90 Loud Rayleigh on the Manufacture and 



On a casual inspection the second, apaft from the greater num- 

 ber of its lines, would be preferred as presenting a more even 

 appearance. The 3000-line grating is divided into three parts, 

 giving spectra of differing degrees of brightness, corresponding 

 no doubt to a variation in the cut of the diamond or other stone 

 employed, a peculiarity which is faithfully preserved in the 

 copies. But on actual trial it is found that the spectra of the 

 3000-line grating are much the best in respect of definition ; 

 and the same difference is observed in the copies. The supe- 

 rior brilliancy of the closer-ruled grating is thus of little or no 

 advantage for the investigation of the solar spectrum. In order 

 to make good use of it, a higher degree of magnifying-power 

 would be necessary than the definition of the spectra will bear. 



The other original grating was engraved by Mr. Rutherford, 

 of New York, and was kindly lent me by Mr. Browning; it 

 contains 6000 lines to the inch. Owing to a change of resi- 

 dence, I have not hitherto had an opportunity of testing either 

 the original or the copies on the solar spectrum ; but I may ob- 

 serve that in respect of brightness they fall far short of Nobert's. 

 This, as I have already remarked, is not always an objection ; 

 and the accuracy of division, on which definition depends, is said 

 to be very superior*. 



In testing gratings I prefer to work in a dark room. The 

 slit is fastened in the window-shutter, outside which is placed 

 the heliostat or porte-lumiere. As slits are frequently required 

 in optical experiments, and as usually made are rather expen- 

 sive, I may be allowed to mention a very simple method by 

 which serviceable slits may often be obtained. A piece of glass is 

 covered with tinfoil, which must be made to adhere well ; I have 

 found a weak shellac-varnish a suitable cement. The alcohol is 

 allowed to evaporate, and the thin layer of shellac softened by 

 heat. In order to make a slit, it is only necessary to lay a 

 staight edge on the tinfoil and draw a line with a sharp knife, 

 afterwards wiping the line of the cut with a rag moistened with 

 alcohol. The width and regularity of the slit may be judged of 

 by holding it close to the eye, and observing the appearance 

 presented by a distant candle. The narrower the aperture the 

 more dilated (in the direction of the width of the slit) the image 

 will appear, Broader slits may be made by removing the foil 

 between two parallel cuts. 



At a distance of 12 feet or more from the shutter are placed 



the grating and the object-glass of the observing-telescope. In 



making the preliminary adjustments, it is convenient to use a 



slit so wide that the spectra and the light reflected from the 



* Draper, " On Diffraction- Spectrum Photography," Phil. Mag. Dec. 

 18j:-i, p. 41.9. 



