29 



of the passage, that is to the square of the diameter. 

 Accordingly, this rate decreases rapidly as the passage 

 becomes smaller. Furthermore the actual passages are 

 very tortuous and frequently constricted by narrowings, 

 small orifices of communication between vessels, etc. 

 All of these tend to reduce the diffusion rate. 'J he assump- 

 tion of no removal of oxygen during passage through the 

 tube is obviously untrue but is probable compensated for 

 by the increased size of passage or by the pfcesence of 

 additional parallel passages in the thicker root-portions 

 nearer the air. Of the factors other than simple dif- 

 fusion, all but one can be discussed as certainly of 

 lower order than the diffusion effect. This one is the 

 effect of turgor changes in the root. On this matter 

 not only is experimental data entirely lacking but there 

 are not even materials for intelligent speculation. 



Although the calculations outlined suggest that 

 oxygen entry through the air-passages of long roots cannot 

 be very great, it is not intended to assert that such 

 oxygen supply is necessarily insignificant in all cases. 

 Indeed certain specific morphological data imply the ex- 

 istence of a real relation between root passages and 

 aeration in the particular cases concerned. Thus it is 

 stated that swamp plants usually have large and more 

 numerous internal air passages than have typical land 



