73 
lidet luftholdige, medens de, der have absorberet sin Luft- 
mængde i de arktiske Egne, maa indeholde meget større 
Mængder, og man vil derfor netop i de paa Expeditionen 
udførte Gasanalyser have et fortrinligt Middel til at con- 
trollere den forhen opstillede Hypothese, ifølge hvilken en- 
kelte Regioner af det iskolde Dyb skulde være opfyldt af 
Vandmasser, der ialfald delvis havde. atlantisk Oprindelse. 
Forat vise Udfaldet af en saadan Control har jeg teg- 
net Kartet No. III. hvorpaa efter samme Princip. som det 
ved Tegning af Kartet No. II befulgte Andes afsat de i 
Dybet fundne Kvælstofmængder udtrykte i CC. pr. Litre 
reducerede til 0° og 760 wm Tryk. ligesom der ogsaa ved Si- 
den af disse Tal Andes opført den Temperatur, hvorved 
Søvandet absorberer denne Ivvælstofmængde. beregnet til 
nærmeste hel Grad efter den af de tidligere bskrevne For- 
søg udledede Formel 
N= 14 . 4 ;— 0 . 23 | 
Det siger sig selv, at disse Temperaturer ikke kunne 
gjøre Fordring paa nogen stor Grad af Nøiagtighed. da en 
forholdsvis liden Feil i Kvælstofbestemmelsen bevirker en 
meget stor Feil i den deraf beregnede Temperatur. Der 
Andes saa ledes flere Observationer, der give Temperaturen 
— 4", en Temperatur, der mig bekjendt. ikke er observeret 
i Havet. Dette vil dog ikke forekomme saa urimeligt, 
naar man tåger Hensyn til. at Søvand ai — 2° ved <8p mm 
Barometerstand’ absorberer en Kvælstofmængde. der paa 
det Nærmeste gaar op til. hvad der i Ydertiliældene er 
fundet. 
Farvelægningen er her foretagen saaledes, at*deStrøg, 
hvor Kvælstofmængden er funden at være 14.4 CC. eller 
derover, ere betegnede med blaa Farve, de Strøg, hvor 
Kvælstofmængden ligger mellem 14.4 og 12.5 CC.. med en 
svag rød Farve, medens et mindre Parti, hvor Kvælstot- 
mængden er funden at ligge under 12.5 CC., er betegnet 
med en noget kraftigere rød Farve. Betydningen at disse 
Farver bliyer ligesom i Kartet No. IT den, at de røde Far- 
ver* bedække de Strækninger, hvor Yandet i mere eller 
mindre Grad besidder atlantisk Karakter, medens den blaa 
Farve tilhører de Yandmasser, der have absorberet sin 
Kvælstofmængde. ved en Temperatur af under 0°, og som 
altsaa nærmest synes at hidrøre fra de arktiske Egne. 
Ved at sammenligne Kartene No. II og III vil man 
strax se. at Farvelægningen i disse i alt \ æsentligt \isei 
en særdeles stor Overensstemmelse, som paa mange Puucter 
endog nærmer sig til Congruents,* om man end 'ed næi- 
niere Betragtning vil Ande. at disse Ligheder ikke gaa 
igjen i alle Detailler, hvad man heller ikke paa nogen 
Maade kunde vente. Grændserne bliver nemlig paa Kartet 
No.* III paa Grund af Observationernes Faatallighed meget 
vanskelige at bestemme, ja der Andes endog her paa om- 
kring den 65de Breddegrad et større Strøg, hvorom man 
intet med Bestemthed kan slutte, da der under Analysen 
tabtes en mindre Del af de Luftprøver, der vare bestemte 
til at udfylde dette Hul. saaledes at den samlede Lult- 
mængde désværre ikke kunde maules. Desuden ere ogsaa 
Observationsfeilene baade for Salt- og Kvælstofbestemmel- 
Den norsko Nordlinvscxpedition. Tornoe: Cliemi. 
with that absorbed in the Arctic regions, wherefore the 
analyses of gas performed on the Expedition furnish an 
excellent means of testing the value of the hypothesis ac- 
cording to which certain sections of the cold area are 
assumed to be made. up of water part of which at least 
would seem to be of Atlantic origin. 
To show the result of such a test, I have annexed a 
third Plate (drawn on the same principle as Pl. II), in 
which are given the different amounts of nitrogen present 
in deep water, expressed in c.centim. per litre, reduced to 
0° and a pressure of760™ m . Along with these Agures will 
be found, too, the temperature at which sea-water absorbs 
such an amount of nitrogen, computed, in whole degrees, 
by means of the formula deduced from the observations pre- 
viously described, viz : — 
14.4 — 0.23C 
These temperatures cannot of course pretend to any 
high degree of accuracy, a comparatively small error in a 
nitrogen-determination involving a very considerable error 
in the temperature. Thus, for instance, several of the ob- 
•servations indicate — 4". a temperature which, so far as 
I am aware, was not anywhere observed in the sea. r I his. 
however, will not appear so strange, if regard be had to 
the fact, that sea-water of — 2°, at a pressure correspond- 
ing to 780™ m , absorbs an amount of nitrogen which agrees 
very closely with the highest found on the Expedition. 
The sections in this Plate are coloured as follows: 
those in* which the amount of nitrogen was found to equal 
or to exceed 14.4% blue; those in which it ranged from 14.4 
to 12.5% light red; a somewhat deeper red serves to indi- 
cate a small tract in which the amount of nitrogen did 
not reach 12.5. Moreover, as in Plate II. the red coloui 
indicates water more or less distinguished by Atlantic 
characteristics; the blue, -water in which the nitrogen was 
absorbed at a temperature below 0°, and which, therefore, 
would seem to have derived its origin' from some part of 
the Polar Seas. 
A comparison of Plates II and III will at once 
show considerable agreement in the distribution oi colour, 
many of the sections almost coinciding; though, on closer 
inspection this approach to congrulty is not found to char- 
acterise all details, which indeed there was no reason to 
expect. In PI. III. the limits proved exceedingly difficult 
to define, owing to the limited number of observations; 
nay, respecting an extensive tract near the 65th parallel 
of latitude nothing definite can be inferred, part of several 
samples of air. the analysis of which would have served 
to All up the blank, having been unfortunately lost, and 
the total amount of air could not, therefore, be measured. 
Besides, the errors of observation both in the salt and the 
nitrogen determinations, are so considerable, when com- 
pared with the minute differences in amount, that, in some 
n 
