Il2 THE PLANT 
and he has cited-the work of Kréber, and of Eberdt in proof. This state- 
iment holds, however, only for short periods of a few hours, or more rarely, a 
day, and even here its truth still remains to be conclusively demonstrated. 
The discrepancy between absorption and transpiration for a short period is 
often greater than for a longer time, but it is evident that a transient change 
in behavior or a small error in the method would inevitably produce this 
result. Eberdt found the discrepancy for a few hours to be 1-2 ccm. in an 
entire plant of Helianthus annuus, while for a whole day the. water absorbed 
was 33.57 ccm. and the water lost 33.98 ccm. _Krdéber’s experiments with 
cut branches of Asclepias incarnata showed a maximum difference for 12 
hours of 2.5 cem., but the discrepancy for the first 24 hours was I ccm. 
and for the second 1.9 ccm. In both cases, the potometer was employed. 
Consequently, as will be shown later, Eberdt’s results are not entirely trust- 
worthy, while those of Kréber, made with cut stems, are altogether unre- 
liable. Hence, it is clear that the discrepancy is slight for a period of several 
days or weeks, and that it may be ignored without serious error, except in 
a few plants that retain considerable water as cell sap, in consequence of 
extremely rapid growth. Accordingly, the amount of transpiration, which 
may be readily and accurately determined, can be employed as a measure 
of absorption that is sufficiently accurate for nearly all purposes. The 
truth of this statement may be easily confirmed. It is evident that the 
amount of water absorbed equals the amount transpired plus that retained 
‘by the plant as cell-sap, or used in the manufacture of organic compounds. 
In plants not actively growing, the amount lost equals that absorbed, as 
already shown in the experiment with Helianthus. According to Gain’, 
Dehérain has found that a plant rooted in-ordinary soil transpired 680 kg. 
of water for each kilogram of dry substance elaborated. In Helianthus 
annuus, the dry matter is Io per cent of the weight of the green plant. A 
well-grown plant weighing 1,000 grams, therefore, consists of 100 grams 
of dry matter and goo of water. The iength of the growing period for 
such a plant is approximately too days, during which it transpires 68 kilo- 
grams of water. Assuming the rate of transpiration and of growth to be 
constant, the plant transpires 680 grams daily, adds 9 grams to its cell-sap, 
and 1 gram to its dry weight. The amount of water in a gram of cellulose 
aud its isomers is about'3/5. Consequently, the total water absorbed daily 
by the plant is 689.6 grams. The 680 grams transpired are 98.6 per cent 
of the amount absorbed; in other words, only 1.4 per cent of the water 
absorbed is retained by the plant. From this it is evident that the simplest 
1Recherches sur le Réle Physiologique de !'Eau dans la Végétation. Ann. Nat. 
Sci., 7:20:65, 1895. 
