i88 THE MECHANISM OF GASEOUS EXCHANGE 



From these co-operating factors (absorption and gaseous diffusion) it follows that 

 the less soluble hydrogen will pass through more rapidly than the heavier oxygen 

 gas, and Devaux found this to be the case during diffusion through layers of turgid 

 cells '. Obviously the processes of diffusion, filtration, &c., may co-operate in various 

 other ways. It is, moreover, easy to understand that in the streaming of gases 

 through capillaries precisely the same relationships will not be found to exist in 

 rigid capillaries* as in the intercellular capillaries of tissues, the diameter and 

 dimensions of which may be altered by pressure. 



Methods. A detailed account of all the different researches and experimental 

 methods employed can hardly be given. In some cases portions of the upper 

 epidermis of a leaf free from stomata were employed (N. J. C. Miiller, Lietzmann, 

 Mangin), in others entire leaves were used (Wiesner, Blackman). The object was 

 fixed by cement, or by screwing up the two halves of a tube across which it was 

 stretched. A difference of pressure was then created between the two sides 

 by evacuation or compression, or a different gas led into one end. The rapidity 

 of the diosmosis or diffusion may be determined either 

 by noticing the differences of level, &c., on the two 

 sides, or the percentage composition of the gas present 

 on either or both sides may be estimated. To prevent 

 rupture, Miiller used a plate of gypsum as a support, 

 while Lietzmann and Wiesner attained the same end by 

 using small fragments only. When the isolated epidermis is 

 employed, the diosmotic transference takes place through 

 a layer of cells, just as when a unilamellar leaf of a moss 

 (Mniunf) or of a fern (Hymenophylluni) is used, or even when 

 an ivy leaf is employed, for in the last case gases pass readily 

 through the stomata and intercellular spaces to the upper 

 FlG ' epidermis, which is free from them. In certain researches 



the stomata were closed by smearing with fat (Garreau, 

 Boussingault, Stahl, Blackman) or with gelatine (Mangin). 



In Devaux's experiments with submerged water plants (Elodea, Ceratophyllum, 

 &c.) the arrangement shown in Fig. 19 was used. The plants were so imbedded in 

 gelatine in the filter funnel, that by evacuation at r, gases were obtained which 

 have diffused through the exposed outer epidermal layers, and have passed through 

 the aeriferous canals to the cut surface of the stem (s) projecting above the gelatine. 

 Devaux mentions the extent to which errors, due to the existence of gases in the 

 plant previous to the commencement of the experiment, and to the evolution of 

 the carbon dioxide produced by respiration, may be avoided or adjusted. 



SECTION 3 1 . The Communications of the Aeriferous System with 



the External World. 



The aeriferous system communicates with the external world, neglecting 

 accidental fissures in the peripheral tissue layers, mainly by means of the 



1 Devaux, Ann. d. sci. nat., 1889, vii. ser., T. IX, p. 95 ; also Wiesner u. Molisch, 1. c., p. 713. 

 3 Wiesner und Molisch, 1. c., p. 707. 



