542 Mr. E.J. Rendtorffem 



The apparatus is so adjusted that the two spectra are con- 

 tiguous with their Fraunhofer lines in exact coincidence, and 

 no parallax between the interference-bands and the scale. 



The wedges were now adjusted until the coincidence, at 

 equal periods on either side of the Na line, was the best 

 obtainable. The cross-hair was then set midway between the 

 Dj and D 2 lines, where coincident interference-bands were 

 always formed, and the orders of the two crystals determined 

 as previously described. 



The wedges and auxiliary plates were cut from crystals of 

 Iceland spar, quartz, and of selenite and mica parallel to the 

 principal cleavage-planes, and were of sufficient thickness to 

 give orders of several hundreds or a thousand. On account 

 of the great absorption of mica, high orders could not be 

 used. This, however, was unnecessary in the comparison 

 with other crystals, since its differential dispersion differs 

 more from the crystals examined than any other, and rela- 

 tively low orders are sufficient for determining the orders of 

 achromatism. The best coincidences were obtained for the : — 

 40" order for Na light of mica when combined with 35" 



order of selenite ; 

 45 7 order for Na light of mica when combined with the 



40" order of quartz ; 

 222" order for Na light of quartz when combined with the 



212'' order of Iceland spar; 

 210" order for Na light of selenite when combined with 



the 198" order of Iceland spar ; 

 606" order for Na light of selenite when combined with the 



594" order of quartz. 

 From the orders obtained for mica when combined with 

 quartz and with selenite respectively, it follows that the 64" 

 order of quartz, for Na light, produces the best achromatism 

 when combined with the 63" order of selenite. These relative 

 orders give a fair coincidence of bands in the red, but leave 

 the green uncompensated. The orders 101 selenite to 99 

 quartz give a far better compensation throughout the visible 

 spectrum, especially between the yellow and blue, but achro- 

 matize poorly in the red. With this combination of crystals, 

 as also with selenite and Iceland spar, it is possible to achro- 

 matize for certain periods only. Mica and selenite, as well 

 as Iceland spar and quartz, give a good coincidence of bands 

 throughout the entire spectrum. 



In order to study the disposition of the interference-bands 

 with respect to the wave-length for various orders of selenite 

 and quartz, the refracting-prisni Gr, fig. 1, was replaced by a 

 plane grating. By means of an achromatic lens K, fig. 2 y 



