The usage of early measurements to investigate long-term trends in the solar diameter commenced with the pioneering work of Eddy and Boornazian (1979). These authors deduced from the Greenwich meridian circle observations between 1836 and 1953 that the solar diameter had shown a secular decrease at a mean rate of . However, this result was disputed by Parkinson et al. (1980), who criticised Eddy and Boornazian's interpretation of the Greenwich data. In particular, Parkinson et al. demonstrated that different observers obtained discordant results with the same instrument. They also showed that analysis of series of timings of both Mercury transits and total solar eclipses since 1715 revealed no evidence of a secular change in the solar diameter. Soon afterwards, Gilliland (1981), by combining meridian circle measurements since 1836 with Mercury transit data since 1715, suggested that a secular decrease in the solar diameter by was "likely". Débarbat (1982) deduced that determinations of the horizontal diameter of the Sun at noon by Picard between 1666 and 1673 revealed irregularities of periodical nature. Subsequently Ribes et al. (1987) made an analysis of 50 years of measurements by Picard and La Hire between 1666 and 1718 which were in the form of annual means. These results indicate that around 1700 the solar diameter was some 4" greater than it is now. However, in reply Morrison et al (1988) demonstrated that careful observations of the edges of the shadow at the total solar eclipse of 1715 implied that at that date the solar diameter had essentially the present value. In order to try to resolve these difficulties, I have analysed extensive series of measurements of the solar diameter covering a long time scale - from 1660 to the present day - with the object of establishing a reliable database for the diameter of the Sun (Toulmonde 1995). The present paper is a summary of this work.
© European Southern Observatory (ESO) 1997
Online publication: April 28, 1998