DR. A. SCHUSTER ON THE SPECTRA OF METALLOIDS. 
49 
in the spectrum of the gas in different cross sections of the spark seems to me to 
be that the lines of flow spread out more midway between the poles than they do 
at the poles, so that the temperature is not so high some distance away from the 
poles. 
6. The Elementary Line Spectrum. 
This is the spectrum which is seen when a strong spark passes through oxygen 
at atmospheric pressure. It can be seen at all pressures in vacuum tubes, when a 
Leyden jar and air break are introduced into the circuit. Plucker gives a drawing 
of the lines in his paper on “ Double Spectra” (Phil. Trans., 1865) ; he does not give, 
however, his measurements. Huggins gives a measurement of the oxygen lines 
as seen at atmospheric pressure; but these are referred to an arbitrary scale, and 
some of the weaker lines are invisible at atmospheric pressure owing to the continuous 
background. Thalen gives the wave-lengths of the air lines. 
When the spark is taken in air, however, the oxygen lines are weak compared to 
the nitrogen lines, so that even some strong oxygen lines do not appear in Thalen’s 
list. In their last paper on the spectra of metalloids, Messrs. Angstrom and Thalen 
gave a drawing of the lines of oxygen. 
This drawing is pretty complete, but the numbers which have been obtained by 
measurement are not given separately, so that we are dependent on the accuracy 
of the engraver, which cannot be relied on to any great extent. As far as can be 
judged from the map, Thalen’s measurements agree fairly well with mine. Salet 
also has given some measurements, but the small dispersion which he used did not 
enable him to obtain great accuracy. There are two weak lines on Angstrom and 
Thalen’s map (A.= 5163 and X—4964) which I have never been able to see, 
although I have been constantly on the look-out for them. On the other hand I 
see a number of weak lines not included in that map. Most of them are given by 
Salet. Some of these lines have a different appearance from the oxygen lines; 
they are not so well defined, and are sometimes bands. This may lead some 
observers to think that they are due to impurities. Greatly struck with the 
difference in appearance of these lines, and the ordinary oxygen lines, I have paid 
some attention to them, but I find they appear with electrodes of very different 
metals, and cannot be due to the glass, as I have seen them in the receiver, six 
inches in diameter, which I have mentioned. I have finally come to the conclusion 
that we have no evidence for rejecting them. I have tried to represent in the 
spectrum in Plate 1, marked A, the relative intensities of the oxygen lines. The 
intensities were carefully estimated, but cannot, of course, be greatly relied on 
except as far as the relative intensity of a group is concerned. It seems hardly 
possible to compare the intensity of a yellow and a blue line. The following table 
contains my measurements. I have not carried them beyond the group of lines 
which is near the solar line G. Photography will do the work much better in the 
extreme violet than I could have done. 
MDCCCLXXIX. 
H 
