218 Mr. F. Soddy on the 



Two series of experiments were performed, the first with 

 100 grams o£ the thorium-contaminated urauyl nitrate and 

 the second with 50 g. of the purest uranyl nitrate, both as 

 used by Fleck. Only the latter series need be considered in 

 detail here. The method was to dissolve the crystals in 

 ether in a tap-funnel, separating the aqueous layer, and 

 washing a few times, each time with a few drops of water. 

 The first aqueous extract containing the uranium X was 

 rejected. The uranium was left to itself for periods varying 

 from 20 hours to 4 days and the washing with a few drops 

 of water then repeated, the uranium X being separated, from 

 the aqueous layer of uranium solution obtained, by addition 

 of a trace of ferric chloride, pouring into warm N/2 

 ammonium carbonate in the usual way, digesting a few 

 minutes on the water bath and filtering through an ashless 

 filter, which was ignited and spread over the ground-glass 

 plate as described. From the ethereal solution, alter settling, 

 a few c.c. were siphoned off from the top into a platinum 

 crucible, and ignited at once to oxide, U 3 8 , and spread over 

 another plate. In all, the decay curves of five uranium X 

 preparations and the recovery curves of five uranium oxide 

 preparations were measured. The last uranium X prepara- 

 tion from the purest uranium was prepared by the barium 

 sulphate instead of by the ferric method. The whole ethereal 

 solution of uranium, after complete removal of uranium X, 

 was ignited to uranium oxide. This weighed 15'25 grams 

 and was dissolved in dilute nitric acid, the solution, almost 

 neutralised with ammonia, being left to reaccumulate 

 uranium X for 23 hours. Excess of sulphuric acid and 

 a dilute barium nitrate solution was then added and the 

 precipitated barium sulphate, weighing 89 milligrams, was 

 got on to a glass plate in the usual way, as a film weighing 

 about 0*35 mg. per cm. 2 All of the five uranium X prepara- 

 tions showed decay curves similar to those described by 

 AntonofT, so that probably the reason why AntonofF did not 

 observe the initial decay tor uranium X separated by barium 

 sulphate, is to be ascribed to some secondary cause, such as 

 absorption in the material itself. For the barium sulphate 

 and ferric oxide films, in which about one day had elapsed 

 from the previous separation, the curves showed that about 

 30 per cent, of the total initial radiation was due to soft 

 /3-rays of the rapidly decaying radiation (fig. 1, PI. II.). In 

 those for which the time of accumulation was four days the 

 proportion was about 16 per cent. The decay of the hard 

 /3-radiation was normal (Curves AIL, A IIJL, B II.). The 

 decay of the total radiation was abnormal (Curves A I., B L). 



