548 



NA TURE 



{April -], 1887 



first great exponent of Dniton's 'generalisation, was torn by con- 

 flicting beliefs until he found peace in the hypothesis of Proiit 

 and Meinecke that the atojiic weights of all the so-called 

 elements are multiples of a coaimon unit, and which he sought 

 to establish by some of the very wor4 quantitative determina- 

 tions to be found in chemical literature. It is curious to note 

 the bondage in which the old metaphysical quibble concerning 

 the divisibility or indivisibility of the atom held the immediate 

 followers of Dalton. Graham, however, never felt such tram- 

 mels. To him the atom meant something which is not divided : 

 not something which cannot be divided. With Graham, as 

 with Lucretius, the original atom may be far down. 



Every philosophic thinker to-day has, I should imagine, come 

 to be of this opinion. Not many years ago it was the fashion to 

 maintain that Stas's great work had for ever demolished the 

 doctrine of the primordial///, and that Roger Bacon's aphorism 

 that " barley is a horse by possibility, and wheat is a possible 

 man, and man is possible wheat," was henceforth .an idle saying. 

 Stas's work is a monument of experimental skill, and it has 

 furnished us with a set of numerical ratios which are among the 

 best determined of any physical constants. It may be that it 

 demolished Prout's hypothesis in its original form, but it has not 

 touched the wider question. Whether indeed the wider question 

 is capable of being reached by direct experiments of the nature 

 of those of Stas is very doubtfid, unless the weight of the common 

 atom is some very considerable fraction, say one-half or one- 

 fourth, ofth.at of the hydrogen atom. Dumas has, as you know, 

 modified Prout's hypothesis in this sense, by assuming as the 

 common divisor half the atomic weight of hydrogen, but there is 

 no a priori reason why we should stop at this particular sub- 

 division. The exact relation of Stas's work to Prout's law has, 

 I think, been fairly stated by Prof. Mallet at the conclusion of 

 his admirable paper on the atomic weight of aluminium, in the 

 Philosophical Transactions for 1880 (vol. clxxi. 1033). Stas's 

 main result, says Mallet, " is no doubt properly accepted if 

 stated thus, that the differences between the individual deter- 

 minations of each of sundry atomic weights which have been 

 most carefully examined are distinctly less than their difference, 

 or the difference of their mean from the integer which Prout's 

 law would require. But the inference which St.as himself seems 

 disposed to draw, and which is very commonly taken as the 

 proper conclusion from his results, namely, that Prout's law is 

 disproved, or is not supported by the facts, appears much more 

 open to dispute. It must be remembered that the most careful 

 work which has been done by Stas and others only proves by 

 the close agreement of the results that fortuitous errors have been 

 reduced within narrow limits. It does not prove that all sources 

 of constant error have been avoided, and, indeed, this never can 

 be absolutely proved, as we never can be sure that our knowledge 

 of the substances we are dealing with is complete. Of course, 

 one distinct exception to the assumed law would disprove it, if 

 that exception were itself fully proved, but this is not the case. 

 As -suggested by Marignac and Dumas, anyone who will im- 

 partially look at the facts can hardly escape the feeling that there 

 must be some reason for the frequent recurrence of atomic 

 weights differing by so little from accordance with the numbers 

 required by the supposed law. " Prof. Mallet, in tabulating the 

 atomic weights which may be fairly considered as determined 

 with the greatest attainable precision, or a very near approach 

 thereto, and without dispute as to the methods employed, points 

 out that out of ihe eighteen numbers so given ten approximate to 

 integers, within a range of variation less than one-tenth of a 

 unit. And he then proceeds to calculate the degree of pro- 

 bability that this is purely accidental, as those hold who carry to 

 the extreme the conclusions of Berzelius and Stas, and he finds 

 that the probability in question is only equal to I : 1097 '8. And 

 he concludes that not only is Prout's law not as yet absolutely 

 overturned, but that a heavy and apparently increasing weight 

 of probability in its favour, or in favour of some modification of 

 it, exists, and demands consideration. 



It would be impossible for me to attempt to traverse the whole 

 ground of this question which has been opened up during the past 

 fifteen or twenty years. Even if I could cl.aim the time and your 

 indulgence, there is hardly the necessity for such a demand on your 

 patience. Mr. Crookes, only so recently as September last, gave 

 an admirably complete exposition of the present state of the case 

 in his address to the Chemical Section of the British Association 

 at the Birmingham meeting, and for me to go over the ground 

 again with you would be simply to plough with Mr. Crookes' 

 heifer. Some years ago Mr. Norman Lockyer, as you doubtless 



know, approached the subject from another point of view, and 

 in his recent work, " The Chemistry of the Sun," you will find 

 a summary of the evidence which the spectroscope has afforded us 

 concerning the dissociation of "elementary" matter at such 

 transcendental temperatures as we have in stars like the sun. 



Now, when we pa«s in review all this evidence ; when we 

 reflect upon the mode of distribution of the elements, and 

 especially their tendency to associate in correlated groups ; when 

 we bear in mind the absolute analogy which exists in the general 

 behaviour and mode of action of the radicles which are con- 

 fessedly compound with those which are assumed to be simple ; 

 when we have regard to the phenomena of allotropy, isomerism, 

 and homology, — the mind insensibly appeals to the principle of 

 continuity, and refuses to believe that the seventy and odd " ele- 

 mental " forms, to which our processes of analysis have reduced 

 all the kinds of matter we see around us, differ in essence from 

 bodies which are known to be compound. 



The connexion between the properties of the " elements " and 

 the relative weights of their atoms, as developed by Newlands, 

 Mendelejeff, Lothar Meyer, Carnelley, and others, has served to 

 strengthen this conviction. The discovery that the physical and 

 chemical properties of the elements are as periodic functions of 

 their atomic weights, is unquestionably the most important 

 generalisation we have had in chemical philosophy during 

 the last five-and-twenty years. Its bearings upon the 

 question of the origin of the "elements" have been 

 worked out in the Presidential address I have already 

 referred to. Mr. Crookes, like Mr. Lockyer before him, 

 in seeking to apply to this question of the genesis of the ele- 

 ments the same principles of evolution which Laplace has 

 already applied to the creation of the heavenly bodies, and 

 which Lamarck, Darwin, and Wallace have applied to that of 

 the organic world, is again appealing to the law of continuity. 

 The mind which holds that Nature is one harmonious whole is 

 fain to believe that the probability that the elements have 

 originated by chance and are eternally self-existent is just as 

 remote as that the animals and plants of to-day are primordially 

 created things. I think in what I am now saying I may fairly 

 claim to be reflecting the opinion on this matter of every philo- 

 sophic thinker of to-day. Nay more, you must .allow that the 

 germ which has been kept alive for so many centuries, and 

 which has come down to us through the brains of a succession 

 of thinkers like Leucippus, Aristotle, Lucretius, Bacon, Newton, 

 Dalton, and Graham, has become quickened and endowed by 

 the light which modern science has shed upon it from all sides, 

 with a vitality which will persist and strengthen. 



Having thus traced the development of the idea held by 

 Graham of the essential oneness of matter, let us spend the few 

 remaining moments in considering, in the most general way, 

 how the science of the last twenty-five years has worked out and 

 extended his conceptions concerning the properties of the atom 

 and its mode of motion. 



The treatment which "the few grand and simple features of 

 the gas," to quote Graham's phrase, have received at the hands 

 of Clausius, Clerk Maxwell, Helmholtz, Sir William Thomson, 

 and a score of workers in this country and on the Continent 

 who have been actuated by their influence, has served to dispel 

 much of the metaphysical fog which has enshrouded the notion 

 of the atom, and to-day we are able to reason about atoms, as 

 physical entities, having extension and figure, and of their 

 number and dimensions and peculiarities of movement, with the 

 confidence which is based on well-ascertained facts. 



We have, of course, not yet attained to a complete molecular 

 theory of gases. But we know the relative masses of the mole- 

 cules of various gases, and we have calculated in miles per 

 second their average velocity. The phenomena of diffusion 

 indicate that the molecules of one and the same gas are all equal 

 in mass. For, as was pointed out by Clerk Maxwell, if they were 

 not, Graham's method of using a porous septum would enable us 

 to separate the molecules of smaller mass from those of greater, 

 as they would stream through porous substances with greater 

 velocity. We should thus be able to separate a gas, say 

 hydrogen, into two portions, having different densities and other 

 physical properties, different combining weights, and probably 

 different chemical properties of other kinds. As no chemist has 

 yet obtained specimens of hydrogen diftering in this way from 

 other specimens, we conclude that all the molecules of hydrogen 

 are of sensibly the same mass, and not merely that their mean 

 mass is a statistical constant of great stability (see art. "Atom," 

 " Encyclopedia Britannica," 9th edition). This line of argument 



