i68 



NATURE 



Syune 23, i8'8i 



coefficient -^ of the general form above mentioned will still 

 d t 

 xist for pressures below 4'6 mm. and temperatures higher than 

 .-0"007S'' C, where it is now supposed to have a point d'arict, 

 to be "naturally erroneous," an assertion which I hope the 

 learned English inquirer may feel inclined to withdraw after the 

 preceding explanation. Of course the necessity of my conjecture 

 may be disproved by facts — if there really should exist no "hot 

 ice " — but nevertheless it deserves to be discussed as well as the 

 opinion of Dr. Lodge, who considers the existence of hot ice to 

 depend entirely on an irrever.sible process of vaporisation from 

 the ice resembling the evap iration of water in an atmosphere 

 which is not saturated with damp. Tnis ob ervation only regards 

 the experiment, not the theory. I fully admit that Dr. Car- 

 nelley's experiment is carried on iu an irreversible way, but that 

 is the case with every distillation or sublimation which is piacti- 

 cally performed. Nevertheless there exists a line m n where the 

 liquid water changes gradually and reversibly into saturated 

 vapour, and that may be the case also with the hot ice at tlie 

 limit 7H k. Any irregularity iu the operation will not exclude 

 the possibility of the existence of an equilibrium established by 

 nature. The difference of temperature between the hot ice in 

 the experiment of Carnelley and the cooled vacuum bottle is no 

 objection to this, because we might carry on the operation in 

 quite another way, dispense with the vacuum bottle and the 

 cooling mixture, and Iveep up the necessary minimum of pressure, 

 which is the only sin,: quA non, by means of a powerful air- 

 pump. In a similar experiment (with HgClj) Dr. Carnelley 

 operates in that way. Dr. Lodge, on the contrary, is con- 



?o 0/ n^.sije 



JUctr^irtet: 



KJ 



vmced that ice, which has once' passed the triple point ;;; can 

 sustain whatever augmentation of pressure and temperature 

 may be applied to it. We may destroy the vacuum, so 

 scrupulously kept up by all experimenters, and allow the air to 

 enter through hot pipes, nevertheless the ice will not melt. I 

 interpret this in the following way : — Dr. Lodge admits thzt the 

 surface v =■ fSf, t) extends over the limit I m, and even sur- 

 passes the critical pressure 4"6 mm., the vaporisation of the h ,t 

 ice going on irreversibly the whole time.^ This is indeed an 

 interesting hypothesis, which well deserves to be tested by 

 experiments, but yet lacks any foundation from facts. I there- 

 fore think that the proper method of resolving the entire 

 problem would be : — 



1. To try (by experiment) if ordinary ice under low pressure 

 by sufficient supply of heat can be made to pass over the limit 

 Im and assume higher temperatures than those coiTCsponding to 

 the equation x (/, t) = o (or Regnault's table). 



2. If this should be the case it remains to ascertain if the 

 vaporisation of the "hot ice "tends towards any limit (mk), 

 where this process becomes reversible, saturated vapour being 



^ Or : Dr. Lodge supposes that the volume of ice which has once passed 

 the limits, beyond which liquid water cannot exist, is totally independent of 

 the temperature and pressure. In this case no theory can be applied to 

 account for the existence of hot ice, because every tlieory must start from 

 the assumption that there exists a certain relation between the variables 

 v,p, and t, and that the volume of ice. as long as it is ice. is not af^itiary^ 

 but regulated by an equation v =/J./>, t). Therefore I do not think that 

 this explanation can be in accord with the views of Dr. Lodge. 



formed [my conjecture], or if tliere is no such limit [theory of 

 Dr. Lodge]. 



The apparatus employed (see Fig. 2) differs from those 

 recently used by Messrs. Boutlerow, McLeod, L. Meyer, &c., 

 only by its combination with a barometer, by means of which 

 the variati m of the pressure of the vapour given up by the ice 

 during the whole process could be exactly measured. The only 

 drawback to this was that the barometer of the apparatus did 

 not instantly indicate the variation of the pressure, because the 

 upper part of the.barometer was made of a very wide glas ; tube 

 to avoid the influence of capillarity. The effect of the vacuum, 

 which consisted of a 4-litre glass bottle, was very powerful, 

 since the full heat of two strong gas-lamps, each furnished 

 with three pipes, must be employed on the outside of the 

 glass tube in order to raise the temperature of the ice 

 covering the bulb of the thermometer from - 15° C. or 

 - 11° C, up to 0° C. The result of the experiment was 

 (the ice being heated only by radiation from the glass tube) : — 

 By intense heating the temperature of the ice sHwly (in about 

 six minutes) rose from — 11° C. to 0° C, when it became 

 constant for half a minute. Then tlie ice melted, and the first 

 drop of water falling upon the bottom of the heated glass tube 

 was sufficient to crush the apparatus. During the process of 

 heating the niveau of the mercury in the barometer-tube con- 

 stantly fell, the internal pressure augmenting as tlie temperature 

 of the ice ro e. It was quite impossible to raise the temperature 

 of the ice without simultaneously augmenting the pressure. 



Experiment I. 

 The ordinary barometer showed = 756S mm. 



The barometer of the apparatus showed = 755 'o mm. 

 The initial pressure in the apparatus = i"8 mm. 

 The initial temperature of the ice ^ — ii°'o C. 



By heating the temp. 



the press. =-i*5 



„ =2-90 



=3'5 



The ice melted. 



Experiment II. 



= - 15° C. 



'=- 9°P-= 1-8 mm. 

 t = - 6-p.= 2-6 mm. 

 The mercury in the 

 stem of the thermo- 

 meter separated by 

 the heat. 



Table of Regnault. 



Tension of saturated 

 vapour at — 

 t=-id'C.;p = 0-927 

 / = — 15" C. : / = I '700 

 / = - io° C. ; / = 2 093 

 t = - 5'C. ; /= 3-113 

 ^ = — o C. ; ^ ^ 4-600 



I also repeated the experiment of Mr. Haunay by substituting 

 a little sealed tube c^intainiug frozen water under atmospheric 

 pres.sure, instead of the bulb of the thermometer. I found, 

 in accordance v\ ith Mr. Hannay, that the enveloping ice melted 

 before the ice in the tube. 



Aftertheexperiments published by Messrs. McLeod, Boutlerow, 

 L. Meyer, v. Hasselt, de la Riviere, and Hannay, I think it may 

 be considered as a matter of fact that ordinary ice under low 

 pressure cannot be heated over 0° C. In the experiments I. and 

 II. I vainly tried to raise the temperature of the ice without 

 simultaneously augmenting the tension of the vapour in the 

 apparatus. It seems probable therefore that the area correspond- 

 ing to V =/; (/, t) dues not extend farther than to the limit !m 

 [equation x (/> '') = o], since the temperature of the ice and the 

 tension of its vapour vary almost exactly in the ratio given by 

 Regnault's table, which in Fig. I is represented by the line 

 I ni. We may conduct the heating of the ice so as to follow 

 almost continuously the line tnl [Experiment I.] without ever 

 being able to pass over it or to reach temperatures situated 

 beyond /w (i.e. in the area /?« ,f). Still I think these experi- 

 ments to be strictly convincing only in the case of ordinary ice. 

 Nobody has yet repeated Dr. Carnelley's experiment exactly 

 in the same way as Dr. Carnelley himself. In his experiment 

 the ice on the bulb of the thermometer is formed not by the 

 freezing of a quantity of water, but by the sublimation and con- 

 densation of icy vapour to thin layers. It may be possible that 

 ice, by sublimation under low pressure, changes into another 

 allotropic modification, just as the red modification of Hgl^ is 

 changed into yellow iodide by sublimation. In this case we may 

 foresee the existence of a new surface, v —f„ [p, t) on the other 

 sideof /;«. For, according to the principles of the theory of 

 mechanical heat, there ought to be a new function, v = f„{p, t] 

 for every new allotropic modification of a body which, geo- 

 metrically, is represented by a surface (?? in the figure). We are 

 scarcely autliorised to deny the possibility of the existence of hot 

 ice, since Dr. Carnelley has obtained several pieces of ice, which 



