392 Astronomical arid Nautical Collections, 



progress of the two pencils, we shall find that it always 

 preserves equal intensities in the two images, as we turn the 

 rhomboid, and that its light acts as if it had been completely- 

 depolarised. Passing now to the point which answers to the 

 difference of half an undulation between the two systems of 

 waves, we shall find it perfectly dark in the ordinary image, 

 and at its maximum of brightness in the extraordinary, when 

 the principal section of the rhomboid is horizontal, and when 

 it is vertical wholly dark in the extraordinary image, and at 

 its maximum of brightness in the other ; so that all the light 

 collected in this point is polarised vertically. Continuing to 

 go through the different points of interference of the two 

 luminous pencils, we find, in general, that their union pro- 

 duces a light completely polarised in a horizontal plane ; 

 that is to say, in the primitive plane of polarisation, when 

 the difference of their paths is either nothing, or an even 

 number of semiundulations : that the whole light is polarised 

 ver1:ically, or in the azimuth 2 z, when the difference of the 

 paths is an uneven number of undulations; and that the 

 whole light, on the contrary, is entirely depolarised when 

 this difference is a whole uneven number of quarters of an 

 undulation; and that in all intermediate cases there is only 

 a partial polarisation. The perfect polarisation is only ob- 

 servable in the fringes of the first three orders, but it is 

 clear that if the middle points of the dark and bright stripes 

 of the other orders appear only partially polarised, this cir- 

 cumstance depends on the want of homogeneity of the light, 

 which cannot be more simplified without weakening it too 

 much. 



Mr. Arago has invented a valuable method of consider- 

 ably augmenting the intensity of the light in the experi- 

 ments on diffraction, which may be applied to experiments 

 of this kind. It consists in substituting for the lens em- 

 ployed, a portion of a cylindrical surface, which forms a 

 linear focus instead of a point, and when this line is turned 

 into the direction of the fringes, they become extremely dis- 

 tinct, the situation being easily found by turning the cylin- 

 drical lens while the fringes are viewed with a common mag- 

 nifier. 



In order to study with convenience the kind of polarisation 



