HYDROGRAPHY. 14g 



If, on the other hand, the second assumption being the case, that is to say, if we suppose that 

 a cold fresh layer of water is flowing and expanding over warm and salt water, such a thing will 

 have the effect of producing an unstable equilibrium at the boundary-plane; the two strata of water 

 will then be mixed together, and part of the mixture will sink on account of its higher specific gra- 

 vity. This mixture may, as has been pointed out before, be supersatured with nitrogen, but it does 

 not follow absolutely that such a thing is the case, as we cannot consider it for certain that the two 

 strata of water that are to be mixed together, should be saturated with nitrogen, before the process 

 of mixture is taking place. The lower surface of the uppermost layer of water must of course consist 

 of water that has sunk down in consequence of cooling, while, on the other hand, we cannot know 

 anything about the state of matters at the surface of the lower stratum of water. 



It will be seen, when looking at the chart, that all the water west of 17 Long. W., with 

 exception of station 129, is very near the point of saturation, while the water east of this longitude is 

 supersaturated with nitrogen. This proves that the water south-east of Iceland is subject 

 to heating during the first half of the month of May, while, on the contrary, such a 

 heating does not take place in the seas south-west of Iceland during the first half of 

 June. That a heating may be traced at the beginning of May in the first-named part of the seas 

 that have been subject to investigations, is not unlikely, but that a similar thing should not be the 

 case at the beginning of June with the seas that are situated more westerly, seems at the first glance 

 to be rather strange, as the heat of the sun is constantly increasing at this time of the year. The 

 last-named circumstance can only be accounted for by the fact of the water southwest of 

 Iceland coming from more southerly or warmer regions of the sea, nor does the chart 

 in the annual publication from the Meteorological Institute «Meteorologisk Aarbog for 1896*, 3d part, 

 say anything to the contrary. It must be admitted that the considerable supersaturation at the sta- 

 tions 46 — 55 may be indicative of a southerly motion during the month of May in the region of the 

 sea south-east of Iceland, but it appears from the chart on plate XXXIII that such a motion is hardly 

 appreciable, and the supersaturation must therefore be due to the increasing heat of the sun. 



In the northern part of the Denmark Strait, the state of matters is in a hydrographical point of 

 view similar to what we meet with in the tract of the sea south-west of Iceland, which furthermore 

 serves as a proof of the correctness of the general conception concerning the currents west of Iceland. 

 At the first glance station 95 seems to make an exception to this, but it can be seen by the salinity 

 (35-24, stations 96 and 97 have respectively the salinities 35*30 and 35*35,) and the heating (0-3, 

 for the stations 96 and 97 this latter is relatively — o*i and — o-i,) that the influence of the ice-water 

 can be traced at this station. Station 94 holds an exceptional position, as it is lying entirely in the 

 ice-water on the shores of Greenland, or, if we prefer to call it so, in the East Greenland Polar 

 Current. 



The stations east, north-east, and north of Iceland are all remarkable by the great supersatura- 

 tion, which is inseparable from the great heating of the ice-water. At station 112, the super- 

 saturation seems to be unusually great in proportion to the heating. 



Instead of reducing to 7oo mm , we might in this case, in so far as the surface-water is concerned, 

 reduce to the average height of barometer (the quantity a); but by the adoption of this method, the 



