SOLUTE DISTRIBUTION AND BEHAVIOR 249 



influenced the activity of the meristematic cells which, in 

 turn, influences the diffusion gradient and perhaps also 

 the activity of the streaming protoplasm which connects 

 the receiving cells with the source of supply. Effects of 

 gravity seemingly less related to a gradient have been 

 reported by Loeb (1920) who has given evidence showing 

 that if isolated leaves of Bryophyllum are suspended ver- 

 tically in a moist chamber, certain materials, w^hich he 

 calls ''sap," tend to accumulate on the lower side. This is 

 indicated by greater dry weight, greater pigment formation, 

 and greater development of plantlets on this lower side. 

 The pigment would point to greater sugar accumulation. 

 Inadequate data obtained by MacDaniels and myself 

 indicate that the upper sides of horizontal branches have 

 relatively high nitrogen contents while the lower sides 

 have higher carbohydrate concentrations (see MacDaniels, 

 1923). It is true, however, that mass analyses for total 

 carbohydrates or total nitrogen are of little use in indicating 

 diffusion gradients. This is well indicated by the observa- 

 tions of Mason and Maskell (19286) that the diffusion 

 gradient of sucrose in the cotton plant is toward the boll, 

 whereas that of glucose is away from the boll. 



That something other than food supply and initial 

 activity influences meristematic activity or food movement 

 is clearly indicated by the cambial activity of woody stems. 

 It has been widely observed since very early investigations 

 of ringed stems that diameter growth above a ring, if 

 leaves are present above, is likely to become excessive, 

 whereas cambial growth below a ring is greatly diminished. 

 This decrease immediately below a ring may be marked 

 even if there are leaves or side branches nearby from which 

 a fairly adequate food supply would seem available. In 

 fact, sufficient food may move into this region below a 

 ring to allow for considerable starch deposition or con- 

 siderable starch and sugar may already be present, as 

 observed by Proebsting (1925), and yet cambial growth 

 may be very weak. Some attribute the lack of cambial 

 activity below a ring or on a defoliated branch to the lack 



