SPECTRA OF SOME OF THE BRIGHTER STARS. 
703 
The complete discussion of this region is reserved until better and more numerous 
photographs have been obtained. 
Detailed Discussion of the Spectrum of a. Orionis. 
As a Orionis is the brightest star given in the Table under discussion, a special 
investigation has been made of its spectrum. Several photographs have been taken 
at Kensington, and one at Westgate, with the 3-prism slit spectroscope. The latter 
was exposed for one hour on November 30, 1891, and, as the slit was very narrow, 
the spectrum is well-defined. All the lines shown in the Westgate photograph have 
been reproduced in a later photograph taken at Kensington with the instrument B. f 
The intensity of the violet is somewhat feeble, and the reduction has been limited 
to the region F to X 4029. On reference to the photograph (Plate 28), it will be seen 
that the part of the spectrum extending from the violet to X 4860 (F) consists almost 
entirely of lines, over 100 being distinctly visible between F and G. There are 
distinct groupings of lines, however, which correspond with the flutings seen in 
a Herculis and other stars of the same type. 
For the reduction of the photograph a spectrum of daylight photographed with the 
same instrumental conditions has been employed. 
It is easy to recognise well-marked lines which serve for the construction of the wave¬ 
length scale by means of the usual projection curve. 
The hydrogen lines at F, G, and h, are almost as strong as in the spectrum of the 
Sun. 
The investigation of the origins of the various lines includes also an inquiry into 
the probable temperature of the absorbing regions in the star. Thus, when v'e come 
to investigate the lines due to iron, we find that they do not correspond in relative 
intensities with those which appear in the arc spectrum. This will be clear on reference 
to the map, which shows the strongest lines in the arc spectrum, and all the lines of 
the flame spectrum of iron. Particular attention may be drawn to the four lines in 
the arc spectrum at wave-lengths 420P6, 4203‘5, 4206'3, and 4209'8. It will be seen 
that there is a complete inversion in intensities if we take the arc spectrum of iron, 
and many other cases are equally evident. 
When compared with the flame spectrum of iron, which has been specially 
photographed for this investigation, a smaller number of lines is explained, but there 
are no inversions of importance. The feeble line at 4132 in the flame spectrum is 
comparatively strong in the spectrum of the star, but this is in all probability due to 
the superposition of the line of lithium which has the same wave-length. The only 
important divergence is in the case of the line 4143, which is much stronger in the 
star than in the flame spectrum. It is, however, a strong line in the arc spectrum of 
iron. 
These comparisons lead to the conclusion that the temperature of the most 
