Chap. III.] DOWNWARD GROWTH OF THE ROOT. 
109 
placing three chimney-pots one on the other. 
In the autumns of 1850, 1851, 1852, and 1853, 
I knocked off each year about a foot of the upper 
part of the chimney-pots, and denuded the roots 
to that extent: they are, however, branching. 
The ends of the branches, on reaching the sides, 
had apparently died, and new shoots from them 
had struck downward. There was not the least 
circular growth round the sides, as with plants 
in flower-pots; possibly, because the chimney¬ 
pots were encrusted with soot. Two of the 
upper roots were entirely cleared from the earth 
in 1850. The woody parts of them are still 
alive (1853), though it would puzzle us to say 
whence their wood is supplied with sap: I sup¬ 
pose, from the stem; if so, the upward sap, in 
this case, is a downward sap. 
The “herbaceous envelope” of these roots 
(that is, the surface of the bark immediately 
below the outer bark) is perfectly green where 
they are exposed to light and air, and perfectly 
white from the very spot where the earth pro¬ 
tects them from the light and air. When de¬ 
nuded, the herbaceous envelope begins to turn 
green in about a fortnight.* 
* I have omitted to observe whether the under bark of 
the old roots of plants grown in water is green or white ; if 
