170 HOW CROPS GROW. 



a maize plant, five and three quarters feet high, and equal 

 hi every respect, as regards size, to plants from similar 

 seed, cultivated in the field. The ears were not, however, 

 fully developed when the experiment was interrupted by 

 the plant becoming unhealthy. 



With the oat his success was better. Four plants were 

 brought to maturity, having 46 stems and 1535 well-devel 

 oped seeds. ( Vs. St., VIII, 190-215.) 



In similar experiments, Nobbe obtained buckwheat 

 plants, six to seven feet high, bearing three hundred plump 

 and perfect seeds, and barley stools with twenty grain- 

 bearing stalks. ( Vs. St., VII, 7:2.) 



In water-culture, the composition of the solution is suf 

 fering continual alteration, from the fact that the plant 

 makes, to a certain extent, a selection of the matters pre 

 sented to it, and does not necessarily absorb them in the 

 proportions in which they originally existed. In this way, 

 disturbances arise which impede or become fatal to growth. 

 In the early experiments of Sachs and Knop, in 1860, they 

 frequently observed that their solutions suddenly acquired 

 the odor of sulphydric acid, and black sulphide of iron 

 formed upon the roots, in consequence of which they were 

 shortly destroyed. This reduction of a sulphate to a sul 

 phide takes place only in an alkaline liquid, and Stohmann 

 was the first to notice that an acid liquid might be made 

 alkaline by the action of liA ing roots. The plant, in fact, 

 has the power to decompose salts, and by appropriat 

 ing the acids more abundantly than the bases, the Litter 

 accumulate in the solution in the free state, or as carbon 

 ates with alkaline properties. 



To prevent the reduction of sulphates, the solution must 

 be kept slightly acid, best by addition of a very little free 

 nitric acid, and if the roots blacken, they must be washed 

 with a dilute acid, and, after rinsing with water, must be 

 transferred to a fresh solution. 



On the other hand, Kiihn has shown that when chloride 



