L 278 ] 



ferent substances, which though in quantities scarcely 

 perceptible in the water, may accumulate in the plant, 

 which probably perspires only absolutely pure water. 



In 1801 I made an experiment on the growth of 

 cats, supplied \vith a limited quantity of distilled wa- 

 ter in a soil composed of pure carbonate of lime. The 

 soil and the water were placed in a vessel of iron, 

 which was included in a large jar, connected with the 

 free atmosphere by a tube, so curved as to prevent the 

 possibility of any dust, or fluid, or solid matter from 

 entering into the jar. My object was to ascertain 

 whether any siliceous earth would be formed in the 

 process of vegetation; but the oats grew very feebly, 

 and began to be yellow before any flowers formed: 

 the entire plants were burnt, and their ashes compar- 

 ed with those from an equal number of grains of oat. 

 Less siliceous earth was given by the plants than by 

 the grains; but their ashes yielded much more carbon- 

 ate of lime. That there was less siliceous earth I 

 attribute to the circumstance of the husk of the oat be- 

 ing thrown off in germination; and this is the part 

 which most abounds in silica. Healthy green oats ta- 

 ken from a growing crop, in a field of which the soil 

 was a fine sand, yielded siliceous earth in a much 

 greater proportion than an equal weight of the corn 

 artificially raised. 



The general results of this experiment are very 

 much opposed to the idea of the composition of the 

 earths, by plants, from any of the elements found in 

 the atmosphere, or in water; and there are other facts 

 contrary to the idea. Jacquin states that the ashes of 



