270 M. R. Ruhlmann on the Alteration -produced by Heat 



the application of the phenomena of interference in thick plates. 

 A plate with parallel faces is cut into two parts. One of these, P 1? 

 is fastened in a stand and receives a pencil of parallel rays. A 

 ray of light from A (fig. 1, Plate V.) is partly reflected upon the 

 front face of plate P 2 , and partly on the back one, and is thus 

 divided into two pencils, which are delayed in regard the one to 

 the other by the passage in the glass plate, and the loss of phase 

 caused by the reflexion at the back surface. At some distance, 

 and in the path of the two parallel rays> the second half, P 2 , of 

 the glass plate is set up in a position parallel to the first one. 

 Then the original ray (always apart from manifold internal re- 

 flexions) is divided into four, two of which appear perfectly coin- 

 cident both in phase and direction. One of these two is the one 

 which is reflected at the front surface of the first glass and at 

 the back surface of the second glass; the other is reflected at the 

 back surface of the first and at the front surface of the second 

 glass (fig. 1). In the intermediate space the two rays are sepa- 

 rated by a distance which depends upon the thickness of the 

 plates and their inclination to the rays, which can be increased 

 at pleasure. If P 2 and P 2 are the two parallel plates, we can 

 imagine T and T, to be introduced media, and we can measure 

 the displacement of the bands from B. If, for instance, follow- 

 ing Arago's proposition and Jamiii's example, we introduce tubes 

 of equal length containing liquids of different temperature, we 

 can follow the alterations of the indices of refraction with great 

 facility. In this method certain difficulties are to be noticed : 

 these consist, first, in the necessity for placing tubes of different 

 temperature close to one another ; and secondly, that these tubes 

 must remain of equal length, or that the alterations in their 

 length must be taken into account. This latter is very inconve- 

 nient, while the required condition can scarcely be secured by 

 mechanical means. If, with Jamiu, homogeneous instead of 

 white light be employed, then, as Stokes* has pointed out, the 

 results got from the displacements of the interference-fringes are 

 somewhat too high, because, in consequence of the greater 

 breadth of the band for rays of little refrangibility, the middle 

 appears to be somewhat more displaced than it is in reality. 



By this method Jamin obtained his above-mentioned determi- 

 nations of the alteration with the temperature of the refractive 

 index of water, and proved that the refractive index varied pro- 

 portionally to external pressure f. 



A method has often been brought forward for determining 

 the index of refraction in liquids which is founded upon the 

 measurement of the magnitude of the lateral parallel displace- 



* Memoir es de Tlnstitut, 1856, p. 453.* 



t Ann. de Chim. et de Phys. vol. lii. p. 163. 



