176 



NATURE 



[June 



1904 



parents. ... It was the original intention," of the 

 author, " to summarise all the principal child-study in- 

 vestigations that have been made." But this plan was 

 evidently abandoned at a very early stage, and we have 

 Instead the present popularly written volume, which we 

 can heartily recommend both to teachers and parents. 

 Its style is pleasing, and its matter fairly correct, em- 

 bodying the experience of fourteen years' study and 

 teaching in the subject. Were the contents as widely 

 read as thev deserve to be, the immense importance of 

 child-study, as a basis for methodical teaching and 

 rational education, would bo more generally realised. 



The greater part of the book is devoted to the develop- 

 ment of instincts — a word used in an extended sense by 

 the author to embrace the phenomena of imitation, 

 curiosity, migration, and even aesthetics, morality, and 

 expression. These nine chapters, together with those 

 on heredity, individuality, and on the development of 

 the intellect, are all admirably written, containing ex- 

 cellent food for the parent's reflection and stimulating 

 the interest of the teacher in her work. It seems 

 strange that the subject of fatigue should be relegated 

 to the chapter entitled "Abnormalities." This latter 

 contains some useful hints on the mental and physical 

 defects of children, but the accompanying pathological 

 and anatomical remarks are in several instances in- 

 accurate and misleading. C. S. M. 



LETTERS TO THE EDITOR. 



[The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 tn return, or to correspond with the writers of, rejected 

 iiwniiscripls intended for this or niiy other part of Na ruKE 

 No notice is taken of anonymous communications.] 



Residual Affinity. 



There appears to be a tendency among chemists to 

 abandon their own doctrine of definite valency, and to 

 recognise an indeterminate and fluctuating number of links 

 connecting atoms with each other. 



The electron theory of the physicist, which assigns one 

 indivisible unit of charge to a monad, two to a dyad, &c., 

 has therefore encountered some opposition, inasmuch as it 

 seems to tend to harden the old doctrine of " bonds " 

 whereby atoms were supposed to be linked only in a simple 

 definite and numerical way, no fraction of a bond being 

 contemplated. 



Assuming this rough statement to represent something 

 like historical truth, I have a few remarks to make on the 

 subject. 



First, the possession by an atom of a definite charge, 

 numerically specifiable as a simple multiple of an indivisible 

 unit, must be accepted as a physical fact. 



Second, this fact corresponds with those other facts which 

 originally led chemists to assert, for instance, that nitrogen 

 was a triad or pentad, carbon a tetrad, &c. — a position which 

 it would seem absurd to abandon. (Incidentally it may be 

 noted that a monad must be either electro-positive or electro- 

 negative, but that a tetrad need not be either, since its 

 pairs of charges may be opposite in sign.) 



Third, there is nothing in these doctrines inconsistent 

 with the e.^istence of fractions of a bond and any required 

 amount of " residual affinity." 



It is this last thesis that I wish briefly to develop. 

 Indeed, in 1902, in a paper on electrons published in the 

 Journal of the Institution of Electrical Engineers, vol. 

 xx.xii., p. 103, I showed how it was possible to regard 

 ordinary mechanical cohesion on the electric theory ; and 

 likewise that it was easy to regard molecular combination 

 from the same point of view. 



In a short conversation with Prof. Armstrong, at the 

 Mansion House recently, I realised more clearly than before 

 where the imaginary difficulty now lies. 



It has been an occasional habit with physicists when 

 speaking of lines of force to think of a single line of attrac- 

 tion or elastic thread joining each negative electron to its 



NO. 1808, VOL. 70] 



corresponding positive charge; each unit charge, in fact, 

 being regarded as the cut end of a line of force and nothing 

 else. But so far as 1 know it has never been considered 

 that these lines of force so interpreted were physical reali- 

 ties, and that one and only one line really appertained to 

 each unit charge ; though in his recent remarkable book 

 reviewed in these columns on May 26, p. 73, Prof. J. J. 

 Thomson goes near to assigning so great a physical reality 

 to the lines of force as would make the number issuing 

 from any charge a commensurable number ; that is to say 

 he begins hypothetically to regard each line of force as a 

 discrete physical entity. But even so there is no evidence 

 that eacli unit of charge ought to have assigned to it one 

 solitary line of force, it might have a great number; though 

 it is true that on that view it becomes a definite question 

 how many lines of force a unit charge possesses, whereas 

 on the ordinary vaguer view of a centre of force the influence 

 of which is felt in all directions, any specification of number 

 of lines is either meaningless or a mere question of con- 

 venience of measurement, like the number of miles in the 

 circumference of the earth, or the number of cubic feet in 

 a room : a number which is necessarily and always in- 

 commensurable. 



On any view electrons are supposed to repel and to be 

 attracted with a force varying as the inverse square of the 

 distance, and this is only consistent with a very large 

 number of lines of force radiating from each and starting 

 out in every direction equally. 



When opposite charges have paired off in solitude, every 

 one of these lines start from one and terminate on the 

 other constituent of the pair, and the bundle or field of 

 lines constitutes a full chemical " bond " ; but bring other 

 charges or other pairs into the neighbourhood, and a few 

 threads or feelers are at once available for partial adhesion 

 in cross directions also, the quantitative distribution of the 

 force being easily calculable from geometrical data. 



Briefly, the charge is indivisible, it is an atomic unit 

 (up to our present knowledge) ; but the lines of force eman- 

 ating from it are not indivisible or unified at all. The bulk 

 of them may be occupied with straightforward chemical 

 affinity while a few strands are operating elsewhere ; and 

 the subdivision of force may go on to any extent, giving 

 rise to molecular combination and linking molecules into 

 complex aggregates, so that a quite gradual change of 

 valency is conceivably possible, the number of wandering 

 lines being sometimes equal to, or even greater than, the 

 number of faithful lines — though this would usually represent 

 an unstable condition not likely to persist. 



I state the position in order that physicists who see reason 

 to disagree with it may intervene in good time and prevent 

 any premature acceptance of a harmonising interpretation 

 by chemists ; because so long as there is any real out- 

 standing diflficulty it is clearly best for the progress of 

 science that diverse views should continue. 



Oliver Lodge. 



On a Dynamical System illustrating Spectrum Lines. 



I DESIRE to express to Prof. Nagaoka my regret at my 

 misinterpretation of his letter to Nature of February 25, 

 which was due simply to my failure to find any mention 

 there of the larger system of which be speaks. No doubt 

 his ring is quasi-stable if the central positive charge is 

 large enough ; but is it allowable to leave out of account 

 the rest of the system? Waiving this objection, I would 

 point out that there are upper limits to the central charge 

 which cannot be exceeded without making the whole 

 system positive, or the velocity of the ring greater than 

 that of light. It may very well be that either limit is too 

 low to allow a stable system to be reached ; the discussion 

 of this point must be reserved for another time. 



G. A. ScHOTT. 



Physical Laboratory, L'niversity College, Aberystwyth. 



A Correction. 



In my letter to Nature of June 16 (p. 151) concerning the 

 source of radio-active energy, I should of course have halved 

 the expressions given for the electrostatic energy of an 

 isolated electron, and for energy set free by annihilation of 

 matter. C. V. Burton. 



Cambridge, June 18. 



