?92 



NATURE 



\yan. 30, 1879 



bodies. In other words, if the molecular energy of a so- 

 called element could be changed, the element would be 

 dissociated, a view of special interest in relation to the 

 researches of Lockyer, The lecture was illustrated by 

 many effective experiments, and concluded with the 

 statement that it had not been instituted from the merely 

 special interest of Graham's researches to the physicist 

 and chemist, but in honour of the labours of a life the 

 memory of which will be as enduring as its work, and to 

 stimulate others to investigate as patiently and earnestly 

 the varied phenomena whose basis is " molecular 

 mobility." 



Sir William Thomson, in proposing a vote of thanks 

 to the lecturer, called attention to a diagram on the wall 

 recording the rates of passage of gases by diffusion, 

 effusion, transpiration, and by the peculiar passage 

 through such "colloid septa" as non-crystalline metals 

 or india-rubber ; and he stated that before Graham's 

 time these valuable physical constants were absolutely 

 unknown. They had listened with much interest to the 

 connection which had been traced between Graham's law 

 of diffusion and the science of molecular physics, as well 

 as to the account of Graham's work generally, so care- 

 fully set before them by Graham's pupil and friend. 



PRELIMINARY NOTE ON THE SUBSTANCES 

 WHICH PRODUCE THE CHROMOSPHERIC 

 LINES'^ 



T_T ITHERTO, when observations have been made of 

 ■*■ -^ the lines visible in the sun's chromosphere, by means 

 of the method introduced by Janssen and myself in 1868, 

 the idea has been that we witness in solar storms the 

 ejection of vapours of metallic elements with which we 

 are famiUar from the photosphere. 



A preliminary discussion of the vast store of observa- 

 tions recorded by the Italian astronomers (chief among 

 them Prof. Tacchini), Prof. Young, and ijiyself, has 

 shown me that this view is in all probability unsound. 

 The lines observed are in almost all cases what I have 

 elsewhere termed and described as basic lines; of these I 

 only need for the present refer to the following : — 



^3 ascribed by Angstrom and Kirchlioff to iron and nickel. 

 ^4 ,, ^ Angstrom to magnesium and iron. 



5268 by Angstrom to cobalt and iron. 



5269 ,, ,, calcium and iron. 

 5235 ,, ,, cobalt and iron. 

 5017 ,, ,, nickel, 



4215 ,, ,, calcium, but to strontium by myself. 

 5416 an unnamed line. 



Hence, following out the reasoning employed in my 

 previous paper, the bright lines in the solar chromosphere 

 are chiefly lines due to the not yet isolated bases of the 

 so-called elements, and the solar phenomena in their 

 totality are in all probability due to dissociation at the 

 photospheric level, and association at higher levels. 

 In this way the vertical currents in the solar atmo- 

 sphere, both ascending and descending, intense absorp- 

 tion in sun-spots, their association with the facula;, and 

 the apparently continuous spectrum of the corona and 

 its structure, find an easy solution. 



We are yet as far as ever from a demonstration of 

 the cause of the variation in the temperature of the sun ; 

 but the excess of so-called calcium with minimum sun- 

 spots, and excess of so-called hydrogen with maximum 

 sun-spots follow naturally from the hypothesis, and 

 afford indications that the temperature of the hottest 

 region in the sun closely approximates to that of the 

 reversing layer in stars of the type of Sirius and a Lyrae. 



If it be conceded that the existence of these lines in 

 , the chromosphere indicates the existence of basic mole- 

 cules in the sun, it follows that as these lines are also 



' Paper read at the Royal Society on January 23, by J. Noiman Lockyer, 



seen generally in the spectra of two different metals in 

 the electric arc, we must be dealing with the bases in the 

 arc also. 



ON A THEORY OF THE VISCOSITY OF THE 

 EARTH'S MASS^ 



IN these two papers the investigation is continued of 

 the physical results which follow from the theory that 

 the mass of the earth is either viscous or imperfectly elastic. 

 In the first paper of the series (which was read before the 

 Royal Society on May 23, 1878, and of which an account 

 appeared in Nature, vol. xviii. p. 265) the theory of the 

 bodily tides of such a spheroid was considered. In that 

 paper it was shown that the bodily tides would lag, and 

 that this lagging would produce an acceleration of the 

 time of high water of the oceanic tides relatively to the 

 nucleus. The author' s attention was directed to the tidal 

 reports of the British Association by Sir W. Thomson, 

 and he has tried to find whether the tidal observations 

 give any indications of a yielding of the earth's mass. 

 The theory of the semi-diurnal and diurnal oceanic tides 

 is so imperfect that it is impossible to say whether or not 

 high water takes place earlier than it would do on a rigid 

 nucleus ; the long-period tides are those from which alone 

 any indications are to be expected. 



The fortnightly tide is the most marked of these, but 

 its height is very small, and the results in the tidal 

 observations show so much irregularity that it cannot be 

 asserted with certainty that they represent the true fort- 

 nightly tide. Nevertheless, it is interesting to learn 

 that, out of eleven years of observation at Ramsgate, 

 Liverpool, Hartlepool, Brest, and Kurrachee, the fort- 

 nightly tide appears to be accelerated in eight cases and 

 only retarded in three. Although the accelerations are 

 exceedingly irregular, it may perhaps be maintained that 

 these observations give some indications of a tidal yield- 

 ing of the earth's mass. 



The first of the two papers of which we are here speak- 

 ing deals with the effects of the tidal distortion of the 

 spheroid on its rotation, and with the reaction on the 

 tide-raising satellite. An account of some of the results 

 of the investigation was read before the British Associa- 

 tion at Dublin, and an abstract appeared in Nature, 

 vol. xviii. p. 580, and therefore the principal results will 

 be here merely repeated. 



For convenience of diction the spheroid is spoken of 

 as the earth and the tide-raising body as the moon. 



It was found, then, that the obliquity of the ecliptic, 

 the length of day and of the month, become variable, 

 and that, if we look into the remote past, we find the 

 obliquity less, and the day and month very much shorter 

 than at present. When the changes were traced back- 

 wards as far as possible it was found that the whole 

 diminution in the obliquity was about 10°, and that the 

 beginning from which the earth and moon must have 

 started was a state in which they rotated, as though fixed 

 rigidly together, in sh. 40m., the moon being then only 

 10,000 miles distant from the earth's centre. 



In the second paper (read before the Royal Society on 

 December 19) some other problems were considered. 

 The first of these is concerning the secular distortion of 

 the spheroid. Under the attraction of the moon the 

 earth becomes distorted into an ellipsoidal shape, with 

 ' the longest axis in the plane of the equator, but, since 

 the tide lags, this longest axis does not point directly to- 

 wards the moon. The excess of the attraction of the moon 

 on the nearer protuberance above that on the further one 

 gives rise to the tidal frictional couple. This couple 

 tends to retard the earth's rotation; but it is clear that 

 ' unless the tidal protuberance has some special form 



I I An account of two papers, "On the Precession of a Viscous Spheroid, 

 I and on the Remote History of the Earth," and " Problems Connected with 

 the Tides of a Viscous Spheroid," by G. H. Darwin, read before the Royal 

 ( Society en December 19, 1S78. 



