320 



Intelligence and Miscellaneous Articles. 



angles to the path of the incident light, and the lines are on the 

 second face of the plate. 



Observing with the astronomical telescope, two superposed spectra 

 are seen in the field, arising from the two halves of the incident 

 pencil. As long as the deviation is small the intensities are vir- 

 tually equal ; but from 30° the difference becomes appreciable, and 

 goes on increasing in a regular manner ; the feeblest spectrum arises 

 from the pencil polarized parallel to the lines of the grating. 

 The same test, repeated with the spectra diffracted by reflexion, 

 taking care that this time the lines should be on the first face, gave 

 the same result. 



From Mr. Stokes's explanation, assuming that refraction takes 

 place before diffraction, the ratio of the amplitudes of the two dif- 

 fracted pencils should be equal to the cosine of the deflection. To 

 determine this ratio, the two pencils are to be received on a Nicol, 

 arranged so that the two images may become equal. The following 

 are the results of a series of measurements made with the Druramond 

 light, the grating being placed as in the first experiment: — 



Deviation. 



Ratio of the amplitudes. 



Calculated. 



Observed. 



o / 



7 45 

 9 05 

 16 25 

 16 45 

 23 45 

 33 33 

 33 50 

 42 27 

 46 37 



49 28 



50 52 



63 07 



64 02 

 69 55 



099 

 0-99 

 0-96 

 0-96 

 0-91 

 0-83 

 0-81 

 074 

 0-69 

 065 

 0-63 

 045 

 044 

 0-34 



101 



0-90 

 0-84 

 0-81 

 0-82 

 0-81 

 0-85 

 0-79 

 0-69 

 075 

 0-69 

 0-49 

 0-52 

 0-48 



The agreement is not very close ; but, allowing for the difficulties 

 attendant on measuring the intensities, the feebleness of the light 

 used, and the enfeeblement of the light diffracted at a great angular 

 distance, these differences are seen to be due to errors of observa- 

 tion. In any case the more rapid enfeeblement of the pencil polarized 

 parallel to the lines appears discordant with M. Holtzmann's views, 

 and confirmatory of the ideas of Mr. Stokes. 



The explanation of diffraction by gratings is complicated in theory 

 by the longitudinal vibrations, whose action is not well understood, 

 and in practice by phenomena of polarization by interference, which 

 are produced by striated lamina, as has been shown by M. Fizeau. — 

 Comptes Rendus, December 10, 1866. 



i i>p 



MX 



R. 20 



