74 



NA TURE 



[May 26, 1904 



The next chapter (v.), on the constitution of the 

 atom, begins still on familiar ground, and exhibits 

 the evidence for the great intrinsic energy of electrically 

 constituted atoms, and estimates the rate of radiation 

 of energy from corpuscles variously distributed inside 

 an atom, showing that the radiation from a single 

 corpuscle is far greater than that to be expected from 

 two or more, especially when their speeds are not 

 excessive. [It may be observed that a pair of revolving 

 electrons, at opposite ends of a diameter, will be 

 equivalent to two equal opposite currents, and hence 

 will tend to neutralise each other's influence at a dis- 

 tance, especially along the axis of revolution ; unless 

 indeed they are moving with the speed of light, in 

 which case they could start the crest and trough of an 

 advancing ellipticall3--polarised wave.] Hence a cer- 

 tain number of corpuscles are essential to the stability 

 of an atom, a few would soon radiate their energy 

 away ; and this fact, on the doctrine of the evolution 

 of matter, suggests a reason for the non-existence 

 of permanent elements of lower atomic weight than 

 hydrogen. 



But a great deal more than that is made out towards 

 the end of the chapter, vi'here the electric constitution 

 of the atom is applied to chemistry, and a beginning of 

 the explanation of " the periodic law " is made in this 

 chapter which is further developed by the author in 

 a great paper in the March number of the Philosophical 

 Magazine for the present year. -Also the homologous 

 series of lines in the spectrum, investigated by Ryd- 

 berg, Runge and Paschen, and Kayser, is shown to 

 be a fairly natural, or at least plausible, consequence 

 of the groupings of various numbers of corpuscles or 

 electrons inside an atom ; though it must be admitted 

 that unless attention is paid to the modern view that 

 concussions and not regular motions are the real 

 cause of perceptible visible radiation and line spectra, 

 the theory remains obviously very incomplete. The 

 mode in which the corpuscles would statically dis- 

 tribute themselves, if they were limited to a plane, 

 under the combined action of a mutual inverse 

 square repelling force and a direct-distance central 

 attracting force, is treated much as Lord Kelvin had 

 alreadv treated it in his remarkable paper called 

 " ^4ipinus atomized " (reprinted as an appendix in his 

 Baltimore Lectures), and is illustrated by Alfred 

 Mayer's experiments on floating magnets; and the 

 deductions, although plainly only the nucleus of an 

 investigation, are already very suggestive and pro- 

 mising. In the March Philosophical Magazine the 

 investigation into the stability of moving electrons is 

 carried much further, and a distinct step is made in 

 mathematical chemistry. 



One very simple and important remark is made near 

 the end of chapter v. concerning the " bonds " of the 

 chemist, which, being I suppose universally recog- 

 nised as Faraday lines of electric force, must differ 

 from ordinary stretched elastics in having opposite 

 properties at the two ends. The bonds, in fact, must 

 have " sense " as well as direction, the\- are not simple 

 links; and hence when carbon atoms are linked 

 together those atoms cannot really have identical 

 properties, unless indeed they are linked to each other 

 NO. 1804, VOL. 70] 



by an even number of bonds. Incidentally it becomes- 

 clear — at any rate in the March Phil Mag. it becomes 

 clear to me — why carbon or tetrad atoms can link 

 themselves into complex molecules : the foundation of 

 organic chemistry. 



.•\ suggestion is also made concerning certain 

 " additive " properties, such as the refractive power 

 of different substances for light, which students of 

 physical chemistry will do well to take up and press- 

 further. 



The obvious question as to how an electrlcallv con- 

 stituted atom can acquire an additional charge, so as 

 to become an ion, either positive or negative, and 

 either monad, dyad, or triad, is discussed, and a guess- 

 is made as to the nature of molecular combination. 

 It is also attempted to explain why a liquid, say liquid 

 mercury, is so immensely better a conductor than 

 mercury vapour, and why some gases may be con- 

 ductors and others not. The violent motion of 

 corpuscles going on inside an atom is styled by J. J. 

 Thomson the " corpuscular temperature " of the atom, 

 which may or may not be a convenient term; the 

 ordinary temperature of the gas, called " molecular 

 temperature," is generally very much smaller, and has 

 no apparent relation with the corpuscular temperature. 

 By the interaction of these two temperatures on one 

 another, an attempt is made to account for certain 

 chemical facts of combination. The whole of the 

 chapter is so concentrated and full of suggestion that 

 it is impossible effectively to abstract it further. What 

 I wish to indicate to students is the desirability of 

 studying the original. 



Prof. Pointing has on this point made a remark to 

 me which he permits me to incorporate, viz., that at 

 high molecular temperatures there must be some dis- 

 tinct correspondence between molecular and corpus- 

 cular temperatures. For in the sun corpuscles are set 

 free bv collisions [as they may also be set free by the 

 clash of chemical combination at more ordinary tem- 

 peratures, a bright line spectrum resulting in both 

 cases from the perturbation of those corpuscles which, 

 although shocked, escape separation]. There would 

 appear to be some high temperature at which the 

 atoms go to pieces — a limiting molecular temperature 

 bevond which they cannot exist (an atomic dissoci- 

 ation temperature), not much higher probably than 

 the solar temperature. It is worthy of remark that 

 no star is much hotter than the sun : possibly none so 

 high as ten thousand degrees centigrade. If not, why 

 not? unless it be because there is a natural limit at 

 which matter goes to pieces. 



The final chapter, on radio-activity and radio-active 

 substances, emphasises the way in which atomic 

 collapse, or re-distribution of corpuscles — a sort of 

 atomic earthquake — may occasionally occur, after the 

 radiation of a certain amount of energy has gone on 

 for some time, by spontaneous re-arrangement of the 

 constituents into a more stable form. F'or since orbital 

 motion plainly tends to increase stability, enabling a 

 greater number of corpuscles to resist central attrac- 

 tion than could hold out if they were stationary, it 

 follows that, as the corpuscles slow down, they may 

 ar certain critical stages find it necessary to fall into 



