eller 37 Procent om Natten. Luftens Temperatur var 
ce. 25°. Vindhastigheden liden — 3 til 4 Meter pr. Sekund. 
Fordunstningens Hurtighed afhænger af det fordun- 
stende Vands Temperatur, dets Saltholdighed, Luftens 
Fugtighed, Lufttrykket og Vindens Hastighed. I Fordunst- 
ningsmaaleren kan Temperaturen vanskelig holdes saa lav, 
som den er i Hayoyerfladen, og Luftens Temperatur er 
ogsaa i Regelen højere ved Apparatet end over Havet. 
Vandets Saltholdighed er, praktisk taget, den samme paa 
begge Steder, da Søvandet skiftes 1 Fordunstningsmaaleren 
hver Dag. Luftens absolute Fugtighed og Lutttrykket er 
ogsaa ligestort paa begge Steder. Men Vindens Hastighed 
er forskjellig, dels fordi Skibet, paa hvilket Apparatet staar, - 
bevæger sig gjennem Vandet, dels fordi Vmden paa Deæk- 
ket altid er svagere end paa Sven. Denne Omstændighed 
gjør, at den ombord maalte Fordunstning i Regelen vil 
være mindre end Fordunstningen over selve Havfladen. 
Og hertil kommer endvidere den Omstændighed, at Hay- 
fladen med sine Bølger og end mere med Sprøjtet af disse 
frembyder en i Forhold til den rolige horizontale Vand- 
flade, saaledes som den er i Fordunstningsmaaleren, i hoj 
Grad forøget Fordunstnings-Overtlade. Den med vore Ap- 
parater maalte Størrelse af Havvandets Fordunstning kan 
derfor ikke repræsentere den fulde Fordunstning paa Ha- 
vets Overflade, men har i det højeste kun en relativ Verdi. 
De Forsøg, som vi have gjort paa at beregne de 
ovenfor meddelte Fordunstningsforsøg efter Weilenmann's 
Formler*), have ikke givet noget gyldigt Resultat. For 
1877, i hvilket Aar vi havde den roligste Sø, stemmer 
Formelen bedst med Maalingerne, men der bliver en Mid- 
delafvigelse mellem de (for 12 Timer) beregnede og obser- 
verede Værdier af Fordunstningshøjden af + 0.6. Som 
ovenfor omtalt, ansloges, efter den Noiagtighed, hvormed 
Apparatet kunde indstilles paa Merke, denne Størrelse til 
+ 01, og de to forskjellige Apparater stemme i sine 
Resultater paa + 0.””87. For 1878 kom Coeffieienten for 
det Led, der indeholder Vindens Hastighed, ud med nega- 
tivt Fortegn. For Observationerne fra Vestindien fandtes 
Constanterne ganske forskjellige fra de for 1877 fundne og 
en Middelafvigelse mellem observeret og beregnet Fordunst- 
ningshøjde for 12 Timer af + 1.775, De Betingelser, un- 
der hvilke Apparatet har virket og de Blementer, med 
hvilke Beregningen er gjort, synes saaledes ikke at stemme 
med Formelens Forudsetninger. Naynlig skal jeg i den 
Anledning bemerke, at ombord 1 ,,Vørmgen* maaltes den 
i Formelen benyttede Vindhastighed oppe paa Hyttedækket, 
medens Fordunstningsmaaleren stod meget mindre udsat for 
Vindens fulde Styrke. De paa begge disse Steder sted- 
findende Vindhastigheder staa heller neppe altid i samme 
Forhold til hverandre. TI 1878 var Vejret ofte meget 
uroligt, og Apparatet bevægede sig ikke rigtig godt i sine 
1) Schweizerische Meteorologische Beobachtungen 1875. 
4""90, or 63 per cent during the day (8 am. to 8 
p.m.), and 2.”"83, or 37 per cent during the night. The 
temperature of the air was about 250 C. Velocity of the 
wind inconsiderable, 3 to 4 metres per second. 
The rate of evaporation is dependent on the tempera- 
ture of the evaporating water, the amount of salt, the 
humidity of the air, the pressure of the air, and the velocity 
of the wind. Inthe atmometer, the temperature can hardly 
be kept so low as it is at the sea-surface, and moreover, 
the temperature of the air is as a rule higher around the 
apparatus than above the sea. The proportion of salt in 
the water is, practically speaking, the same for both, the 
sea-water in the atmometer having been changed every 
day. The velocity of the wind. however, is different, partly 
since the ship in which the apparatus is mounted moves 
through the water, and partly because the velocity of the 
wind is less on deck than on the sea-surface. Owing to 
this circumstance, the evaporation as measured on board 
will generally be less than the evaporation going on at the 
surface of the sea. And moreoyer, the surface of the ocean. 
by reason of the waves, and more especially of the spray, 
presents, as compared with the calm, horizontal surface of 
the water in the atmometer, a much greater evaporating 
surface. The evaporation of-sea-water. as measured with 
our apparatus, cannot, therefore, represent the full amount 
of evaporation at the sea-surface, but has, at most, only a 
relative value. 
Our computations, made according to Weilenmann’s 
formule,! of the evaporation experiments described above. 
did not give a trustworthy result. For 1877, the year in 
which we had the. calmest sea, the formula agrees best 
with my measurements; there is, however, a mean differ- 
ence (for 12 hours) between the computed and the observed 
values for the depth of the water evaporated of + 0.”"6. 
As stated above, this difference was estimated, from the 
accuracy with which the buoy could be adjusted to the 
mark, at + 0.771, and the results found with the two 
different apparatus agree withm + 0.7"37. For 1878, 
the coefficient for the term containing the velocity of 
the wind came out with a negative sign. For the observa- 
tions from the West Indies, the constants were widely 
different from those found in 1877, and the mean difference 
between the observed and the computed depth of the water 
evaporated in 12 hours was + 1.”"5. Hence, the condi- 
tions under which the apparatus gave its results, would 
not appear to agree with the assumptions of the formula. 
In particular, I will observe, that on board of the “Vorin- 
gen,” the velocity of the wind introduced into the formula 
was measured from the roof of the roundhouse, 
the atmometer was far less exposed to the full force of 
the wind. Nor do the velocities of the wind in both places 
at all times bear the same relation to each other. In 
1878, the weather was often very boisterous, and the ap- 
whereas 
1) Schweizerische Meteorologische Beobachtungen, 1875. 
