Dixon and MA&SHA^.h—Exmninafion of the Wood of Trees. 363 



was made iu November, when the fresh-cut wood is quite dry, and contains 

 so much air that it floats in water, and since no precautions were taken to 

 prevent the formation of air-bubbles in the trachesa on cutting, it is certain 

 that this fresh bit of wood contained at least the normal amount of bubbles. 

 In the second experiment the area of the cross-section of the wood was 

 2"35 sq. em. According to the measurements given in Table I this would 

 include 0'66 sq. cm. of cells, of wliich 0'33 sq. cm. miglit be engaged in raising 

 water. We would have to assume protoplasmic streaming tlirough this cross- 

 section with a velocity of 0"970 cm. per minute, merely to overcome the 

 downward percolation, which amounted to 0'321 g. per minute. Such a velocity 

 is 32 times that observed by Janse, and is, of course, quite improbable. 



Table II. 



Cotoneaster frigida. 



In Table II are given the determinations of the areas of the cross-sections 

 of the elements of the wood of Cotoneaster frigida, arrived at in the same 

 manner as those for Acer jjseiidoplatanus. 



The " leak " for this wood was also determined in the fresh and in the 

 injected condition, and the amount of downward percolation through a piece 

 of stem having a cross-section of 1 sq. cm. under unit head was found to be 

 0'080 g. per minute, for a similar piece with the same cross-section, but 

 injected, was 0"103 g. under the same conditions. Of the total cross-section 

 the effective area of the cells would be 0*037 sq. cm., and to overcome the 

 leak in the fresh wood protoplasmic streaming having a velocity of 2'173 cm. 

 per minute would have to be assumed, and in the case of the injected 

 2'788 cm. per minute. Such a velocity is again of course quite inadmissible. 



It should be noticed that in the piece of Cotoneaster stem examined, 

 many of the vessels, perhaps one-sixth of all, were choked with a pale-brown 

 transparent substance. If this had been absent, a still greater rate of " leak " 

 would have been observed. 



