384 



NAl^URE 



[Fed. 24, 1 88 1 



I'o obtain the rapid evaporation and superheated state of ice, 

 the U part of the tube was placed in a freezing-mixture of ice 

 and salt till the condensed water coated the tube infernally witli 

 a hollow sleeve of ice extending in the long and short parts of 

 the U to a total length of about eight inches, and the flask itself 

 was thereupon placed in a sawdu-t-bedded tin pail containing a 

 large freezing-mixture of 5 lbs. of salt and 10 lbs. of pounded 

 ice. Hydrochloric acid was at the same time added to the 

 _frigorific mixture round the U-tube, lowering the temperature of 

 that part of the cryophorus to - 29° C. ( - 20° F.), which had tlie 

 effect of cracking the ice-sleeve (apparently by contraction) in 

 all directions, giving rise at first, from its appearance, to tlie 

 apprehension that the glass tube was completely splintered ! Now 

 came the critical operation. Would the hard frozen ice-coat bear 

 the application of the heat ordeal \iithout melting? 



The freezing mixture round the U-tube was replaced by cold 

 water in a water-bath, which was heated rapidly with a Bun- en- 

 flame. About two inches of tlie ice-sleeve in the long part of 

 the tube stood above the surface of the water; the re-t soon 

 melted, while the teaoperature of the water in the bath rose to 

 70° C, and a thick rime of white fi-ost from the condensed 

 vapours of this water formed round the tube containing tlie 

 projecting part of the ice-tongue, which must have been intensely 

 cold, since the rime almost touched the surface of the warm 

 water. Below the surface of the water the tube appeared always 

 to contain some snowy-looking solid, along with what resembled 

 water proceeding from its liquefaction. When the temperature 

 of the bath reached 70° C. ihe projecting ice-tongue was stil] 

 unmelted ; it had replenished itself by condensing and freezing 

 the vapours rising up to it from belou-, and formed a snowy 

 pluJ in the tube an inch or more long. 



The water-bath was then removed in order to heat this snow- 

 plug with a Bunsen-flame, and to our extreme surprise a similar 

 snow-lining was found still to remain coaling, quite dry, the 

 whole long part and much of the short part of the U'ljend. 

 The flame was ajiplied, and the whole tube was heated vi ilently 

 without for some time appearing to have the least effect upon 

 the white crust within, notwithstanding the tube was too hot to 

 be touched ! A small flake weighing at most a few tenths of a 

 grain, at the bottom of the U, withstood the strongest heat tliere 

 for several minutes, wasting insensibly away, and unchanged in 

 shape, until it vanished with perhaps a moment's collection to a 

 drop, as it dis^'.ppeared. As it slowly grew thinner tire -.ihite 

 coating in the tube seemed in general to be no more affectel by 

 the heat than wliite feathers would have been ; but in particular 

 parts there often occurred partial liquefaction or pasty fusion 

 allowing pieces of the film to turn over by their weight and roll 

 or slip down the glass while still adhering to it, quite in the 

 manner of drops. That the liquid itself, completely freed from 

 air, refuses to boil, and may be described in that respect as practi- 

 cally oily, was evident enough in the preparation of the flask, 

 when except by sudden bumps the distillation of vapour, how- 

 ever much it was urged by heat, and seen to be passing copiously 

 through the bent worm-tube, proceeded almost entirely from the 

 surface of the water in the flask. 



In these drops at least, adhering to the heated glass, it seems 

 difficult to believe that the solid ice which makes them angular, 

 jagged in form, and pasty, can be anything but superheated ; 

 and it seems also scarcely credible that the latent heat of sub- 

 limation of the insignificant weight of a few grains of ice-crust 

 in the tube resisting its strong heat for many minutes can reach 

 the large amount of its gain of heat in that time by conducli n\ 

 and radiation from the surrounding tube, if the ice-film is at no 

 higher temperature than its natural melting-point. The^e are 

 questions however which the calorimetric methods devised and 

 pursued by Prof. Carnelley are be^t designed to ansner, and to 

 which replies without such positive determinations can only be 

 urged at present as probable conjectures, or as clear preceptions 

 and presentiments on the other hand, of the action and opera- 

 tion of some hitherto undiscovered thermal laws. 



Whether by direct sublimation, or by evaporation accomjianied 

 by liquefaction, the slender snowdrift with its enlargement at 

 the top was gradually reduced in thickness, and subdivided into 

 remaining parts along its length, which all adhered to the lube 

 until they wasted away, and the largest top piece never fell by 

 withdrawal of support below, but, like the rest, it clung lii.e a 

 thin scale to the last to its original place on one side of the tube. 

 This power of adhesion of hermetically volatilised ice to hot 

 bodies with which it is in contact forms, as Prof. Carnelley 

 observes, one of its most marked, although not at all one of its 

 most prognosticable, properties, and it certainly prompts, if it 



does not unequivocally substantiate, the supposition that the 

 ice in such close contact with extremely hot bodies must be 

 superheated, or at a temperature seu'^ibly higher than its natural 

 melting-point. 



The perfect success of the experiment and the beautiful ap- 

 pearance of the ineffaceable white snow-lining of the tube 

 afforded us unqualified pleasure and surprise, and the simple 

 preparation and preservation of the cryophjric apparatus needed 

 for its exhibition will, I have no doubt, cause the experiment to 

 be often reproduced and shown wherever freezing-mixtures and 

 the cryophorus, and Boutigny's and Leidenfrost's phenomena, 

 and Prof. Carnelley's theory and researches, are subjects of 

 lecture-demonstrations. A. S. Herschel 



College of Physical .Science, Newcastle-on-Tyne, February 15 



Dust, Fogs, and Clouds 



I NEED not say that the information cmtained in M. H. J. 

 H. Groneman's letter in Nature, vol. xxiii. p. 337, was a most 

 unexpected surprise. Nothing whatever seems to have been 

 known in England about the results obtained by Messrs. Coulier 

 and Mascart ; and my astonishment was not diminished when I 

 considered that their important investigation had borne no fruit, 

 never having been practically applied by meteorologists or 

 others. 



I have just seen the. article by M. Coulier in the yoiirnal de 

 Pharmacie et de Chimie, and will (with your permission) make a 

 few remarks on his paper and on my own. M. Coulier was led 

 to the discovery of the important part played by dust in the con- 

 densation of vapour, by making experiments on the well-known 

 cloudy candeii-^ation produced by expanding saturated air. " In- 

 stead of the ordinary air-pump arrangement M. Coulier's appa- 

 ratus consisted of a glass flask, in which was placed some water. 

 This flask was connected by a tube to a hollow india-rubber ball. 

 He first compressed the india-rubber ball, thus compressing the 

 air in the flask. The pressure was then removed from the ball, 

 when the air in the flask expanded, and the condensation became 

 visible. 



In making repeated experiments with this apparatus M. 

 Coulier noticed that the action of the air was capricious. After 

 the air had remained in the flask some days he found it quite 

 inactive. He also found it inactive after shaking it up with the 

 water in the flask, and that on expelling S3ine of the inactive air 

 and replacing it with the air of the room the cloudiness again 

 appeared. He then found that if he filtered the air before 

 admitting it to the flask it did not give any cloudiness when 

 expanded, and he explains with perfect clearness that the dust in 

 the air formed the nuclei rn \^hich the vapour condensed. 



Having explained the manner in which M. Coulier approached 

 the subject I shall now give the history of the corresponding 

 part in rfy work and show the direction from which I approached 

 it. I had been studying the action of "free surfaces" in water 

 and other forms of matter, when changing from one state to 

 another. I knew that water could be cooled below the freezing- 

 point ; I was almost certain ice could be heated above the melting- 

 point, and I had shown that water could be heated far above the 

 boiling point ; that the material of which the vessel holding the 

 water was composed had no influence on the result — and all that 

 was necessary to prevent the change of state taking place, at the 

 freezing and boiling points, was an absence of " free surfaces " at 

 wliich the change could begin. Arrived at this point, the pre- 

 sumption was very strong that vapour could be cooled below 

 the " condensing point " without changing to water, if no " free 

 surfaces " were present. I first intended to experiment with 

 steam at the pressure of the atmosphere, but found difficulties in 

 operating at so high a temperature. I then made arrancrements 

 to conduct the experiments at a lower temperature, and for con- 

 venience experimented on steam mixed with air. I then saw 

 that dust in the air would form "free surfaces" on which the 

 vapour would condense. I therefore attached a cotton-wool 

 filter to the apparatus, and filtered the air before it entered the 

 receiver. When this was done I found that the steam on entering 

 the receiver was perfectly invisible, and gave rise to no cloudy 

 condensation, the air remaining supersaturated. The experiment 

 was immediately afterwards repeated somewhat in the same way 

 as was done by M. Coulier, the air being supersaturated by 

 expanding it by means of an air-pump. 



Though the two investigations approached the subject from 

 very different points, they seem afterwards to have flowed in 

 almost parallel directions. Starting from his first experiments, 



