1. Introduction The open cluster NGC 7092 (M39, C2130+482; l=92.5 deg, b=-2.3 deg) which belongs to the group of so-called poorly-populated open clusters has been previously studied by this author (Platais, 1984b). A new set of astrometric and photometric data have proven the suspected anomaly of the luminosity function - genuine deficiency of intrinsically faint cluster members. Besides the new data for the cluster members, more than 7,500 field stars in the area ~2 x ~2 deg also have homogeneous astrometric and photometric data which were used to investigate the interstellar absorption as a function of distance (Platais, 1988). These data might be useful in galactic structure studies and also could provide a quick reference for individual stars and help to make an easy star selection. It must be noted that the Catalogue is restored from the only existing computer-typed version deposited at the VINITI institute in Moscow (Platais, 1983) since the tape version was lost. 2. Data description Although a detailed description of the data reduction can be found in a somewhat obscure publication (Platais, 1984a) a brief summary of basic reductions, attained accuracies and other essential information is provided here. The most important difference between the previously published parts of the catalogue and the present Catalogue is a slightly changed treatment of the magnitude equation in the proper motions. These differences are small and do not change conclusions regarding the NGC 7092 cluster membership. 2.1 Proper motions The relative proper motions have been determined from measurements with the Zeiss-ASCORECORD machine. This measuring machine provides only manual-driven human-eye centering and, taking into account all precautions, the measuring standard error for well-exposed images is about 2 microns. Surprisingly, it was possible to measure many very faint and/or heavily overlapping images which may have failed using a PDS microdensitometer. In total 7931 stars were measured on 11 Pulkovo normal astrograph plates (scale 59.57"/mm) in a 110'-diameter circle around the cluster. The Catalogue of proper motions is complete down to B=16.0m, while the limiting magnitude is B=16.7m. The final proper motion for each star is calculated from a maximum of four plate pairs. The mean standard error of the Catalogue proper motions derived from plate pairs 1 to 4 in the central portions of the field for stars with B<14.5m is +/-1.5 milliarcseconds per year (mas/yr). However, for a given star the mean standard error is a function of individual plate pair mean error, the distance from the field center and the star's magnitude. The unit weight errors are as follows: +/-2.8 mas/yr (pairs 1 to 4, at B=14.0m) and +/-8.0 mas/yr (pairs 5 to 9, at B=15.8m). The plate pairs 1 to 4 all have weights equal to 1.0. The pairs 1 to 9 have the following weights with the pair number indicated in parentheses: 1.0 (5), 2.8 (6), 6.2 (7), 1.7 (8) and 3.0 (9). An estimate of the formal mean standard error of proper motions is given in the Catalogue for each individual star. In the previous study (Platais, 1984a) the magnitude equation was examined for each plate pair separately using the field-star secular parallaxes. After that a small residual magnitude equation correction was applied to the Catalogue proper motions using provisionally identified cluster members. However, it was noticed that the linear corrections derived from the field stars were systematically too large at faint magnitudes, (B>14m). Over this approximately 2-magnitude interval, the data were insufficient to confidently derive a magnitude term and thus it was decided to introduce no corrections for stars fainter than B=14.0m. Though, the final proper motions are corrected using all 81 identified probable cluster members (Platais, 1984b). After these revisions, the mean proper motion of NGC 7092 is MUX=-5.2 and MUY=-15.4 mas/yr. In general, the extreme differences between the present proper-motion Catalogue and its earlier version do not exceed 1 mas/yr and do not affect the conclusions about the NGC 7092 membership. 2.2 UBV photometry The photographic UBV magnitudes were derived from plates taken with the Schmidt telescope (80/120/240 cm) at Baldone (Latvia). The photoelectric sequences in NGC 7092 and NGC 7067 were used for the plate calibration. For the stars measured on plate pairs 1 to 4 the magnitudes were derived from four Schmidt plates in each passband. For the fainter stars (pairs 5 to 9) three plates in each passband provided the magnitudes. The standard error of the UBV magnitudes is +/-0.05m for isolated, non-overlapping star images. As pointed out by Turner et al. (1994) some small systematic differences (about 0.07m or less) exist between the recent photoelectric and photographic photometry which definitely can be attributed to the Schmidt telescope photometric field error. On the Schmidt plates some stars (Nos. 3368, 3740, 4036, 4292, 4716, 6140, 7013) are not visible in the U-passband. Only for the sake of convenience a false U-B colour has been assigned for these seven stars. In addition, a number of stars are known or suspected variables, namely: Nos. 1083, 1357, 1588, 1821, 2513, 3368, 3456, 4624, 4638, 4892, 5382 and 7186. No attempt has been made to get mean magnitudes or colours for these stars. 2.3 Spectral classification Spectral types have been assigned for 511 stars based on three spectral plates taken with the Abastumani meniscus telescope (70/98/210 cm) in Georgia and 8-degree objective prism giving 166-angstroms/mm dispersion at H-gamma. The spectral types have been assigned to virtually all stars brighter than B=12.0m. B- to K-type stars can be classified as faint as B=13.3m but M-stars are identified down to B=15.0m. The luminosity classes are reliable for stars brighter than B=11.4m. A previous publication (Platais, 1988) contains a more detailed description of the classification procedure, peculiar stars, interstellar absorption and also has a finding chart for all spectroscopically classified stars. 2.4 Cluster membership The area covered by the Catalogue contains two open clusters: NGC 7092 and Anon (C2128+488). In order to recognize the probable cluster members among the overwhelming majority of field stars all membership probabilities are given, even those which have not been included in the earlier paper (Platais, 1984b). Recently, an oddity has been noticed by the author in the NGC 7092 membership. A small number of supposed cluster members are grouped together at MUX=-9.0 and MUY=-20.0 mas/yr, slightly offset than the cluster. The offset, for the same stars, is consistent in several of the proper motion studies. However, no effort has been made to include an interpretation of this phenomenon or to re-calculate the membership probabilities in the current version of the Catalogue. The membership calculation technique is presented in Platais (1984b). Since the probabilities were calculated in three different magnitude intervals there exists a discontinuity between the probabilities in each of the intervals. Finally, a comment ought to be made regarding the study by Glushkova (1991) which reports finding 17 relatively bright, previously unknown cluster members. None of these stars shows up as a cluster member in the present Catalogue. Taking into consideration some irregularities and the lower proper-motion accuracy in the Glushkova study, it is suspected that these stars are most likely field stars. The very loose open cluster Anon C2128+488 has been scrutinized twice, by Platais (1986) and Turner et al. (1994). A rigorous calculation of the cluster membership is not possible due to the scarcity of members. All highly probable members of the Anon C2128+488 selected by Turner et al. (1994) are marked in the Catalogue with fake probabilities equal to "-99". 2.5 Positions Almost all X,Y-positions in the Catalogue refer to the deepest plate B187 taken on August 25th, 1909. However, a small fraction of stars was not measured on that particular first-epoch plate. In such cases, the coordinates have been transformed into the system of plate B187 and, consequently, should be used with some caution. For the sake of convenience, J2000 equatorial coordinates are calculated for all Catalogue stars using 70 reference stars from the PPM catalogue. In contrast to the X,Y-coordinates, the epoch of the equatorial coordinates is also 2000.0. Only linear plate constants were used in the transformation to celestial coordinates. The r.m.s. scatter in the PPM star- position residuals is +/-0".4 in both coordinates. 3. Other proper-motion catalogues Since NGC 7092 has been studied on several occasions, it is worthwhile to intercompare the determined proper motions. The differences in proper motions between a selected study and the present Catalogue can be expressed by simple equations of condition which include linear (or linear and quadratic) X,Y-coordinate-dependent and magnitude terms (for reference, see Platais (1991)). The right-hand side of the equations is the difference between the proper motions from the present Catalogue and the study in question. The derived solutions of the equation system by least squares for various individual studies relative to this Catalogue lead to the following conclusions which are summarized in Table 1. The notation for the coordinate-dependent errors is as follows: "-" - negligible error; "+" - notable error; "++" - significant. The value of estimated magnitude equation is presented in mas/yr per unit B-magnitude. Table 1. Comparison of the proper-motion catalogues ------------------------------------------------------------------------ Source Coord.-dep. error Magnitude equation in X in Y X (mas/yr/magn) Y ------------------------------------------------------------------------ Ebbighausen (1940) + ++ - -0.6 van Schewick (1957) + + - - Lavdovsky (1961) - - 0.5 0.4 Artiukhina, Kalinina (1970) + + -1.9 0.8 McNamara and Sanders (1977) - - - 0.7 Glushkova (1991) - + -0.5 0.5 ------------------------------------------------------------------------ As can be seen from Table 1, almost every catalogue is affected by magnitude equation. The discovery of lower main sequence members in NGC 7092 is the principal reason these magnitude effects can now be detected. All cross-identifications of common stars in different catalogues were found by coordinate coincidence. Multiple "hits" were resolved by examining the magnitude differences as well as by checking the finding chart when available. It must be noted that the finding chart provided by McNamara and Sanders (1977) has a number of mis-labeled stars. In Table 2 is given a list of incorrect or dubious star labelings in the finding chart. Also, in the case of an unresolved double star in the McNamara and Sanders catalogue, a colon (:) is added after the star number. A few stars which have correct labeling but show unusually large proper-motion discrepancies with respect to the Catalogue are marked with "pm?". In total, out of 1710 McNamara and Sanders catalogue stars, 22 cannot be identified in the Catalogue. Obviously, some of them are intrinsically fainter than the limiting magnitude of this Catalogue, although a few others (e.g. Nos. 47, 402, 798, 919, 1198) may not have the reported large proper motions. Table 2. A list of McNamara and Sanders (1977) stars ---------------------------------------------------- 1 187pm? 258 446 641 1113 1510 2 195: 270 448 652 1209 1607 9 202 277 459 712 1259 1720 19 207 291 464 868: 1298 1733 35 210: 320 527 877 1339 36: 213 413pm? 561 880 1361 109 218 421 570 976 1483 138 226 428 589: 1019 1484 155 235 429 614: 1072 1485 184 246: 435 641 1086: 1491 ------------------------------------------------------ Acknowledgments I am grateful to Drs. Zdenka Kadla and Vladimir Frolov from Pulkovo Observatory for invaluable help they provided when I was just entering the thorny realm of open clusters. References [1] Artiukhina, N.M., Kalinina, E.P.: 1970, Trudy Sternberg Astron. Inst., vol. 40, 3. [2] Ebbighausen, E.G.: 1940, Astrophys. J., vol. 92, 434. [3] Glushkova, E.V.: 1991, Soviet Astron., vol 35, 466. [4] Lavdovsky, V.V.: 1961, Trudy Pulkovo Obs., Ser. 2, vol. 73, 5. [5] Maevers, F.-W.: 1940, Astron. Nachr., vol. 270, 201. [6] McNamara, B.J, Sanders, W.L.: 1977, Astron. Astrophys. Suppl., vol. 30, 45. [7] Platais, I.: 1983, "Catalogue of relative proper motions and photographic UBV magnitudes in the field of NGC 7092 (M39)". VINITI manuscript depository in Moscow No. 3126-83, 229 pp. [8] Platais, I.: 1984a, "On the Catalogue of proper motions and UBV magnitudes in the field of NGC 7092 (M39)". VINITI manuscript depository in Moscow No. 1318-84, 75 pp. (in Russian) [9] Platais, I.: 1984b, Soviet Astron. Lett, vol. 10, 84. [10] Platais, I.: 1986, Nauchnye Inform., vol. 61, 89. (in Russian) [11] Platais, I.: 1988, Nauchnye Inform., vol. 65, 119. (in Russian) [12] Platais, I.: 1991, Astron. Astrophys. Suppl., vol. 87, 69. [13] Turner, D.G., Mandushev, G.I., Forbes, D.: 1994, Astron. J. (in press) [14] van Schewick, H.: 1957, Veroeff. Univ.-Sternw. Bonn, No. 47, 1. ================================================================================