122 



KEPOKT — 1882. 



A few years ago Dr. Hnggins succeeded in obtaining photographs of 

 ten ultra-violet lines observed in the spectra of stars, which in the visible 

 part give chiefly the hydrogen lines. These ultra-violet lines are most 

 likely all due to hydrogen, and we know that this is the case with the 

 four least refrangible ones. Mr. Johnstone Stoney has pointed out 

 several harmonic ratios connecting them together ; the hydrogen line Hy 

 near G, for instance, has a wave-length which is very nearly in the ratio 

 of 35 : 32 with the one which is nearly coincident with the solar line H.' 



Mr. Johnstone Stoney ^ has also examined the absorption spectrum of 

 chlorochromic anhydride. The bands of that spectrum seem to be dis- 

 tributed with remarkable regularity. Mr. Stoney considers them to be 

 all harmonics of one fundamental vibration. The measurements do not, 

 however, seem to possess that degree of accuracy which is desirable, and 

 can be obtained by our present methods. We must, therefore, suspend 

 our judgment for the present on the reality of the coincidences pointed 

 out by Mr. Stoney. Other writers, as, for instance, Soret,^ have fi'om 

 time to time drawn attention to harmonic ratios in various spectra, and 

 the author of this report'' has during the last ten years collected a lai-ge 

 quantity of material bearing on the question. The results have, on the 

 whole, not been favoui-able to the theoiy of harmonic ratios. In an}^ 

 spectrum containing a large number of lines it is clear that, owing to 

 accidental coincidences, we shall always be able to find ratios which agree 

 very closely with the ratios of small integer numbers. It is only by 

 means of a systematic investigation that we can find out whether these 

 coincidences are due to any real cause. We must, by means of the theorj- 

 of probability, calculate the number of the coincidences which we might 

 expect to find on the supposition that the lines are distributed at random 

 throughout the whole range of the visible spectrum. If on calculating 

 out all fi-actions which can be formed in a spectrum by any pair of lines 

 the number of ratios agreeing within certain limits with ratios of integer 

 number greatly exceeds the most probable number, we should have reason 

 to suppose that the lines are not distributed at random, but that the law 

 suggested by Messrs. Lecoq de Boisbaudran and Stoney is a true one. 

 The following two tables exhibit the results of an investigation which has 

 been conducted on these lines. 



Table II. 



' A photograph taken hy Captain Abney shows conclusively, as lias already been 

 pointed out by Vogel, that the hydrogen line is a little less refrangible than H, and 

 it is very likely coincident -ndth the line 3969 ? which, according to Young, falls 

 within the broad shadow of H, and is always present in the chromosishere. 



2 Phil. Mag. xlii. p. 41 (1871). ^ Phil. 3Iag. xlii. p. 464 (1871). 



* Proc. Roy. Soc. xxxi. p. 337 (1881). 



