records, etc. These are all the 

 same as for water-cultures with wheat. 



The Plant Records. These are the same as for water-cultures, 

 excepting that the root weights require special manipulation. 

 The wax seal is removed and the contents of the crock are placed 

 on a wire sieve (about ten meshes to the inch) , through which 

 the free sand is washed with a stream of water. Then the roots 

 are severed from the tops, as in the case of water-cultures, and 

 the tops are treated as before described. The root systems of 

 all plants from the same crock (including the adhering sand) 

 are then brought together, dried at 102° C, and the dry weight 

 for each culture is determined. Then each mass of roots is 

 ignited in a weighed porcelain crucible till all organic matter has 

 been removed, the weight of the remaining mineral matter is 

 determined, and this value is subtracted from the original dry 

 weight of roots and sand, to give approximately the dry weight 

 of the roots alone. The errors introduced by this method are 

 negligible in this sort of work. (See McCall, Soil Science 2: 

 223. 1916.) 



CULTURES WITH SOY BEAN. 



Experimenters who wish to work with soy bean are requested 

 to communicate immediately with the committee. While the 

 general procedure for this plant will be similar to that for wheat, 

 yet some details must necessarily be different, and many of these 

 are not yet worked out by preliminary experimentation. 



THE SUB-DIVISION OF THE PROJECT. 



Introduction. — The following notes may serve as suggestions, 

 to aid cooperators in choosing their problems in the project. It 

 is hoped that cooperators will choose for their work those por- 

 tio^is of the plan in which they feel most interested. At the same 

 time, the committee will be glad to aid in this connection, and it 

 may be necessary (in order that all parts of the plan may receive 

 attention) for the committee to make special requests at a some- 

 what later time. 



In the first place, it is strongly urged that every experiment 

 include at least three cultures with Shive's best solution for the 



50 



