and Differential Double Refraction. 359 



Iceland spar with selenite gave nearly as dark a band as 

 with quartz at approximately the same orders. 



Mica and quartz gave the blackest line for the orders nine 

 and eight respectively, very little coloration being present for 

 these orders. 



Mica and selenite also gave the best compensation for the 

 orders eight and seven respectively, less coloration being 

 observable than with the previous pair. These results all 

 agree with those obtained by the previous method. 



It should be noted that when the comparisons are made 

 between wedges beginning with the zero order, say, and in- 

 creasing, that the black band changes into a bright colourless 

 band after a gain of half an order of one of the wedges over 

 the other, and then changes into a black band after a gain of 

 one order, and so on. By counting the number of bands 

 gained it is evident that we can determine the ratios for 

 achromatism for the entire spectrum or for a part, by counting 

 these passages of either the dark or bright bands, the wedges 

 having been previously graduated for the bands of sodium 

 light, say, and the position of either of these bands with 

 respect to the latter determining the ratio. For example, 

 with wedges of selenite and of mica the bright bands occurred 

 at the fourth order in the mica, and the three and one half 

 order in the selenite wedge ; and at the eighth and seventh 

 orders respectively, for the first black band gained. Similar 

 results were noted for the other specimens examined. 



It is evident from the above that in determining the order 

 of a plane plate by means of a wedge, as is usually done, that 

 the order will depend upon the comparison wedge used. For 

 example, using white light, a mica plate tested with a mica 

 wedge gave 9"2 as the order. The same plate tested with a 

 quartz wedge gave 8'3 as the order, and with a selenite wedge 

 8*4. A selenite plate gave 24' 6 as the order when tested with 

 a selenite wedge, and 28 with a mica wedge. The same 

 results would also follow with measurements made with the 

 ordinary Babinet's compensator. 



Of the crystals examined above combinations of mica and 

 selenite, and of mica and quartz, are the most available for 

 retardation plates, such as achromatic quarter-wave plates. 

 Both of these combinations were tried. In the construction 

 of a mica-selenite 1/4A. plate, for example, a film of mica was 

 split off of as near the second order for sodium as possible. A 

 piece of selenite was then cemented over this in subtractive 

 position and then polished and tested until the resultant 

 retardation was reduced to exactly \/4 for sodium light. On 



