Surron—Study of Evaporation from Water-Surfaces. 139 
press firmly into the rubber and so prevent any access of air 
below. The water-vessels are silver-plated cups set into holes 
bored into the block of wood. These cups are 23 inches deep 
inside, about 22 inches diameter near the bottom, and about 23 
inches diameter near the top; their rims are turned out to a 
diameter of 2? inches. The holes are about 24 inches diameter, 
and 24 inches deep. Thus the rims of the cups project about 
4 inch above the base; the water-surface being therefore nearly 
level with the base. For comparison, a similar cup (called here 
the “open cup”) is hung by a thin wire from the top of the 
apparatus. The loss by evaporation was determined by weighing. 
At 8 a.m. each day the cups were nearly filled with clear rain- 
water, the weight of the cup and water together being nicely 
adjusted in a sufficiently delicate balance to 3750 grains, of which 
the cups themselves averaged about 1500 grains. The tempera- 
ture of the water in the cups was carefully observed with a 
delicate thermometer at 5, 8, and 11 pm., and 8 am., and 
occasionally also at ll a.m. and 2 p.m.; and at the same times the 
loss by evaporation was determined. The open cup was retained 
in the same position throughout the experiment, but the other 
cups were permuted from day to day from one pipe to another, so 
that individual differences in the cups might be smoothed out in 
the averages.* 
Since it is likely that the evaporation between 8 a.m. and 
5 p.m. would not be quite the same when the temperature and 
evaporation were only read at 5 p.m. as it would be if these 
quantities were read also at 11 a.m. and 2 p.m. (on account of the 
disturbance possibly caused by taking the temperature of the 
water and removing the cups from the stand to weigh them), a 
separate series was formed for the more frequent readings. We 
thus have one series of observations made upon 15 days, with 
readings at llam., 2 p.m., 5 p.m., 8 p.m., 11 p.m., and 8 a.m. ; 
five series, averaging 23 days each, with readings at 5 p.m., 8 p.m., 
11 p.m., and 8a.m.; and there is also a series of observations 
made upon 5 days in which the readings were only taken at the 
end of the 24 hours, namely, at 8 a.m. 
We have also nearly simultaneous observations made with the 
Piche atmometer; with the tub at 11 a.m., 2 p.m., 5 p.m., 8 p.m., 
1 It has not been thought necessary to reduce grains to grams, nor degrees F. to C. 
T 2 
