Chap. I.] BY THE SURFACE OF THE ROOTS. 
29 
imbibe from all mature parts which are exposed 
to moisture, and excrete from all parts which are 
exposed to drought. 
The radishes had been grown in very light 
mould, so that, by passing the fingers into it in 
taking the radishes up, I believe not a rootlet, 
main or lateral, was broken. The whole of the 
roots were left on No. 3., and much earth ad¬ 
hering to the side-roots of some. The weather 
was very favourable for the first week, the wind 
south or south-west, the thermometer about tem¬ 
perate, with constant heavy showers. • For this 
reason, none of the experiments were watered for 
the first fortnight. On the 21st the wind got to 
the north, and the weather was sunny and dry. 
But, ere this, nearly all No. 3. were defunct. 
On the 29th, after a week’s drought, some of the 
leaves of Nos. 1. and 2. looking flagging, the ex¬ 
periments were all watered. Care was taken that 
the water should fall only on the earth, without 
touching a leaf, in order to make certain that the 
revival (which took place) was caused solely by 
the absorption of moisture by the rootless bulbs, 
— nay, in the case of No. 2., by the rootless 
HALF-bulbs. 
On the 28th of May, when the experiments 
were between six and seven weeks old, I turned 
