MR. A. FOWLER ON THE SPECTRUM OF MAGNESIUM HYDRIDE. 453 



grating of greater resolving power afterwards became available and the results were 

 discarded. 



The concave grating was ruled by ROWLAND, and has a radius of curvature of 

 10 feet, and a ruled space 3|- inches by If inches. It was mounted temporarily as a 

 Littrow spectrograph by Mr. EAGLE, and the photographs were taken on plates 

 12 inches by 2^ inches. In the first order, the dispersion is about 5j tenth-metres to 

 the millimetre, but in this method of using the grating the spectrum is not a normal 

 one. Excellent photographs of the first-order spectrum were obtained with exposures 

 ranging From 5 to 90 minutes, and the definition was very satisfactory. Photographs 

 have also been taken in the second order, showing the principal lines of the green and 

 blue flutings, but giving no impression of the fainter fluting in the yellow green. 



A general record of the ultra-violet spectrum, extending to 2300, has been obtained 

 by the use of a Hilger quartz spectrograph having one prism of 60 degrees. 



General Description of the Spectrum. 



When observed with small or moderate dispersion (Plate 12A, No. 1), the spectrum 

 of magnesium hydride, as remarked by LIVEING and DEWAR, is somewhat similar in 

 general appearance to the carbon (" Swan," or candle flame) spectrum, and the 

 respective flutings occupy nearly the same parts of the spectrum. In each case the 

 brightest fluting is in the green, and there are fainter ones in the yellow-green and 

 blue, the first heads of the carbon flutings being at 5635'4, 5165'3, and 4737'2, while 

 those of magnesium hydride fall at 562T6, 521 1'l, and 4844'9. Further, there is a 

 fluting of simple structure at 4311 in the hydrocarbon spectrum, and a somewhat 

 similar fluting of magnesium hydride at 4371 '8. 



On more minute examination, however, the resemblance is less close than might at 

 first appear. Even under moderate dispersion the blue flutings in the two spectra 

 are clearly of different structure, and they have practically nothing in common when 

 the dispersion is great. The green and yellow-green flutings in the two spectra 

 resemble each other in showing three or four heads, but whereas in the case of carbon 

 the secondary heads persist under high dispersion, those of magnesium hydride are 

 soon lost in the general mass of lines as the dispersion is increased. 



From the extreme visible red to the beginning of the yellow-green fluting at 5621 

 the spectrum consists of a great number of faint lines, showing no heads sufficiently 

 pronounced for measurement as such, though the lines are by no means equally spaced. 

 (The carbon arc shows a very similar appearance in this region.) The yellow-green 

 fluting begins with a series of rather faint lines, but stronger lines appear in the more 

 refrangible part, and the components can be traced to the head of the next group. 

 The green fluting beginning at 5211 also has its strongest lines a considerable distance 

 from the first head, and becomes very faint as it approaches the blue band. Still 

 greater complexity is shown by the blue group of flutiugs, which has its greatest 



