1900- 1.] Effects of Water on Foliage Leaves. 305 



shows that in all coloured solutions used, water ascends faster than the 

 substance in solution ; he cites sixteen solutions in which it is said the 

 water is known to rise more rapidly than does the substance in solution, 

 and he states that three substances, colourless in solution, ascend as high 

 as the water. Lithium nitrate is one of the latter. It should be stated 

 that Sachs' experiments were chiefly for the purpose of determining 

 whether solutions ascended on the surface of the cell walls (by capil- 

 larity), or in the substance of the wall (by imbibition). Lithium, he 

 said, ascended by capillarity and reached a point as high as the water 

 and so there was no decomposition. Sachs' results, therefore, are not 

 contradictory to those of the writer. 



It is shown by using such solutions as potassium ferro-cyanide and 

 ferric chloride, or mercuric chloride and potassium iodide, or any pair of 

 solutions which produce a precipitate soluble in excess of either solution, 

 that the solution loses in concentration as it ascends. 



The solutions used to demonstrate the fact that water ascends more 

 rapidly than the substance in solution were : — 



Potassium ferro-cyanide and ferrous sulphate. 



Potassium sulpho-cyanide and ferric chloride. 



Potassium iodide and mercuric chloride. 



Potassium iodide and lead acetate. 



Copper sulphate and potassium ferro-cyanide. 



Silver nitrate and hydrogen sulphide solution. 



Silver nitrate and potassium iodide. 



Salicylate of soda and ferric chloride. 



To test the rate of ascent of water by the rate of ascent of lithium in 

 solution will lead to error, as has been proved by actual experiment, and 

 as is inferred from experiments with other solutions which produce a 

 precipitate (see Fig. 9). 



A leaf whose petiole is in a solution does not absorb the solution in 

 quite the same way as does a filter paper. The solution in the case of 

 living leaves extends over the leaf so that the area affected is sym- 

 metrical to the whole leaf (Fig. 9 B, and Photo. 1-6). With the filter 

 paper the solution extends equally in all directions without any regard 

 to the outline, commencing at what represents the base of the leaf 

 (Fig- 9)- 



