MR. A. FOWLER ON THE SPECTRUM OF MAGNESIUM HYDRIDE. 463 



yet be given, but there is sufficient evidence that many of the band lines in these 

 regions are certainly due to magnesium hydride. 



To determine exactly the amount of detail in the spot which is accounted for by 

 magnesium hydride will necessarily involve a comparison of wave-length measurements. 

 In this investigation, identity of numerical values must not in every case be expected, 

 in consequence of the presence of so many other lines in the spot spectrum. Thus, 

 the line 5186'59 is evidently combined in the spot with a titanium line at 5186 - 50, 

 giving rise to an apparent discordance with the wave-length 5186'53 assigned to the 

 spot line by HALE and ADAMS. Such combinations will likewise produce apparent 

 small displacements of the metallic lines. It should be noted also that the apparent 

 intensification of some of the metallic lines, as for example the faint iron line 5177'41, 

 must be due, in part at least, to superposed lines of magnesium hydride, and similarly 

 that in some cases the weakening of metallic lines may be obscured. Further, some 

 of the magnesium hydride lines appear as mere fringes to comparatively strong metallic 

 lines, and are likely to be overlooked if not specially sought. These are some of the 

 points that must be taken into consideration when a minute comparison with the 

 spot spectrum is undertaken. 



There is another feature of the spot spectrum which is to some extent explained by 

 the identification of magnesium hydride, namely, the bright interruptions whicli occur 

 in the dark background of the spot spectrum. YOUNG observed that these inter- 

 spaces, some of which are quite narrow and look like bright lines, were of the same 

 order of brightness as the xmdimmed spectrum outside the spots, and the observation 

 has been repeatedly confirmed by others. The account given by HALE, ADAMS, and 

 GALE,'" from observations made in the 3rd and 4th orders of a Littrow grating 

 spectroscope of 18 feet focal length, may be usefully quoted : " Although an immense 

 number of fine lines can be seen in the spot spectrum, they nevertheless seem to lie 

 on a continuous dark background, which we have not been able to resolve into lines. 

 This background, however, is interrupted at certain points by lines or breaks, which 

 seem to be nearly as bright as the spectrum of the adjacent photosphere. They do 

 not appear to us like genuine bright lines, and we are unable to offer an adequate 

 explanation of them, unless the dark background is resolvable." 



As no wave-lengths were given by the Mount Wilson observers, it is not quite 

 certain that the bright interruptions which they observed are identical with those 

 seen in the Kodaikanal photograph of the spot spectrum shown in Plate 13, where the 

 resolving power is not so great. The latter, however, represents very closely the 

 appearance and positions of the bright spaces as observed with the Evershed solar 

 spectroscope employed in my own observations of spots, which gives a purity of about 

 25,000 near b, and a dispersion of 60 from A to H. The great majority of these, in 

 the region about b, are coincident with clear spaces in the absorption of magnesium 

 hydride (see p. 454), except, of course, that the latter are sometimes sub-divided, 



* ' Astrophys. Jour.,' vol. 24, p. 185 (1906). 



