May 2, 1892.] 



KNOWLEDGE 



85 



have gone far enough, however, to show the peculiar and 

 novel character of the relation which Mendelejeff discovered. 

 "Whereas, the properties of the Elements of any one family 

 vary cdittiniioiixli/ with the atomic weight, the properties of 

 all Elements are a pi'i-iodu- fimction of the atomic weight ; 

 a certain increase of atomic weight being accompanied by 

 a recurrence of certain properties possessed by an element 

 lower in the scale. This may be made clearer to the eye 

 by writing the above list of 16 elements in a somewhat 

 different way. As there is a return to the metallic 

 character after the sixth element from Lithium, we will 

 begin in our table a second line with Potassium on the 

 left, thus : — 



Li. 7, Be. 9, B. 11, C. 12, N.14, 0. 16, F. 19, 

 Na. 23, Mg. 24, Al. 27, Si. 28, P. 31, S. 32, CI. 35^, 

 K. 39, Ca. 40, etc. 



We see that after a /leriod including seven elements, 

 there begins a second period of seven. The diti'erence of 

 atomic weight between the first and eighth Element, 

 between the second and ninth Element, and so on, is in 

 every case, as will be seen on examining the table, sixteen 

 imits or very nearly so. Two successive members of any 

 family are separated by six intervening elements, and 

 differ fi-om one another by 16 units of atomic weight. 

 The addition of mass or weight to the chemical atom, 

 from the mass 7 of the Lithium atom to the mass 39 of 

 the Potassium atom, is not accompanied by a continual, 

 imbroken increase of certain properties, but by a periodii- 

 variation of those properties. The interval from Lithium 

 to Potassium comprises two periods, each of which contains 

 seven elements. The vertical rows contain the natural 

 families (as Lithium, Sodium, Potassium, and Beryllium, 

 Magnesium, Calcium). Previous to Mendelejeff's work 

 relationships could only be clearly traced between members 

 of the same family (or homologous elements). The 

 Periodic system of classification enables us to trace the 

 connection between the lutcrohiiinu-s elements. Among the 

 following — 



Be. 9, 



Na. 23, Mg. 24, Al. 27, 



Ca.. 40, 



Sodium (Na.), Magnesium (Mg), and Aluminium (AL), 



are termed heterologous elements, and Beryllium (Be.), 



Magnesium i Mg.), and Calcium ( Ca. I homologous elements. 



The elements Sodium, Aluminium, Beryllium, Calcium 

 are termed the four analoijues of Magnesium. Mendelejeff 

 showed that the properties of any element are completely 

 determined by that of its four analogues. Thus, supposing 

 the properties of the element Magnesium were wholly un- 

 known, those properties could be deduced from the pro- 

 perties of the analogues. Thus, the atomic weight will be 

 the mean of those of the four analogues. Now, 



9 + 40 -f 23 + 27 



= 24i 



which gives (approximately) the atomic weight of Mag- 

 nesium. Again, take the specific gra\-ities. They are as 

 follows : — 



Sodium ... specific gravity -97 



Aluminium ... ,, 2-56 



Beryllium „ 2-10 



Calci»m . ... ... ,, 1'58 



Mean of these values = 1'8. 



Specific gravity of Magnesium = 1-75. 



It win be noticed that although the difference between 



the weights of neighbouring elements in the horizontal 



rows (heterologous elements) is not absolutely constant, 



yet the variations are small, the interval rarely exceeding 



one or two units in the foregoing table, except in the case 

 of the interval Fluorine — Sodium and Chlorine — Potassium. 



At the time when Mendelejeff" first di'ew up his table of 

 the elements, it was found that in several cases the 

 neighboming heterologous elements did not fall into place, 

 that is to say, did not come into the same vertical row with 

 other members of the same natural family. Thus the 

 element next to Zinc (Zn. 6.5 1, which belongs to the same 

 family as Magnesium and comes vertically below it, was 

 followed by Arsenic (As. 75), which thus comes vertically 

 below Aluminium, although its properties are similar to 

 those of Phosphorus not to those of Aluminium. 



Thus we have in the second and fourth horizontal rows : 



Mg. 

 2i 



and 



Na. 

 23 



Cu. Zn. 

 63 65 



Al. 



27 



As. 



Li. 

 28 



Se. 

 79 



P. 

 31 



Br. 

 80 



S. 

 32 



CI. 

 35J 



Selenium (Se.) and Bromine (Br.) have properties 

 similar to those of Sulphur and of Chlorine respectively. 

 These facts led to an idea entirely novel in chemistry, that 

 of ijaps among the elements. Hitherto, the existence of an 

 element with any particular atomic weight and particular 

 properties had been regarded as an isolated and, so to 

 speak, an accidental fact in Nature, but Mendelejeff's 

 generalization introduces the idea of the necessity for the 

 existence of elements with such and such atomic weights, 

 and such and such properties. 



It appeared extremely probable that there existed two 

 elements intermediate in atomic weight between Zinc and 

 Arsenic, between which there is an interval of ten uuits. 

 Supposing two such elements to exist (called provisionally 

 Eka- Aluminium andEka-SUicon), Mendelejeff arranged the 

 elements in the fourth horizontal row thus : 

 2ud Row— :!fa. 23 ilg. 2t Al. 27 Si. 28 P. 81 S. 32 CI. Zb\ 

 4thRow— Cu. 63 Zq. 65 Efca-Al. Eka-Si. As. 75 Se. 79 Br. 80 

 Reasoning from the assumption that the properties of an 

 element are the mean of those of its four analogues, 

 Mendelejeff' drew up a table representing the properties of 

 the hypothetical elements Eka-Aluminium and Eka-Silicon. 

 Two elements having the atomic weights required by the 

 position of Eka-Aluminium and Eka-Silicon in the table 

 have since been discovered, and named respectively 

 GalUum and Germanium. Their properties agree very 

 closely with those predicted by Mendelejeff'. 



This power of prediction of hitherto unobserved elements 

 was an enormous advance in chemical science. The dis- 

 covery of Gallium holds in the history of chemistry a 

 similar place to the discovery of Neptune in astronomy. 

 The Periodic system of classification enables us not 

 merely to say with every confidence that such and such 

 elements exist though yet unobserved, but it puts us 

 in a position to limit the number of possible, or at all 

 events probable, elements. It enables us to predict with 

 considerable accuracy the properties of chemical com- 

 pounds before these compounds have been actually investi- 

 gated, audit has in numberless ways proved of the greatest 

 ' service to systematic chemistry. The philosophical interest 

 j of Mendelejeff''s great generahzation is not inferior to that 

 I of the discovery of the laws of planetary motion. The 

 Russian chemist, like Copernicus and Kepler, has shown 

 the existence of law or order in one of the great depart- 

 ments of Nature's administration. Newton showed that 

 the laws of planetary motion discovered by Kepler were 

 the necessary outcome of the property ol universal 

 gravitation. We yet await the discovery of a law which 

 wiU account for the coimection between the weights or 

 masses of the atoms and their properties. Such a discovery 

 would be of surpassing interest. Previous to Mendelejeff, 



