430 



THE AMERICAN NATURALIST 



[Vol. XL 



one liter from the whole tree. Since this is distributed over the 

 whole tree, and the numerous vessels expose an immense surface 

 to the wood fibers, an almost complete impermeability would be 

 necessary to affect the pressure markedly. But the gas expansion 

 must later be supposed to overcome this same resistance. Con- 

 sidering the weak pressure of the gas and its compressibility it 

 would be impossible for more than a small fraction of the whole 

 volume of gas expansion to be transmitted to the tap-hole. The 

 volume of flow is only barely accounted for by the total gas ex- 

 pansion, hence under these conditions only a small fraction of 

 the flow would occur. Again, if gas expansion in the fibers causes 

 flow, the latter would rise to a maximum slightly after the air 

 temperature. The water expansion in the vessels would tend, 

 however, to cause maximum flow at maximum pressure. The 

 actual maximum would be a resultant of the two, so the curve of 

 flow would reach its maximum later than the pressure curve and 

 fall much more slowly. This is not the case with curves prepared 

 from tables in the Vermont report. The curves of pressure and 

 flow are almost exactly coincident. 



It seems evident therefore that this theory also must be laid 



Combination of Wood-kxi-w^iox and (i.\s TtiKoinKs 



It has been suggested that although the expansion of the wood 

 is not sufficient to account for the total flow of sap, still, if com- 

 bined with the gas expansion, the two together might account for 

 both pressure and flow\ Granting that the vessels are nearly if 

 not quite saturated with water and that the gas is mainly within 

 the wood fibers, both of which conditions seem to be true, then 

 when the tree warms, the conditions would l>e as follows. The 

 rapid warming of the outer layers of wood when tiie sim Hivst falls 

 upon them in the early morning would ])r(Mhi(c pressure as aheady 

 outhned. "With the vessels saturated, this j)ressure uiioht l)e very 



the sap would be constantly but slowly filtering through the walls 



