17 
eredes en Kurve. Denne nærmer sig asymptotisk til at 
blive en ret Line parallel med Abscisseaxen. Det Punkt, 
fra hvilket af denne Parallelisme begyndte; afmerkedes og 
ligeledes det tilsvarende Tidspunkt efter Uret, ved hvilket 
altsaa den fuldkomne Accommodation kunde ansees for at 
være naaet. Fra Accomodationspunktet afsattes tilbage 
hele Grader, og de tilsvarende Tidspunkter efter Uret 
afmerkedes. Der toges Forskjellen mellem Tidspunktet for 
den indtraadte Accomodation og disse Tidspunkter. Disse 
Tal udviste da, hvor lang Tid Dybvandsthermometret be- 
høvede for at accommodere sig, naar dets Temperatur var 
1°, 20 3°, o. s. v. forskjellig fra det omgivende Vands 
Temperatur. Jeg fik saaledes følgende Tabeller, i hvilke 
A T betegner Temperaturforskjellen og Z\ t Tidsforskjellen 
i Minuter og Tiendedele deraf. 
of time and seconds; and with ordinates: computed 
difference between readings of thermometers, I con- 
structed a curve. This tends asymptotically towards a 
right line parallel to the axis of the abscisse. The 
point from which this parallelism commences, was marked 
off, as also the corresponding point of time, by which 
therefore perfect accommodation may be assumed to have 
been attained. From the point of accommodation were 
set off, backwards, whole degrees, and the corresponding 
points of time marked off. he difference was taken be- 
tween the point of time for the accomodation fully set 
in and the said points of time These figures indicated 
how long a time the deep-sea thermometer required to 
accomodate with its temperature different 1°, 2°, 3°, &e., 
from that of the surrounding water. I was consequently 
able to work out the following Tables, in which Z\ T 
denotes the difference in temperature and /N t the differ- 
ence in time, computed in minutes and tenths of minutes. 
Negretti & Zambra 89. 
Kviksølvpiezometer. 
(Mercury-Piezometcr.) 
Casella-Buchanan 49. 
AR AG AT 
m. | 
Ig BG | The 
2 2.7 | 2 
SG 2.8 | 3 
4 2.85 | 4 
5 2.87 | 5 
ry) 7 6 
br ? 7 
? y) | 8 
’ ” | 9 
; on IO 
” p) I I 
b) 29 12 
” 27 13 
14 3-0 IA 
Efter disse Undersøgelser lod jeg Indexthermometrene 
ikke hale ind, førend de havde været 10 Minuter i Van- 
det, og Vendethermometrene 3 Minuter. 
metrenes Bevægelse med Loddet ned igjennem Vandet vil 
Strømningen bevirke en meget raskere Åceomodation end 
Tilfældet var under Forsøgene, i hvilke Thermometrene 
stode stille i Vandet. Den Tid, som blev givet Dybvands- 
thermometrene til at accommodere sig, var saaledes altid 
fuldkommen betryggende. 
Jeg gaar nu over til at redegjøre for de enkelte Dyb- 
vandsthermometres Correctioner. 
Miller-Casella No. I. 
Nulpunkt-Correction. 
Under Thermo- - 
At 
m. 
7+7 
9.0 
9-7 
10.1 
10.3 
10.5 
10.6 
10.7 
10.8 
10.8 | 
10.9 
10.95 
I 1.0 
I 1.0 
> 
(=) 
=) 
So 
to Oo dn LIS) Ko) 7 Wo) @H AS 5 
OO ON OM f GW NY & 
— 
OO OM own DAM 
After these investigations, the index-thermometers were 
not hauled in till they had been 10 minutes in the water, 
nor the inverting-thermometers till they had been 3. On 
the downward-passage of the thermometer with the lead 
through the water, the current will occasion a much more 
rapid accomodation than was the case with the experiments 
during which the thermometers remained stationary in the 
water. Hence, the time given the deep-sea thermometers 
to accomodate was always amply sufficient. 
I shall now pass on and explain the corrections for the 
several deep-sea thermometers. 
Miller-Casella No. 1. 
Zero-Correction. 
1877. Juli 10. — 0.17 Tromsø. 1877. July 10. + 0%17 Tromsø. — 
1878. Maj 8. + 0.07 Christiania. 1878. May 8. -+ 0.07 Christiania. 
- Juli 23. + 0.13 ved Beeren Biland. » July 23. +- 0.13 off Beeren Niland. 
å Nov. 29. + 0.05 Christiania. ” Nov. 29. + 0.05 Christiania. 
Middel + 00.105 Mean + 00.105 
Den norske Nordhavsexpedition. H. Mohn: Nordhavets Dybder, Temperatur og Strømninger. 3 
