ATMOSPHERIC AIR AS THE FOOD OF PLANTS. 37 
lower part in a hole in a glass cover,* that was cemented 
air-tight upon the vessel. The stem itself was cemented 
by soft wax into the hole, so that the interior of the ves- 
sel was completely cut off from direct communication with 
the external atmosphere. The plant thus situated had its 
roots in an atmosphere as nearly as possible saturated with 
vapor of water, while its leaves were exposed to the ex- 
ternal air. After four days had expired, the entire appa- 
ratus, plant included, had lost 1.823 grm. Thereupon the 
plant was removed from the vessel and weighed by itself; 
it had lost 2.188 grm. The loss of the entire apparatus 
was due to vapor of water, which had escaped through 
the leaves. The difference between this loss and the loss 
which the plant had experienced could be attributed only 
to an exhalation of water through the roots, and amount- 
ed to (2.188 — 1.823) 0.365 grm, 
This exhalation of water into the confined and moist at- 
mosphere of the glass vessel is explained, according to 
Sachs, by the fact that the chemical changes proceeding 
within the plant elevate its temperature above that of the 
surrounding atmosphere. 
Knop, in experiments on the transpiration of plants, 
(V. St, VI, 255,) obtained similar results. He found, 
however, that a moist piece of paper or wood also lost 
weight when kept for some time in a confined space over 
water. He therefore concludes that it is nearly impossible in 
the conditions of such experiments to maintain the air sat- 
urated with vapor, and that the loss of weight by the roots 
is due, not to the heat arising from internal chemical 
changes, but to simple evaporation from their surface. In 
one instance he found that a portulacca standing over 
night in a bell-glass with moistened sides, did not lose, but 
gained weight, some dew having gathered on its foliage. 
* The cover consisted of two semicircular pieces of ground glass, each of 
which had a small semicircular notch, so that the two could be brought tapes 
by their straight edges around the stem. 
