88 TIMBER AND SOME OF ITS DISEASES. [CHAP. 



radial walls) from tracheide to tracheide, would go by 

 steps spirally round the stem as up a spiral stair-case. 

 It is this last-mentioned position which renders pos- 

 sible an ascent of the water from lumen to lumen 

 of the tracheides. 



Now, as a mere matter of observation, determined 

 by finding the quantity of water, &c. in all parts of 

 the wood at regular heights in the tree, the tracheides 

 always contain some water and some air, but the 

 quantities of each differ considerably both according 

 to the part of the tree and according to the 

 season. In all cases the walls must be saturated 

 with imbibed water. In the Dicotyledons, &c. the 

 liquid water in the lumina of the vessels and tracheides 

 occupies at least one-third and often two-thirds of the 

 volume of the lumina : in the Conifers, the tracheides 

 may have only two-thirds of the lumina occupied by 

 water, but as much as nine-tenths may occur. The re- 

 maining portions of lumen are filled with air, and 

 perhaps the most valuable of Hartig's contributions to 

 the question was his patient and ingenious determina- 

 tion of the fact that the actual air-contents of the tra- 

 cheides, &c., decreases from below upwards : that is to 

 say, if we suppose the air to be at the same pressure 

 throughout, the amount of water in the lumina of the 

 tracheides, &c increases as we ascend the tree. 



