The age–metallicity relation of the solar neighbourhood is studied using a sample of 552 late-type dwarfs .
This sample was built from the intersection of photometric catalogues with chromospheric activity surveys of the Mount Wilson group .
For these stars , metallicities were estimated from uvby data , and ages were calculated from their chromospheric emission levels using a new metallicity-dependent chromospheric activity–age relation developed by Rocha-Pinto & Maciel ( [ 1998 ] ) .
A careful estimate of the errors in the chromospheric age is made .
The errors in the chromospheric indices are shown to include partially the effects of the stellar magnetic cycles , although a detailed treatment of this error is still beyond our knowledge .
It is shown that the results are not affected by the presence of unresolved binaries in the sample .
We derive an age–metallicity relation which confirms the mean trend found by previous workers .
The mean metallicity shows a slow , steady increase with time , amounting at least 0.56 dex in 15 Gyr .
The initial metallicity of the disk is around -0.70 dex , in agreement with the G dwarf metallicity distribution ( Rocha-Pinto & Maciel [ 1996 ] ) .
According to our data , the intrinsic cosmic dispersion in metal abundances is around 0.13 dex , a factor of two smaller than that found by Edvardsson et al .
( [ 1993 ] ) .
We show that chromospheric ages are compatible with isochrone ages , within the expected errors , so that the difference in the scatter can not be caused by the accuracy of our ages and metallicities .
This reinforces some suggestions that the Edvarsson et al. ’ s sample is not suitable to the determination of the age–metallicity relation .