318 M. H. Becquerel's Experimental Investigations 



I preferred to utilize all the bobbins I possessed, and to as- 

 certain the correction of the glasses by special experiment, sub- 

 mitting them to a known magnetic action of considerable power, 

 and then reducing the results observed to what they would be 

 under the conditions of the measurements with the gases. 



The presence of the glasses in the tube in the course of the 

 luminous rays gives rise to another inconvenience which com- 

 plicates the regulation of the system. Each surface acts as a 

 mirror, and a large number of reflected images appear in the 

 field of vision. However, they cannot be confounded with the 

 images reflected in the mirrors, because of their faint lustre ; 

 and they are nearly obliterated when the analyzer is turned so 

 as to obtain equality of tint. 



Nevertheless it sometimes happens that these accidental 

 images encroach on the images to be studied ; and as the parts 

 which are superposed have not the same plane of polarization, 

 the optical measurements become very difficult, and sometimes 

 impossible. It is then necessary, by manipulation, either to 

 displace the mirrors a little, or to wedge the tube in a slightly 

 different position, so that the accidental images do not interfere 

 with the measurements to be effected ; usually the accidental 

 images are situated a little above or a little below the images 

 reflected by the mirrors. 



When the apparatus was regulated, it was left as long as 

 possible in the same position. At the end of each day of ex- 

 periments the mirrors, which would be affected by the air, were 

 simply removed, and then replaced in exactly the same position 

 when a fresh series of observations was begun. 



The gases were introduced into the tube without altering its 

 position, by simple displacement under the influence of a slight 

 pressure. The greatest care was taken to avoid any pressure 

 which would either distort the glasses or cause them to exhibit 

 the phenomena of chromatic polarization. 



Almost the whole series of experiments below recorded 

 were made without displacing either the tube or the glasses 

 from their position, which remained the same for several 

 months. 



The difficulties of construction presented by the very delicate 

 regulation of all the instruments, as well as by the workman- 

 ship of the surfaces of the glasses and mirrors, have been sur- 

 mounted with rare skill by M. T. Duboscq. who made the 

 apparatus. 



On the Employment of Piles of Amplifying -glasses. — In order 

 to augment still further the magnetic rotations observed, the 

 polarized rays might be received on one or more piles of 

 glasses placed in front of the analyzer. 



