VE6A-EX1'EDITJ0NEN S V ET EN S K A PhlC! A ARBETEN. 



271 



i.e. an increase of O.ooou m the spec. gravity of sea-water 

 makes its freezing point sink O^^.oi C. We might therefore judge 

 with sufficient accuracy of the specific gravity and the percen- 

 tage of salt in sea-water from its freezing point, if determined 

 under due precautions with a sensible thermometer. 



All determinations of specific gravity were made with the 

 Sprengel pycnometer. 



CHAPTER 3. 



On pure water. 



Ermann^ and Kopp-, in their celebrated researches on 

 the volumes of solid and liquid bodies, first called attention 

 to the fact, that solids are sometimes liable to irregularities 

 of volume in the vicinity of their melting point. Thus, for 

 example, the volume of sulphur was found (by Kopp) to in- 

 crease extraordinarily from + 70'' C upwards. The melting 

 process, which ordinarily is confined to a fixed temperature, 

 here may be considered to begin more than 40 centigrades 

 below the ordinary melting point. Other bodies, for example, 

 phosphorus and ice, did not show any similar irregularity. 

 Neither Pl ii eker and Geissler, nor the aulhor and H. Lar- 

 son were able to discover any signs of premature melting of 

 pure ice below zero. The cause of the anomaly of sulphur 

 and still more of selenium is ordinarily attributed to the for- 

 mation of allotropic modifications of these elements. Also in 

 the case of compound bodies, as, for example, wax, stearine 

 etc. showing similar properties, this explanation is commonly 

 adopted. I think, however, that in some cases another inter- 

 pretation may be possible. 



From my previous experiments on the thermic and vo- 

 lumetric properties of the hydrates of acetic and formic acid 

 I had learned, that the whole melting process of a solid body 

 may be essentially altered, in case a slight quantity of another 

 substance is present, tchiclt is capaUe of entering Info solid ion 

 with the liquid body. I found that traces of water, too slight 

 to be detected by any chemical analysis, are able to disturb 



1 Pogg. Ann. IX. 



3 Liebig'8 Ann. XCIIT. 



