Dixon — The Cohesion Theory of the Ascent of Sap. 55 



through it is at least thirteen times as great as that of wood. 

 Taking into consideration the presence of heart-wood and the 

 progressive clogging alluded to, we may, with greater probability, 

 assume that the effective wood of Taxus baccata is at least 15-20 

 times more permeable than plaster of Paris. 



The narrow bore of the tube in Copeland's experiment would 

 emphasize this property of plaster of Paris. For, if any local 

 differences of pressure arose, the rate of equalization of pressure 

 would be proportional to the square of the diameter of the tube. 

 So that, with a tube only 3 mm. in diameter, and filled with 

 piaster of Paris, local differences of pressure would be maintained 

 for long periods. 



In order to arrive at a more definite idea as to how great will 

 be the flow of water through Copeland's tube under the difference 

 of gas-pressure he observed — viz. 122 mm. of mercury — we will 

 apply the results obtained in the experiment just quoted. The 

 amount of water passing through the tube will be proportional to 

 the pressure urging it, inversely proportional to the length, and 

 directly proportional to the square of the diameter. In the 

 experiment just quoted, the pressure was a head of 35 cm. of 

 water ; the length 5 cm. ; the diameter 1*2 cm. ; and the amount 

 of water transmitted in twenty-four hours was l - 54 c.c. In 

 Copeland's experiment the pressure was equal to a head of 122 mm. 

 of mercury = 165*92 cm. of water ; the length 840 cm. ; and 

 the diameter 0*3 cm. Therefore the amount transmitted in 

 twenty-four hours 



165-92 x 5 x (-3) 2 



35 x 840 x (1-2) 2 

 •0027 c.c. 



x 1*54 c.c. 



In other words, with the pressure observed, it would take over 

 a year to transmit 1 c.c. 



We now come to consider how differences of pressure may 

 have arisen in Copeland's tube. 



Evaporation at the beginning of the experiment may have 

 been in part responsible for pressure-differences ; but, owing to the 

 impermeability of plaster of Paris, it seems improbable that its 

 influence would have been appreciably felt at lower levels in the 

 tube. 



