12 Hall, Hoiv docs the Plant obtain its nutriment ? 



concentration. Exactly the same results were obtained 

 when the experiment was repeated, but the nutrients 

 were placed in narrow porous pots sunk in the sand, in 

 order to ensure that the nutrients must diffuse before 

 they could reach the roots. 



There was thus no depression of growth due to slow- 

 ness of diffusion of the nutrients along the water film on 

 sand particles, but it might be supposed that such lag 

 would become operative in the extended film that must 

 exist on the far finer particles found in an ordinary soil. 

 Accordingly a large quantity of sandy .soil was graded 

 into coarse sand as before, fine sand consisting of particles 

 between 02 and 0*04 m/m. in diameter, and silt between 

 004 and 001 m/m. Pure kaolin was taken to represent a 

 clay material largely constituted of still finer particles. 

 The same solution was diffused over all the materials. 

 1 he experiment was twice repeated, and the growth in all 

 the solid media was superior to that in the same volume 

 of solution in a bottle, so that the possibility of a retar- 

 dation of growth due to slowness of diffusion may be 

 dismissed. The photograph, however {PI. IV.), shows that 

 the growth in the sand was the best, next that in the 

 kaolin, while those in the fine sand and silt fell much 

 behind. 



Why should the growth in the sand cultures be so 

 much better than in the equivalent solutions contained in 

 bottles ? We were led to suspect that differences in the 

 aeration of the roots might be the disturbing factor. The 

 sand when properly wetted remains in a very open state, 

 with large air spaces between the aggregates, and the roots 

 could be observed traversing the whole medium freely. 

 The kaolin preserved a very similar structure, whereas 

 the fine sand and silt quickly settled down to a close 

 mass. The appearance of the roots after they had been 



