We have used the recently reduced Hipparcos data and the Washington Double Star catalogue to derive two volume- and magnitude-limited samples of single stars in the Galactic plane. The star count method has been used to compare synthetic to empirical samples of stars within both the main sequence and the evolved regions of the Hertzsprung–Russell diagram. Corrections have been made for unrecognized binaries and the residual incompleteness of the Hipparcos catalogue. Two methods of binary reduction have been compared, a flat 71 per cent binary fraction (including multiples) and a...

We have used the recently reduced Hipparcos data and the Washington Double Star catalogue to derive two volume- and magnitude-limited samples of single stars in the Galactic plane. The star count method has been used to compare synthetic to empirical samples of stars within both the main sequence and the evolved regions of the Hertzsprung–Russell diagram. Corrections have been made for unrecognized binaries and the residual incompleteness of the Hipparcos catalogue. Two methods of binary reduction have been compared, a flat 71 per cent binary fraction (including multiples) and a mass-dependent binary fraction, which decreases with primary mass from 71 to 57 per cent.

A semi-empirical relation between the vertical scaleheight and stellar age has been derived from kinematic data provided by the OSACA data base. This relation is used to prescribe the loss of stellar content vertically from the plane due to heating.

When employing the fixed binary ratio, we find that the best fit is provided by a Scalo initial mass function (IMF) with an exponent of 1.85 ± 0.15, together with an average Galactic thin-disc star formation rate (SFR) of 618(±15) stars Myr⁻¹ kpc⁻³ with 1211(±30) M_⊙ Myr⁻¹ kpc⁻³ for single stars with M_* > 0.9 M_⊙. The application of the mass-dependent variable binary ratio yields more single stars at lower mass and hence a steeper (Γ= 2.2, …, 2.3) IMF and an increased SFR.