4i8 



NA TURE 



[August 30, 



1883 



persist without sensible mixture. As an example may be 

 cited some observations made on board the Willem 

 Barents in the Kara Sea on August 3, 18S1 : — 



Depth, fathoms. Temp. f C. Density. 



+8'2 IT>o6 



1 ... ... +62 ... ... 1-009 



2 +'7 1020 



3 -i"o 1-0236 



5 -I'S 1-0247 



Here, while we have what is practically fresh warm water 

 at the surface, and to a depth of a fathom from it, at two 

 fathoms we have cold Arctic Ocean water. Looking 

 therefore to the great variety in the composition of the 

 waters exposed to the winter cold and therefore likely to 

 produce the ice met with in Arctic regions, Dr. Petterson 

 has studied separately the change of heat and volume 

 by the freezing of (1) pure water, (2) brackish water of 

 little saltness, and (3) ocean water of ordinary saltness. 

 With regard to the freezing of brackish or salt water no 

 previous investigations of a quantitative character exist, 

 and the author's results are all new. With regard to the 

 freezing of pure water the most important investigations 

 were those of Plucker and Geissler. While verifying 

 their result as to the average coefficient of dilatation of 

 ice, the author made the important discovery that the 

 volume of ice decreases as the temperature rises, in the 

 vicinity of the melting point. In extending his researches 

 to brackish and salt waters he found this anomaly more 

 and more accentuated the more salt was contained in 

 the ice formed. Rightly seizing the importance of this 

 very remarkable observation the author makes the be- 

 haviour of pure ice in the vicinity of its melting point one 

 of the main objects of the investigation. The"dilato- 

 meter" used was a glass vessel of peculiar construction 

 and of a capacity of 41 cubic centimetres. The water to 

 be experimented on was frozen in it, so that it formed a 

 cylinder of ice surrounded by mercury, which extended 

 also into a capillary tube and indicated changes of 

 volume. As the accuracy of the results depends, 

 amongst other things, on the correctness of the deter- 

 minations of the absolute dilatability of mercury ; and 

 as this is somewhat uncertain, and indeed variable, 

 at low temperatures, the author adopted the device of 

 Plucker and Geissler for producing a practically undil.n- 

 able envelope for his experimental substance. The prin- 

 ciple of it is very simple. The envelope is of glass with 

 a coefficient of expansion 0-000028 : that of mercury is 

 0'oooiSi. If the volume of the glass envelope is to that 

 of the mercury contained in it in the inverse proportion 

 of their coefficients of expansion the residual volume will 

 be constant even though the temperature vary. If the 

 volume at o' C. of the glass be iS'l cc. and that of the 

 mercury 2'S cc the residual volume is 1 Si — 28= 15-3 cc. 

 If the temperature is t c the volume of the glass is i8 - i 

 ()-|-o 000028/) and that of the mercury 2 - 8 (i-j-°°°oi8i/), 

 and the residual volume is 7'= iS'i —2-8 as before. The 

 effect of variation in the coefficient of expansion of the 

 mercury is thus reduced to a minimum 



When a cylinder of ice had been frozen in the instru- 

 ment, it was immersed in a mercury bath, and subjected 

 to variations of temperature, either with freezing mixtures, 

 or, in winter, by exposure to the atmosphere. 



These series of experiments were made with distilled 

 water. The first series was made with water taken from 

 the store jar in the laboratory. It gave a slight opalescence 

 with nitrate of silver, and cannot therefore claim to have 

 been pure. The ice formed by its congelation expanded 

 with rise of temperature from — 20 C. to — o° - 3 C. Here 

 it began to contract until it melted at 0° C. Two other 

 series of experiments were made with water repeatedly 

 distilled. The ice from it did not begin to contract 

 till the temperature had risen to -0-03° C. 



There can be no doubt, especially in view of later expe- 

 riments with brackish waters, that the not chemically 



pure distilled water did contract at a measurable distance 

 from its melting point. With regard to the other two 

 samples, the temperature at which the ice began to expand 

 with heat is so close to its actual melting point, that it is 

 impossible to have implicit reliance in the result claimed. 

 The author's own view will be best judged from the fol- 

 lowing paragraph (p. 282):— 



" It is impossible to decide if absolutely pure water 

 would be entirely free from this weakness or not. since 

 we cannot assume that water which has boiled for a 

 quarter of an hour or more in a glass vessel is absolutely 

 free from minimal quantities of foreign substances as, for 

 example, sodium salts, silica, &c. For my own part I 

 am rather inclined to think that absolutely pure water, if 

 it could be tested, would show an absolutely fixed melting 

 point, but I think that this problem very much resembles 

 another question still undecided, viz. is absolutely pure 

 water a conducting or non-conducting substance for 

 electricity?" 



It would be well to repeat the experiment with purt 

 freshly distilled water, freeing it from air by boiling in 

 vacuo, which Dr. Petterson's apparatus would easily admit 

 of. There would then be very much less risk of the glass 

 being attacked. 



Experiments made with sea-water ice proved that the 

 property of contracting with heat, as the melting point is 

 reached, becomes more and more marked the greater the 

 quantity of salt in the ice. Three series of experiments 

 were made. In the first, the ice when melted had a 

 specific gravity of 1-0003, and contained 0-014 per cent, 

 chlorine. It began to contract at -4 C C. The second 

 had a specific gravity of 1-00534, and contained 0-273 P er 

 cent, chlorine. It began to contract at - 14° C. The 

 third had a specific gravity of 1-0094, and contained 0-649 

 per cent, chlorine. It was contracting at the lowest 

 observed temperature, - 19° C. 



In connection with these remarkable results it must be 

 mentioned that at the same temperature, as, for instance, 

 — 1 5° G, the volume of the ice which on being melteJ 

 furnishes 1 cc. water at o° G is less the greater the 

 amount of salt contained in it. Sea water being an ex- 

 ceedingly complex body, it is to be hoped that Dr. Petter- 

 son will extend his research so as to examine in the same 

 direction the ice formed by simple solutions of each of the 

 more important ingredients of sea water. How different 

 ice produced by the freezing of sea water must be from 

 what we are accustomed to see on our lakes and ponds in 

 winter, will be evident when we read (p. 286) : — ". . . 

 The new ice which arises by sudden freezing of the calm 

 surface of the Arctic sea is a tough substance, which can 

 be wrinkled and folded by external pressure without 

 breaking. Although it may be thick enough to bear the 

 weight of a man, it is so plastic that a footstep makes a 

 deep impression as in mouldable clay.'' 



The physical part of the work closes with the investiga- 

 tion of the latent heat of fusion of fresh and salt ice. 

 The result is that " the latent heat developed by the 

 freezing of sea water is extraordinarily inferior to that oj 

 pure water." 



Hardly less interesting than his physical experiments, 

 are the investigations into the chemical composition of 

 sea water ice. 



It has been very generally believed that sea-water ice 

 owes its salinity to mechanically entangled brine, and 

 that all that is really solid in it is pure ice. Scoresby, 

 probably the most acute observer amongst Arctic voyagers, 

 referring to this subject, says :— ' 



" Although I have never been able to obtain from the 

 water of the ocean, by experiment, an ice either compact, 

 transparent, or fresh, yet it is very probable that the reten- 

 tion of salt in ice may arise from sea water contained in 

 its pores ; and, in confirmation of this opinion, it may be 

 stated that if the newest and most porous ice be removed 



1 "An Account of the Arctic Regions," Edinburgh, 1820. vol. i. p. 230. 



