65 



Af denne Tabel fremgaar det, at Differentserne mel- 

 lem de ved Hjælp af Bgenvægt og Chlorinængde beregnede 

 Saltmængder i Regelen ore meget smaa, kun de 3 samtidig 

 udførte Bestemmelser i Vandprøverne No. 261, 2G2 og 

 263 danne i saa Henseende en Undtagelse. De store 

 ber optrædende Differentser skyldes uden Tvivl en Peil 

 ved Aflæsningen af Egenvægterne, som for disse Vandprø- 

 vers Vedkommende eve fundne altfor lave. til at de 

 kunne bringes i Harmoni med andre paa Steder i Nærhe- 

 den udførte Observationer. Det er saaledes i hoi Grad 

 paafaldende for Vandprøven No. 262, optagen fra et Dyb 

 al' 95 Favne (174 Meter) i ca. 8 Miles Afstand fra Land, 

 at tinde Egenvægten 1.0254, medens man i de indenfor 

 liggende' Fjorde. hvor Saltgehalten ellers overalt er mindre 

 end paa Havet, i lignende Dyb tinder en meget storre 

 Egenvægt. Selv i den indelukkede Skjærstadfjord, hvor 

 Overfladevandet er særdeles fattigt paa Salte, er dogEgen- 

 vsegten paa Bunden funden at være 1.026, kort sagt, Egen- 

 vægter som de i de omtalte Tilfælde observerede staa paa 

 dette Strog af Kysten fuldstændig uden Sidestykke. Natur- 

 ligst lade disse Urimeligheder sig forklare ved at åntage 

 Egenvægterne aflæste med 0.001 for lavt. da de ved denne 

 Antagelse paa det Nærmeste kan bringes i < hærensstem- 

 melse saavel med de i de samme Yandprover udførte Chlor- 

 bestemmelser som med de andre Observationer fra nærlig- 

 gende Puncter. 



Bortser man fra disse 3 nævnte Observationer og af 

 de øvrige beregner den gjennemsnitlige halve Differents 

 mellem to paa samme Yandprove ved Hjælp af Chlortitrering 

 og Aræometer udførte Saltbestemmelser, resulterer som 

 Udtryk for denne 0.00904. eller man erholder under For- 

 udsætning af, at Feilene i lige høi Grad skyldes Chlor- som 

 Egenvægtsbestemmelserne. for den gjennemsnitlige Feil af 

 en Egenvægtsbestemmelse Yærdien ( l.l •• " M 169 og af en Chlor- 

 bestemmelse 0.005. Differentserne falde, som man ser, 

 snart til den ene snart til den anden Side. idet det dog 

 maa bemærkes, at Cblormængden gjennemsnitlig giver lidt 

 over 0.008 u / høiere Saltgehalt end Egenvægterne. hvad 

 der næsten udelukkende skyldes de nordenfor den 75de 

 Breddegrad udførte Observationer. 



Førend jeg nu gaar over til at give en Oversigt over 

 de Resultater, som af disse Observationer lader sig tullede, 

 vil det være nødvendigt parenthetisk at indskyde nogle Be- 

 mærkninger om Dybde- og Temperaturforholdene i det 

 norske Hav i sine groveste Træk. Hvad der til den Ende 

 her meddeles, er hovedsagelig hentet fra en af Professor 

 Dr. Mohn forfattet Afhandling. som tindes trykt i C. F. 

 Schubelers -'Yæxtlivet i Norge." 



1 lybden i det af den norske Expedition undersøgte 

 Hav. forsaavidt det ligger vestenfor en Linie fra Spitsber- 

 gen til det nordlige Norge-, er i større Afstand fra Land 

 overalt over 1000 Favne (1829 Meter) og gaar i Regelen 

 op til mellem 1500 og 2000 Favne (2743 og 3658 Meter) 

 eller endog derover. Paa Strøget mellem Beeren Eiland 

 og .lan Mayen hæver sig en Ryg. hvor Dybden ikke naar 

 L500 Favne (2743 Meter), medens der saavel søndenfor 



Den norske Nordhavsexpedition. ' Tornne: Chemi. 



This Table shows the differences in the amount of 

 salt computed from specific gravity and the proportion et 

 chlorine to be, as a rule, exceedingly small, the '■'< determ- 

 inations performed successively with samples Nos. 261, 262, 

 and 263 constituting the sole exception. The great dif- 

 ferences observed here must unquestionahly arise Ironi er- 

 roneous readings of the specific gravity. which. as found for 

 these samples, is much too low when compared with that 

 determined for others obtained from adjacent localities. 

 Tlius. for instance. the specific gravity of sample No. 262, 

 drawn at a depth of 95 fatboms (174 metres), about s 

 geographica] miles from land. is stated to be 1.0254, 

 wbereas that determined for the water of the neighbouring 

 fjortis, in which the amount of salt at equal depths is 

 invariably less than in the open sea. was much greater. 

 Even for a frith locked in as is the Skjærstadfjord. 

 where the surface-water is remarkably deticient in salts. 

 the specific gravity of bottom-samples was found to be 

 1.0260; in short, such excéptional specitic gravities 

 are without a parallel on this line of coast. The most 

 natura! explanation of these incongruities. is afforded by 

 assuming the specitic gravity in each Case to have been read 

 0.001 too low; the results could then be made to agree 

 pretty closely both with the chlorine-determinations performed 

 with the same samples of water and with observations tåken 

 in adjacent localities. 



Now, if we disregard the 3 excéptional observations. 

 and for the others compute the average half-difference be- 

 tween two salt-determinations performed with the same 

 sample of water by means of the areometer and titrating 

 with chlorine. this will be expressed by 0.00904; or, as- 

 suming the errors to lie equally in the chlorine and the 

 specific gravity determinations, the mean error of a spec- 

 ific ' gravity determination is 0.000069, and of a chlorine- 

 determination 0.O05. As will be seen, the differences be- 

 tween the 2 right-hand columns of the Table are sometimes 

 positive, sometimes negative; but the amount of salt ind- 

 icated by the proportion of chlorine exceeds on an average 

 that denoted by the specific gravity by a trine over 0.008 

 per cent. which must be referred almost exclusively to the 

 observations tåken north of the 75th parallel of latitude. 



Before proceeding to review the results deducible 

 from these observations, it will be necessary to interpolate 

 a few general remarks on the dejoth and temperature of 

 the Norwegian Sea. To this end. I shall merely recapit- 

 ulate what Professor Mohn has stated on the subject 

 in a Memoir printed in C. F. Schubelers "Væxtlivet i 

 Norge." 



The depth of the Sea investigated by the Norwegian 

 North- Atlantic Expedition was found to be as follows: — 

 Throughout the tract extending west of an imaginary line 

 drawn from Spitzbergen to the northern extremity of Nor- 

 way, it is never less than 1000 fathoms (1829 metres) 

 some considerable distance from land. and generally ranges 

 from 1500 to 2000 fathoms (2743—3658 metres); nav. 

 in some places it is even greater. Between Beeren Ei- 



10 



