Photoelectric radial velocities, paper XIV. Variation of the radial velocity of epsilon Cygni.

McMillan et al. have shown from a large number of very precise measurements that the radial velocity of ɛ Cyg has changed by about 0.3 km s^-1 over the past five years. We now discuss, on the basis of observations which, although fewer and far less precise, span nearly a century and include continuous coverage by the photoelectric method over the last 25 years, the behaviour of the velocity over the longer term. The star is clearly a spectroscopic binary; there is an unfortunate gap in the radial-velocity data at a critical time, but it seems likely that the period is about 55 yr. It has been possible to derive preliminary orbital elements, which include a high value (0.9) of the eccentricity. The next periastron passage is expected to occur in about 2016; it may be possible to verify the occurrence of the last one if suitable spectrograms taken in the early 1 960s can be located. The companion star is likely to be a main-sequence object of type G, K or early M, and is presently expected to be something like 1 arcsec from the primary; it should be no fainter than twelfth magnitude and may be substantially brighter, so it may be directly detectable.