DR. A. SCHUSTER ON THE SPECTRA OF METALLOIDS. 
41 
The homogeneousness of the phenomena seen in vacuum tubes is greatly increased, if we 
consider the one line spectrum of oxygen as the representative of the band spectrum of 
nitrogen, and I shall therefore consider it as such. I have called the spectrum the 
compound line spectrum of oxygen, because, according to the theory of molecular com¬ 
bination, it would be due to a more complicated molecular structure than the other 
(the elementary) line spectrum. I must, however, add that I do not at all feel certain 
that the two line spectra are necessarily to be explained the same way as the line and 
band spectra. I have found it necessary, for the sake of clearness, to explain the 
results of the investigation in the language of a definite theory. It will be easy for 
any one to separate the two. The fact that nearly all lines which make their appear¬ 
ance at low temperatures only are lines which widen easily seems to be suggestive. 
Spectra of the negative pole. —-It has long been known that in many gases the glow 
surrounding the negative electrode has a different colour from the remainder of the 
discharge. This colour appears generally more blue to the eye than that of the posi¬ 
tive discharge, and this fact was the cause that a certain blueness was often associated 
with the idea of negative electricity. When examined with the spectroscope, it is 
found that the negative glow has a spectrum of its own, different, of course, in different 
gases. Here, also, I think, it will be found simpler to assume that a definite molecular 
combination is formed at the negative pole than to suppose that exterior forces pecu¬ 
liar to the pole modify the period of vibration. The following experiments seem to me 
to support strongly, if not to prove, this view. When the pressure is very small, the 
spectrum of the negative pole is seen throughout the tube.* In vacuum tubes of the 
form usually employed for purposes of spectrum analysis, it is easy to obtain a state of 
exhaustion in which they show the spectrum of the negative pole throughout that half 
of the tube which contains the negative electrode (which I shall call the negative half)) 
and also in the capillary part, but not in that half of the tube which contains the 
positive electrode (which I shall call the positive half). The experiments are supposed 
to be made at that stage. 
Experiment 1.—Allow the spark to pass some time in one direction ; the spec¬ 
trum of the negative pole is seen in the negative half and in the capillary part only. 
Now reverse the current suddenly, and the spectrum will now be seen at first through¬ 
out the tube, but will gradually disappear in the positive half, so that finally the same 
permanent state as before is established, and the spectrum is only seen in the negative 
half and capillary part. 
Experiment 2.—In order to show that it is the reversal, and not the interruption 
of the current which produces the appearance of the spectrum in the positive half, 
break the current and make contact the same way : no change can be observed. 
Experiment 3.—But if, after the spark has passed some time, you break the con¬ 
tact and wait a little time, say one minute, so as to allow any compounds' which might 
* Goldstein, ‘Berliner Monatsberielite,’ May, 1876. 
MDCCCLXX1X. 
a 
