Ascent aiul Descent of Water in Trees. 119 



Traces of copper were, however, detected in the outer wood at the 

 top of the tree, after incinerating and treatinp: the residual ash with 

 a drop of nitric acid and an excess of auniionia. At the same time, 

 however, it was found that the posin solution had also travelled down 

 the stem, and it was traced out for distances up to 7 feet along the 

 lateral roots, up to roots less than | inch in diameter. Judging from 

 the depth of colouration, a greater flow had taken place down the 

 stem than up it from the point of section. We are so accustomed to 

 think of the sap as always ascending, that this result seems at first 

 sight somewhat surprising. As a matter of fact, tne water would flow 

 more easily downwards, particularly if the central intact wood cylinder 

 exercised a suction upon the roots below the point of injury. The 

 fact that this is possible, however, makes this method of investigation 

 unsuitable fur application to the largest trees, unless two conditions 

 can be fulfilled, which are --(a) it must l)e possible to cut away the 

 whole of the alburnum, leaving only perfectly non-conducting 

 duramen ; (b) the lower cut surface must be blocked with wax or some 

 siuiilar substance. 



The total amount of liquid absorlied by the upper and under cut 

 surfaces from 10 a.m. to 4.30 p.m. on the first day, was no less than 

 33-i litres, or 5.1 litres per hour. A large proportion of this, how- 

 ever, may have been absorbed either to bi-ing the wood to full satura- 

 tion point, or a portion may have actually oozed out from the roots 

 into the soil. That pressure applied to a cut stem will cause water 

 to exude out through the roots, is easily shown, and the pressure of 

 continuous columns of water in the tallest trees, if it were trans- 

 mitted directly to the absorbing roots would certainly cause an out- 

 ward flow through them. Whether the forces preventing this are 

 derived solely from the suction of the leaves, or are partly the result 

 of agencies acting in the stem along the path of the transpiration 

 current, is still an open question, to which the foregoing experiment 

 affords no decisive answer. 



