472 



NA TURH 



[October 14, 1909 



6 is the geometric mean between the mean P.M. of stars 

 moving in the direction e and the mean P.M. of stars 

 moving in the direction e+i8o. A paper by Prof. H. T. 

 Barnes dealt with the variation of the specific heat of 

 mercury at high temperatures. Prof. Barnes, employing 

 the continuous-flow method, has now determined the 

 specific heat up to a temperature of 268° C. with an 

 «rror (for the higher temperatures) of not more than one 

 or two_ parts per thousand. The results are specially 

 interesting inasmuch as they show that the specific heat 

 of mercury passes through a minimum at about 140° C, 

 and then increases fairly rapidly up to the boiling point. 

 The minimum for water occurs at about the same relative 

 position between the freezing and boiling points. During 

 the discussion on the paper Prof. Perrv emphasised the 

 great need there is for an accurate determination of the 

 specific heat of superheated steam. Dr. T. Proctor Hall 

 exhibited an apparatus for making enlarged tracings of 

 sound waves from a cylindrical graphophone record, and 

 showed some of the results obtained with it. Prof. k. W. 

 Porter exhibited some transparencies of electric discharges 

 upon photographic plates. A paper bv T. Kinoshita was 

 read m his absence by Prof. Rutherford, on the photo- 

 graphic action of a. rays, and Prof. A. S. Eve con- 

 tributed one on secondary radiation bv 7 rays on 

 ■different metals. Prof. J. C. .McLennan then' read 

 one by Mr. W. T. Kennedv, on the active deposits from 

 actinium in uniform electric fields. The deposits were 

 obtained on both of two plates 2 mm. apart, under a 

 field of 250 volts at various atmospheric pressures. As 

 the pressure is decreased from atmospheric the amount of 

 active deposit on both electrodes gradually increases, passes 

 through a maximum value, and then 'rapidlv decreases. 

 The maximum on the kathode is about 27 times that for 

 the anode, and is formed at a different pressure. The 

 total deposit is independent of the electric field until 

 potentials higher than the sparking ones are applied. The 

 coefficients of diffusion of the emanation from actinium 

 mto CO,, air, and hydrogen are in the proportion 

 1/I-3/42. On Dr. O. Hahn stating that in the case of 

 actinium emanation it is difficult to know what we are 

 dealing with, the reply was elicited that in all probabilitv 

 a number of products are concerned in the experiments 

 described. A paper by Mr. F. W. Bates, on the effect of 

 -ight on sulphur insulation, was read by Prof. A. S. Eve. 

 The author concludes from some electroscope experiments 

 that sulphur in the presence of light becomes to a slight 

 degree a conductor of electricity to an extent depending 

 upon the intensity of the incident light. In the discussion 

 the resemblance to selenium was pointed out. Dr Hahn 

 elicited the statement that the effect exhibits no time-la«. 

 Dr. P. Pnngsheim mentioned that Rontgen had found "a 

 similar effect for paraflln. A paper followed bv Dr. T 

 l-ranck and Dr. W. Westphal on the charge upon gaseous 

 ions. Ihe authors consider from their experiments that 

 lownsend s doubly charged ions are onlv a small part of 

 the total lonisation, and that generation in an electric 

 held, contrary to Townsend's view, has nothing to do 

 with their formation. The two kinds can be separated 

 ty fractional diffusion if a piece of wire gauze be inter- 

 posed in their path. The coefficient of diffusion of the 

 double ions is found to be half that of the single ones 

 With a, 3, Y rays and point discharge no doubly charged 

 ions were found, though in the last case big charged 

 Clusters, due to chemical processes, were proved to occur 

 A paper on the re-combination of ions in air at different 

 temperatures, by Dr. P. Phillips, was read bv the recorder, 

 in the experiments outlined in this paper Langevin's 

 niethod IS adopted. The ravs produced bv a single dis- 

 charge in a Rontgen bulb ionise a layer 'of air between 

 two paralle electrodes, one of which ' is connected to a 

 Dolezalek electrometer and the other raised to anv desired 

 potential. These plates are 3 cm. apart, and the' laver of 

 ionised air is 15 cm. thick : in these circumstances diffusion 

 IS probably negligible. The whole is surrounded bv a 

 vapour jacket. The experiment consists in measuring" the 

 charges received by the electrometer with different electric 

 helds and temperatures. The following values for a, the 

 coetticient of re-combination, were obtained, and are put 

 alongside Prof. Erikson's values recently obtained [VhW. 

 Mag., August) for ions produced by radium:— 



NO. 2085, VOL. Si] 



'l'emper.iture a trickson 



15" C I 00 fOO 



100 o'5o o'5i 



15s 0'40 0405 



178 0-36 0-38 (extrapolated). 



The day's session was concluded with a paper by Prof. John 

 Zeleny and Mr. L. A. McKeehan, on the terminal velocity 

 of fall of small spheres in air. Experimenting with 

 spherical spores, the authors find divergences from 

 Stokes's law for the limiting velocity as follows : — 



Substance Radius Velocity Theoretical 



Lycopeidon ... 0'0OO2O7 cm. ... o "04 65 cm. /sec. ... 00757 



Polytiicum ... 0-000478 ... 0228 ... 0'417 



Lycopodium ... 000158 ... 177 ... 3-52 



These experiments have an important bearing upon deter- 

 minations of the numbers of nuclei in cloud condensations. 

 Sir Joseph Larmor attributed the divergences to the in- 

 applicability of the usual theory to particles comparable 

 with the length of the mean free path, and recommended 

 that e.xperiments be made in air of different densities. 

 Prof. Hull agreed that the theory must fail, but expected 

 that the divergences should tend the other way. In 

 answer to Prof. E. W. Brown, it was stated that no 

 Brownian motion or rotation was visible, and the fall was 

 quite steady. 



On Tuesday, August 31, the section began with a dis- 

 cussion on earth tides, opened by Prof, h, E. H. Love. 

 Prof. Love stated that Lord Kelvin had shown (1S63) that 

 if the earth could be regarded as homogeneous and abso- 

 lutely incompressible, and possessed of the same degree of 

 rigidity as steel, the oceanic tides of long period would be 

 reduced, owing to the yielding of the earth, to about two- 

 thirds of the theoretical heights which they would ha\'e 

 if the substance were absolutely rigid. Sir G. Darwin 

 (1881) estimated the actual height of the fortnightly tide 

 as about two-thirds the theoretical height. .Attempts to 

 measure directly the lunar disturbance of gravity were 

 made by several observers ; and recently Dr. O. Hecker, 

 by using two horizontal pendulums mounted in an under- 

 ground chamber, has demonstrated the existence of the 

 corporeal tide, and has shown that the deflection of such 

 pendulums is about two-thirds what it would be if the 

 earth were absolutely rigid. This means that, besides the 

 tide-raising force, F, of the moon, there act on the pen- 

 dulum other forces arising from the deformation of the 

 earth. These forces are (i) the component of undisturbed 

 gravity tangential to the deformed surface, denoted by 

 hV ; (2) a genuine disturbance of gravity, consisting in the 

 attraction of the tidal protuberances and other related 

 changes of the attraction of the mass of the earth, denoted 

 by k¥. The results obtained by Darwin and Hecker, and 

 confirmed by Schweydar, show that the two numbers 

 h and k are connected by the equation /i — fc = i/3. To 

 find h and k separately we must have recourse to hypo- 

 thesis or to new observations. If we adopt Kelvin's hypo- 

 thesis we find fe = 3/!/5, and thence /i = 5/6, fc=i/2, and 

 the corresponding estimated height of the corporeal lunar 

 tide is about 46 cm. If, however, we bring in the fact 

 of observation, discovered by Dr. S. C. Chandler, viz. 

 that the period of variation of latitude (about ten months 

 if the earth were absolutely rigid) is actually about four- 

 teen months, we can determine k in terms of known 

 quantities. Variations of latitude imply an adjustment of 

 the earth's figure to rotation about an instantaneous axis 

 which does not quite coincide with a principal axis. The 

 corresponding inequality of " centrifugal force " has the 

 same effect as a certain external force producing a de- 

 formation of the earth and a genuine disturbance of 

 gravity. If the force in question is denoted bv F, the 

 genuine disturbance may be denoted by /;F, where the 

 coefficient k is necessarily the same as in the tidal problem. 

 It has been proved independently by Sir J. Larmor and 

 Prof. Love that k is about 4/15- It thence appears that 

 ''=3'.'i approximately, and that the height of the corporeal 

 lunar tide is about 33 cm. The earth would therefore 

 rppear to be more rigid than Lord Kelvin estimated it to 

 be, a result confirmed by the interpretation of seismo- 

 graphic records. In the discussion which followed Sir J. 

 Larmor asked whether there was any evidence for 

 Wiechert's theory of the constitution of the earth, viz. 



