THE METHOD OF MOVEMENT 143 



an interchange, either absorption or loss, between a Uving 

 cell and neighboring dead cells or the external environment 

 is extremely slow. This movement through plasmodesma 

 was previously suggested by Kienitz-Gerloff (1891), though 

 the latter did not emphasize the low permeabiUty of sur- 

 face membranes to sugars. This is a highly important 

 suggestion of Maugham's, even though the mechanism 

 that he proposed to account for the actual movement is 

 untenable. Although Mangham implied adsorption on a 

 stationary colloidal surface, it requires no great alteration 

 of his adsorption hypothesis to adapt it to adsorption 

 at an interface between liquids where the material is 

 actually carried in the moving film. As thus modified, the 

 hypothesis is far from untenable and would agree with 

 the protoplasmic streaming hypothesis or a modification 

 of it suggested by Van den Honert (1932) (see Sec. 37). 



28. The Hypothesis of Munch.— Munch (1926, 1927, 

 1930) has recently revived the hypothesis that there is a 

 unidirectional mass flow of sieve-tube contents which is 

 due entirely to a fall in pressure in one direction. Though 

 there are some very serious and perhaps insurmountable 

 weaknesses in the hypothesis, there are a number of points 

 that strongly favor it. His hypothesis is based funda- 

 mentally on the fact first stated by Pfeffer that if solutes 

 are distributed unequally in an osmotic chamber, there 

 will be an absorption of water by that side of the chamber 

 where the concentration is high and a secretion of water 

 from that side where the solution inside of the membrane 

 is weakest, and that this will result in a mass flow of 

 solution within the chamber from the region of high 

 concentration toward that of lower concentration. Recog- 

 nizing the fundamental principle but not reahzing that 

 Pfeffer had already proposed it, I set up a series of osmotic 

 cells, consisting of atmometer shells impregnated with 

 copper ferrocyanide membranes, connected with small 

 tubes, and by this apparatus I demonstrated this fact of 

 unilateral secretion through a closed system, where there 

 was a high concentration at one end of the cell or system 



