58 3Ir. William CrooJces [Feb. 18, 



Pass along the curve and in nearly every case the same law holds 

 good. Thus the last element of the first complete vibration is 

 chlorine. In the corresponding j)lace in the second vibration we 

 have not an exact repetition of chlorine but the very similar body 

 bromine, and when the same position recurs for the third time we 

 see iodine. I need not multiply examples. I may, however, point 

 out that we have here a phenomenon which reminds us of alter- 

 nating or cyclical generation in the organic world, or we may 

 perhaps say of atavism, a recurrence to ancestral types, somewhat 

 modified. 



In this evolutionary scheme it cannot be expected that the 

 potential elements should all be equal to each other. On the con- 

 trary, many degrees of stability will be represented, and if we look 

 with a scrutinising eye we shall see our old friend the " missing 

 link," coarse enough to be detected in the groups comprising such 

 bodies as iron, nickel, and cobalt ; palladium, ruthenium, and 

 rhodium ; iridium, osmium, and platinum : whilst in a more subtile 

 form these missing links present themselves as representatives of the 

 diifcrences which I have suggested between the atoms of the same 

 chemical element. 



On the even or paramagnetic half of the swing the energy 

 appears to have acted in a very irregular manner, whilst on the odd, 

 or diamagnetic half, there is considerable regularity. Thus, be- 

 tween the extreme odd elements, silicon (28), germanium (73), 

 tin (118), a missing element (163), and lead (208) there is a diiier- 

 ence of exactly 45 units, rendering this half of the curve remarkably 

 symmetrical. On the even side the differences are 36, 42, 61, 39 

 and 53 (assuming an atomic weight of 180 for a missing element 

 between cerium and thorium). At first sight these differences 

 appear to follow no law, but they gain interest when we see that the 

 mean differences of these figures is 44*2 — almost exactly the same as 

 that on the odd side of the curve. 



From this uniformity of difference — actual on the one side and 

 average on the other — we may fairly infer that whilst on the odd side 

 there has been little or no variation in the force symbolised by the ver- 

 tical line, minor irregularities have been the rule on the even side. 

 Or, in other words, the fall of temperature has been very uniform on 

 the odd side — where, accordingly, we see that every original element 

 represents a well marked group, sodium, magnesium, aluminium, 

 silicon, j^hosphorus, and chlorine ; whilst on the even side the tem- 

 perature has fallen with considerable fluctuations, thus preventing 

 the formation here of any well-marked groups of elements, excepting 

 those of w^hich lithium and glucinum are the leading types. 



Having thus detected irregularities in the fall of temperature in 

 the protyle, we may next ask is there any fluctuation in the force re- 

 presented by the pendulum-movement ? This movement I have 

 assumed to be connected with electrical energy. The earliest-formed 

 elements are those in which chemical energy is at a maximum ; as 



