PART I: EMISSION LINES IN EARLY TYPE SPECTRA - GENERAL SURVEY
The present state of the study of emission line stars
of early spectral types is briefly reviewed, mainly from the
observational point of view.
PART II: THE BE- VARIABLE SPECTRUM OF O PERSEI
Radial velocities of central hydrogen absorption lines
were measured on 23 quartz -prism spectrograms, covering the
star's period of variation. The new observations agree well
with the period given by Dustheimer, but it is probable that
the decrease in the velocity range is real.
The secondary variation is discussed. It is likely that
the extent of this variation increases for lines of shorter
wave -length. The observations are compared with Schiefer's
The variation of the number of Balmer lines with sharp
central cores is examined. It is shown that this variation
is not likely to occur in the same atmosphere, and therefore
the binary hypothesis seems necessary.
PART III: LINE INTENSITITES IN SOME BE- STARS
Line intensities in seven plates of `. Per., covering
about 30 days of the star' s period were measured. A clear
correlation is found between emission line intensity and the
colour of the star as given by Roach. The widths of emission
lines in some Be -stars are compared with the value computed
from Curtiss' formula. The possibility of measuring the
hydrogen-decrement in Be -stars is discussed and the previous
Equivalent widths of helium and hydrogen lines in K Cas. and B Ceph. are also given. H a is definitely variable in K Cas., while HB is apparently constant. The most probable
turbulent velocity is obtained for the two stars.
PART IV: TOTAL INTENSITIES OF A TRIPLET OF SiIII IN STELLAR SPECTRA
Total intensities of the three Si III lines, 4552, 4567 and 4574 were measured on 81 plates of 32 -early type stars. Deviations from the square roots of the multiplet intensities are small, but nevertheless real. Previous measures are included in a discussion of the differences between high and low luminosity stars and between wide and narrow line stars.
The formation of the lines, mainly through damping, is examined. A definite discrepancy is observed between observation and theory. This discrepancy may be removed by assuming
a negative turbulence velocity in the atmosphere, an incipient emission filling the lines, or a damping constant more than 440 times the classical value. This last assumption may be
nearest the truth.