ON THE EVOLUTION OF THE ELEMENTS. 129 



spectroscope. Large quantities of materials are required ; these materials 

 are the more likely to be changing the rarer they are, which makes the 

 research very costly, and, in the end, it is by no means certain that a single 

 lifetime is long enough to give to the investigation to secure a positive 

 result. In this connection the suggestion made by our President, 

 Principal Griffiths, is most welcome. If we do not live long enough for 

 the purpose ourselves, it is, as he remarked, our duty to provide our 

 successors with data on which to base their conclusions. 



A possible research along these lines, which I started a year ago, but 

 have not yet pushed very far owing to lack of time and material, is the 

 examination of old coins, medals, and ornaments of undoubted antiquity, 

 for evidence of occluded helium or similar gases. The age or date of 

 manufacture of such articles can often be accurately fixed by the expert, 

 and there is a tolerable certainty that the metals were melted during 

 manufacture, and have Tiot been heated since. It seemed that if helium 

 or other gas were found a definite idea of the magnitude of the rate of 

 change might be arrived at, as a preliminary to more direct experiments. 

 But a good deal of material is required, which, however, need not 

 necessarily have any great antiquarian value, so long as its antiquity is 

 above suspicion. An application for help to the British Museum failed, 

 although individual officials have assisted me in the most generous manner 

 from their own collections. But certainly the line of work looks promis- 

 ing enough to justify an attempt if the requisite quantity of material was 

 forthcoming. In the meantime, I have been engaged in a determination 

 of the minimum quantity of helium detectable by the spectroscope, and 

 with some new methods find I can be sure of ^y\j-{yth part of a cubic 

 millimetre. This is by far the smallest quantity of any element that 

 has with certainty been detected with the spectroscope. 



Taking the possible methods in the order of their directness and 

 certainty, we come next to the method, to which our President has 

 already alluded, by which valuable evidence has been secured to prove 

 that radium is being produced from uranium. If two elements can be 

 shown to occur in constant relative quantity together in all cases in 

 nature, this of itself is the strongest indirect evidence that the one is the 

 parent of the other. If the parent element is the longest lived member 

 of the series, the relative quantities of all the successive members of a 

 disintegration series present with the parent element attain with lapse of 

 time a constant equilibrium value, when as much of each is being formed 

 as disappears in each unit of time. The relative amounts of each member 

 when the equilibrium state is reached is inversely proportional to the 

 relative rates of change, or directly proportional to the average life of 

 each. Thus the researches of Strutt and Boltwood have proved- that the 

 amount of radium in all the natural uranium minerals is proportional to 

 the quantity of uranium. This ratio is 1 : 2,500,000, and represents the 

 ratio of the life of radium to that of uranium. 



The importance of this result is that it is not confined to radioactive 

 changes. It is a simple deduction which holds good in any series of 

 successive changes where the first change is the slowest and each member 

 changes into the next at a rate proportional to its quantity. If the rate 

 of change of any member is I'apid, the equilibrium amount that can 

 accumulate is small ; if the rate of change is slow it will be large. Thus 

 polonium, which is formed from radium, has never yet been obtained in 

 quantities greater than the fraction of a milligram, although many tons of 



190G. K 



