August 17, 1905 j 



NA TURE 



365 



in a very agreeable style ; moreover, he does not 

 assume the reader to possess any knowledge of 

 advanced mathematics. For those who wish to study 

 any of the more important topics in detail he gives 

 ample references ; for those merely in search of 

 diversion he provides a mine of amusement, in ex- 

 ploring which many pleasant hours may be spent. 

 ,\nd there are some unsolved problems mentioned 

 which the amateur with a mathematical turn of mind 

 may attack with nearly as much chance of success 

 as the expert ; for instance, to give a strict proof that 

 only four different colours are necessary to colour a 

 map distinctly. Altogether this is an excellent work 

 of its kind, and ought to find a large number of 

 readers ; even those who have a former edition will 

 be likely to buy this one, if only for the sake of the 

 verv interesting account of the vicissitudes of the 

 mathematical tripos. 



LETTERS TO THE EDITOR. 



\riie Editor does not bold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for tliis or any other part of Nature. 

 No notice is taken of anonymous communications.] 



The Rate of Formation of Radium. 



The production of radium from uranium has now been 

 observed experimentally ; the rate of production is not, how- 

 ever, in accordance with the quantitative theory. Mr. 

 .Soddy's observations {PItil. Mag., June, 1905) gave a rate 

 of production of only one-thousandth of the theorttical 

 amount. .\n experiment which I made on a specimen of 

 uranium salt, known to be at least thirty years old, has 

 confirmed Mr. .Soddy's conclusion so far as to show that 

 the mean rate of production of radium could not have 

 exceeded a hundredth part of the theoretical amount. 

 It may, of course, have been much less, since the amount 

 of radium initially present is unknown. 



The explanation of this discrepancy, which has been 

 suggested by Mr. Soddy and others, is that thtre may 

 be a transitional product. If this is the case, it is 

 to be expected that the rate of production of radium 

 from uranium initially purified will be found to accelerate 

 as time goes on. In the meantime, I am trying an ex- 

 periment which promises to give the required information 

 more easily. 



The transitional product, if it exists, must be contained 

 in pitchblende. If, therefore, we could remove all the 

 radium, but as little else as possible, from a solution of 

 pitchblende, the increased rate of production of radium 

 might be apparent. 



Fifty grams of the best pitchblende were dissolved in 

 nitric acid. The insoluble residue was fused with sodium 

 carbonate and added to the solution. The whole was 

 evaporated to small bulk to render silica insoluble ; more 

 dilute acid was added, and the silica filtered off and re- 

 jected. The metallic bases were thus got into solution. 



The solution was freed from radium so far as possible 

 by adding barium nitrate solution in small portions, al- 

 ternately with equivalent quantities of potassium sulphate. 

 Four and a half grams of the barium salt were thus 

 added. .After this the amount of radium remaining was 

 determined by its emanation ; three determinations gave, on 

 an arbitrary scale, 69, 585, 61-5, mean 63-0. .After an 

 interval of three and a half months the amount was 

 again determined. The values were 735, 745, 720, 750, 

 725, mean 735. It appears probable that this increase 

 is significant, since each of the second series of numbers 

 is larger than any of the first series. 



.Assuming that the difference is significant, the rate of 

 production per gram of mineral per year would be, on 

 the same scale, 0-723. The equilibrium quantity of radium, 

 the amount, that is, in the untreated mineral, was found 

 to be, per gram, 10,100. If radium decays to one-half its 

 initial quantity in a thousand years, as theory indicates, 

 then the production in one year from a gram of the 



NO. 1868, VOL. 72] 



mineral should be 10,100/ 1-45 X 1000 = 0-9, about ten times 

 the observed amount. 



The increase is insufficient to inspire complete confidence. 

 It seems most probable, however, that there is an increase 

 much greater than in Mr. Soddy's experiments with pure 

 uranium salts. It would not have been difficult to remove 

 all traces of radium, and then the increase (if real) would 

 have been unmistakable. It was feared, however, that 

 the barium precipitation might remove part of the hypo- 

 thetical intermediate product. It seems likely that this 

 is ihc- I a^ ■, vim e the rate of production is still less than 

 theor\ r.quirrs. 



.\ little longer interval will, it is hoped, give a conclusive 

 result. It is intended to try other methods of separating, 

 the radium, in the hope of avoiding all loss of the inter- 

 mediate product. R. J. Strutt. 



The Effect of Radium on the Strength of Threads. 



In a note which appeared in Nature on February 4, 

 1904, Lord BIythswood announced his observation of the 

 destructive action exerted on cambric by the radiation 

 from radium. Having at our disposal recently twenty 

 milligrams of radium bromide which had, for a time, 

 nothing better to do, we investigated the progressive de- 

 crease of strength of threads exposed to its influence. In 

 order to have examples of both animal and vegetable 

 fibres, we used unspun silk and ordinary bleached cotton 

 thread. 



Ten pieces of thread were exposed at a time. The 

 threads were folded round a strip of writing paper and 

 held in place by being caught in notches cut in the edges 

 of the strip. The paper was laid on the top of the capsule 

 containing the radium, so that the ten threads were ex- 

 posed to the bare radium at a distance of about half a 

 centimetre. The whole was enclosed in a lead box. After 

 a certain period of exposure the average breaking strength 

 of the threads was taken and plotted against the time. 

 The points obtained lay closely on a smoothly descending 

 curve. 



In the case of the silk fibres the loss of strength went 

 on at a practically uniform rate from the beginning up 

 to the longest duration of exposure given (seven days). 

 The initial strength was 78 gnis., and this decreased by 

 about 4 gms. per day. The cotton threads, on the other 

 hand, gave a curve which fell more rapidly in the early 

 than in the later stages. The strength began at 370 gms., 

 and decreased at first by about 60 gms. per day. -After 

 ten days the rate of weakening was about half this. The 

 longest exposure given was seventeen days ; at the end 

 of this time the strength was reduced to 50 gms. The 

 different behaviour of the two kinds of fibres may be 

 due to the much greater thickness of the cotton threads. 



The effect seemed to be due entirely to the a rays. .A 

 piece of paper was interposed between the threads and 

 the radium, and three days' exposure was given. In the- 

 subsequent test none of the thre?^s broke at the exposed 

 part, and the strength was not decreased. 



We tried the effect of moistening the cotton threads, 

 the two ends of each thread being left, during a three 

 days' exposure, dipping into a vessel of water. On open- 

 ing the lead box, in which the whole arrangement was 

 enclosed, it was found that the radium bromide, being 

 hygroscopic, was wet and partially dissolved. The strength 

 of the threads was found to be higher than when exposed 

 in a drv condition for the same period. The difference 

 was too great to be attributed to the increase of strength 

 imparted to threads by moisture, and was plainly due to 

 the decreased emission of rays accompanying the solu- 

 tion, and the consequent removal of the emanation from 

 the radium. We traced the course of the recovery of 

 activity by the dried radium by making a series of three- 

 day exposures of dry threads. The effectiveness of the 

 radiation as measured by the weakening of the threads 

 came back by regular steps in about a fortnight to a value 

 slightly greater than its original one. This may have 

 been due to a re-arrangement of the upper surface of , the 

 powder, which was not, at the beginning, very regularly 

 spread over the bottom of the capsule. 



Hilda P. Martin. 

 W. B. Morton. 



Queen's College, Belfast, .August 8. 



