Jan. 3,1916 
Hourly Transpiration Rate on Clear Days 
643 
An inspection of the curves in figure 21 will show that the computed 
graph agrees with the observed transpiration graph much better in the 
morning than in the afternoon. 1 The computed graph always reaches its 
maximum in advance of the observed graph. The greater departures 
occur during the early afternoon and early evening. The agreement is 
by no means as good as is to be desired, and the graphs show clearly 
that transpiration can not be completely accounted for on the assumption 
7ransp/netion computed from vertice/ relation Transpiration computed from rerti'ca/ racf/ation 
an<f temperature. anct saturation-c/efsc/f 
Fig. 21.—Graphs showing the observed transpiration with that computed from vertical radiation and 
temperature (on the left) and from vertical radiation and saturation deficit (on the right). 
that the vertical component of radiation and the rise in temperature 
are the controlling factors. 
The relative values of the computed coefficients are of interest. In 
the case of alfalfa, the radiation is weighted 0.97 relative to temperature; 
amaranthus, 1.23; and rye, 0.60. In this connection it should be recalled 
that rye shows a sudden change in the slope of the transpiration graph in 
the morning, differing markedly from alfalfa and amaranthus in this 
respect. 
1 Since preparing figures 21 and 22 a recalculation based on more exact determinations of the vertical 
component of radiation has given computed values of transpiration and evaporation which are in some¬ 
what closer agreement with the observed values during the daylight hours than those indicated in the 
charts. The coefficients in the equations are based upon the revised calculation. 
