A NEW EVAPORATION FORMULA 119 
may be seen by a comparison of the values derived in Solution V x with V 2 and 
BB t with BB 2 . The only difference between these solutions was that Solutions 
Vi and BBi contained the run-off terms, —Riri—R 2 ri — R 3 r, . . . — R t r,, and 
Solutions Vi and BB 2 did not. The comparison is given in the accompanying 
tabulation, from which it may be seen that the change in these constants caused by 
ignoring entirely the variable part of the run-off into the lake is of the order of their 
own probable errors. Although these changes appear to be systematic with refer- 
ence to each lake, they are essentially accidentally distributed with reference to 
both lakes. 
SUMMARY OF GENERAL CONCLUSIONS TO PART I 
The principal conclusions are here briefly summarized for the convenience of 
anyone who wishes to get a general view of the results of this part of the investiga- 
tion. Page references enable those interested to look up the basis of particular 
conclusions. 
The best expression of the laws of evaporation from large water surfaces under 
natural conditions which could be determined in this investigation is equation 
(22) (page 82). 
The probable errors of the constants in that equation were computed directly 
from the residuals on the assumption that all the errors involved are accidental in 
character. As there is some evidence of systematic errors, these probable errors 
should be slightly increased to represent the actual uncertainties in the values 
(pages 85 to 90). 
As indicated by the probable errors, the evaporation on the lake on any day 
may be computed from the formula developed in this investigation with an error 
of about =^10 per cent (pages 121 and 122). 
The evaporation from the surface of Lake Michigan-Huron apparently varies 
between the limits 0.000 foot and 0.029 foot of depth on the lake area per day, with 
a mean value for the months May to October, inclusive, of 0.0054 foot of depth per 
day. The equivalent of 0.001 foot of depth per day on Lake Michigan-Huron is 
14,650 cubic feet per second (pages 84-85). 
Wind velocities below 10.8 miles per hour or 260 miles per day are too feeble 
to have an appreciable effect upon evaporation from the surface of water, ground 
or vegetation. The wind velocity here referred to is that as measured about 100 
feet above the ground or water surface (pages 90 to 99). 
For winds above 260 miles per day, evaporation is a linear function of wind 
velocity, with increasing evaporation corresponding to increasing wind velocity 
(pages 98 to 102). 
The rate of increase of evaporation from the surfaces of the Great Lakes with 
increasing wind velocity for winds above 10.8 miles per hour, as measured at the 
Weather Bureau Stations, is many times greater than the corresponding rate of 
increase as determined from observations on small-scale apparatus (pages 122 to 130). 
The run-off into the lake when the ground-water is at its average level, and 
the change in storage in the ground, i.e. (rainfall on land) minus (evaporation from 
land) minus (run-off), has been zero for a long time, is equivalent to 0.006 foot of 
depth per day on the area of Lake Michigan-Huron, or 87,900 cubic feet per 
second, and 0.004 foot of depth per day on Lake Superior, or 41,400 cubic feet per 
second (pages 16 to 19, 117 to 118). 
