THE PROCESSES OF ABSORPTION AND EXCRETION 105 



The special properties of the plasmatic membrane render it possible that 

 a dissolved colloid substance may pass through it, but not through the 

 water-saturated cell-wall, although the micellar interstices of the plasmatic 

 membrane are certainly much narrower than those of the latter, and 

 although in the cell-wall the micellar interstices probably form minute 

 canals, across which the molecular influences radiating from the micellar 

 walls are unable to extend. 



In dealing with the selective powers (Sect. 22), still further instances will 

 be brought forward to show how external agencies may by inducing metabolic 

 changes, directly or indirectly, cause the diosmosis of a particular substance. 

 From this point of view, the peculiarities which Nageli J observed in the yeast of 

 alcoholic fermentation are worthy of further investigation. Nageli found that 

 peptone and albumin are excreted by yeast plants in an alkaline medium, but 

 that, in an acid medium, albumin alone is excreted, and that only when fermenta- 

 tion is active. It is possible that in this case the existence of the solvent for 

 albumin becomes of importance by extracting albumin from the protoplast, or 

 rendering the diosmosis of the former through the cell-wall possible 2 . The 

 manner in which tartrate of iron acts in favouring the nutrition of algae with sugar 

 is quite uncertain, though Klebs 3 suggested that it may possibly aid in the absorp- 

 tion of sugar. 



Diosmosis. This term applies to the diffusion which may take place through 

 a solid or fluid membrane. The nature of these phenomena as well as the imbibi- 

 tory processes to which they are directly due have already been described, in so far 

 as they affect, or are related to physiological processes (Sect. 12). A more complete 

 physical account is unnecessary here, since a full and rational description of the 

 phenomena of osmosis is now given in all physical text-books 4 . 



Reasons have already been advanced for the conclusion that in semi-permeable 

 membranes the interstices filled with water can have only a relatively very smal 

 diameter (Sect. 16). When a particle of dissolved material passes through the 

 axis of an interstitial canal, only the margins of which are influenced by the mole- 

 cular forces radiating from its walls, we can speak of capillary diosmosis, as 

 contrasted with molecular diosmosis, in correspondence with the distinction already 

 pointed out between capillary and molecular imbibition (Sect. 12). Probably 

 both forms of diosmosis take place simultaneously through cell-wall, although 

 the small diameter of the capillary spaces, together with the relative rigidity of the 

 substance of the wall, allows the large molecules or molecular complexes of 

 colloid substances to diosmose only with difficulty, or not at all. By the diosmotic 



1 Theorie d. Gahrnng. 1879, pp. 79 and 105. See also Gayon et Dubourg, Compt. rend., 1886, 

 T. cil, p. 978. Sufficient attention has not been paid to the possibility of the proteid substances 

 being partly or entirely derived from dead cells. 



3 Nageli, Sitzungsb. d. Bair. Akad., 1878, p. 169. 



3 Klebs, Unters. a. d. Bot. Inst. z. Tubingen, 1886, Bd. II, p. 544. 



* Ostwald, Lehrb. d. allgem. Chemie, 1891, 2. Aufl., Bd. I, pp. 651, 674; Winkelmann, Handb. 

 d. Physik, 1891, Bd. I, p. 618. Also Pfeffer, Osmot. Unters., 1877. 



