72 PHOTOSYNTHESIS 



It was also found that the rate of diffusion of water vapor is con- 

 trolled by the linear dimensions of the apertures. This applies equally 

 when the water is evaporating from a surface of water through an aper- 

 ture or is absorbed by, e.g., sulfuric acid. 



In order to explain their experimental results on the rate of diffusion 

 of carbon dioxide through apertures and the "diameter law" Brown and 

 Escombe pictured "lines of creep" of the carbon dioxide as it passes 

 through the air towards the disc to replace that absorbed. Thus, the 

 simplest case would be a circular disc, capable of absorbing carbon dioxide, 

 freely exposed to the air and the disc surrounded with a rim, in the 

 same plane, three or four times the diameter of the disc. If the air is 

 perfectly still there will be established a steady gradient density of car- 

 bon dioxide surrounding the disc. If lines are drawn through all the 

 points of the same carbon dioxide-density above the disc, curved surfaces 

 are formed. These surfaces will be in the form of shells surrounding 

 the disc. If the latter is a perfect absorbent of carbon dioxide each shell 

 will represent a carbon dioxide-density varying from zero at the absorb- 

 ing surface to the maximum density which will be that of the carbon 

 dioxide in the air. This maximum is at a distance of 5 or 6 diameters 

 from the disc. The gradient of density will thus be a line perpendicular 

 to the shells of equal density. The problem is very similar to that of 

 the lines of force of an electric field. 



Stefan and others studied the exact converse case, that of the evapora- 

 tion from a circular surface. A mathematical analysis of this problem 

 showed that the amount of evaporation is proportional to the linear dimen- 

 sions of the liquid surface and not to the area. The shells in this case 

 form an orthozonal system of ellipsoids having their foci in the edge of 



the disc. 



Larmor worked out the following formula for the absorption of 

 atmospheric carbon dioxide by a perfectly absorbing circular disc : 



Q = 2kpD, 



in which Q is the amount absorbed in a given time, k the coefficient of 

 diffusion of carbon dioxide in air, p density of atmospheric carbon dioxide, 

 and D the diameter of the disc. 



By using discs of very small diameter and perfectly quiet air Brown 

 and Escombe were able to obtain results which showed an absorption de- 

 pending upon the linear dimensions of the surface according to the formula 

 given above. In Figure 5 are shown Brown and Escombe's conception 

 of diffusion shells. A represents the shells in the case of an absorbent 

 disc with rim; the convergent hyperbolic lines of flow, representing the 

 carbon dioxide gradient, terminate in the surface of the disc. 



In the case of perforations in a septum which divides two regions 

 of dift'erent density the conditions are very different. In the former case, 

 as the experiments show, the absorption, on the basis of linear dimen- 

 sions, is exceedingly sensitive to any disturbances which affect the hypo- 



