April 15, 1922J 



NATURE 



483 



Fig. 2. 



egins as a single delicate ring surrounding the minute 

 Central nucleus. It can be measured from a stage 

 )rdering on invisibility to a stage when its central 

 rea is beginning to darken up and the first shadowy 

 of the outermost ring of all — that due solely to 

 idium C — appear. A large number of readings on these 

 ibfyonic haloes, made recently by various observers, 

 )nfirm the mean value of is radius as cited in a paper 

 communicated to the Royal 

 Society in 1916. The dis- 

 crepancy with the theoretic 

 curve is small; 10 or 12 

 pel cent, of the external 

 radius. The allowance for, 

 and measurement of, the 

 nucleus is sufficiently 

 difficult to introduce some 

 uncertainty. 



This misfit may be of 

 considerable significance. 

 I have already reminded 

 you that the range of 

 the a-ray emitted by a 

 transforming element is re- 

 lated to its rate of break- 

 up. The range is longer for the shorter lived ele- 

 lents. Now here the first ring of the uranium halo 

 mica shows a longer range than we would expect 

 )m the air-curve as observed to-day. The agreement 

 jtween the two in other cases appears to show that 

 lis is not due to any unknown effect influencing the 

 ^retardation in mica. The location of the first uranium 

 ring is mainly referable to those short-range a-rays 

 arising from the initial transformations of the uranium 

 series. We infer that one or more of these rays must 

 have had a longer range in past times and, of course, 

 that the corresponding transformation periods must 

 have been shorter. A specially influential ray is that 

 slowest of all the rays — that which is emitted in the 

 break-up of uranium i. The discrepancy might be due 

 to this ray possessing a greater range in early geological 

 times. But, whatever the cause, the nature of the 

 misfit suggests evidently that formerly the rate of 

 transformation of uranium to lead was faster than it 

 is to-day. 



It is with some reserve that I refer here to measure- 

 ments made lately on haloes of comparatively recent 

 and of very remote geological ages. I say " with 

 reserve," for not only are the results of a nature calling 

 for very adequate confirmation, but the measurements 

 present considerable difficulty. The point at issue 

 may be stated in a few words : — Is the abnormality 

 observed in the dimensions of the uranium halo de- 

 pendent in amount upon the antiquity of the rock in 

 which the halo is developed ? 



I had sought occasionally for uranium haloes in 

 rocks younger than the Leinster granite — which is of 

 early Devonian age. The granite of Mourne, which is 

 of Eocene or early Tertiary age, for long refused to 

 reveal any haloes suitable for measurement. However, 

 recently, I was so fortunate as to find a few of these 

 early halo rings which I was able to measure. Further 

 search has revealed a few more ; but htey are excess- 

 ively scarce and rather difficult to detect. The nuclei 

 of these haloes are only rarely zircon — they seem to be 



N30. 277, VOL. 109] 



apatite — possibly allanite — and their average size is 

 greater than the zircon nuclei of the Carlo w mica. 

 Both the mineral nature of the Mourne nuclei and their 

 dimensions involve, therefore, a bigger subtractive 

 correction on the observed radius than is required in 

 the case of the Carlow haloes. But in addition to this, 

 there appears to be a small difference in the external 

 radius of the Eocene halo and that of the Devonian 

 halo. According to a large number of readings by 

 several observers, some of whom were not acquainted 

 with the question at issue, the external radius of the 

 Eocene halo-ring — no allowance being made for the 

 nuclear radius — is 0*0135 mm. The same observers 

 obtained for the Devonian halo 0-0146 mm. — without 

 allowance for the nucleus. The nuclear correction, as 

 I have said, would have increased the discrepancy, but 

 the correction is a difficult one. There is no reason to 

 believe that more than i per cent, of this differ- 

 ence can be ascribed to fhe chemical composition or 

 density of the micas, both of which have been 

 investigated. 



Still more recently I have found these primary ring- 

 haloes in the micas of Arendal and Yitterby, which 

 are said to be of Archaean age, and which are certainly 

 extremely ancient. These haloes appear to possess a 

 radial dimension of o-oi6o mms., or a little less. Here, 

 again, the nature of the mica does not appear to be 

 responsible. According to these measurements it 

 would appear that the radius of the Eocene halo-ring 

 must be increased by about 7 per cent, to attain the 

 size of the Devonian halo-ring, and that this is, in 

 radial dimension, about 10 per cent, smaller than the 

 Archean. It would seem as if we might determine 

 a geological chronology on the dimensions of these 

 halo-rings ! 



The foregoing results, if confirmed, would give strong 

 support to the view that some factor, variable over 

 geological time, had affected the ranges and periods of 

 certain elements concerned in building up the uranium 

 halo. However, too much stress must not be placed 

 on these measurements till they are confirmed by 

 haloes in yet other micas. Pending further investiga- 

 tions, I return to the fact that the uranium halo of 

 Devonian age does not conform to the ionisation curve 

 of the uranium family as determined on present-day 

 measurements. Serious discrepancy seems confined 

 to the shorter ranges, more especially with that 

 primary range which is most influential in determining 

 the rate of production of uranium lead. 



We do not appear to be in a position to deny the 

 possibility that uranium i may have slowed down in its 

 rate of decay over geological time. Such laboratory 

 observations as can be extended to the case of short- 

 lived elements would not, probably, shed any light on 

 the matter. It is a possibility long ago suggested by 

 Rutherford. But if this is the explanation we must 

 admit that in the case of thorium any corresponding 

 effect must have been much smaller. On the whole 

 the former influence of one or more isotopes of uranium 

 — which possibly may almost have disappeared — 

 seems the more probable explanation. Hypothetical 

 isotopes of uranium have been invoked by highly 

 competent authorities to meet the difficulties affecting 

 the ionisation accounts of the uranium family of 

 elements. Boltwood suggests as " not impossible " 



