662 report — 1884. 



it is also equally impossible to look on the opposite view as tending towards progres r r 

 inasmuch as to ascribe to an element as many valencies as it possesses compounds 

 with some other element, is only expressing by circuitous methods what the old 

 Daltonian law of combination in multiple proportions states in simple terms. Still 

 we may note certain generally-accepted conclusions: in the first place, that of the 

 existence of non-saturated compounds both inorganic and organic, as carbon-mon- 

 oxide on the one hand, and malic and citraconic acids on the other. Secondly, 

 that the valency of an element is not only dependent upon the nature of the element 

 with which it combines, but that this valency is a periodic function of the atomic 

 weight of the other component. Thus the elements of the chlorine group are always 

 monads when combined with positive elements or radicals, but triad, pentad, and 

 heptad with negative ones. Again, the elements of the sulphur group are dyads in 

 the first case, but tetrad and hexad in the second. The periodicity of this property 

 of the atoms, increasing and again diminishing, is clearly seen in such a series as 



AgC^, CdO\„ InCL, SnCl 4 , SbH 3 , TeH,, III, 



as well as in the series of oxides. The difficulties which beset this subject may ho 

 judged of by the mention of a case or two : Is vanadium a tetrad because its highest 

 chloride contains four atoms of chlorine ? What are we to say is the valency of 

 lead when one atom unites with four methyls to form a volatile product, and yet 

 the vapour-density of the chloride shows that the molecule contains one of metal to 

 two of chlorine ? Or, how can our method be said to determine the valency of 

 tungsten when the hexchloride decomposes in the state of vapour, and the penta- 

 chloride is the highest volatile stable compound? How again are we to define 

 the point at which a body is volatile without decomposition Y — thus sulphur tetra- 

 chloride, one of the most unstable of compounds, can be vaporised without decompo- 

 sition at all temperatures below —22°, whilst water, one of the most stable of known 

 compounds, is dissociated into its elements at the temperature of melting platinum. 



But, however many doubts may have been raised in special instances against a 

 thorough application of the law of valency, it cannot be denied that the general 

 relations of the elements which this question of valency has been the means of 

 bringing to light are of the highest importance, and point to the existence of laws 

 of nature of the widest significance ; I allude to the periodic law of the elements 

 first foreshadowed by Newlands, but fully developed by Mendelejeff and Lothar 

 Meyer. Guided by the principle that the chemical properties of the elements are 

 a periodic function of their atomic weights, or that matter becomes endowed with 

 analogous properties when the atomic weight of an element is increased by the 

 same or nearly the same number, we find ourselves for the first time in possession 

 of a key which enables us to arrange the hitherto disjecta membra of our chemical 

 household in something like order, and thus gives us means of indicating the 

 family resemblances by which these elements are characterised. 



And here we may congratulate ourselves on the fact that, by the recent experi- 

 ments of Brauner, and of Nilson and Pettersen respectively, tellurium and beryl- 

 lium, two of the hitherto outstanding members, have been induced to join the 

 ranks, so that at the present time osmium is the only important defaulter amongst 

 the sixty-four elements, and few persons will doubt that a little careful attention to 

 this case will remove the stigma which yet attaches to its name. But this periodic 

 law makes it possible for us to do more ; for as the astronomer, by the perturbations 

 of known planets, can predict the existence of hitherto unknown ones, so the 

 chemist, though, of course, with much less satisfactory means, has been able to predict 

 with precision the properties, physical and chemical, of certain missing links 

 amongst the elements, such as ekaluminium and ekaboron, then unborn, but which 

 shortly afterwards became well known to us in the flesh as gallium and scandium. 

 We must, however, take care that success in a few cases does not blind us to the 

 fact that the law of nature which expresses the relation between the properties of 

 the elements and the value of the atomic weights is as yet unknown ; that many 

 of the groupings are not due to any well-ascertained analogy of properties of the 

 elements, and that it is only because the values of their atomic weights exhibit 

 certain regularities that such a grouping is rendered possible. So, to quote Lothar 



