1072 
where S is the salinity in parts per thousand. Over the 
oceans the salinity differs so little from 35 per mille 
that with sufficient accuracy e; = 0.98ea. 
The observed evaporation values could be repre- 
sented with good approximation by the equation, 
E = 0.40 P (i + 0.40w) (3) 
where H represents evaporation in centimeters per 24 
hours, and e, which enters into the term P, is measured 
in millibars, and w in meters per second. 
MARINE METEOROLOGY 
reduction factor. Somewhat arbitrarily he selected a 
height of 0.2 m above the sea surface, calculated what 
the evaporation from a pan at that height should have 
been, and assumed that a pan placed 0.2 m above the 
sea surface would have given values that would repre- 
sent the evaporation from the sea surface itself. The 
numerical values are based on only a few series of 
measurements of wind and humidity gradients carried 
out at low wind velocity and on assumptions, the 
validity of which is not obvious. It is therefore not 
Taste I. EnerGy AVAILABLE FoR EVAPORATION IN AREAS BETWEEN PARALLELS oF LATITUDE, AND AVERAGE EVAPORATION 
Rates DETERMINED BY DirrEeRENT METHODS 
Evaporation (mm day) 
Latitude range Ocean area (Qs — Or)* Rt Oct 1 
(deg) (1018 cm?) (cal cm™ day-?) (cal cem™ day) (cal) Extrapolation Obs. at sea From From met. obs. 
(Briickner) (Wiist) Qe (Jacobs) 
70°-60°N 5.6 45 0.45 31 593 0.55 0.35 0.52 Mike 
60°-50° 10.9 80 0.31 61 592 1.10 1.18 1.03 1.78 
50°-40° 15.0 113 0.21 93 590 1.80 2.11 1.58 2.00 
40°-30° 20.8 164 0.15 143 586 2.75 2.90 2.44 3.32 
30°-20° 25.1 239 0.11 215 583 3.55 3.50 3.69 3.56 
20°-10° 31.5 253 0.10 230 582 4.10 3.61 3.95 3.66 
10°-0° 34.0 239 0.10 217 582 4.35 3.20 3.73 3.11 
0°-10°S 33.7 218 0.10 198 582 4.35 3.36 3.40 
10°-20° 33.4 228 0.10 207 583 4.10 3.60 3.55 
20°-30° 30.9 229 0.11 206 585 3.55 3.36 3.52 
30°-40° 32.3 191 0.14 168 587 2.75 2.69 2.86 
°-50° 30.5 117 0.17 100 591 1.80 1.76 1.69 
50°-60° 25.4 67 0.20 56 594 1.10 0.71 0.94 
60°-70° 17.1 35 0.23 28 597 0.55 0.22 0.47 
* Radiation surplus according to W. Schmidt (see p. 1073). 
+ Bowen ratio, equation (8). Adjusted values from Jacobs (see p. 1074). 
tQ. = (Q: — Q,)/(1 + R). 
§ Latent heat of vaporization, taking average sea-surface temperature into account. (L = 596 — 0.563). 
In discussing the accuracy of the evaporation values 
observed on board ship, Wiist [29] arrived at the con- 
clusion that the values represent the actual evaporation 
from the pan with an accuracy better than +8 per 
cent. The pertinent question is, therefore, whether 
these measurements permit any conclusions as to the 
evaporation from the sea surface. This question was first 
dealt with by Schmidt [21], who showed that the 
evaporation from the sea surface was probably only 
half that from the pans. Later on, the matter was dis- 
cussed in great detail by Wiust [29, 30] and by Cheru- 
bim [5]. 
In attempting to reduce the pan values to values of 
evaporation from the sea surface, Wiist pomts out 
that: (1) the temperature of the sea surface deviates 
from that at the surface of the water in the pan, (2) 
the vapor pressure of the air at a short distance from 
the sea surface differs from that observed on board 
ship, and (3) similarly the wind velocity at a short 
distance from the sea surface differs from the wind 
velocity recorded on board ship. By means of special 
observations carried out in the Baltic in 1919 Wiust 
established the variation of the vapor pressure and 
wind with height above the sea surface and used the 
values thus obtained for a computation of a combined 
surprising that somewhat different approaches give 
different results. In 1920 Wist found a factor of 0.46, 
in 1931 Cherubim obtained a value of 0.58, and in 
1936 Wiist finally adopted the value of 0.53, which, as 
will be shown, gives evaporation values that agree 
remarkably well with those obtained on the basis of 
energy considerations. 
Results. From the evaporation measurements which 
have been carried out on board ship along different 
routes and by means of climatological data from which 
the evaporation on board ship can be computed using 
equation (3), Wist derived average values of the an- 
nual evaporation from all oceans between 70°N and 
70°S. These, based on a reduction factor of 0.53, are 
entered in Table I. Wiist estimates the probable error 
in his values to be +12 per cent. The average value is 
96 + 12 cm per year. 
Wiist claims only that the average values observed on 
board ship can be reduced to values of evaporation from 
the sea surface. He does not claim that the relationship 
between evaporation, evaporation potential (equation 
(8)), and wind velocity, which was established for the 
pan measurements, is applicable to the evaporation 
from the sea surface after multiplication by the factor 
0.53. It does not appear probable that the equation 
