5() Scientific Proceedings, Royal Dublin Society. 



bubbles in the conducting tracts. And if by any chance a bubble 

 does arise in one of the vessels or tracheids, it can only throw the 

 lumen of that vessel or tracheid, in which it has arisen, out of 

 function. For the free gas of this bubble cannot pass through the 

 wet walls, but is confined within the compartment where it was 

 formed. So the transpiration current is by no means brought to 

 a standstill, but is merely deviated from that one element of the 

 wood in which the rupture in the water (i.e. the bubble) has 

 occurred. 



So far, then, from the occurrence of bubbles being a difficulty 

 to our Theory of the Ascent of Sap, it seems to be a happy con- 

 firmation of it. For it furnishes a reason why natural selection 

 has favoured the development of conducting tracts provided with 

 very numerous longitudinal and transverse partitions, which in 

 themselves must act as obstacles to the free flow of water upwards. 

 This fact we pointed out clearly in our original paper. 



II. 



Quite recently another criticism on the Cohesion Theory has 

 appeared. It is published by Copeland in the Botanical Gazette. 1 

 His criticism is based on his own experimental work and on a 

 discussion of the work of others. 



The experiment on which Copeland lays most stress was one 

 which he set up with the intention of illustrating the Cohesion 

 Theory more strikingly than had been done before. The inter- 

 pretation of his results, however, has led him to believe that 

 this theory is inadequate. At the very outset, he admits that he is 

 unable to explain the facts observed in his own experiment ; and 

 it would appear that his experimental methods are by no means 

 free from error, nor is his way of interpreting them convincing. 



In this discussion of his work, I shall first describe his experi- 

 ment, as I understand it, and then try to show that his results are 

 not inexplicable ; and further, that the nature of the experiment 

 does not allow it to be used as fairly illustrating the Cohesion 

 Theory. 



A tube, 12*4 m. high and 3 mm. in diameter, filled with 

 plaster of Paris and water, and also containing some undissolved 



1 Bot. Gaz., Sept., 1902, 



