go-v.orgdoi

The spectrum and characteristics of the long-period eclipsing binary 32 Cygni.

Edwin B. Weston · The Astronomical Journal · 1953

  • DOI 10.1086/106936
  • Publisher Institute of Physics
  • Type article
  • Issued 1953

Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.

The Astronomical Journal · Institute of Physics · article · 1953

Abstract

The spectrum of 32 Cygni at recent eclipses has been studied on plates taken at the David Dunlap and the University of Michigan Observatories. Reference has also been made to Wright's published description of the Dominion Astrophysical Observatory (Victoria) spectra,1 to McLaughlin's description of the Mount Wilson Observatory spectra,2 and to Wellmann's description of the Hamburg-Bergedorf Observatory spectra. Atmospheric eclipse intensities have been estimated for all plates and radial velocities measured for some of them.

For the most recent eclipses the general appearance of the spectrum was studied on all of the plates available and compared with spectrograms at corresponding phases of the systems of ~ Aurigae and 31 Cygni. From this it was estimated that mid-totality occurred 1952 December 27.3 UT. Totality lasted about 13 days and the partial phases about four days each. The atmospheric eclipse intensity estimates for the eclipses since 1940 require a period between eclipses in excess of 1147 days, and 1148.0 days has been adopted as the most probable period.

However, the description of the Harvard spectra of 1890 and 1896 requires a period about 1141 days,1 which is impossible to reconcile with the observations at all eclipses since 1940. Any attempt to interpret the disagreement as arising from a change of period or of the velocity of the system is not supported by the published Lick1, ~ and Victoria1 radial velocities covering a total interval of 45 years, which themselves favor a period of 1147 or 1148 days. The atmospheric eclipse intensities are, on the whole, closely correlated with the changes in the equivalent widths of the chromospheric lines.

The atmospheric eclipse intensities show, and the spectrophotometric analysis confirms, that irregularities occur in the development of the atmospheric effects at different eclipses, between ingress and egress, and even from day to day. These irregularities differ for the various chromospheric lines and are probably produced by the light of the B star passing through large gaseous clouds or streams in the atmosphere of the K star. These clouds and streams may be analogous to the giant prominences observed on our sun. The behavior of the hydrogen lines in the atmospheric phase in 32 Cygni differs conspicuously from ~ Aurigae and 31 Cygni.

The cores of He and higher members of the Balmer series show some enhancement in the atmospheric phase; shows little change; and H~ becomes slightly weaker relative to neighboring lines. The interpretation has been considered by Wellmann.5 Comparison of the composite spectrum out of eclipse and the nearly pure spectrum of the K star, the primary, during totality has been made with stars of standard spectral type and luminosity class on the Morgan-Keenan-Kellman scheme of classification. The primary is of spectral class K3 or K~ and of luminosity class Ib-tI, or perhaps brighter.

The strong lines of the K star are visible far into the ultraviolet even though the filling in of the strongest lines of the K star by the spectrum of the B star can be detected near H~. Lines of hydrogen and Hei are present in the B star, but Cail is very doubtful, indicating a spectral class of B~ or earlier. The K line has variable emission structure during totality. The spectrum of the K star is also suspected of other slight variability. Near the time of the last eclipse, the lines of the K star that should not have been displaced by chromospheric effects did, in fact, show small residuals from the radial velocity curve.

Small residuals of opposite sign were exhibited by lines showing maximum chromospheric effects. It is not possible to interpret these displacements in terms of direct rotation of the K star. It is probable that they represent some form of currents in the atmosphere of the K star. The orbital elements determined for the primary by Wright1 from the Victoria radial velocities are found to be adequate for e, K1, and V0 when combined with the new period of 1148.0 days, but w and T need small corrections. The Lick radial velocities are also in very good agreement except for a slight difference in V0.

The Ottawa radial velocities,6 although numerous, show too much scatter to be of value in the orbit determination, but they `suggest the possibility of secondary oscillations in the velocity curve; however, this is not supported by the Lick and Victoria observations. Irregular, short-term deviations from the velocity curve are probably present, but the observations are not of sufficient frequency and quality to permit study of these deviations in detail. The radii of the stars are probably larger, and the inclination of the system less, than the values published by Wright.

Wellmann has considered the numerical values of these.5 The eclipse is nearly grazing; thus, if the K star is even slightly variable in size, then the durations of the total and partial phases must be variable, and at some conjunctions totality may even fail to occur. The geometry of the system is such that the star constitutes an interesting and unusual case where no secondary eclipse is possible. I.Pub. Dom. Astroph. Obs. 9, 189, 1952. 2.Ap. J. III, 449, 1950. 3.Harvard Announcement Card No. 1046, 1949. 4.Pub. Lick Obs. ~6, 298, 1928. 5.A. J. ~8, 233, 1952. 6.J. B. Cannon, Pub. Dom. Obs. 4, 151,6]1918. DavidDunlap Observatory, Richmond Hill, Ont.

Open the resolved locator · 10.1086/106936

Metadata licence: OpenAlex metadata (CC0)

Registry
OpenAlex DOI index + API
Fetched
2026-11-20 00:46 UTC
Cache state
not cached yet (bulk-index+openalex)
Locator
https://doi.org/10.1086/106936
Metadata licence
OpenAlex metadata (CC0)
Scope
Metadata as returned by the resolver; abstracts are never generated here.