75 
pegecle, og det er derfor helt naturligt, naar Undersøgel- 
serne ikke i Henseende til Studiet af tidligere ukj endte 
Eiendommeligheder ved Havet kunde føre til saa fyldige 
Resultater, som ønskeligt kunde være. Men om end disse 
Undersøgelser paa. Grund lieraf nærmest faa Karakteren af 
forberedende Arbeider, saa vil det døg, som jeg lumber, 
indrømmes, at de desuagtet kunne have sine maaske ikke 
uvigtige Følger, idet de vise, at man gjennem de chemiske 
Observationer, der tidligere i Sammenligning med Tempe- 
ratur- og Dybdebestemmelser har spillet en mindre frem- 
tredende Rolle ved Studiet af Havets Pliysik, vil kunne 
skaffe Oplysninger om mærkelige Forholde i Havet, som 
man ad anden Vei vanskelig skulde falde paa at søge op- 
klarede. Man vil ved Hjælp af de her erholdte Resultater 
med Lethed i Fremtiden kunne udkaste en detailleret Plan 
for en fornyet Undersøgelse af det norske Hav, der i mine 
Øiue stiller sig som særdeles ønskelig, da man ved at gjøre 
et hidindtil ukjendt Hav til Gjenstand ' for Bearbeidelse 
vanskelig turde gjøre Regning paa at træffe et, der i Hen- 
seende til Studiet af Strømforholdene er saa iustructivt 
som det norske Hav. 
Ved saadanne fremtidige Undersøgelser kunne de paa 
den norske Expedition benyttede Arbeidsmethoder ikke i 
alle Retninger blive optagne i uforandret Form. og det vil 
derfor ikke være ubeføiet til Slutning med faa Ord at paa- 
pege de Mangler, der klæbe ved disse. 
De til Saltbestemmclserne tidligere . benyttede Metho- 
der, ifølge hvilke alle herken hørende* Observationer an- 
stilles ombord, bør utvivlsomt ior Fremtiden ikke komme 
til Anvendelse, da man ad den Vei tiltrods for al anvendt 
Moie ikke vil kunne opnaa den Nøiagtighed. som tiltrænges 
for med ønskelig Sikkerhed at kunne paavise de i Havet 
forekommende ofte meget smaa Differentser. Paa den 
norske Expedition blev denne Frenlgangsmaade benyttet, 
fordi man med ældre Iagttageres 1 dtalelser lor Øie maatte 
befrygte. at Søvand ikke lod sig opbevare i længere Tids- 
rum uden at undergaa forskjellige Forandringer, en Frygt, 
der imidlertid efter min Erfaring kun forsaavidt er be- 
grundet, som man til Opbevaring af Våndet benytter Kar, 
der ere forsynede med Korkeprop. Jeg har nemlig under- 
søgt llere- Vaiulprøver, der have været opbevarede paa 
denne Maude i omkring 2 Aar og fundet, at de alle uden 
Undtagelse have undergaaet Forandringer al saadan Art, 
at man turde være berettiget til at anse dem uskikkede 
til Egenvægtsbesteinmelse, hvorimod jeg hos Vandprøver, 
der i lignende Tidsrum havde henstaaet paa Flasker lor- 
synede med isleben Glasprop, ikke kunde opdage nogen- 
somhelst Eiendommeligheder, der kunde adskille -dem fra 
friskt øste Vandprøver. Ved denne Opbevaringsraaade 
risikerer man dog ganske sikkert Fordunstning al en Del 
af Våndet, og man maa derfor beskytte sig mod denne 
Feilkilde ved at lijemføre det til Saltbestemmelser be- 
stemte Vand paa tilsmeltede Glasrør. 
I de saaledes conserverede Vandprøver vil man senere- 
hen efter Hjemkomsten kunne bestemme Egenvægten ved 
Sprengels Pyknometer og Ohlorgehalten ved fijælp at \ ei- 
ningsanalyser med saadan Skarphed, som man ved. Arbeide 
servations, • and hence the results of the work done, em- 
bracing as it did the investigation of phenomena unknown 
before, were naturally less comprehensive than might other- 
wise have been attained. But, though such labours must 
to a certain extent, be regarded as preliminary, they will, 
1 trust, prove of considerable importance, showing as they 
do, that chemical observations, which, as compared with 
determinations of temperature and depth, previously held 
quite a subordinate rank among the means employed for 
studying the physical conditions of the ocean* will serve to 
throw light upon many remarkable phenomena, that without 
such data would be extremely difficult to explain. On 
I the basis of the results here set forth, a detailed plan 
might be easily laid down for the further exploration of 
the Norwegian Sea, — in my opinion a most desirable un- 
dertaking, since of ocean tracts as yet unknown, there are 
probably few that, in regard to the study of ocean cur- 
rents, would so well repay investigation as that section of 
the N Orth-Atlantic. 
As several of the methods of investigation practised 
on the Norwegian Expedition, will not admit of being 
adopted on future occasions in a wholly unmodified form, 
it will not be out of place in conclusion briefly to point 
out their defects. 
The methods previously devised for determining the 
amount of salt in sea-water, by which all observations with 
this object in view were taken on board, should unquestion- 
ably cease to be adopted, since they will not suffice, with 
the greasest care even, to attain the high degree of accu- 
racy requisite for detecting such minute differences as are 
frequently found to occur. These defective modes of op- 
eration were, however, adopted on the Norwegian Expedi- 
tion. there being reason to believe from the statements ol 
earlier observers, that sea-water could not be preserved 
for any length of time without undergoing chemical change, 
a supposition which, so far as my experience goes, is con- 
tinued only in the event of its being kept in corked vessels. 
I have examined, for instance, various samples ot sea-water 
that had been preserved for about 2 years in corked bottles, 
and found all without exception to have undergone a change 
sufficient to render them unfit for specific gravity deterin- 
■ inations; whereas, on the other hand, sea-water which had 
. been allowed to stand over for the same space ol time in 
bottles furnished with ground glass stoppers, was not to be 
distinguished from freshly drawn samples. There is, how- 
ever, a risk of loss from evaporation.’ the Stoppers being 
seldom, if ever, tight-fitting; and to guard against this 
source of error, the water lor salt-determinations must be 
brought home in hermetically sealed glass tubes. 
With water thus preserved, the specific gravity may 
be determined bv means of Sprengel s pycnometer, and the 
ainoifnt of chlorine by weighing, on the return ol the Ex- 
pedition, far more accurately than would be possible on 
n * 
r 
