H YD ROGRAPH Y. x * l 



out of consideration that the determinations for station 15 were made in 1895, while those for station 99 

 were made in the following year; for there is a possibility of the temperature at this place, even in the 

 depth, having been subject to a considerable change during the course of a year. Or, what is more 

 likely, it may be supposed that station 15 is lying in a recess at the bottom of the sea, in which 

 very cold water has accumulated; what is in favour of this supposition is that so low a temperature 

 at so small a depth, has not been found at any other place in the under-layer as at this station. 



It is seen that the o° isotherm is stretching round the northern part of Iceland, and then in 

 a south-easterly direction along the Faroe-Iceland ridge, north of which it constantly remains. 

 In about io° longitude west of Greenwich, a tongue of water with a temperature that is less than 2 , 

 and a salinity below 35'co, is expanding over the ridge in a southerly direction. A very broad belt 

 that is stretching over the ridge, has a lower temperature than 3 , and a salinity below 35'io, the 

 same state of matters as mentioned under the sections page 104. The series of stations south of 

 Greenland does not give any good determination of the 2° isotherm, which, like the isotherms in the 

 Denmark Strait only gives a rough sketch of the continued course of the curve. 



TEMPERATURE AND SALINITY AT THE SURFACE. 



It will be seen from the table of determinations at the surface, page 40—61, that the salinity 

 of the sea-water, south and southwest of Iceland, only varies very little, and that the variations which 

 appear, do not show any regularity in so far as regards the position of the place of 

 observation. Temperatures taken at short intervals in the same tract of the sea, are nearly equally 

 large, whether the observations were made at the beginning, or at the completion of the voyages. Nor 

 do the almost inappreciable deviations, in so far as the temperature is concerned, show any regular 

 dependence of the place of observation. Owing to this, there is every reason to believe that the 

 variations in temperature and salinity which cannot be attributed to observational 

 errors, must be due to causes pertaining to meteorology. 



It will be seen that where we have determinations south and south-west of Iceland at one 

 place during the two years 1895 and 1896, the difference of salinity has been so trifling that 

 it can scarcely be attributed to other causes than those to which the irregular variations are due 

 in so far as regards the position of the place of observation. At such places where the ice-water 

 west, north and east of Iceland manifests its presence by the slight salinity from 32-00 to 35 - oo, we 

 meet also with irregular variations in so far as the distribution of the salinity is concerned, which 

 variations, moreover, are much greater than those we meet with in the Atlantic. It is not difficult to 

 find the cause of this, as the ice at the time that it is melting, does not form a single connected 

 sheet, but is broken into large hummocks, often floating about at a great distance from one another. 

 Owing to this, it is very likely that the water has a very small degree of salinity at 

 those places where such a hummock of ice just has been subject to melting. 



If we consider the 35-00 isohaline as a line of demarcation between Atlantic-water and Polar- 

 water, it will be seen that at the north-west coast of Iceland, this line is lying at a greater distance 



