No CrossRef data available.
Published online by Cambridge University Press: 04 August 2017
Standard stars should normally be constant in the characteristic for which they have been chosen to be standards. Individually they should be capable of testing the instrument used in their measurement, and collectively (usually together with a particular type of instrument) they must define a system of measurement. A star should not be adopted as a standard until several years' observation have demonstrated its constancy. The number of independent observations in that interval is as important as its length. Many of these points are illustrated by particular reference to standard-velocity stars. Even 100 observations of such a star may fail to reveal a detectable variation that is present. A distinction is drawn between “primary” standards, meeting the above criteria, and “reference” or “comparison” stars which need not be so severely tested since their use is more limited. Standard stars must be calibrated in some fundamental way. Since random errors of observation can introduce systematic errors in calibration, the use of the intermediate step of standard stars may decline with increasing internal precision of observations.