A NEW EVAPORATION FORMULA 7 
on the non-routine parts by Dr. John F. Hayford, amounts to over 32,000 man-hours 
of labor. A total of 41 persons has been engaged on the computations from time to 
time since the investigation began in 1913. Of these, not more than 10 persons have 
been engaged upon it at any one time. 
The preceding statements indicate that this was an extensive investigation. 
It is probably safe to say that there have been few times, if any, when an equivalent 
amount of time and energy has been expended in the derivation of an evaporation 
formula, including the by-products, the method of evaluating the effects of winds 
and barometric pressures on free, open surfaces of water, and including the studies 
toward formulating the laws of stream-flow. 
OUTCOME OF THE INVESTIGATION 
The outcome of this part of this investigation may be briefly stated as follows : 
(1) An evaporation formula has been derived, based upon over 700 observa- 
tions independently on each of two of the Great Lakes, Michigan-Huron and Su- 
perior. This formula enables one to estimate the daily rate of evaporation from 
any free, open surface of water (such as from the surface of a lake, reservoir or 
river) anywhere in the world, in terms of the meteorological elements of air tem- 
perature, vapor-pressure and wind velocity as ordinarily observed at the standard 
U. S. Weather Bureau stations. 
(2) An estimate has been made of the constant part of the run-off into each of 
Lakes Michigan-Huron and Superior from their respective land draingage areas, 
and certain general knowledge with reference to the variable parts of the run-off in 
each case has been obtained. 
(3) As a direct sequel to the results already published in Publication No. 317, 
and as a necessary precursor of Item (1), the following two items of knowledge have 
been derived: 
(a) A reasonably accurate numerical expression has been obtained for the 
effects of barometric pressures on the elevation of the water surface at the Marquette 
station on Lake Superior. With this expression one may compute, from the dis- 
tribution of the barometric pressures ordinarily shown on the forecast maps of the 
Weather Bureau, the disturbances in elevation of the water surface thereby pro- 
duced at Marquette. 
(6) The numerical constants (the S's) in the general formula for wind effects 
stated as equation (69), page 63, Publication No. 317, have been derived for the 
Marquette station on Lake Superior. These constants enable one to compute the 
hourly or daily effect of a wind of any velocity and direction upon the elevation of 
the water surface at Marquette, Michigan. 
(4) The relationship of the new knowledge gained in (1) and (2) to various 
important problems in science and engineering has become increasingly evident. 
Some of these problems are : 
(a) The problem of estimating the evaporation from the surface of each of the 
Great Lakes as a basis for more accurate estimates of the hydrological phase of the 
problem of regulating the surfaces of the lakes by controlling works in their con- 
necting channels, and the use of this available knowledge of the daily evaporation in 
such regulation after the controlling works are constructed. 
