New BV(RI)(C) observations of 77 stars in the Hyades are reported and discussed. The new observations are used to test published magnitudes and color indices for that cluster. For values of (V-R)(C) and (R-I)(C) published previously by Taylor & Joner, the tests reveal no detectable scale-factor problems. In addition, the tests show that possible zero-point corrections to the published data can be no larger than a few millimagnitudes. These test results indicate that future studies requiring precision photometry for Hyades stars would be well served by selecting data samples from sources as close as possible to the native Cousins system. Tests of B-V photometry published by Johnson & Knuckles reveal a zero-point ambiguity of approximately 8 mmag in the new data that will require further measurements to resolve.

In this paper, the first statistically rigorous test is performed to determine whether there are dwarfs in the solar neighbourhood with [Fe/H] > +0.2 dex. The possibility that no such stars exist is rejected with a confidence limit exceeding 99.994 per cent. If the threshold value is raised to +0.38 dex, the corresponding hypothesis is rejected with a confidence limit of 98.5 per cent. It is tentatively found that this upper limit does not depend on the presence or absence of planets orbiting metal-rich stars. In a result that is unchanged from a previous paper, the upper limit for the known metallicities of giants is found to be about 0.2 dex lower than the limit for dwarfs. Stars that might be observed to improve the upper metallicity limit for dwarfs are listed.

In this paper, an updated catalog containing averaged values of [Fe/H] is presented for FGK stars on and near the main sequence. The input data for the catalog are values of [Fe/H] derived from weak and moderately strong lines and published before 2005 July 1. Those data are corrected to a uniform temperature scale, and a statistical analysis is then applied to a subset of the data that did not contribute to a previous version of the catalog. In this way, it is found that an accurate zero point for the catalog can be established with an rms error of 0.005 dex. After corrections are applied to a number of the newly added data, it is shown that those corrections help to produce satisfactory zero-point coherence among the catalog entries. Standard errors that are derived for the catalog data are shown to be accurate. It is also shown that those standard errors are based on pervasive scatter in the input data, as is expected if those data are affected by genuine random effects. Samples of the metallicity catalog and an accompanying temperature catalog are displayed and discussed. To make possible an effective extension of the catalog to stars without catalog entries, a database containing photometric metallicities derived by Nordstrom et al. is considered. Standard errors for those metallicities are derived, and zero-point corrections required to put those data on the catalog zero point are presented.

Using Hyades photometry published by Mendoza and other authors, Pinsonneault et al. have recently concluded that Cousins V - I photometry published by Taylor & Joner is not on the Cousins system. Extensive tests of the Taylor- Joner photometry and other pertinent results are therefore performed in this paper. It is found that in part, the Pinsonneault et al. conclusion rests on ( 1) a systematic error in Mendoza's ( R - I)(J) photometry and ( 2) a small error in an approximate Johnson- to- Cousins transformation published by Bessell. For the Taylor- Joner values of ( V - R)(C), it is found that there are possible ( though not definite) differences of several mmag with other results. However, the Taylor- Joner values of ( R - I)(C) data are supported at the 1 mmag level. Using the ( R - I)(C) data and other published results, an ( R - I)(C) catalog is assembled for 146 Hyades stars with spectral types earlier than about K5. For single stars with multiple contributing data, the rms errors of the catalog entries are less than 4.4 mmag. Temperatures on the Di Benedetto angular- diameter scale are also given in the catalog and are used to help update published analyses of high- dispersion values of [ Fe/ H] for the Hyades. The best current mean Hyades value of [ Fe/ H] is found to be + 0.103 +/- 0.008 dex and is essentially unchanged from its previous value. In addition to these numerical results, recommendations are made about improving attitudes and practices that are pertinent to issues like those raised by Pinsonneault et al.

For dwarfs near the Sun, published metallicity distributions commonly have peak metallicities that are a few tenths of a dex below the solar value. However, Haywood has recently used data from photometric calibrations to obtain a peak metallicity of about -0.05 dex. Haywood argues that uncorrected sampling biases explain the difference between his result and previous results. To check Haywood's peak metallicity, a statistical analysis is applied to Haywood's sample and also to a set of averaged high-dispersion metallicities for nearby dwarfs. In addition, a magnitude-limited sample of evolved stars is considered. Care is taken to make sure that all the data in these samples have a common zero-point which there is good reason to regard as reliable. In addition, sampling biases are duly considered, and full allowance is made for the contribution of the thick-disc population. Only statistical analysis is used; no conclusions are drawn by inspecting data or histograms without the aid of statistical analysis. The analysis yields a best-fitting Gaussian for dwarfs whose peak metallicity is fully consistent with Haywood's results. The 2 width of that Gaussian is 0.37 dex, and its peak falls at a metallicity of -0.041 +/- 0.013 dex. Although that peak differs detectably from zero, the difference is not appreciable when compared with a number of previous results. For evolved stars, a problem appears: their mean metallicity turns out to be significantly lower than that for dwarfs (by 0.054 +/- 0.016 dex). In addition, the metallicity distribution for evolved stars is found to be narrower than its counterpart for dwarfs. It is suggested that these discrepancies can be traced to the present lack of knowledge about giants with [Fe/H] > +0.2 dex. This suggestion is supported (although not proven) by analysing an augmented test sample which includes data for a number of high-metallicity dwarfs. It is suggested that this problem should be examined further when more has been learned about the numbers of high-metallicity giants in the solar neighbourhood.

This paper is one of a pair in which temperatures and metallicity catalogs for class IV-V stars are considered. The temperature catalog described here is derived from a calibration based on stellar angular diameters. If published calibrations of this kind are compared by using color-index transformations, temperature-dependent differences among the calibrations are commonly found. However, such differences are minimized if attention is restricted to calibrations based on Johnson V - K. A calibration of this sort from Di Benedetto (1998) is therefore tested and adopted. That calibration is then applied to spectroscopic and photometric data, with the latter predominating. Cousins R - I photometry receives special attention because of its high precision and low metallicity sensitivity. Testing of temperatures derived from the calibration suggests that their accuracy and precision are satisfactory, though further testing will be warranted as new results appear. These temperatures appear in the catalog as values of theta = 5040/T(effective). Most of these entries are accompanied by measured or derived values of Cousins R - I. Entries are given for 951 stars.