■12- 



trated stock; nutrient solutions vrere ten times as concen- 

 trated as those actuall:; used. 



The ox-gen and carbonaltfe-dioxide contents of the 

 nutrient solutions used for the tests were assumed to be 

 alike; ft is feature of nutrient solution experimaat at ion 

 has not yet attracted the serious attention/ of physiologists 

 "but the assumption here made was probably safe , especially 

 in view of the fact that the -very early stages of growth 

 here dealt v/ith gave no clearly defined diffarences in growth 

 that mi^ t be related to the chemical properties of the 

 solutions. 



The cultures were arranged as follows: Glass 

 tumblers ■ f capacity about 7\00 c.c.) were prepared, each with 

 a tightly stretched! cover of thoroughly washed mosquito 



netting ( cotton thread, with open meshes about S mm. 

 sqiiare ) tightly sealed to the outside of the wall of the 

 tumbler by means of paraffin. Each of 13iBse net- covered 

 tumblers was filled with the proper nutrient solution and 

 tv/enty-five selected seeds were distributed uniformly over 

 the netting (the area being about 58 sq.cm,). All seeds 

 were in contact with ih e solutions, but they were not sub- 

 merged. This simple method was adopted as the best of 

 several methods that were compared in preliminary tests. 



Each single series of cultures involved but one type 

 of solution; it comprised twenty-one different solutions, 



