VEGA-EXPEDITIONENS VETENSKAPLIGA A II HET EX. 



291 



the next series of determinations another specimen of ice-water 

 from the sea-ice in Kattegatt, obtained in December 1881 at 

 Marstrand. 



When the outer \i\yer of the frozen ice in the dihatometer 

 is thaAved, in order to get the sohd cyHndre surrounded by a 

 rind of mercury (see the discription of this operation on page 

 263 and fig. 2 on plate 1), one or two cc. of water are ordhiarily 

 expelled through the capillary. In this case the water was 

 found to be of an extreraely bitter salt taste. The original 

 ice-water bad thus separated in two parts of different chemical 

 constituents. The remaining quantity" of ice in the dilato- 

 raeter was naturally deficient in saltness relatively to the ice 

 formed at first. After the close of the experiment, the melted 

 ice-water of the dilatometer was examined as usual with 

 regard to its latent heat, its percentage of chlorine and specific 

 gravit3^ The water of sample V was found to possess: 



, Original sample. Ice-water V. 



spec. gr. atjT^ C .- ^.oiio I.0053 



p. c. of Cl. 0.727 0.273. 



The important question of the latent heat of the ice-water will 

 be discussed in the last chapter. 



Finally I repeated exactly the same series of operations 

 and nieasurements with a real specinicn of salt water from 

 the deeper strata of the Siberian sea, in order to stiidy the 

 formation of ice from real ocean water. The ice, which bad 

 formed at first in the instrument, on superficial thawing, gave 

 off a few cc. of a verv concentrated solution of bitter taste. 



Origiii!il sample. Ic -^vat^r VI. 



spec. gr. at ^7j^ c I.0278 l.o 



094 



p. C. of Cl. 1.91 0.649. 



The numerical results of the determinations on the samples 

 IV, V, VI are recorded in tables IV, V, VI and are graphi- 

 cally represented on plate 21 (the volumes of the ice) and on 

 plate 23 (the vol. of the waters). In the calculation of the vol- 

 umes I have started from the assumption 



q ^= 0.00017905 

 g = 0.00002753 



to avoid unnecessary complication. All the numbers given 

 in this chapter refer to special kinds of sea-ice and cannot 

 be considered to be fixed constants. of nature, like the vol- 

 umes of pure ice and pure water, which must be calculated 

 with the utmost accuracy attainable in the present state of 

 physical science. 



