1898] A NEW TRANSPIRATION MACHINE 347 
the latter are rather more convenient since there is only the top 
of the containing vessel from which evaporation must be pre- 
vented. Several attempts were necessary before a large enough 
tube was used on the float; 2. ¢., the apparatus was at first set 
up so as to be too sensitive. A slender tube is appropriate when 
the intervals of time are short; but it sank so fast that it reached 
bottom within a few hours. The results introduced below were 
obtained by the use of a tube whose cross section was 54%"; 
one, therefore, which would sink 8™ while the plant lost 5°. 
in weight. As the areometer sank, the water it floated in rose 
a very little, but this is not a source of error because the water 
was in the same vessel when the value of a unit of movement 
was determined. I did not guard against evaporation from this 
water, as would be advisable if the clover plant instead of the 
apparatus were the real subject of experiment.’ For the same 
reason, the temperature and relative humidity do not concern 
us here. The plant was a red clover with about ten leaves, 
grown in soil, but transferred to water two days before the 
experiment began. 
The data are derived from an experiment on June 15 and 
16, and are given in measurements of vertical distances on the 
smoked cylinder, and in grams of water. Hours in light-face 
figures are a.m., those in bold-face figures P.M.: 
How | Reow | Weict  om | Beam | We 
9-10 25.6 0.800 1-2 2.9 0.091 
I0O-II 20.3 0.635 2-3 a7 0.116 
I-12 14.6 0.456 3-4 pe 35 papinbd 
ti aa | 22:3 0.697 4-5 3.9 0.122 
| ae | 29.5 0.922 5-6 4.6 0.144 
2-3 4.0 0.125 6-7 14.2 . 0.444 
3-4 3.8 0.119 7-8 22.9 6.716 
45 4.0 0.125 8-9 26.6 0.831 
5-6 5.9 0.184 g-I0 3LI 0.972 
6-7 4.2 0.131 10-11 23.0 0.719 
7-8 3-9 0.122 II-12 9.9 0.309 
8-9 3-7 0.116 12-1 8.7 0.272 
9-10 3.0 0.094 I-2 11.9 0.372 
10-11 2.7 0.084 2-4 15.2 0.473 
II-12 23 0.072 4-5 6.3 0.197 
12-1 3.0 0.094 5-6 5.5 _— 
etc —_—, 
