r 
rays increased very greatly by its metallic reflection the ex- 
tinguishing ones. I could not see any difference between the 
spectrum of the light that had come through this film and that — 
ore the face was silvered, but the reflected light was incom- 
parably more brilliant. The complete obliteration of the entire — 
spectrum presented now no difficulty. 4 
Nothing need be said about collateral contrivances, which 
would suggest themselves to any one: A strip of wood a 
metre long, and bearing divisions served to keep the extinguish- — 
ing lamp in the proper direction as regards the ground glass, _ 
and indicated its distance. I may add, however, that satisfac- _ 
tory observations can be made very conveniently by keeping 
the extinguishing flame at a constant distance, and varying its _ 
intensity by opening or closing its stop-cock. This avoids the — 
trouble arising from moving the flame. In one instrument 1 
caused an index attached to the head of the stop-cock to move 
over a graduated scale, and so ascertained how much it was 
opened. This, though permitting of pleasant working, had not 
the exactness of the method of distances. q 
32 J. W. Draper—New form of Spectrometer. 
gray light I had seen when a strip of platinum is ignited by 
ee pat 1 
C rough one of them, by a suitable arrangemen 
of a heliostat, slit, direct-vision prism, and convex lens, a sola 
‘spectrum was formed on the ground per: Through the se 
ond aperture, which was about an inch square, covered with 
