Preston—-Radiation Phenomena in a Strong Magnetic Field. 389 
cases the two inside lines are absent altogether, so that we are presented with 
what appears to be a doublet instead of a triplet (the light being still viewed 
across the lines of force). 
This, at first sight, extraordinary modification, or variation, of the expected 
triplet may be accounted for in more than one way. Thus the central line of the 
triplet might be almost wholly, or partially, obliterated by absorption in the outer 
layers of the colder vapour surrounding the source of light. That is, the basis of 
this modification of the triplet may be what is known as reversal. If the central 
line is completely absorbed, we are left with a doublet; and if the middle part 
only of the central line is absorbed, we are left with two strong outside lines 
enclosing two weak inside lines—that is, a quartet with a weak middle, such 
as the line 4800 appears. Further, since the absorbing vapour is itself in a strong 
magnetic field, it is possible that the outside lines of the triplets may also be 
reversed; that is, each line may be replaced by two bright lines, separated by a 
dark space. The appearance then presented, when each line of the triplet is 
reversed, should be, not three, but six bright lines separated by dark spaces. If, 
however, the central line of the triplet should be completely absorbed, while the 
outside lines are reversed, we are presented with two pairs of bright lines instead 
of a triplet. 
Now, in carefully examining this phenomenon, I have repeatedly observed all 
the modifications just mentioned, that is to say, while some lines are resolved into 
sharp distinct triplets, having the middle line the brightest, others in the same 
field are resolved into what appear to be quartets and sextets. It is interesting to 
notice that the two D lines of sodium, and the blue line 4800 of cadmium do not 
belong to the class which show as triplets. In fact, the blue cadmium line belongs 
to the weak-middled quartet class, while one of the D lines (D,) shows as a sextet 
of fine bright lines, 7.e. four sharp and equally intense lines enclosed by two some- 
what less sharp on the outside. On the other hand, the other D line (D,) shows 
as a quartet, not of the weak middled class, such as 4800 of cadmium, but of the 
doublet type, that is, of the type which would result from the complete absorption 
of the central line combined with the sharp reversal of each of the side lines. 
Thus the modifications introduced by absorption in the vapour of the outside 
layers of the source of light sufficiently explain the alteration in the number of the 
component lines which make up the image of the modified line as viewed in the 
eye-piece, or recorded on the photographic plate. The appearance of the image 
however, is not that which are ordinarily associates with reversal, for, in a reversed 
line, there is generally a grading of the light towards the outside and a sudden 
fall of intensity, or sharp edge, on the borders of the inside dark line or reversal. 
This characteristic appearance is absent in all the cases cited above, and for this 
reason I have spent a considerable amount of time in investigating whether the 
