Nov. 15, 1 888] 



NA TURE 



cording to the writer. Humboldt found it 56 kilometres long, and 

 Herrvon Ilesse-Warteggonly 49. Yet.while the former estimated 

 the area of the lake at 424 square kilometre^, the latter gives it 

 at 587. The author gives many interesting detail?, not only 

 about the lake, but also about the region in which it is situated. 

 To the same number Dr. O. Kriimmel contributes a paper, in 

 which he endeavours to solve the old problem of the Euripus. 



The last supplementary issue (No. 91) of Pdennanns 

 MitteiluHgen contains, according to its title, an account of a 

 journey from Hankow to Soochow, and of journeys in Central 

 and Western China between 1879 and 1881. The contents of 

 this particular paper are misdescribed, for it contains only the 

 record of a journey in 1875 from Shanghai to Hankow on the 

 Yang-tsze, thence by the Han River through the Hupeh, Honan, 

 and Shensi provinces to tanchow in Kansu, and thence to 

 Soochow, close to the Great Wall and the Mongolian deserts, 

 where Herr Michaelis, the writer, remained for Fome time, and 

 carried out certain explorations in the neighbourhood. Possibly 

 another part or other parts are to follow, of which there is at present 

 no indication. Herr Michaelis was employed in 1874 as a mining 

 expert by the late Viceroy and General Tso Tsung Tang, who 

 had just then chased the Mahommedan rebels out of the Shensi 

 and Kansu provinces, and was about to begin his famous march 

 to Kashgar. He was to investigate the region both within and 

 without the Great Wall for mineral deposits, and especially for 

 gold. Herr Michaelis met Count Szchenyi and his party in 

 Soochow, and naturally a good deal of the ground he traversed 

 has already been described by Lieut. Kreitner, who was surveyor 

 to the Szchenyi Expedition, in his well-known book, " Im Fernen 

 Osten." The paper is accompanied by three excellent route maps. 



MOLECULAR PHYSICS: AN ATTEMPT AT A 

 COMPREHENSIVE DYNAMICAL TREAT- 

 MENT OF PHYSICAL AND CHEMICAL 

 FORCES} 



IV. 

 § 1 6. Electrical Act ions. 

 T T follows from the principle of the conrervation of energy 

 that the processes which give rise to electrical excitation 

 can themselves be called into play by electrical action. 



The heating of a conductor by the passage of an electric 

 current is easily explained on the author's theory that electrical 

 conduction is effected by means of molecular vibrations. The 

 electric spark he considers to be due to the separation of par- 

 ticles of the conductor heated in this manner. 



The author explains the Peltier effect in the following manner. 

 Let a closed metallic circuit be formed, con-istingof two metals, 

 soldered together at the points L and H., and suppose the 

 circuit to be traversed by a current flowing through the junction 

 L, from the less easily excited metal A to the more easily ex- 

 cited metal B. The molecules of the metal A will then, by 

 hypothesis, easily be thrown into vibration ; the metal A will 

 therefore be more heated than B, and will, moreover, be a worse 

 conductor of heat than B. The heat excited in A at the junction 

 L will therefore be carried to warm the junction H. in the same 

 direction as the current ; it will then accumulate at this junction, 

 for A, being the worse conductor, will carry away less heat from 

 the junction H. than is carried to it through B. The junction H. 

 will therefore be heated, while the junction L will be cooled. "■' 



The direct production of light by electrical action has already 

 been considered in § 14 (October 11, p. 581). It is clear that 

 secondary luminous phenomena may also come into play. 



Both cliemical combination and decomposition may be effected 

 l)y means of electrical action. The author selects, as an example 

 of the former, the combination of a mixture of oxygen and 

 hydrogen lo form water when traversed by electric sparks, which 

 he considers to be due to the absorption by the molecules of the 

 radiant electrical energy proceeding from the positive pole. The 

 motion of the atoms would be acce'erated, and the number of 

 impacts increased, giving rise to a series of phenomena similar to 

 those described in § 8 » (September 6, p. 460). The internal 



" A Paper read before the Physico-Economic Society of KSnigsberg, by 



Prof. F. Lindemann, on April 5, 1888 Continued from vol. xxxviii. p. 581. 



In the original, some confusing mi-prints occur in this paragraph, viz. 



p 38, second line, i^fjit-r shouLi be sjilcchter, and in the third and fourth 



lines A and B should be interchanged.— G. W. UE 1'. 



3 Since the oxygen and hydrogen molecules are electrically excited to 

 different degrees, they will attract one another. A hydrogen molecule will 

 therefore impinge upon an oxygen molecule more often than upjn another 

 hydrogen molecule, thus increasing the chemical action. 



vibrations of the newly-formed molecule will tend towards a 

 steady state, in which the internal energy is as small as possible. 

 Hydrogen and oxygen will unite to form water, supposing the 

 molecules of the latter to be less electrically sensitive than those 

 of its constituents. We should therefore conclude, from the fact 

 that combination occurs under these circumstances, that water 

 is only very slightly sensitive to electrical excitation, which is in 

 agreement with the observed fact that pure water is an exceedingly- 

 bad conductor of electricity. 



The decomposing action of electricity is exhibited in electro- 

 lytic phenomena. These occur in the inverse order to the 

 chemical actions which serve to produce the current. The 

 action is supposed by the author to take place as follows. The 

 fluid receives electrical energy from the positive electrode, which 

 excites electrical vibrations in the molecules immediately sur- 

 rounding it. These vibrations are transmitted through the fluid 

 according to the ordinary laws of hydrodynamics, but this would 

 not necessarily give rise to an electric current through the liquid,, 

 for an accumulation of electricity may take place even in non- 

 conductors. In consequence of these vibrations, however, the 

 molecular impacts will occur more frequently, and a new steady 

 state will be set up, provided such is possible, in which the 

 internal energy has a smaller value than before. Decomposition 

 will therefore take place if the separate constituents are less, 

 sensitive to electrical vibrations than when in combination, as 

 their electrical energy will then be less than that of the com- 

 pound. One of the constituents will, however, be excited to- 

 a greater extent than the other, and the one which is least ex- 

 cited will be attracted more strongly by the anode than the more 

 highly excited one. The latter constituent will not, therefore,, 

 move towards the kathode with a definite velocity, but will 

 remain where it is, and reenter into combination with the oppo- 

 site constituent of a neighbouring molecule. This would appear 

 at first to be in contradiction with the assumption that the 

 compound is more sensitive to electrical vibrations than its con- 

 stituents, but the apparent contradiction is explained by the 

 consideration that the internal energy lost during the decompo- 

 sition of the first molecule must reappear in the form of external 

 energy— that is to say, in the form of heat ; and the heat thus 

 set free will supply the elec.rical energy necessary to cause the 

 recombination. This give? an explanation of the "migration 

 of the tons." From particle to particle during this migration, 

 alternate transformations of electrical energy into heat, and of 

 heat into electrical energy, take place. A certain amount of 

 electrical energy will be lost during the process — namely, the 

 amount transformed into heat during the decomposition of the 

 first molecule, and the heat developed in the solution will raise 

 its temperature to such an extent as to cause a recombina- 

 tion between the products of the decomposition set free at the 

 electrode — a result which is in agreement with observation. ' 



§ 1 7. Rotation of the Plane of Polarization. 

 One of the principal arguments in favour of Maxwell's electro- 

 magnetic theory of light is, that it gives an explanation of the 

 rotation of the plane of polarization by an electric current on 

 the assumption of the existence of molecular vortices. It is 

 therefire of considerable importance to determine how far the 

 author's theory is capable of explaining the same phenomenon. 

 Suppose a right-handed spiral to be wound round the axis of X> 

 proceeding from the origin in the positive direction. A current 

 flowing through the spiral away from the origin will then pro- 

 duce a north pole at the origin and a south pole at the other 

 extremity of the spiral. Let a ray of plane-polarized light 

 traverse the solenoid in the direction of the axis, then every 

 point on the axis will move in a short rectilinear path perpen- 

 dicular to it. Now an electric current has been defined as 

 consisting in a disturbance of the molecular equilibrium of the 

 conductor, propagated along the conductor with great velocity 

 by radiation from molecule to molecule through the intervening 

 ether. The electrical vibrations may be assumed to take place 

 in the conductor in every direction, as in the case of heat-waves ;. 

 and, as a special case, a disturbance of equilibrium taking place 

 in a single direction only must give rise to an electric current, 

 so that every motion of the ether in a definite direction must be 

 equivalent to an electric current. Motions of any great extent 

 do not come into consideration, for every disturbance in the 

 equilibrium of the ether must consist in vibrations; but, how- 

 ever small the light-vibrations may be, they must be coiisidered, 



' See von Hel.-nKohz's " Wissenschaftliche AbhanJlungen," vol. ii. p; 958, 

 et ieq. 



