2i8 SCIENCE PROGRESS 



careful measurements showed the masses of the two isotopic 

 constituents of neon to be 20-00 and 22-00 to an accuracy of 

 tV per cent, on the oxygen scale. In order to give the accepted 

 density, the relative quantities required are 90 per cent, and 

 10 per cent., which is in good agreement with the estimated 

 intensity of the lines. 



It is as well to state here that the photographic intensity 

 of the line is not, in general, to be regarded as a measure of 

 the quantity of the substance present. With two isotopes of 

 the same element, however, the chemical identity makes it 

 practically certain that the intensity is a reliable criterion — ^this, 

 of course, only refers to the intensity of lines on the original 

 negative, not to reproductions. 



The two first order lines of neon can be seen on Spectrum I 

 between the two carbon groups. The above results obtained 

 for their masses prove the complex nature of atmospheric 

 neon beyond dispute. Faint indications of a third constituent 

 of mass 21 were obtained, but this can only be regarded as 

 doubtful, and in any case only present in the minutest propor- 

 tions. 



Chlorine {At. Wt. 35-46). — The very marked fractional 

 atomic weight of chlorine has been the subject of much in- 

 terest and discussion in the past. It was naturally the next 

 element to be investigated. The question whether its accepted 

 atomic weight represented the true mass of its individual atonis 

 or was merely a statistical mean possessed a far deeper chemi- 

 cal significance than that at issue in the case of neon, which 

 is inactive chemically. It was not long in doubt, for the 

 photographs obtained showed that the introduction of chlorine 

 into the discharge tube gave rise to a group of strong lines 

 corresponding to masses 35, 36, 37, and 38, all of which were 

 whole numbers to the accuracy of experiment. There was no 

 indication whatever of a line corresponding to a mass 35*46. 

 Spectra II, III, IV were obtained with chlorine ; on the first, 

 which was taken with the smallest magnetic field, two faint 

 lines can be seen at lys and 18-5. These are the second order 

 lines of 35 and 37, and are strong testimony that the latter are 

 elementary, for it is extremely rare for compounds to give 

 second order lines at all. Further evidence on the same point 

 is given by the two fines 63, 65, Spectrum IV, which are doubt- 

 less due to two compounds, COCl. 



These figures leave no escape from the conclusion that the 

 element chlorine is complex, and two of its constituents have 

 atomic weights 35 and 36. It might be argued that 36 and 38 

 are elementary lines, and at present there is no evidence to 

 deny this ; it is much more likely that they are due to the two 

 hydrochloric acids. In all spectra taken with chlorine present, 



