Jtine 23, 1 881] 



NATURE 



167 



came under my notice at Brown's Town, viz. two perfectly black 

 parents having a family all pure albinos. 

 lOngstou, Jamaica, May 26 Thomas Capper 



Singular Behaviour of a Squirrel 

 A. NEIGHBOUR of mine, «hose cottage is thickly surrounded 

 with trees, observed a squirrel, during the severe weather of 

 winter, occasionally stealing food from the troughs set out for 

 the poultry. At first it caused great commotion among the 

 bird^-, but latterly they were less uneasy in its presence. Taking 

 an interest in the wild creature he began to lay out refuse food 

 for it, including bits of ham, which it greedily appropriated. 

 Getting more courageous, it ventured within doors. After a time 

 it got caught in a trap set for rats underneath the bed. Being 

 freed from its irksome ptsition it was thought that the squirrel 

 would venture no more within doors. Neither the incident of 

 the trap nor confinement for some time within a cage availed to 

 restore to it its original shyness. With the coming of summer 

 its visits have been less regular, but occasionally it looks in still. 

 May not a habit like this, affecting only one out of many, he 

 looked upon as corresponding to a "sport" in the vegetable 

 world, and shed some light on the subject of the domestication 

 of animals ? The squirrel seems to have been quite a wild one 

 to start with, for there is no one in the district %\ ho had been 

 in the habit of keeping one as a pet. J. Sn.wv 



Dumfriesshire 



Hot Ice 



In reply to a very interesting letter on this subject recently 

 published in Nature (vol. xxiii. p. 504) by Dr. Oliver J. Lodge, I 

 wish to express my views of the theoretical and practical possibiliiy 

 of the experiment of Dr. Carnelley. I « i>h to start from some 

 well-known principles accepted by everybody acquainted with the 

 mechanical theory of heat and its applications. According to these 

 principles the volume "z^" (and also the total amount of internal 

 energy) of water can be expressed as a function of its pressure 

 "/" and temperature '7"; v=f{p,t). The form of this 

 function, which we need not discuss here, will change with 

 the state of aggregation, so that we shall have three different 

 equations expressing the volumes of water in the solid, liquid, 

 and gaseous form. 



^ ~J} Y' I "" , I fi and t being considered inde- 



^=A\P''\ "■'^'^'^ r pendent variables. 



i'=y[.j(/, vapour) '^ 



Geometrically the volumes of ice, water, and vapour will 

 belong to three different surfaces extending between certain 

 limits. Thus the surface v =f,[p, /)> which represents the 

 volumes of ordinary ice, is situated between the limits ?/, /?«, 

 >ii d ; the surface representing liquid water lies between m n and 

 m d, though it may be extended a little on either side of these 

 limits, if it applies to water heated or cooled over its regular 

 boiling or free2ing temperatures, A\hich are situated along the 

 lines 1)1 d and m'n?- The values of/ and /, which belong to m d 

 and III n, will satisfy two equations — <^{p, = o and if (/>,/) = o. 

 At these points the water will change its form of aggregation 

 and pass over in the state of saturated vapour along the line 

 mn [equation i/{p, t) = o\ or into ice along md [equation 

 (^ (/>, t) = o] in a continuous and reversible way. At any other 

 point, which is not situated on m n or ind, water may also be 

 liable to change of aggregation, but this process will not be 

 rtversibh: The line in n, where the surface v =/,i{/', t] breaks 

 up and liquid water changes into vapour, is the ciiife of ttnsion 

 of saturated vapour contained in the renow ned table of Regnault. 

 The boiling-points of water under varying pressure are situated 

 on in II, and may be found by solving the equation ij/ (/, t) = o. 

 At the point in (p = 46 mm., t = ~ 0°'007S C.) the line in n 

 terminates, but is continued by / in [equation x (/> ') = °]i along 

 which the vaporisation of ice takes flace in a reversible way. 

 According to the table of Regnault there is no sudden rupture 

 at the point m, the pressure of saturated vapour r.t O C. being 

 identically the same, if the vapour is in contact with water or 

 dp 

 dt 



^ The surface corresponding to the volumes of aqueous vapour v =yijiC/. /) 

 is not sketched in the figure, which gives only the projection of the surfaces 

 on the plane of co-oidinates p and ;", not the real situation of these surfaces 

 in space. The reader will also observe that the limiting lines ;« «, nd, Im, 

 ink are the intersections of vertical cylindrical surfaces ('' Uebergangs- 

 flachen ") with the plane '', /. 



"/> (/> ')> + (/' '). *Q<i X (A ')) Of 'he tangents to the lines in d, 

 III H, and nt I are found by application of Carnot's Theorem to 



be of the general form -£- — ^-^ \r = latent heat ; 



dt \s - s{\\2.^l + tY 

 s and ^1 = the specific volumes of water in two different forms 

 of aggregation]. 



The point m, where m n, in d, and in I unite, is of particular 

 interest. J. Thomson called it " the triple point," and Guldberg 

 the " Fiillespunkt " of water. Lately (in Berichte, 18S0) I 

 ventm'ed to call it the "absolute point of sublimation," not 

 because I wished to introduce a new term for a well-known 

 scientific object, but only to point out some important conse- 

 quences of the phenomenon just then announced by Carnelley, 

 of which Prof. Lothar Meyer of Tiibingen had published an 

 interpretation different from mine. This point m, situated 



-o°'0078 C. below the ordinary freezing-point of water, is really 

 the upper limit of sublimation, because at any higher tempera- 

 ture ice first changes into water before it evaporates. At 



-o°'oo78 C, where the boiling- and melting-point of water 

 coincide, a real sublimation of ice begins, provided that the 

 barometric pressure does not exceed 4'6 mm. (="the critical 

 pressure " of Carnelley). 



Now according to the discovery of Dr. Carnelley, ice at 

 pressures lower than 4"6 mm. would exist by temperatures up to 



-f 178- C. Thus the surfsice v - f^ (/>, /), which we have hitherto 

 supposed to be inclosed between the limits qp, q I, Itn, in d would 

 extend far beyond / in nearly up to k, but always at pressiures 

 smaller than 46 mm. Geometrically this new and unforeseen 



Sar, 



■■p •^•< 'mm 



■ Ai'C 



-i^+ra'C 



extension of the surface of ice is represented by the area hni. 

 Here the process of Carnelley, whereby ice of low presstu-e is 

 heated to astoundingly high temperatures, would go on. The 

 area. /»ii would of course be entirely a terra incognita to the 

 science of the present day, but there is nevertheless no theoreti- 

 cal objection why the surface of ice v = f,{p,t) should not 

 extend farther than to the limiting line I in pointed out by 

 Regnault. Confiding in the experimental proofs already furnished 

 by Dr. Carnelley, I concluded (Beriehte, 1880): if the surface 

 of ice really extends upwards to about + 178° C. there must be a 

 limiting line in k to the area I ink, since this area caimot extend 

 so far as to the dotted line in the figure indicating the critical 

 pressure =4'6 mm. At this new limit, ink, corresponding to 

 an equation |(/, t)—0, the vaporisation of the "hot ice" may 

 go on in a reversible way, just as liquid water gives up saturated 

 vapour at those pressures and temperatures which belong to the 

 line inn (equation if< (/, /) = o). The line ink would in many 

 respects be the continuation of in d (just as m I forms the con- 

 tinuation of in «), but naturally the syml ^Is entering the equation 



of its differential coefficient ' c. = ^.^ 



it (i-^i) (273 +/) 



must change 



then- signification on the other side of the point in, so that r 

 here would represent the latent heat of vaporisation of the hot 

 ice, s its specific volume, &c. I did not expressly mention this 

 in my paper in the Beriehte, because I thought it unnecessary. 

 This omission on my side may probably have misled Dr. O. 

 Lodge as to the real meaning of my words, since he declares 

 my opinion that an equation i(p,t)-a having a differential 



