K-dwarfs have life-times older than the present age of the Galactic disc, and are thus ideal stars to investigate the disc's chemical evolution. We have developed several photometric metallicity indicators for K dwarfs, based an a sample of accurate spectroscopic metallicities for 34 disc and halo G and K dwarfs. The photometric metallicities lead us to develop a metallicity index for K dwarfs based only on their position in the colour absolute-magnitude diagram. Metallicities have been determined for 431 single K dwarfs drawn from the Hipparcos catalog, selecting the stars by absolute magnitude and removing multiple systems. The sample is essentially a complete reckoning of the metal content in nearby K dwarfs. We use stellar isochrones to mark the stars by mass, and select a subset of 220 of the stars which is complete in a narrow mass interval. We fit the data with a model of the chemical evolution of the Solar cylinder. We find that only a modest cosmic scatter is required to fit our age metallicity relation. The model assumes two main infall episodes for the formation of the halo-thick disc and thin disc respectively. The new data confirms that the solar neighbourhood formed on a long timescale of order 7 Gyr.

We present a comparison of the predictions of stellar models with the luminosity of the lower main sequence ($5.5 < M_V < 7.3$) using K dwarfs in the Hipparcos catalog. The parallaxes of our comparison stars are known to better than 15% and metallicities have been recently determined from photometry. A major advantage of our comparison is that distances in our sample are known with good accuracy, while tests that involve open and globular clusters are constrained by potentially inaccurate distances. We show that the luminosity of the lower main sequence relative to a fiducial (solar metallicity) isochrone is a simple function of metallicity: $\Delta M_V = 0.84375 \times {\rm [ Fe/H]} - 0.04577$. We compare the data with a range of isochrones from the literature. None of the models fit all the data, although some models do clearly better than others. In particular, metal rich isochrones seem to be difficult to construct. The relationship between luminosity, colour and metallicity for K dwarfs is found to be very tight. We are thus able to derive metallicities for K dwarfs based on their position in the Hipparcos colour-magnitude diagram with accuracies better than 0.1 dex. The metallicity-luminosity relation for K dwarfs leads to a new distance indicator with a wide range of possible applications.

The so-called ``G-dwarf problem'' emerged in the 1960's when it was noticed that the observed metal distribution of nearby stars differs from what is expected in simple models of the evolution of the Galaxy. We have determined metallicities for a sample of over 100 K dwarfs drawn from the Hipparcos catalog, using R-I colours to estimate effective temperatures and Geneva photometry to estimate [Fe/H]. We show there is a K-dwarf problem. K-dwarfs have life-times older than the Galactic disc, so these are an ideal star with which to investigate the disc's chemical evolution. Further photometric observations of K dwarfs are under way.