94 Ewart and Rees .— Transpiration and the Ascent of 
optimal conditions. In the case of the tree A the area of the stem at the 
narrowest point below the main branches was 1,256 sq. mm. and the vessels 
averaged 29 per sq. mm., so that the total number in the stem at that 
point was 36,440. The diameter of the largest was 0*13 mm., and the 
average diameter 0-083 mm. Hence the total area of cross-section of the 
vessels would be 1*93 sq. cm. At a rate of 1,230 cm. per hour the vessels 
if filled with water would discharge nearly 2,361 c.c. per hour, whereas the 
actual amount of water absorbed by the tree was 920 c.c. in twenty minutes. 
At the base of the stem, however, there were approximately 55,800 vessels 
with a discharging rate of over 3,600 c.c. per hour. The eosin was fairly 
evenly distributed throughout the wood from the base to half-way up, and 
hence not only were the whole of the vessels conducting, but their effective 
conducting area was little or not at all reduced by the presence of air in 
them, and the tracheides also may have played some part in the conveyance 
of water. 
Even in the case of a capillary vessel occupied by a continuous column 
of water and with perfectly smooth internal walls a rate of flow of 10-12 
metres per hour entails considerable friction and resistance. Thus, in 
a tube 0-008 cm. diam. and to metres long the frictional resistance would 
be in the case of water at 20° C. equivalent to a head of 1,250 cm. of water 
approximately, or slightly over an atmosphere. Hence in such a case the 
viscosity factor alone more than doubles the head due to the height of the 
tree, and to this must be added the resistance due to the transverse 
partitions, to the irregularities on the internal walls, and to the usual 
presence of more or less air. 
The Diameter and Length of the Vessels in Branches and 
in the Main Trunk. 
The experiments detailed below were performed by driving mercury 
into the stem under pressure and noting when it escaped from one or more 
vessels as pieces were cut from the distal end. For the most part each 
experiment meant the sacrifice of an entire young tree, and the quantity of 
mercury used was naturally great. For demonstration purposes, however, 
it is sufficient to use air driven into the stem under pressure from a cylinder 
of compressed air, and if the wood is fresh and the trunk quite sound this 
gives as accurate results as the mercury method for determining the length 
of the longest vessels (Exp. II with Acacia mollissima), since air only 
escapes freely from vessels uninterrupted by partition walls, even when 
a pressure of two or three atmospheres is used. The mercury method, 
however, enables the open vessels to be counted until the number increases 
considerably, whereas with air, as soon as more than three or four are 
opened, it is difficult to count the number further back with any accuracy. 
