i !i Dibination. 



233 



water in the Mo and Lyse Fiords had a temperature corresponding 



to the average annual atmospheric temperature of the place. Also 

 in the latter fiords, submarine ridges prevent the warm bottom 

 water from penetrating in. and, moreover, the supply of fresh water 

 in the winter is so great that the vertical current is prevented from 

 assuming any large proportions. The transmission of heat down- 

 wards takes place principally by conduction. A maximum in the 

 autumn and a minimum in the spring find their way gradually 

 downwards, and finally an average is reached, which corresponds 

 to the annual average atmospheric temperature. 1 ) The Skjerstad 

 Fiord must probably be classed with the Mo and Lyse Fiends. 

 The bottom water is homotermic (cf. Hydrography, p. 14). the 

 supply of fresh water is sufficient to prevent the winter cold from 

 penetrating down into the depths. 



The station Sulh\jelma, in the Vatnbygd river district, had an 

 average fall of 1 097 mm. in the years 1896—1902, and the station 

 Graddis, in the Salt river district, during the same period had an 

 annual average of 533. 2 ) According to Mohn, 3 ) the annual average 

 temperature at Bodo is 4".i C. and at Ranen 3.5. On August 17th 

 1877, the Norwegian North Atlantic Expedition found that the 

 temperature in the Skjerstad Fiord at a depth of about 500 mtrs. 

 was 3.2. ( )n April 4th 1900, I registered at the same depth 3°.1S ('. 

 If one now considers these two atmospheric averages, it would 

 seem that o.2 is a probable value for the annual average tem- 

 perature of the air in the Skjerstad Fiord. 



If we, however, imagine the large basin of the Skjerstad Fiord 

 moved to the inner part of the Porsanger Fiord, and with the same 

 connections with the latter as it now has with the Salten Fiord, 

 we should certainly find that the bottom temperature would be 

 considerably lower than that of the annual average of atmospheric 

 temperature in the inner part of the Porsanger Fiord. For, from 

 what has already been said, it will be seen that the supply of 

 fresh water at the latter place is not sufficient to prevent an 

 evening out of its salinity in the winter, thus allowing the winter 

 cold to exert its influence on deeper layers of water. 



In some of the lesser fiords adjacent to the Vest Fiord, 1 have 

 also noticed that the bottom temperature has been lower than the 

 annual average atmospheric temperature, which would imply that 

 somewhat of the winter cold has found its way downwards. 



Examples of this fact may be found in Rombaken, Skjomen, 

 the Og-s Fiord etc. Such places excel in many arctic forms. These 

 fiords have already been referred to, and I would call attention to 

 what has been said about them in previous pages and also to PI. 

 19, where the curves VI, VII, VIII represent the conditions of 

 temperature in Skjomen, the 0gs and Skjerstad Fiords. The curves 

 for Skjomen and the 0gs Fiord are especially characteristic on 

 account of their slight bend, which is a sign of a uniform salinity. 



The temperature curves for Tranodybet, the Tys Fiord and 

 Oxsund are given so that comparisons may be made, (PI. 19,' 

 curves I, II, IV, V), all the curves are very much bent in the 

 upper layers, where there is great variation in salinity, but in the 

 layers where there is uniform salinity they become almost a straight 

 line. Curves III and IV, which represent the conditions of tem- 

 perature in March 1899 in the sea off Rost and in the Tys Fiord, 

 show plainly that a higher temperature prevails in the deeper 

 layers in the fiords than in the corresponding depths in the Nor- 



') Cf. Studier over naturforholdene i vestlandske fjorde, p. 46. 

 '-) Cf. Nedberiagttagelsev i Norge, aarg. VIII, p. 125. 

 3 j Klimatabeller for Norge I, p. 18. 



wegian Sea. As 1 have already stated, the reason for this is to 

 lie found in the fact that such fiords as the Tys Fiord are of such 

 a formation as excludes the arctic bottom water from the Nor? 

 wegian Sea. hut gives admittance to the warm Atlantic waters 

 which till the basins. Then too the fresh water which flows into 

 the fiord from the land, is sufficiently large in quantity from the 

 Bokn Fiord to the Vest Fiord to prevent the winter cold from 

 penetrating downwards to any considerable depth. 



There is another thine- which one might suppose to lie. to 

 some extent, dependent upon the variation in downfall, I mean the 

 height of the water on the coasts. 



From „Vandstandsobservationer", published by ..den norske 

 gradmaalingskommission" I have on id. 20 drawn some curves, 

 which give the monthly average height of the water at Kabelvaag 

 and Vardo in the years 1882, L88J and 1885. 



The measurements have been made with selfregistering instru- 

 ments at 0, 1, 2, 3 etc. hours after the moon's culmination ami 

 from these results the average has been calculated. Taking it for 

 granted that the points have been unaltered, and that the instru- 

 ments in other respects too have been quite reliable, one must be 

 able, by help of the data thus obtained, to form a wellfounded 

 opinion of the variations in the rise and fall of the water, in the 

 course of the year, on the northern coasts. A glance at the curves 

 (I — VI, PI. 20), will show that there is at any rate one thing 

 which cannot possibly lie accidental, the curves show a definite 

 tendency to a minimum in April. Similarly, too. a maximum can 

 be arrived at for the months November— January, while the re- 

 maining variations suggest accidental causes. 



On the same plate the curves representing the average monthly 

 downfall at Svolva?r, which is situated near Kabelvaag and Vardo, 

 are given. Both these curves show a decrease during the first 

 months of the year up to May, in which month the year's minimum 

 downfall is reached. The Svolvaer-curve shows a maximum in 

 November, and the Vardo-curve in October. There is this point 

 of resemblance between the water-heights and downfall curves, that 

 they generally show a decrease during the first months of the year, 

 respectively up to April and May, but it cannot at all be said that 

 the decrease in heights is caused solely by the decrease in down- 

 fall. Of course the variations in downfall exert some influence on 

 the height of the coast water, but as regards the north of Norway, 

 it will easily be seen on comparing the curves that this influence 

 is by no means sufficient to account for the great differences in 

 height. It should he remembered that 1 cm. is taken as the unit 

 for the height, and 1 mm. for the downfall curves. 



At Svolvaer, the amplitude of the curve representing the aver- 

 age, monthly downfall is 8.3 cm., the corresponding value at Vardo 

 is 5.3 cm. The observations made of heights have not been so 

 complete that it has been possible to calculate the normal average 

 for each month, but on the basis of the amplitudes of the Kabel- 

 vaag and Vardo curves, we get: — 



Amplitude. Amplitude. 



Kabelvaag 1882 61 cm. Vardo 1882 52 cm. 



1884 60 „ — 1884 35 „ 



1885 45 ., L885 47 „ 



These figures show, with all desirable clearness, that the 

 variations in the course of the year are so considerable, that they 



30 



