IN QUARTZ OF LIGHT IN THE VISIBLE REGION OF THE SPECTRUM. 
277 
be read with greater ease than even the least refrangible of the bright blue zinc and 
cadmium lines. 
(8) Copper. 5782-30, 5700-39, 5218-45, 5153-33, 5105-75. 
Unlike the alloys used to produce the spectra of cadmium and zinc, the copper arc 
can be used with heavy currents up to about 10 amperes, although the highest current 
used in the present experiments did not exceed 5 or 6 amperes. The spectrum is 
much more complex than those of the preceding metals and some care is needed to 
ensure spectral purity. Values are now given for the two yellow and three green 
lines but occasional readings have been made with four of the blue lines (4705, 4651, 
4587-19, 4378-40), and by using a heavy current tliere is little doubt that some at 
least of these could be added to the list of 24 wave-lengths tabulated below. 
(8) (a) Copper Yellow. Cu 5782-30, Cu 5700-39. 
The yellow copper lines are of good spectral purity and are easy to read, specially so 
in the case of the less refrangible line. They were read with the extra dense 30° prisms 
on the spectroscope (without a grating), a Rutherford prism at the eye-piece and a 
half-shadow angle of only 4°. 
(8) (b) Copper Green. Cu 5218-45, Cu 5153-33, Cu 5105-75. 
The two more refrangible lines were read without difficulty under the same conditions 
as the yellow lines. In spite of the presence of two satellites, 5158-53 and 5144-35, 
the central line proved to be the best of the three copper greens ; it may be noted 
that these satellites are of weak intensity and lie one on either side, so that even if 
they were not completely eliminated by the dispersive system they would probably 
have very little influence on the optical mass-centre of the line. The least 
refrangible line has a strong satellite at 5220-25 ; when using the system of highest 
dispersion it could be seen distinctly, although separated by only 1-80 Angstrom units; 
but as it could not be eliminated by the methods then available, a direct-vision prism 
was substituted for the Ptutherford prism and the readings were taken with line and 
satellite superposed ; this is the only case in the whole series in which a line with a 
visible satellite was used as if it were a single line. 
5. Quartz. 
With the exception of sodium and thallium, complete series of readings of all the 
lines referred to in the preceding section had been taken as long ago as the spring of 
1908. The delay of four years in the completion of the work may be traced almost 
entirely to the extreme difficulty of securing quartz of sufficiently high quality to 
withstand the more and more stringent tests that were found to be necessary in the 
course of the work. Much time and many thousands of readings might have been 
saved if these new tests had been discovered before the work of measuring the 
rotatoiy dispersion of the specimens was commenced. 
