August 4, 1910] 



NATURE 



139 



LETTERS TO THE EDITOR. 

 [The Editor does not hold 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 this or any other part of Nature. 

 No notice is taken of anonymoiis communications.] 



X-Ray Spectra. 



It was shown by Barkla and Sadler (Phil. Hag., 

 I'obruary, 1907, and October, 1908) that many elements, 

 when subject to a suitable beam of X-rays, emit a homo- 

 geneous beam of secondary X-rays of penetrating power 

 characteristic of the radiating element. One of the writers 

 (Barkla, Proc. Camb. Phil. Soc, May, 1909) showed that 

 various groups of these characteristic radiations exist, and 

 that each element most probably emits a line spectrum of 

 X-rays, each line moving to the more penetrating end of 

 the spectrum, with an increase in the atomic weight of the 

 radiating element. For no single element, however, was 

 the homogeneity of more than one radiation proved, or the 

 penetrating power accurately determined. As all the prin- 

 cipal phenomena accompanying the transmission of X-rays 

 through matter are determined by the spectra of the con- 

 st'tuent elements, it became a matter of considerable 

 theoretical interest to confirm the theory by demonstrating 

 the homogeneity of various radiations from some particular 

 element. The writers therefore chose several of those 

 elements the characteristic radiations of which w^ere 

 expected to be well within the range of penetrating power 

 comparatively easy to experiment upon. 



First, by using a penetrating primary beam, a mixture 

 of the various secondary radiations characteristic of a par- 

 ticular element was obtained. After absorbing the softer 

 constituents, a homogeneous beam of the penetrating 

 secondary X-radiation belonging to Group B was left, its 

 honiogeneitv was proved, and the coefficient of absorption 

 i" aluminium determined. 



In order to isolate one of the more absorbable con- 

 stituents, a very " soft " primary beam was used — too 

 " soft " to excite the radiation of Group B iust referred to. 

 After the effect of the scattered radiation was determined 

 by separate experiment and eliminated, this secondary 

 X-radiation was also found to be homogeneous, and its 

 absorption was determined. This radiation belonged to 

 Group .\. 



Thus two of the lines of the spectra of antimonv, iodine, 

 and barium were determined. The following values of 

 A'p are the results of the most accurate measurements so 

 far made (A is defined by the equation I = I„e-^i in trans- 

 mission through aluminium of density p) : — 



Sb : ((iroup B) 1-21 ; (Group A) 435 

 I : (Group B) 092 ; (Group .'\) 306 

 Ba : (Group B) oS ; (Group A) 224 



.\ more absorbable radiation belonging to Group A has 

 also been found to be emitted by silver in addition to the 

 penetrating radiation of Grouo B, thus accounting for 

 what appeared to Mr. Sadler (Phil. Mag., March) to be an 

 exception to the law connecting the emission of secondary 

 corpuscular and secondary X-radiations emitted bv an 

 element. 



There is indirect evidence of other spectral lines besides 

 those of Groups .A and B. Whether or not the r.adiation. 

 more absorbable than that of Groun .A — in hypothetical 

 Group X — has the properties of ordinarv X-ravs is a 

 question to 6c decided experimentallv. 



C. G. Barkla. 

 J. Xrol. 



King's College, London, July 2Q. 



Pwdre Ser. 



I\ cTse no other reader of Xatl'ri; should do so, may I 

 direct Prof. McKenny Hughes's attention -to a oaper by 

 M. Melsheimer on " Meteorgallerte," published in the 

 Jahresher. d. uwstfal. Provinziak'er. f. W'issensch. u. 

 Kunst (Bd. xxxvi., 1907-S, Miinster, 190S, pp. 5-i-i), an 

 abstract of which appeared in the Centralblatt t Ba'kterio- 

 logie (Abt. ii., Bd. x.xvii., Nos. 10-12. p. 237). published 

 on June 22 of this year? The author appears to have paid 

 NO. 2127, VOL. 84] 



attention to these masses of jelly, which are to be found 

 in winter on meadows and other open places, for a period 

 of years, and has come to the conclusion that they are the 

 swollen oviducts of frogs. Herons eat female frogs in 

 winter, and the oviducts become mixed in the crop with 

 fish remains, wliich may become luminous. The contents 

 are thrown up undigested, and become gelatinous when 

 moistened. It is also possible that the heron may, during 

 flight, discharge the gelatinous mass in a luminous con- 

 dition, a'nd hence the idea that the jelly is of meteoric 

 origin. Geo. H. riLTiivERiDGE. 



Royal College of Science, Dublin, July 4. 



THE PRESSURE OF LIGHT.' 

 "T^HE earliest attempts to detect the pressure of light 

 ■'■ were made in the eighteenth century. The 

 corpuscular hypothesis was then almost universally 

 accepted, and to the believers in that hypothesis the 

 idea that light should exert a pressure upon a body 

 against which it fell was perfectly natural. Regard- 

 ing the atoms and molecules of a luminous surface as 

 a battery of minute guns firing off a continuous 

 stream of still more minute shot — the corpuscles — they 

 inevitably supposed that any body bombarded by the 

 shot would be pressed back. Many experiments were 

 made to detect this bombardment by directing a power- 

 ful beam of light on to a delicately suspended disc, 

 sometimes in air at ordinary pressures, sometimes in 

 a vacuum, but with quite inconsistent and inconclu- 

 sive results. They were met with the disturbances 

 which still beset experiments on light forces — disturb- 

 ances partly due to convection in the surrounding gas, 

 and partly due to the radiometer action which Sir 

 William Crookes discovered and investigated a hun- 

 dred years later. 



.According to the corpuscular theory, the pressure 

 sought should be equal to twice the kinetic energy 

 of translation per unit volume in the beam used. If 

 the earlier experimenters had known the principle of 

 the conservation of energy and the mechanical 

 equivalent of heat, they would no doubt have mea- 

 sured the energy of the beam, and would then have 

 found that the pressure to be looked for was far too 

 minute for detection by the apparatus which they em- 

 ployed. 



With the abandonment of the corpuscular theory 

 and its replacement by the wave theory, the idea of 

 pressure of light disappeared, for the form which the 

 wave theory took at first did not suggest a pressure, 

 and it was not until 1874 that a definite and exact 

 theory of light pressure was given by Clerk Max- 

 well. According to his theory of stresses in the 

 medium, both electric and magnetic tubes of force 

 press out laterally. If, then, light waves consist of elec- 

 tric and magnetic tubes of force transverse to the direc- 

 tion of propagation, these tubes should press on any 

 surface against w'hich they impinge, and the pressure 

 should be equal to the energv in unit volume of the 

 light. Maxwell calculated that the pressure w'hich 

 should be exerted by full sunlight amounted to about 

 i/2;iOoo of a dyne per sq. cm. 



Twentv-five vears later Prof. Lebedew succeeded in 

 detecting and measuring the pressure. He allowed 

 the concentrated rays of an electric lamp to fall on a 

 thin blackened plal;inum disc delicately suspended in 

 a vacuum so high that there was probably no con- 

 vection, and even the radiometer action was compara- 

 tively small. By the ingenious device of using discs of 

 different thickness with radiometer action proportional 

 to the thickness, he was able to calculate the force 

 acting on an infinitely thin disc on which there would 



1 Based upon Ihe Bakerian lecture on 'The Pre«ure of Llaht ngainst 

 tSe Source : the Recoil from Light.' by Prof I. H. Poynting, F.R.S., and 

 Dr. Guy Barlo«', delivered at the Royal Society on March i;. 



