396 



NATURE 



{Sept. 12, 1873 



MELTING AND REGELATION OF ICE 



IN Nature of January 4tli of this year, there is a most 

 interesting account of some experiments on melting 

 and regelation of ice by Mr. James T. Bottomley. These 

 experiments of Mr. Bottomley's suggested the possi 

 bility of passing large bodies through ice in the same 

 way as he caused the wires to pass. I accordingly placed 

 a sixpence on a block of ice, and applied pressure to it 

 by means of a fine steel wire about one-sixteenth of an 

 inch in diameter. On examining the block of ice some- 

 time afterwards, I found the sixpence had passed into the 

 centre of the block, and that the space through which it 

 had passed, except the small part occupied by the steel 

 wire, was again solid ice. I tried the same experiment 

 with a shilling, and found that it also easily passed through 

 the ice, the experiment was then repeated with a half-crown 

 with the same result. I did not attempt anything larger, 

 but have no doubt much larger discs of metal might be 

 made to pass through ice if sufficient pressure were applied. 

 The ice in the parts of the blocks through which the 

 coins had passed did not look very solid, but was rather 

 full of air- bubbles ; on breaking the block, however, it did 

 not seem much weaker than the rest of the ice. Another 

 form of the experiment was then made, a block of ice 

 was supported on two boards placed near each other. A 

 loop of fine wire was passed over the ice, and hung down 

 between the two boards and a weight attached to it, as in 

 Mr. Bottomley's experiments, pieces of wood were placed so 

 as to slop the wire when it had passed half way though 

 the ice. After the wire had passed into the centre of the 

 block, the weight was removed, the wire cut, and a disc of 

 metal half an inch in diameter was attached to one end 

 of the wire, and a weight to the other end. In this manner 

 the disc was drawn through the ice, leaving apparently 

 perfect solid ice behind. The path of the disc could only 

 be traced by its slightly cloudy appearance, it looked as if 

 the few air-bubbles passed through by the disc had been 

 broken up into a great number of small ones. On 

 breaking the ice afterwards it seemed quite as strong 

 where the disc had passed as elsewhere. 



The explanation of these experiments is of course 

 the same as for the experiments with the wires ; Pro- 

 fessor James Thomson showed that the freezing 

 point of water is lowered by pressure, and also that 

 ice has a tendency to inelt, when forces are applied 

 which tend to change its form. So that the ice 

 under the coins has a tendency to melt, and has its 

 freezing point lowered by the pressure. The under side 

 of the coin will thus have a lower temperature than 

 the upper ; there will therefore be a transference of 

 heat from the upper to the under side of the coin, this 

 heat melts the ice under the coin, the water so formed 

 passes round the edges of the coin to the upper side. 

 This water being at a slightly lower temperature than the 

 freezing point at ordinary pressure, a vtry small propor- 

 tion of it will freeze and raise the temperature of the rest 

 to the freezing point. The water arrived at the upper side 

 of the coin, the coin being at a temperature a little below 

 the freezing point, the water will be frozen, giving out its 

 latent heat, which will pass through the coin and melt an 

 equal quantity of ice on the under side, this having ab- 

 sorbed its latent heat of liquefaction will in turn pass 

 to the upper side, and will there be converted into ice, 

 giving out its latent heat to melt another quantity, and 

 so on. 



A slightly different form of the experiment was then 

 made, a small metal cup was filled with water and laid 

 en a piece cf ice, and a heavy weight placed on the 

 cup. After some time the water in the cup was frozen. 

 Tiie freezing point of the ice under the cup being, 

 owing to the pressure, lower than that of the water in 

 the cup, the water in the cup parted with its heat to 

 the ice outside. A quantity of ice outside the cup was 



thus melted equal to the quantity of ice formed in the 



cup. 



At first sight these experiments might seem to have 

 an important bearing on the motion of glaciers. It 

 might be thought, that if large bodies flowed thus easily 

 through ice, why should not ice flow easily in its 

 channel? But when we consider the circumstances, we 

 find they are not so similar as might at first appear. 

 When a body flows in this way through ice, there is 

 not only a displacement of matter but also a displace- 

 ment of heat, and the displacement of the matter cannot 

 take place till there has been a displacement of the heat. 

 In the preceding experiments, circumstances were most 

 favourable for both displacements taking place quickly. 

 The heat easily flowed through the very small thickness 

 of the good conducting silver discs, and the water had 

 only to flow from the one face to the other round the 

 edges of the coins, whereas in glaciers, the ice and the 

 rocks over which it moves are bad conductors of heal, 

 and the distance to which the heat has [to be conducted 

 is so much greater than in the above experiments, that 

 the exchange of heat can take place but very slowly ; and 

 when we further remember the very small difference of 

 temperature between the freezing point of ice under pres- 

 sure and not under pressure — if the lowering of the freez- 

 ing point is the result of hydrostatic pressure alone, a 

 pressure of one hundred atmospheres not lowering the 

 temperature one degree centigrade — we can easily see that 

 there will not be sufficient difference in temperature be- 

 tween the different parts of the glacier to cause the heat 

 to flow quickly from one part to another, through such bad 

 conductors. 



In the explanation given of the passage of the coins 

 through the ice, it has been assumed that the passage 

 depends on the exchange of heat from the freezing ice on 

 the one side of the coin to the melting ice on the other 

 side. If this explanation is correct, then, if the coins had 

 been non-conductors of heat, they would not have passed 

 through the ice. The test was put. A shilling was placed on 

 a block of ice, and over it a disc of a non-conductor (india- 

 rubber),the same size as the shilling and over that another 

 shilling; a weight of 90 lbs. was applied by means of a small 

 steel rod. After four hours it was found that the shillings 

 had only sunk about an eighth of an inch into the ice, most 

 of the heat to sink it this short distance being, in all pro- 

 bability, got by radiation from surrounding objects ; as 

 other two shillings and non-conducting disc placed on a 

 block of ice and similarly situated, but not under pres- 

 sure, had sunk to nearly the same depth. 



There is another point in these experiments in their 

 relation to glacier motion, which requires to be noticed. 

 In all the experiments referred to, ice at the melting 

 point was used. Sir William Thomson showed that the 

 freezing-point of water was lowered o"'i3 C. by a pressure 

 of 168 atmospheres. We should therefore expect that, if 

 we lowered the temperature of the ice by half a degree or 

 a degree below the freezing-point, a much greater pres- 

 sure would be required to cause the coins to pass through 

 the ice. In order to test this, a block of ice was sur- 

 rounded with ice, salt, and water. After it was cooled 

 about a degree below the freezing-point, a shilling was 

 placed on the block of ice, and a pressure of 90 lbs. ap- 

 plied. On examining it three and a half hours afterwards, 

 the shilling was found not to have entered in the slightest 

 degree into the ice. The freezing mixture was then re- 

 moved, and within an hour the shilling had passed some 

 distance into the ice. It would therefore appear, con- 

 sidering the enormous resistance oft'ered by ice at a tem- 

 perature of even one degree below the freezing-point to 

 change of stale, that the motion of glaciers at the higher 

 parts, where their temperature is below the freezing-point, 

 IS, in all probability, not caused by the melting and regela- 

 tion of the ice in the s: me manner as in the experiments. 



Darroch, Falkirk John Aitken 



