MOVEMENTS OF WATER IN THE PLANT. 29 



Experiment 26. (See also Experiment 35.) The suction thus 

 caused would not raise water higher than a suction-pump 

 (about 10 meters). The water, however, is not in a continuous 

 tube like the cylinder of a pump. The rectangular wood-cells 

 are in .the form of a series of chains. The water in each cell 

 is separated from that of the neighboring cells by a thin mem- 

 brane which promotes osmose. Water is transpired from the 

 topmost cells of these chains, the cell-sap becomes concen- 

 trated and draws water from the cells beneath, and they in turn 

 from those beneath them. There is thus formed a series of 

 osmometers extending from the leaves to the roots, and 

 capable of lifting water to any height. 



In passing from the lower to the upper end of the narrow 

 wood-cells, the ascent of sap is greatly retarded by capillary 

 friction. On the other hand, the cavity of a wood-cell con- 

 tains a bubble of gas, which by its expansion and contraction 

 aids in forcing the sap upward. Further, imbibition by the 

 cell-wall allows the passage from one part of the plant to 

 another of a small amount of water which does not enter the 

 cell-cavities. 



It is difficult to account for the rapidity of sap-movements 

 by the action of these physical forces alone. Some investiga- 

 tions tend to show that the protoplasm of the wood-parenchyma 

 has a rhythmic osmotic attraction for water. Some such force 

 is necessary to account for all features of sap-movement in 

 trees. 



EXPERIMENT 34. 



AMOUNT OF WATER FORCED UPWARD BY ROOT-PRESSURE COMPARED WITH 

 THAT TRANSPIRED BY THE LEAVES OF AN HERBACEOUS PLANT. 



With a sharp knife cut off a strongly growing Sunflower plant 

 near the ground. Fasten the upper part with its cut end in a 

 measuring-cylinder containing water. To the stump (lower part) 



