326 Messrs. Hutchins and Holden on the 



by a right-angled prism. Before striking the prism it passes 

 through a cylindrical lens, which condenses it to a band of 

 light about 2 inches long and J- inch wide. The jaws of the 

 slit move equally in opposite directions, so that, however 

 widely they may be opened, no lateral displacement of lines 

 can result from this cause. 



Directly in front of the slit is placed a large tin lantern 

 containing an electric lamp ; the image of the arc can be 

 brought exactly upon the slit by means of an adjustable lens 

 in the front of the lantern. Tn the lower carbon of the lamp 

 is made a cup-shaped cavity, which is filled with the substance 

 a spectrum of which is desired. It is not at all necessary that 

 this be in the form of a metal, for any ordinary compound is 

 at once reduced by the intense heat and the presence of carbon 

 vapour to the metallic state. 



The plan of working has been as follows. The apparatus 

 being arranged as described, the sunlight is admitted and the 

 desired portion of solar spectrum photographed upon the 

 upper half of the plate ; then the sunlight is excluded by a 

 shutter, and the image of the electric arc containing the 

 proper metal is allowed to fall upon the slit, and its spectrum 

 photographed on the lower half of the plate. (Most of the 

 plates used were those made by the M. A. Seed Co., and were 

 cut to the size of 8 inches by 2. The most sensitive plates 

 were obtained, and even then we found the required time of 

 exposure for some parts of the spectrum inconveniently long.) 



In order to effect the exposure of either half of the plate at 

 will, we placed directly in front of the camera an opaque 

 screen, in which was a rectangular opening one half the size 

 of the plate. By turning a handle, this screen is raised or 

 lowered without the slightest disturbance of camera or plate. 

 The metallic spectrum, being thus photographed immediately 

 below the solar spectrum, can be compared with it at leisure. 



These spectra are then examined with the aid of a glass 

 magnifying about ten diameters, and any coincidences be- 

 tween solar and metallic lines carefully noted according to 

 their wave-lengths. In order to eliminate any personal error, 

 they are examined by both observers separately, and their 

 results afterwards compared. 



To eliminate errors arising from suspected impurities of 

 materials, as also from the impurities known to exist in the 

 carbons employed, we took what we called " comparison 

 photographs. " For these, we placed in the carbon cup a 

 portion of the substances known or suspected to be present 

 as impurities in our metal, and then photographed the spec- 

 trum thus given on the upper half of the plate : a piece of 



