708 Intelligence and Miscellaneous Articles: 



shape of the region surrounding the bead which emits light 

 is measured. From this and the velocity of the name gases 

 Prof. Wilson calculates the coefficient of diffusion. 



I may perhaps mention [that the treatment of this question is 

 not quite new. Already seven years ago the possibility of finding 

 the mass of the particles emitting light was the starting-point of 

 such researches on flame diffusion (see P. Jjen&vd, Amidlen der 

 Phi/sik, vol. xvii. p. 198, 1905). The measurements were first 

 carried out by Mr. 0. Jessel, and then continued and improved by 

 the writer together with Dr. J. Laub (published by A. Becker, 

 Heidelberg Acad, of Sciences for 1911). The method used 

 consists in the determination of the distribution of the vapour 

 emitting light by measuring the distribution of the intensity of 

 the light below the bead of salt. From this and the velocity 

 of the flame the coefficient of diffusion has been calculated 

 (I. e. p. 17). AVe believe that this methol, described there in 

 detail, has distinct advantages in comparison with Prof. Wilson's 

 method. 



The value of the coefficient of diffusion found by us for sodium 

 is about 1*0, and lies between the values which are to be expected 

 for the atom of sodium and the molecule of its oxide. From this 

 we conclude that the free state of the atom and the chemically 

 combined state of the same atom alternate at about equal intervals. 

 The much higher values found by Prof, Wilson for (he coefficients 

 of diffusion seem scarcely well supported by theoretical view. 

 Also Prof. Wilson's conclusions as to the charge of the luminous 

 particles agree so badly with the known results of recent investi- 

 gations (see P. Lenard, Amnalen der Physik, vol. ix. p. 612, 1902, 

 and Heidelberg Acad, of Sciences, 1912; E. N. da C. Andrade, 

 Dissertation, Heidelberg, 1911, and Phil. Mag. vol. xxiii. p. 865, 

 & vol. xxiv. p. 15, 1912), that it is hard to believe that they furnish 

 a proof of the. correctness of his results. As our researches are 

 being continued, we may return to these points later on. 



As to the measurement of the velocity of the flame gases, 

 Prof. Wilson uses a vibrating tuning-fork, and observes the wavy 

 outline of the luminous band coming from a bead attached to one 

 prong of the fork. It may be noticed that this is quite the same 

 method as I have published several years ago (Aanalen der Phi/sik, 

 vol. xxiv. p. 823, 1907). There are, however, better methods at 

 hand at present. One of th^se I have given at the same place 

 (Anncd. vol. xxiv.) ; it consists in measuring the force exerted on 

 a small ball by the stream of the flame gases (see also A. Becker,; 

 Annalen der Physilc, vol. xxiv. p. 963, 1907, and S. S. Nehru, 

 Dissertation, Heidelberg, 1911). Another method has been pub- 

 lished later on by Mr. Andrade (Anhateri'der Physik, vol. xxxvii. 

 p. 380, 1912), who has also made comparisons between the different 

 methods named. 



- ... Yours truly, 



Phvsical Laboratory of the A. Bf-CKEE. ' 



University, Heidelberg, 

 July 29, 191:?. 



