?8o 



NA TURE 



[August 20, 1908 



restored v, the contraction so produced c, and the result- 

 ing apparent volume of the ice U = V — c. The values of 

 V are derived from the observed freezing points of specified 

 NaCl solutions. The results calculated for certain values 

 of t are given in Table I. : — 



Table L 



It we study this table we see that between the temper- 

 atures — 23° and 0° the coefficient of apparent dilatation of 

 the ice changes sign three times — namely, tu>icc at the 

 cryohydric temperature, and once at a higher temperature. 

 Between —23° and the cryohydric point — 2i°72 the ex- 

 pansion is uniform, the coefiicient being o-oooib. At the 

 cryohydric point the addition of heat produces contraction 

 without change of temperature ; the coefficient, therefore, is 

 —X. Above the cryohydric ternperature the volume in- 

 creases with the temperature, but at a gradually diminish- 

 ing rate, until at — 7°-o the increase of volume due to 

 simple expansion of the ice is exactly balanced by the 

 contraction due to induced melting. At this temperature 

 the coefficient of expansion changes sign, and between 

 — 7°-o and — o°.i, at which the ice has practically all 

 melted, the coefficient of expansion is negative. 



If the block of ice contained salt in the proportion 29.97 

 grams NaCl to 100 grams ice, it would expand uniformly 

 on being warmed from —23° to — 2i°.72, and would then 

 melt completely at that temperature. In the same way, if 

 it contained no salt or impurity whatever, it would, on 

 being warmed, expand uniformly, while its temperature 

 rose, until, at 0° C, it would melt completely. If the 

 ice contains salt in a less proportion than 1-7164 : 100 by 

 weight, then we witness the three changes of sign in the 

 coefficient of dilatation when the temperature rises from 

 below the cryohydric point to the temperature at which 

 the ice is finally liquefied. When the block contains, per 

 100 parts by weight of ice, less than 29-97 and more than 

 1-7164 parts of NaCl, the coefficient of apparent expansion 

 is negative at all temperatures above — 2i''-72. 



In Table II we have the upper critical temperature (t) 

 at which the coefficient of apparent dilatation changes 

 sign for blocks of ice having volumes ranging from 100 

 cubic centimetres to 100 cubic metres, each containing 

 1-5105 grams NaCl. Under V„ we have the initial volume 

 of the block of ice supposed pure and solid at 0° C, and 

 under v the volume of ice which can be melted under 

 the inducing influence of 1-5105 grams of chloride of 

 sodium at the critical temperature t, at which the ap- 

 parent coefficient of cubic expansion of the ice is equal 

 to o. 



Tarle II. 



A block of 100 c.c. of ice, which contains 1-5105 grams 

 of NaCl diffused through it, furnishes on being' melted 

 91-67 c.c. of water, which contain 0-9167 gram of chlorine, 

 dissolvedin it as chloride of sodium. This water contains 

 chlorine in the proportion i gram to 100 grams of water, 

 and represents a concentration about one-half that of 

 average sea water. When the volume of ice, V„, is 

 1 cubic metre, the water produced by its melting contains 

 chlorine in the proportion of one part to one million parts 

 of water by weight. 



Waters which contain dissolved matter equivalent to no 

 NO. 2025, VOL. 78] 



more than i gram of chlorine in 10,000 grams of water 

 are in the category of ordinary fresh waters, and we see 

 that the critical temperature of ice which furnishes such 

 water lies as low as — 2°-3. When the dissolved matter is 

 equivalent only to i gram of chlorine in 100,000 grams of 

 water, the critical temperature is — o°-725. The other 

 waters are in the category of distilled waters, and it is 

 doubtful if, by any chemical means whatever, we could 

 determine as little dissolved matter as i gram chlorine 

 in one ton of water ; yet the critical temperature of such 

 ice lies nearly a quarter of a degree below the melting 

 temperature of pure ice. The critical temperature of 

 expansion of ice affords a means of detecting impurity 

 equivalent to quantities of chlorine as small as one gram 

 in ten tons, and even one gram in one hundred tons of 

 water. 



Influence of Impurity on the Apparent Latent Heat of Ice. 



This is illustrated bv the numbers in 'I'able I. Thus, 

 at —1° C, the apparent volume of the block of ice is 

 091-644 c.c, and it is made up of 901-49 c.c. ice and 

 90-154 c.c. water. When this is warmed to — o°-i, we 

 may take it that the whole of the ice is melted. Taking 

 the latent heat per unit volume of ice as 66-5 at — o°.i, 

 and its specific heat per unit volume as 0-45, the heat 

 required to raise the ice from —1° to — o°-i is 365-1 gram- 

 degrees (gr.°) ; that required to raise the temperature of 

 the water by the same amount is 81.14 S^°t an^i Ihe heat 

 required to melt the ice at — o°-i is 59949 gr.°, the total 

 heat used being 60395-2 gr.°. If we ignore the possibility 

 of partial melting, and assume that we have 999-84 c.c. 

 solid ice at —1°, and that its temperature is raised to 0°, 

 at which temperature it melts, we have the following 

 expenditure of heat : — for rise of temperature 4499 gr.°, 

 and for melting 66489-3 gr.°, making together 66939-2 gr.°, 

 as against 60391-5 gr.°. If from 60395-2 gr.° we deduct 

 the heat calculated for warming the ice in the second 

 case, 449-9 gr.°, we obtain 59945-4 gr.° as the heat required 

 to melt 1000 c.c, or 916-7 grams, of ice at 0°, whence the 

 latent heat would be, per unit volume, 59-94, ajid per unit 

 weight 65-30. 



This example illustrates also the effect of impurity on 

 the apparent specific heat of ice. 



The nature of the medium is responsible in the case of 

 sea ice for depressions of freezing and melting tempera- 

 tures of thirty, forty, or even more degrees of Celsius's 

 thermometer, while the greatest pressure to which fresh- 

 water ice is exposed in nature cannot produce an altera- 

 tion of freezing and melting point amounting to much more 

 than one degree. 



If we pick up a piece of ice floating in the Polar Sea 

 we know that it will prove to be very far from homo- 

 geneous. It may have a foundation of genuine primary 

 sea ice, but the ice forming the superstructure is sure to 

 consist of snow, frozen spray, and very likely fragments 

 of land ice, all cemented together into a species of con- 

 glomerate. When this is exposed to warmth it begins to 

 melt at a temperature which may be one or two degrees 

 below the melting point of pure ice, and the liquid so 

 furnished is salt water. The further melting takes place 

 in ascending order of temperature, the salt ice of low 

 melting point disappearing first, and the purer ice melt- 

 ing later. We thus see how ice can be cemented by ice, 

 just as metallic objects may be united by solder. In both 

 cases the substance of the binding material differs from 

 that of the objects united, chiefly in being more easily 

 fusible. 



If we have a number of cubes of pure ice which fit each 

 other exactly, and if, after being moistened with salt water, 

 they are exposed to frost, they will solidify to a single 

 block. If this bo exposed to the sun the cementing salt 

 ice will melt first, and, when it ceases to bind, the con- 

 stituent cubes of pure ice will fall asunder, having them- 

 selves suffered practically no diminution due to melting. 



Now this is precisely what happens when a block of 

 sound glacier ice is exposed to the rays of the sun for a 

 short time, and it is one of the most striking and in- 

 structive experiments that can be made. Under the in- 

 fluence of the sun's rays, the binding material melts first, 

 the continuity of the block is destroyed, the individual 

 grains become loose and rattle if the block be shaken, and. 



