THE METHOD OF MOVEMENT 175 



protoplasmic streaming in mature sieve tubes has been 

 due to the fact that, in preparing the material for observa- 

 tion under the microscope, the sieve-tube contents are 

 greatly disturbed by cutting open the system which 

 normally has a high internal pressure. Assuming that 

 streaming does occur in sieve tubes at rates of the order of 

 those found in internodal cells of Nitella or Chara, then 

 it would seem to offer a possible mechanism for the trans- 

 port of solutes in plants. In such a system of tubes 

 containing streaming protoplasm it is conceivable that a 

 diffusion gradient of 1 per cent per centimeter may be 

 maintained over a distance of a meter when the actual 

 difference in concentration at the two points, a meter 

 apart, is only 1 per cent. If streaming is sufficiently 

 rapid, the distance over which actual diffusion takes place 

 may be largely determined by the sum of the thicknesses 

 of the various membranes through which diffusion must 

 take place. If this total thickness of cross walls in a stem 

 a meter long is 2 cm., then the real diffusion gradient 

 would be determined by the actual distance over which 

 movement is limited to diffusion. According to the 

 assumption above, this would be 2 cm. Thus the very 

 slight diffusion gradients reported by Mason and Maskell 

 (19286) which would be meaningless if the movement were 

 left to diffusion alone might well account for considerable 

 movement if the total thickness of the membranes through 

 which diffusion must take place is slight and protoplasmic 

 streaming is carrying the solute from one end of the tube 

 to the other. 



Diffusion across thin membranes is extremely rapid as 

 compared with that over the long distances used by deVries 

 in his calculations, for, as A. V. Hill (1926) has emphasized, 

 the time before appearance of a given amount of solute 

 which is moving by diffusion into a medium lacking the 

 particular solute, is inversely proportional to the square 

 of the distance. Thus taking Stefan's calculation, that 

 it would take 319 days for a milligram of sodium chloride 

 to diffuse past a plane at a distance of one meter, the time 



