GENERAL 179 



on drying, and hence no direct filtration of gas is possible through such 

 a membrane when dry. The diosmotic passage of gases through a mem- 

 brane is determined by their solubility in the fluid it has imbibed, and 

 hence it follows that as the percentage of water present becomes greater, 

 the permeability to soluble gases increases. Gases are, however, transferred 

 much more slowly by diosmotic means than by movements in mass, and 

 hence they can penetrate and pass through a plate of gypsum much more 

 rapidly when its pores are open than when the plate is moistened and the 

 pores filled with water. For the same reason, when water is allowed to 

 block the stomata or intercellular air channels, a marked hindrance is 

 interposed to rapid gaseous exchange, and hence it is of great importance 

 to the plant that any such blocking of the points of entry and the 

 neighbouring passages should be avoided as far as possible. 



The stomata and aeriferous system are indeed admirably adapted for, 

 and of the utmost importance in, gaseous exchange, and usually only the 

 sub-aerial organs are provided with open channels for the entry or exit of 

 gases. The stomata continue to serve as openings for the passage of gases 

 when a leaf immersed in water retains an adhering film of air, giving it a 

 silvery appearance, for by means of this film, dissolved gases are obtained 

 from the water in gaseous form and directly transferred to the plant without 

 any such arrangements being necessary as are exhibited by the respiratory 

 organs of fishes. Submerged leaves may under such circumstances obtain 

 an abundance of carbon dioxide, but when the leaf is actually wetted and 

 the stomata closed by water, the amount absorbed is not sufficient to 

 permit any formation of starch, just as would be the case in similar leaves 

 exposed to air if the stomata were closed and the cuticle thick and im- 

 permeable (Sect. 57). On the other hand, submerged aquatic plants can 

 absorb sufficient amounts of all the gases they require, including carbonic 

 acid gas, in the absence of any intermediary air film, and indeed non- 

 volatile dissolved substances can only be absorbed by parts directly wetted 

 by water 1 . 



From what has just been said, it is clear that direct absorption takes 

 place through the peripheral walls, and that where open gaseous channels 

 are present, they aid in gaseous exchange to an extent determined by the 

 specific peculiarities of the plant and its organs, as well as by the conditions 

 under which these exist 2 . It has already been mentioned that a sub-aerial 



1 Merget (^Coinpt. rend., 1877,1". LXXXIV, pp. 376, 959) erroneously supposed that the presence 

 of a thin film of air was always necessary. Cf. also Devaux, Ann. d. sci. nat., 1889, vii. ser., 

 T. ix, p. 40. 



2 The double power of gaseous exchange through open channels and through cell-membranes 

 \vas perhaps first correctly recognized by Dutrochet (Ann. d. sci. nat., 1832, T. xxv, p. 242). 

 Garreau gave similar correct interpretations in various later works. Merget's conclusion (Compt 

 rend., 1877, T. LXXXIV. p. 376 that gases penetrate almost entirely by means of the stomata, and 



X 2 



