COMMISSION 27 OF THE I. A. U.
            INFORMATION BULLETIN ON VARIABLE STARS
                        Number 2534

                                           Konkoly Observatory
                                           Budapest
                                           13 June 1984

                                           HU ISSN 0374-0676


  HIGH DISPERSION OBSERVATIONS OF epsilon Aur FROM SEPT. 1982 TO MARCH 1983

  Between September 1982 and March 1983 I obtained high dispersion spectra
of epsilon Aur in selected regions using the Palomar 200-inch coude spectrograph
and the Dominion Astrophysical Observatory 48-coude spectrograph. All spectra
were taken with the 90-mm ITT image tube and IIa-D emulsion. Since further
observations during this eclipse are not currently planned, the raw data will
be presented here for others who may find them useful.
  In Table I we list our radial velocities for the sodium D lines, KI lines,

                              Table I
                         Radial Velocity (km/s)

Date (UT)        Disp.    metals   H alpha   H alpha  NaI        KI       OI
                 (A/mm).           (abs)      (em)                  (lambda7772)
 3.53 Sept.1982  6.7    +3.5+-1.1  +11.3+-1.0 +72    +12.8    +21.8+-1.5* +14.3
15.41 Oct. 1982  4.8                              (+19.1+-1.4)%
                                                  (+ 6.7+-0.8)
18.60 Oct. 1982  4.8                                          +22.4+-0.5<
19.17 Mar. 1983  4.8    +1.5       +25.4>            +15.5#
21.14 Mar. 1983  4.8                                 +13.9    + 9.9+-1.5**

* partially resolved component at +8+-2 km/s is present
% two nearly equal components are clearly present but poorly resolved
< partially resolved component at +4 km/s is present
> the deepest point in the line is at +36.6 km/s and an uncertain absorption
  feature may be present at -4.6 km/s . Weak emission further to the violet
  may be present
# the deepest point of the D2 line is at +5.3 km/s
** a little fuzz is visible on the positive side of the KI lines

H alpha, and a few other features. While most of the sodium and potassium features
are surely circumstellar there is certainly an interstellar component present.
In the direction of epsilon Aur interstellar gas is seen mostly with radial
velocities between +5 and +10 km/s . The sodium D lines and potassium lines
show incipient resolution which may involve the interstellar components.
  The equivalent widths of the sodium and potassium lines are given in
Table II. For potassium, blending with atmospheric O2 makes the
line at lambda 7664

                   Table II
Equivalent Widths (in A)
     Date         D1     D2     KI(lambda 7699)
 3.53/9/1982    0.77    0.82          0.36
15.41/10/1982   0.95    1.03          -
18.60/10/1982   -       -             0.40
21.14/3/1983    1.24    1.33          0.50

unmeasurable on the dates of these observations, The increase in equivalent
widths already noted by Pathasarathy and Lambert is evident.
  Blending and partial resolution of the circumstellar and interstellar
features may account for much of the velocity structure which is only partially
discernible at the available resolution.
  This research was conducted by the author as a guest investigator at the
Palomar Observatory, Calif. Inst. of Technology and the Dominion Astrophysical
Observatory, Herzberg Institute for Astrophysics, National Research Council,
Canada.


                GEORGE WALLERSTEIN
               University of Washington
               Seattle, Washington, 98195
               U.S.A.