574 REPORT — 1886. 



potential elements would all be equal to one another. Some would be unable 

 to resist the slightest disturbance of the unstable equilibrium in which they took 

 their rise ; others would endure longer, but would ultimately break down as 

 temperature and pressure varied. Many degrees of stability would be here repre- 

 sented ; not all the chemical elements are equally stable, and if we look with 

 scrutinising eyes we shall still see our old friend the missing link, coarse enough to 

 be detected by ordinary chemical processes, associated in the groups containing 

 such elements as iron, nickel, and cobalt ; palladium, ruthenium, and rhodium ; 

 iridium, osmium, and platinum. Whilst in their more subtile form these missing 

 links present themselves as the representatives of the differences which I have 

 detected and described between the atoms of the same chemical element. 



Dr. Carnelley has pointed out that ' those elements belonging to the even series 

 of Mendeleeff's classification are always paramagnetic, whereas the elements belong- 

 ing to the odd series are always diamagnetic' On this curve the even series to the 

 left, as far as can be ascertained, are paramagnetic, and, with a few exceptions, all 

 to the right are diamagnetic. The very powerful magnetic metals, iron, nickel, 

 cobalt, and manganese, occur close together on the proper side. The interperiodic 

 groups of which palladium and platinum are examples are said to be feebly mag- 

 netic, and if so they form the exceptions. Oxygen, which, weight for weight, is 

 more magnetic than'iron, comes near the beginning of the curve, while the power- 

 fully diamagnetic metals, bismuth and thallium, are at the opposite end of the 

 curve. 



On the odd, or diamagnetic half of the swing, the energy appears to have con- 

 siderable regularity, whilst it is very irregular on the opposite side of the curve. 

 Thus, between the extreme odd elements, silicon (28), germanium (73), tin (118), 

 the missing element (163), and lead (208), there is a difference of exactly 45 units, 

 conferring remarkable symmetry on one half of the curve. The differences on the 

 even side are 36, 42, 51, 39, and 53 (giving the missing element between cerium 

 and thorium an atomic weight of 180) ; these at first sight appear conformable to 

 no law, but they become of great interest when it is seen that the mean difference 

 of these figures is almost exactly the same as that on the other side of the curve 

 —viz. 44-2. 



This uniformity of difference— actual on the one side and average on the other 



brings out the important inference that, whilst on the odd side there has been 



little or no variation in the vertical force, minor irregularities have been the rule on 

 the even side. That is to say, the fall of temperature has been very uniform on the 

 odd side — where every element formed during this half of the vibration is the 

 representative of a strongly marked group — sodium, magnesium, aluminium, 

 silicium, phosphorus, sulphur, and chlorine ; whilst on the even side of the swing 

 the temperature has sunk with considerable fluctuations, which have prevented the 

 formation here of any well-marked groups of elements, with the exception of those 

 of which lithium and glucinum are the types. 



If we can thus trace irregularities in the fall of temperature can we also detect 

 any variation in the force represented by the pendulous movement? I have 

 assumed that this represents cliemical energy. In the early formed elements we 

 have those in which chemical energy is at its maximum intensity, while, as we 

 descend, affinities for oxygen are getting less and the chemism is becoming more 

 and more sluggish. Part may be due to the lower temperature of generation not 

 nermitting such molecular mobility in the elements, but there can be little doubt that 

 the chemism-forming energy, like the fires of the cosmical furnace, is itself dying 

 out. I have endeavoured to represent this gradual fading out by a diminution of 

 amplitude, the cui've being traced from a photographic record of the diminution of 

 the arc of vibration of a body swinging in a resisting medium. 



\Vben we look on a curve of this kind there is a tendency to ask, what is there 

 above and below the portion which is seen ? At the lower end of our curve what 

 is there to be noted ? We see a great hiatus between barium (137) and iridium 

 (192-5), which it seems likely will be filled up by the so-called rare elements. 

 Judging from my own researches, it is probable tliat many of these earthy elements 

 will be found included in one or more interperiodic groups, whilst the higher mem- 



