Chromospheric variations in main-sequence stars

There are several observable indices of the more or less periodic variations in solar activity known generally as the solar cycle. It has long been known that one of these indices is the total flux in the chromospheric emission components of the H and K lines of Ca n. In searching for analogous variations in stars, this same quantity appears to be the most easily observable one, provided suitable instrumentation is available. About ten years ago a coude scanner was completed for the 100-in. telescope on Mt. Wilson. This is a two channel machine, with two photomultipliers and two sets of pulse counting electronics. It is arranged so that one channel receives the signal from a slit whose width can be chosen, and across which the spectrum can be moved in the direction of dispersion. The other channel receives the combined signals from two 25 A windows on either side of the region through which the spectrum can be scanned. In the present work the scan mode was not used. Taking into account the total radial velocity of a star at a given time, the second slit can be set at the centers of the H and K lines with a precision of two or three hundredths of an angstrom. Both the entrance and exit slits were set for one Angstrom width at the focal plane. An observation consists of four integrations, two in K and two in H, with the counts in both channels started by pushing a button and terminated simultaneously by the machine at a pre-set value of the count in the one Angstrom channel. For most stars under normal conditions the latter is chosen to be 10. The counts at the line centers are added and divided by the corresponding counts in the monitor channels, and these numbers are then divided by the same ratios obtained from a standard lamp observed each day. In this way the effect of drifts and changes in the electronics are eliminated. Standard stars are also observed each night. These are stars with minimal H-K fluxes which probably have little or no chromospheric emission and their constancy bears out this assumption. All the stars in the program are main sequence objects. They extend from above the Sun, at spectral types about F5, down to MO. Each star is observed on every night when it is available, and every effort is made to obtain a dozen or more observations per season. This is essential since many of the stars appear to have fluctuations on a variety of time scales and as many observations as possible are necessary to give a good representation of the longer term variations. Figure 1 shows the results on one of the standard stars and the scatter here, of the order of ±2%, is probably due almost entirely to the unavoidable errors of observation. This star is typical of the eighteen standards around the sky, all of which have fluxes in the range 1300 to 1500 instrumental units.