February i8, 19 15] 



NATURE 



671 



laboratory glassware, porcelain filter papers, etc., not 

 hitherto produced in this country. 



The association is not a ring or financial combine, 

 but is merely arranged for the pooling of information 

 and for saving as much time, and preventing as far 

 as possible duplication of correspondence with works, 

 in the way of inquiries and information ; the standard- 

 isation of sizes for moulds where works undertake to 

 make ; and the elimination of difficulties which would 

 occur if all the various firms were working indepen- 

 dently. 



So far, after months of correspondence, interviews, 

 conferences, meetings, and considerable expense, the 

 association has solved the problem in connection with 

 glass ware, and in a few ; weeks' time the various 

 firms constituting the association will be in a position 

 to supply the most useful sizes of glass beakers and 

 flasks in an English-made glass similar to Jena 

 glass, which has been tested by well-known English 

 men of science, and can be definitely and safely 

 recommended. 



The same holds good for porcelain evaporating 

 basins, beakers, crucibles, and covers, etc., and also 

 for an English-made filter- paper equal to the German 

 quality of .Schleicher and Schull No. 595, and further 

 qualities of these filter papers will be forthcoming 

 verv shortly. Also, as time goes on, further items 

 will be added to the list, as negotiations are brought 

 to a successful issue. 



You will therefore see that a great deal of impor- 

 tant and valuable work has been done, and that the 

 problem of glassware has been to a very great extent 

 solved so that an English-made glass equal to the 

 well-known Jena glass will be on the market and 

 can be supplied in a few weeks from the present time 

 by the various firms constituting the British Labora- 

 torv Ware Association. 



C. A. Mercer, 

 Chairman of the British Laboratory 

 Ware .Association, Ltd. 



34 Camomile Street, E.C., London, February 8. 



Problems of Radiation. 



The study of the production of homogeneous Ront- 

 gen radiation reveals some of the most interesting and 

 suggestive facts bearing upon the problems of radia- 

 tion. A few of these will be briefly indicated. 



The emission of a fluorescent X-radiation by an 

 element exposed to a primary X-radiation of shorter 

 wave-length necessitates the absorption of a greater 

 amount of energ>- from the exciting primary beam. 

 It is possible definitely to assign a given portion of 

 the energ>- absorbed from the primary beam as con- 

 nected with the emission of a particular fluorescent 

 radiation (Barkla and Sadler, P/n7. ^lag., 1909.) The 

 element exposed to this primary- radiation emits also a 

 corpuscular radiation, a portion of which is definitely 

 associated with each fluorescent X-radiation. From 

 measurements of these associated quantities, certain 

 broad conclusions have been found to hold at least 

 approximately in all cases hitherto dealt with ; they 

 hold within the limits of experimental error in the 

 one case carefully investigated. These are : — 



(i) The number of quanta of fluorescent X-radia- 

 tion emitted is equal to the number of high-speed 

 electrons in the corpuscular radiation (the associated 

 corpuscular radiation). 



(2) The total energ}- in the corpuscular and fluores- 

 cent radiations is equal to that of the primarj^ beam 

 absorbed. 



These are not merely hypotheses which by trial may 



NO. 2364, VOL. 94] 



be found to be consistent with quantitative measure- 

 ments of the energies of the two secondar)' radia- 

 tions, but are the only rational conclusions which can 

 be drawn from experiments on some substances. TTie 

 deviations which have been obser\'ed in other cases 

 are subjects for further investigation ; the nature of 

 these will be indicated later. 



These conclusions may be described in greater 

 detail as follows : — 



If a primary- Rontgen radiation of wave-length /i, 

 fall upon an element with a characteristic radiation 

 of w-ave-length /*,, then provided /x, is less than ^. the 

 primary radiation is specially absorbed, and the 

 element emits the particular fluorescent X-radiation, 

 together with an associated corpuscular radiation. If 

 we denote the energy of the primary specially ab- 

 sorbed in connection with the emission of the fluores- 

 cent radiation (of Series K, say) by Ek, the energv of 

 the associated corpuscular and fluorescent radiations 

 by Ck and Fk respectively, then Ek=Ck + Fk. Also 

 as the energ\' of each electron in the corpuscular 

 radiation is approximately (possibly accurately) that 

 of a quantum of the priman,' radiation (we make the 

 usual assumption here, that there is only one velocity 

 of ejection ; good reason can, however, be given for 

 this) :— 



Fk _energy of fluorescent radiation _ 



Ck "energy of corpuscular radiation 



energy of quantum of fluorescent radiation _;«| 

 energy of quantum of primary radiation ;*« 



hence : — • 



Ck = 



_ M, 



Ek and Fi 



= _5_E. 



M1 + M2 



So we obtain expressions for the energies of fluores- 

 cent radiation (series K) and the associated corpus- 

 cular radiation in terms of that of the specially ab- 

 sorbed priman,- energy, for primar\' radiation of any 

 wave-length. Plotting energies of secondary radia- 

 tions and wave-length of primary' radiation, we get 

 the curves shown in the accompanying diagram : — 



Thus when ;u, is just less than /Xj, the energy is 

 equally divided between the corpuscular and fluores- 

 cent X-radiations. As ^, diminishes a greater pro- 

 f>ortion of the energy goes into the corpuscular radia- 

 tion, and less into the fluorescent, until for ver\- short 

 waves all the energy is taken up by the corpuscular 

 radiation. 

 The abox'e simple theory agrees exceedingly well with 

 1 the e.xperimental results obtained with the element 



