714 



NATURE 



[February 26, 1920 



renders the first contingency highly improbable. For 

 a rate of change of temperature of i° per million 

 years the change in thermal energy involved is 33 ergs 

 per annum per c.c, while the change in gravitational 

 energy is 20 ergs per annum per c.c. These quantities 

 can be lumped together and regarded as one by 

 assuming the water equivalent of the earth's sub- 

 stance to be 1-6 times its actual value. Thus, neg- 

 lecting radio-active effects, the time taken to cool 

 1° would be 5-4 million years instead of 3-3 million, 

 as calculated without taking gravitational energv into 

 account. From the relative amounts of lead and 

 uranium found in rocks it has been calculated that 

 the time which must have elapsed since the formation 

 of the crust is of the order of 1,000,000,000 years. 

 The temperature of solidification was probably about 

 1300° C, so we have the data necessary to give the 

 present rate of cooling. From this, for any assump- 

 tion regarding the actual distribution of radio-active 

 material, the temperature at points within the earth 

 can be determined. — .Sir .Arthur Schuster : The in- 

 fluence of small ch.-mges of temperature on atmo- 

 spheric refraction. The paper is an investigation of 

 the possible deviation of the light from a star near 

 the sun due to the temperature changes in the atmo- 

 sphere produced bv the passage of the moon's shadow- 

 across the earth during an eclipse. It is shown that 

 while the actual displacements from this cause varv 

 widelv for slight differences in the assumed conditions, 

 they are alwavs negligiblv small compared with the 

 effects observed at the last solar eclipse. 



Royal Meteorological Society, Februarv 18. — Mr. 

 R. H. Hooker, president, in the chair. — Capt. 

 C. J. P. Cave : The status of a Meteorological Office 

 and its relation to the State and to the public (see 

 P- 705)- — W. H. Dines : .\tmospheric and terrestrial 

 radiation. The author endeavours to follow the flow 

 of radiant energy, other than solar, both upward and 

 downward across anv horizontal plane in the atmo- 

 sphere. Certain theoretical assumptions are made to 

 render the calculation possible, and it is shown that the 

 curves that represent the net loss or gain of heat from 

 strata at different heights are all more or less of the 

 same character, whatever possible values are ascribed 

 to the emissivity of the various strata. It is found 

 that over Europe the air from the earth's surface up 

 to about 8 km. is losing heat by radiation, and that 

 from 8 km. to 12 km. it is gaining heat, losing it 

 again at more than 12 km. The validity of the as- 

 sumptions made is then discussed, and it is pointed out 

 that the numerical values agree well with those ob- 

 tained by entirely different means. — D. Brunt : Internal 

 friction in the atmosphere. When a steady state of 

 motion is assumed, anv portion of the atmosphere is 

 in equilibrium under the action of three forces : the 

 gradient of pressure, the deflecting force at riffht 

 angles to its motion, and the frictional force. The 

 first two of these are measurable, and so the third 

 can be evaluated. The paper gives a comparison of 

 the frictional force calculated in this manner, with 

 the values derived from a theoretical discussion of 

 turbulent motion. .\ new derivation of the solution 

 of the equations of motion is given. Use is made 

 of observations at the top and base of the Eiffel 

 Tower to derive the value of the coeflficient of eddv 

 viscosity. .\n additional note shows that a solution 

 of the equation of motion is possible in cases where 

 the coefficient of eddv conductivity varies with 

 height. 



Cambridge. 



Philosophical Society, February q.— Mr. C. T. R. 



Wilson, president, in the chair. — F. W. Aston : The 



mass spectra of the chemical elements. By means 



of a special arrangement of electric and magnetic 



NO. 2626, VOL. 104] 



fields it is found possible to bring positive rays of 

 definite mass to a focus, independent to some extent 

 of their velocit\', so that the dispersion can be made 

 much greater than hitherto without loss of intensity. 

 In this way a mass spectrum is formed on which the 

 values of mass can be compared, in favourable cases, 

 to an accuracy of about one part in a thousand bv 

 comparison with known reference lines such as 

 O (16), C (12), etc. In this way atmospheric neon 

 is definitely proved to consist of two isotopes of 

 mass 20 and 22. Argon gives a line exactly at 40, 

 and if it is a mixed element the other constituents 

 must be present in very small proportion. Chlorine 

 gives a group of four lines exactly at 35, 36, 37, and 

 38, and others from which good evidence can be 

 adduced that this element consists of at least two 

 isotopes, CI" (35) and Cl^ (37), (36) and (38) being 

 the two corresponding hydrochloric acids. Mercury 

 is also found to be a mixture of isotopes, probably 

 three in number, their masses not yet being accurately 

 determined. Very interesting results are yielded by 

 helium and hydrogen ; the former appears to be a 

 " pure " element of mass 400, but hydrogen is very 

 definitely heavier than unity (0=i6. H,, H., and 

 H, all give consistent values in approximate agree- 

 ment with that accepted by chemists : looS for 

 hydrogen. When due allowance has been made for 

 multiple charges, it is found that of more than fiftv 

 atomic and molecular masses so far determined, 

 everv one, with the exception of the three hydrogen 

 lines, falls on a whole number within the error of 

 experiment. — K. Molin : .\n examination of Searle's 

 method for determining the viscosity of very viscous 

 liquids. — H. W. Richmond : Note on the Diophantine 

 equation f + x^ + y' + 0^ -o.—Vxoi. H. F. Baker: 

 Mathematical notes : (i) The stability of rotating 

 liquid ellipsoids ; (2) the general theory of the 

 stability of rotating masses of liquid ; (3) the 

 stabilitv of periodic motions in general dynamics ; 



(4) the invariance of the equations of electro- 

 dvnamics in the Maxwell and in the Einstein forms ; 



(5) a propertv of focal conies and of bicircular 

 quartics ; (6) the Hart circle of a spherical triangle ; 

 (7) a proof of the theorem of a double six of lines by 

 projection from four dimensions; (8) a group of trans- 

 formations of rectangular axes ; (q) transformations 

 with an absolute quadric; and (10) the reduction of 

 homography to movement in three dimensions. 



Edinburgh. 

 Royai Society of Edinburgh, January 12. — Prof. F. O. 

 Bower, president, in the chair. — Prof. W. Peddie : 

 'Ihe atomic space lattice in magnetite. The question 

 of the uniqueness of the determination by the X-ray 

 method was discussed. It appears that the solution 

 is probablv not unique witliin the limits of accuracy 

 in observation. — J. Marshall : .\n unnoticed point in 

 the theorv of Newton's rings. By consideration of the 

 passage of light-waves through three media the author 

 discussed the considerations under which the centre 

 of the rings was a black spot or a bright spot. These 

 depend upon the relative refractive indices of the three 

 materials. It was shown that the reason why 

 Brewster obtained a bright spot at the centre with an 

 appearance of interference rings was because the refrac- 

 tive indices of his materials were approximately in 

 geometrical progression. — .'\. T. Doodson, R. M. Carey, 

 and R. Baldwin : Theoretical determination of the 

 longitudinal seiches of Lake Geneva. The es.sential 

 feature of this oaper was the development of a mathe- 

 matical method (due to J. Proudman) of successive 

 approximations bv which the nodes of seiches in 

 irregular-shaped basins can be calculated. The method 

 was .applied to the seiches of Lake Geneva with satis- 

 factorv results. 



