460 



BOTANICAL GAZETTE 



[DECEMBER 



two functions, absorption and transpiration, are not necessarily 

 proportional; (2) that the absorption is equal to the transpiration 

 when the plant is growing under average and fairly constant con- 

 ditions, as in diffuse light and in average humidity; (3) that 

 when a plant growing under normal conditions is placed in a dry 

 atmosphere, transpiration is greater than absorption; (4) that 

 when a plant growing under normal conditions is exposed to a 

 saturated atmosphere, the absorption, in obedience to the vacuum 

 already existing, is greater than the transpiration; and (5) that 

 when a plant lacks water, the suction produced by transpiration 

 is not lost, but it accumulates to act immediately when the roots 

 come in contact with water, when absorption .is much more 

 energetic than transpiration. 



Clements (4) experimented with potted plants of Helianthus 

 annuus and found by careful weighing of the amount of water 

 absorbed by the roots and the amount transpired by the stem and 

 leaves that at the end of 5 days the one equaled the other. Hi 

 results were as shown in table III. 



TABLE III 



I... 

 II.. 



Wt. of pot 

 and dry soil 



Wt. of pot and wet soil 



I 



1846.0 gm. 2218.0 gm. 

 1886.7 gm. 2253.2 gm. 



II 

 2174.3 gm 



2221.6 gm 



Total H*0 



HzO left 



372.0 gm 

 366.5 gm 



328.3 gm 

 334.9 gm 



H,0 



absorbed 



43 • 7 gm 

 31.6 gm 



H a O 

 transpired 



43 • 7 gm 

 31.6 gm 



To sum up the conclusions of these investigators, the results 

 indicate for land plants growing under natural conditions: 



1. That there may be slight discrepancies between the amounts 

 absorbed and transpired if the periods between measurements are 

 separated by a few hours only. 



2. That for a longer interval of time between measurements the 

 amount transpired is essentially equal to the amount absorbed. 



Hence the amount of water absorbed may be taken as a measure 

 of the amount of water transpired, certainly in view of the fact 

 that any slight error can be scarcely larger than that of any other 

 method applicable to the field. One such error which is known to 

 exist, but which is insignificant when one is dealing with relatively 



