﻿of Glass Diffraction Gratings. 587 



etching is good for a given grating. The intensity appears 

 to rise to a maximum, and then remain constant for some 

 time, further etching having but little effect ; if carried on 

 for halt* an hour or so the grating is usually ruined. 



The following scheme has proved very reliable. One 

 corner of the ruled surface is dipped into the acid to a depth 

 of a few millimetres, for perhaps 15 seconds, washed, dried. 

 and examined by holding it at arm's length up to a distant 

 gas-flame. On bringing any given spectrum into the etched 

 corner, it is easy to see at once whether matters have been 

 improved or not. If there is a decided improvement, the 

 whole grating is immersed for 15 seconds and then examined 

 again. Usually the corner first etched will be found still 

 further improved. This process is to be repeated until the 

 rest of the grating catches up to the corner in intensity — in 

 other words, as long as the corner is brighter than the rest 

 of the grating we can be sure that a further etching will 

 improve things. 



1 believe that all glass gratings can be vastly improved 

 in this way. By testing the grating with a spectroscope 

 after each treatment, the experimenter can satisfy himself 

 that he is not injuring its optical qualities. Personally I 

 have never made a test of this kind, as the gratings did not 

 need to be very perfect for the work for which they were 

 intended. We found no indication, however, that they had 

 lost in flatness of surface, as they copied as well after the 

 etching as before, and it is well known that a grating cannot 

 be uniformly copied by contact printing unless both the 

 grating and plate are plane. The greater part of the gratings 

 experimented with had rulings varying from 3000 to 5000 

 lines to the inch, but it was found that 14,000 line gratings 

 could be greatly intensified in the same way. 



A few experiments were made on speculum gratings with 

 various acids, without much hope of success. These were 

 complete failures. Probably the crystalline structure would 

 etch too unevenly, even if a non-tarnishing etching fluid 

 could be fouud. I shall make a few trials with electrolysis, 

 but am very doubtful of the results. It is much easier. 

 however, to get brilliant gratings on speculum metal, and in 

 many cases it is doubtful if etching would improve them, as 

 they show very dark central images. 



The effect of the etching is to widen the groove, as is 

 clearly shown by microscopic examination of sections cut 

 before and after etching. 



Inasmuch as the diffraction process of colour photographv 

 was first published in this Magazine, a few words about the 



