Dixon — Note on the Role of Osmosis in Tvanspiration. 769 



due to an alteration in surface-tension owing to the presence of the 

 salt. To investigate this question experiments were made in which 

 the tm^gescence of the cells was destroyed without introducing any 

 foreign salt but by the action of heat. A small branch of Helianthus 

 multijlorus was immersed in water at 75°-80° for ten minutes, and 

 then in a very flaccid condition set with its lower end in a vessel 

 of water and exposed to conditions favourable to transpiration. 

 Quickly the surfaces of the wetted leaves dried, and within an 

 hour the edges began to roll up owing to the loss of water, while a 

 control branch of similar dimensions and under the same conditions 

 except that it had not been exposed to the high temperature remained 

 tiu'gescent and fresh. At the expiration of an hour the lower ends 

 of both branches were transferred into a watery solution of eosin. 

 After ten minutes the eosin was plainly visible in all the veins of 

 the leaves of the control branch, but not so in the other. The control 

 branch soon began to flag, and two hours afterwards when the branch 

 which had been immersed in the heated water was examined it was 

 found that the eosin had risen only a short way up the stem and 

 penetrated only into the lower leaves, while in the control branch 

 even the uppermost leaves showed the eosin in the finest veins. 



In order to avoid the contact of the liquid water with the surface 

 of the evaporating cells other experiments were made in which the 

 branches to be experimented were kept, while their cut ends were 

 supplied with water, in a damp chamber raised to a temperature of 

 80°-90°. After they had remained 10-20 min. in the heated 

 chamber they were taken fi'om it, and their rate of transpiration 

 measured and compared with that of a similar control branch. The 

 rate of transpiration was estimated by allowing them to draw the 

 transpired water from a weighed vessel of water from which evapo- 

 ration was prevented by means of a cork fitting closely round the 

 base of the branch and the neck of the vessel. At the end of a 

 certain time the vessel was again weighed, and the difference in 

 the two weighings gave approximately the amount of water tran- 

 spired. 



Prom a number of experiments it appeared that the control branch 

 as long as it remained fresh and turgescent transpired twice to four 

 times as much water as the branch which was raised to the high 

 temperature ; both branches being of course as far as possible similar. 

 If however the experiment is prolonged it is found that when the 

 control branch becomes drooped, it transpires only about half as much 

 water as the other. This sudden diminution in the amount of water 



