156 BULLETIN 1059, U. S. DEPARTMENT OF AGRICULTURE. 
tube dry. Care should be taken to have the filter paper wetted and 
adhering closely to the glass before the instrument is read and left. 
It is unnecessary to calibrate these instruments, because of the fre- 
quent changing of the filter papers and the fact that papers of one 
grade may be quite uniform in their capillary properties. Beyond 
this, the evaporation rate is somewhat controlled by the adjustment 
and the degree to which the paper is wetted. 
The Piche evaporimeter is not now considered so desirable an in- 
strument as some of the other types. Though it is simple and fairly 
easy to operate under most circumstances, it is fragile and hardly 
suited to severe weather conditions ; the adjustment of the feeding for 
changes in the weather is always vexatious and sometimes beyond 
one's power ; a correction for rainfall is out of the question ; and the 
technical point may be raised that the moist surface is somewhat too 
freely exposed to wind, while the white filter paper absorbs only a 
small proportion of incident radiation. The conditions for evapora- 
tion are therefore very different from those within the leaf. 
Porous-cup atmometer. 
The name " atmometer," while describing any instrument for the 
measurement of evaporation from a moist surface, is usually asso- 
ciated with evaporimeters of the porous cup type. A very satisfac- 
tory field instrument of this type has been described in several 
papers by Livingston (159), and may be obtained on the market 
either with or without standardization. Only standardized instru- 
ments should be used in comparative studies. The instrument con- 
sists of a closed cup with porous walls, into which the water is fed 
at any desired pressure by regulating the height of the reservoir. 
The reservoir, connected with the cup by rubber tubing, may be a 
flask of any size or a graduated tube. In the former case, the 
amount of evaporation in any specified period may not be deter- 
mined directly, but rather by measuring the amount of water re- 
quired to fill the flask to its original level. This feature is incon- 
venient, but the use of a large flask so increases the possibilities of 
the instrument for long-period observations that it is, in this re- 
spect, far superior to the Piche evaporimeter. 
The moisture of the cup reaches the outer surface through the 
porous walls, its rate of movement being determined by the internal 
pressure and also by the difference in capillary tension caused by 
the loss at the outer ends of the pores. Presumably capillary 
movement is sufficiently rapid to maintain the supply at the outer 
surface at any reasonable rate of evaporation. Yet there must be 
a limit to this capillary action in both directions, and for this rea- 
son the movement must, under extreme conditions, be governed 
