no 



KNOWLEDGE 



[September 1, 1891. 



P 



He then points out that the wave-lengths of H and the 

 hne near G ai-e connected harmonically, being exactly the 

 85th and 32nd harmonics of a vibration whose fundamental 

 is y2--^-3 when t is the time in which hght travels a 

 millimetre in air — a connection which is rather far fetched 

 and unsatisfactory. 



One of the most remarkable facts with regard to the 

 vibrations of molecules which become visible to us as lines 

 in the spectrum is that no simple harmonics are observed. 

 Very good photographs of the solar spectrum have been 

 taken, extending from above wave-length 3000 tenth-metres 

 to below wave-length 8500 tenth-metres, but no repeti- 

 tions of the lines and groups of lines are found to exist at 

 positions in the spectrum corresponding to double the 

 wave-length, or at two-thirds or any other simple multiple of 

 the wave-length, as would probably be the case if a 

 molecule was constituted like a bell or a tuning- 

 fork, and gave out overtones corresponding to the 

 chief vibrations with which it was pulsating. 



Much interest naturally attaches to any relations 

 that may be noted between the wave-lengths given 

 out by an element, as they may teach us something 

 with regard to the internal architecture of mole- 

 cules. The strange groups of lines in the Solar 

 spectrum and the numerous similar pairs and 

 triplets indicate that there must be many such 

 co-related lines, but the exact law of relationship 

 needs to be worked out and traced back to its 

 probable physical cause. 



Some slight advances have been made in this 

 direction. Prof. Hartley has called attention to 

 a most remarkable relation connecting the lines 

 hi the series of triplets in the spectra of magnesium, 

 zinc and cadmium. He corrected the wave-lengths 

 for atmospheric refraction and then calculated the 

 wave-frequency, and foimd that the difi'erences of 

 these frequencies for each triplet in any one series 

 is a constant quantity within the limits of probable 

 error of the observations used. That some such 

 relation exists between the wave-lengths of the lines 

 of the hydrogen spectrum was evident from the date 

 of the publication of Dr. Huggins" paper. 



In 1885 -J. .J. Balmer''- gave a formula for con- 

 necting the wave-lengths of this group of lines of 

 the hydrogen spectrum which approximately agreed 

 with the positions of the lines as then known, and 

 subsequent observations have shown that the 

 formula is remarkablv accurate. Balmer's formula 



experiments, taking every precaution in the measurement 

 of the positions of the lines, and his places agree remark- 

 ably well with Balmer's formula when they are corrected 

 so as to correspond with the refractive indexes in vacuo. 

 Prof. Ames remarks that he was entirely unable to obtain 

 the Stellar series of lines by itself, although he says " I am 

 confident my hydrogen was pure and I varied the tubes, 

 the current, the vacuum, and the exposure. I also intro- 

 duced large condensers without any noticeable change." 

 We therefore do not seem at present to be able to imitate 

 in the laboratory the conditions imder which hydrogen 

 exists in stars having the first type of spectra, and the 

 same remark applies to nearly all the spectra of elements 

 recognizable in the Solar spectrum. 



The accuracy of Prof. Anies' measures are confirmed by 



O O O O O Q ! 



O p O O C Q c 



lO -^ cc C<1 -^ O t 



^ CD ® CD C ^ 1 



SO o o o o o ; 

 O O Q O O ~ ■ 

 ifCW<-HOOXI . ---- 



? 



G, 



A = 3647-20 



Diagram illustrating the relation of tlie WaTp-LengtlisJof the Stellar Series of 

 Lines in the Spectrum of Hydrogen. 



where m takes in succession the values 3, 4, 5, &c. The 

 value of X, given by making m equal 3, corresponds to the 

 wave-length of the C line ; m = 4 gives the wave-length 

 for the F line, and so on. 



Prof. J. S. Ames, Assistant in Physics at .John Hop- 

 kins University, has, in a very important paper published 

 in the Phil. Mni/. for July, 1890, given an account of his 

 verification of this formula. Coruu + and HasselbergJ had 

 already examined the spectrum of hydrogen in the 

 laboratory, under various conditions of electrical ten- 

 sion, and had succeeded in getting all the hnes of 

 the Stellar series as far as 6, as well as many other 

 lines Which are spoken of as belonging to the " secon- 

 dary " spectrum of hydrogen. Prof. Ames repeated their 



their agreement with Prof. George Hale's measures of the 

 positions of the hydrogen lines in the Solar Prominences, 

 which are given below, i; There is some shght doubt 



• Wied. Ann. xxv. 1855. 

 t Jotir. de Phys. [10] t. 1886. 



X Mem. de VAcad. Imp. St. Fetersb., xxx. p. 7 (lf82), xxxi. p. 14. 

 Bull, de I'Acad. Imp. lit. Petersh. xi. p. 203 (1884). 



