PREFACE In this compilation, every effort has been made to report accurately the observations of a large number of investigators. However, since this will be the first widely distributed edition of the Open Cluster Interstellar Matter Database, for there to be a few misquotations is inevitable. (I would greatly appreciate being informed of discrepancies when they are discovered.) For this reason, but, more importantly because measurements taken out of context are prone to misinterpretation, I recommend using this catalog as a guide to information available about the clusters rather than as a final source of information. The catalog format was designed to facilitate location of original references. My objective was to make the database a useful research tool for optical observers and radio astronomers alike, and thus to help bridge a communication barrier that separates astronomical subdisciplines. The database description that follows was written with that objective in mind. D. Leisawitz March 1988 OPEN CLUSTER INTERSTELLAR MATTER (OCISM) DATABASE David Leisawitz INTRODUCTION Data pertaining to 128 open clusters, and the interstellar matter associated or potentially associated with them, are copiled in this database. Common names, aliases, and galactic coordinates of the catalogued clusters are shown in Table 1. Also tabulated are the names of OB associations of which the clusters are members, and the Sharpless (S) and Westerhout (W) identifications of H II regions ionized by the clusters. Below we describe briefly the criteria used to select clusters for this compilation; known selection effects as a result of which there are differences between the catalogued clusters and the population of all open clusters in the Galaxy; how information for the OCISM database was obtained; the database contents; and the associated processing and analysis software. CLUSTER SAMPLE SELECTION CRITERIA AND SELECTION EFFECTS Clusters in the OCISM database satisfy four criteria: they are relatively well studied (photometrically and, often, spectroscopically); their declinations are >~-20 deg; they are more distant than ~1 kpc but not more distant than ~5 kpc; and they are younger than ~100 Myr. Of the approximately 1200 known open clusters, ~10% satisfy these conditions. The OCISM database was created to document information that characterizes open clusters and their interaction with the interstellar medium and the sample Table 1. Clusters in the Database Number in lII bII Name of Cluster Member of H II Region Database OCL (deg) (deg) Common Alias Association S W ___________________________________________________________________________ 1 19 6.13 -1.36 NGC 6530 M8 SGR OB1 25 29 2 23 6.99 -0.25 NGC 6514 M20; TRIFID SGR OB1 30 - 3 26 7.72 -0.44 NGC 6531 M21 SGR OB1 - - 4 33 11.99 -0.94 MARKAR 38 BYURAKAN 5 SGR OB1 - - 5 34 12.43 -3.22 TRUMP 33 - - 6 35 12.80 -0.80 COLLIN 469 SGR OB1 - - 7 36 12.86 -1.32 NGC 6603 M24 - - 8 40 14.15 -1.01 NGC 6613 M18 - - 9 44 15.09 -0.74 NGC 6618 M17 SER OB1 45 38 10 54 16.99 0.79 NGC 6611 M16 SER OB1 49 37 11 56 18.26 1.69 NGC 6604 SER 0B2 - 35 12 66 21.64 -0.78 NGC 6649 - - 13 67 23.86 -2.92 NGC 6694 M26 - - 14 68 23.95 -0.50 NGC 6664 SCT OB2 - - 15 74 26.28 -0.81 NGC 6683 - - 16 77 27.36 -1.93 BASEL 1 APRIAMASVILI - - 17 82 28.23 -2.23 NGC 6704 - - 18 83 28.29 -0.01 TRUMP 35 - - 19 96 38.55 -1.70 NGC 6755 - - 20 99 39.06 -1.69 NGC 6756 - 50 21 100 42.16 4.70 NGC 6709 - - 22 122 59.16 -0.16 NGC 6820 VUL OB1 86 - 23 124 59.41 -0.15 NGC 6823 VUL OB1 86 55 24 125 60.14 -1.83 NGC 6830 VUL OB1 - - 25 126 60.21 -0.29 ROSLUND 2 VUL OB1 - - 26 134 65.70 1.18 NGC 6834 SNR DA495 - - 27 138 66.96 -1.26 ROSLUND 4 IC 4954/5 - - 28 148 72.64 2.08 NGC 6871 CYG OB3 - - 29 149 72.73 1.74 BYURAKAN 1 CYG OB1 - - 30 150 72.76 1.35 BYURAKAN 2 CYG OB1 - - 31 152 73.29 1.19 NGC 6883 CYG OB1 - - 32 157.1 74.91 3.29 BASEL 6 - - 33 158 75.36 1.31 IC 4996 CYG OB1 - - 34 161 75.71 0.31 BERK 87 - - 35 167 76.66 1.26 BERK 86 CYG OB1 - - Table 1. (continued) Number in lII bII Name of Cluster Member of H II Region Database OCL (deg) (deg) Common Alias Association S W ___________________________________________________________________________ 36 168 76.92 0.60 NGC 6913 M29 CYG OB9 - - 37 177 78.09 2.79 COLLIN 419 SNR GAMCYG - - 38 181 78.66 2.03 NGC 6910 SNR GAMCYG - - 39 197 85.46 -0.47 NGC 6996 NGC 7000 CYG OB7 117 80 40 205 89.93 -2.72 NGC 7062 - - 41 208 91.19 -1.67 NGC 7067 - - 42 210 91.32 2.26 NGC 7031 CYG OB7 - - 43 213 94.39 -5.50 IC 5146 CYG OB7 125 - 44 214 94.41 0.20 NGC 7086 - - 45 218 97.35 0.42 NGC 7128 - - 46 222 99.29 3.73 IC 1396 TRUMP 37 CEP 0B2 131 - 47 224 101.36 -2.20 IC 1442 - - 48 229 102.72 0.78 NGC 7235 - - 49 231 103.10 -1.18 BERK 94 132 - 50 235 103.99 14.27 NGC 7023 VdB 139 - - 51 236 104.02 6.45 NGC 7160 CEP OB2 - - 52 237 104.04 0.86 NGC 7261 CEP OB2 - - 53 - 106.76 5.30 S 140 CEP OB2 140 - 54 244 107.08 -0.90 NGC 7380 CEP OB1 142 - 55 256 110.96 0.05 NGC 7510 IC 1470 CAS OB2 156 - 56 257 111.36 -0.20 MARKAR 50 BYURAKAN 3 CAS OB2 157 - 57 260 112.76 0.46 NGC 7654 M52 - - 58 273 116.21 -0.37 HARVARD 21 - - 59 275 116.43 -0.79 NGC 7788 CAS OB5 - - 60 276 116.59 -1.01 NGC 7790 CAS OB5 - - 61 286 118.25 4.95 BERK 59 CEP OB4 171 1 62 291 119.80 -1.38 NGC 103 CAS OB4 - - 63 294 120.25 -2.54 NGC 129 - - 64 297 120.72 0.36 KING 14 CAS OB4 - - 65 299 120.87 0.49 NGC 146 CAS OB4 - - 66 306 122.09 1.33 KING 16 CAS OB7 - - 67 307 122.29 1.54 BERK 4 - - 68 313 123.13 -6.24 NGC 281 184 - 69 314 123.99 1.10 BERK 62 CAS OB7 - - 70 316 124.68 -0.59 NGC 366 CAS OB1 - - Table 1. (continued) Number in lII bII Name of Cluster Member of H II Region Database OCL (deg) (deg) Common Alias Association S W ___________________________________________________________________________ 71 319 125.90 -2.60 NGC 433 - - 72 320 126.07 -3.91 NGC 436 - - 73 321 126.56 -4.35 NGC 457 - - 74 322 127.19 0.75 NGC 559 - - 75 326 128.02 -1.76 NGC 581 M103 CAS OB8 - - 76 328 128.22 -1.14 TRUMP 1 - - 77 329 128.55 1.70 NGC 637 - - 78 330 129.09 -0.35 NGC 654 CAS OB8 - - 79 332 129.34 -1.51 NGC 659 CAS OB8 - - 80 333 129.46 -0.94 NGC 663 CAS OB8 - - 81 339 130.74 2.65 STOCK 5 - - 82 345 132.39 -6.16 NGC 744 - - 83 349.1 134.21 -2.64 BASEL 10 PER OB1 - - 84 350 134.63 -3.72 NGC 869 h PERSEI PER OB1 - - 85 352 134.74 0.92 IC 1805 CAS OB6 190 3,4 86 353 135.08 -3.60 NGC 884 X PERSEI PER OB1 - - 87 356 135.67 2.31 CZERNIK 13 - - 88 357 135.78 1.48 NGC 1027 CAM OB1 - - 89 359 135.79 -1.58 CZERNIK 8 - - 90 360 135.84 0.27 BERK 65 - - 91 361 136.02 -1.20 KING 4 - - 92 362 136.34 -2.66 NGC 957 - - 93 364 137.19 0.92 IC 1848 CAS OB6 199 5 94 383 143.65 7.62 NGC 1502 CAM OB1 - - 95 394 148.16 -1.29 NGC 1444 CAM OB1 - - 96 403 155.35 2.58 NGC 1624 212 - 97 404 157.08 -3.65 BERK 11 213 - 98 406 158.61 -1.58 NGC 1605 - - 99 409 160.43 -17.74 IC 348 PER OB2 - - 100 429 168.88 -2.00 NGC 1778 AUR OB1 - - 101 439 173.59 -1.70 NGC 1893 IC 410 AUR OB2 236 - 102 441 173.90 0.28 NGC 1931 AUR OB1 237 - 103 445 174.52 1.04 NGC 1960 M36 AUR OB1 - - 104 465 186.45 1.25 IC 2157 - - 105 467 186.61 0.13 NGC 2129 - - Table 1. (continued) Number in lII bII Name of Cluster Member of H II Region Database OCL (deg) (deg) Common Alias Association S W ___________________________________________________________________________ 106 476 190.20 0.42 NGC 2175 GEM OB1 252 - 107 481 195.63 -2.92 NGC 2169 - - 108 495 202.94 2.20 NGC 2264 MON OB1 273 - 109 499 203.60 0.13 NGC 2251 MON OB1 - - 110 502 204.62 13.96 NGC 2395 - - 111 512 206.18 -2.29 NGC 2239 MON OB2 - - 112 515 206.42 -2.02 NGC 2244 ROSETTE MON OB2 275 16 113 518 207.14 -0.91 COLLIN 107 MON OB1 - - 114 523 207.97 -3.38 COLLIN 96 - - 115 527 208.56 -1.80 V D BERGH MON OB1 - - 116 548 215.32 -2.30 NGC 2286 - - 117 - 218.13 -0.39 MONOCEROS 287 - 118 559 221.67 -1.24 NGC 2323 M50 - - 119 562 223.62 -1.27 NGC 2335 - - 120 565 224.32 -1.16 NGC 2343 - - 121 567 224.72 0.38 NGC 2353 CMA OB1 - - 122 575 226.57 -2.30 NGC 2345 - - 123 594 230.82 1.00 HAFFNER 10 - - 124 595 230.80 0.93 CZERNIK 29 - - 125 598 231.41 1.96 NGC 2414 - - 126 626 236.24 0.08 NGC 2421 - - 127 633 238.18 -5.53 NGC 2362 310 - 128 668 243.14 0.41 NGC 2467 PUP OB2 311 - selection criteria were specified accordingly. Because they are relatively young, most OCISM clusters can be expected to be surrounded by the interstellar gas and dust from whence they came, or at least by what remains of that material. The ages of clusters are gauged by comparing photometric and/or spectroscopic data with stellar evolution models and the theoretical H-R diagram. Clusters can be sorted or binned by age to evaluate the progress of the stellar-interstellar interaction. One virtue of this method, which has been employed with success by Gordon, Howard, and Westerhout (1968), Schwartz (1971), Bash, Green, and Peters (1977), and Leisawitz, Thaddeus, and Bash (1988), is that the chronometer is independent of the interaction. It is feasible to observe interstellar matter in a large number of regions, each ~50 pc in size, with several arcminute spatial resolution, and possible to resolve interesting interstellar structure, if the regions are at distances >~1 kpc and <~5 kpc. For comparison, a typical open cluster diameter is ~3 pc (~10' at 1 kpc); a 50 pc molecular cloud is considered to be large; and the radiation energy density due to a typical young cluster is equal to the energy density of the background interstellar radiation field at ~40 pc. Clusters in the OCISM database can be observed from the northern hemisphere. Our decision to include a cluster in the database was based primarily on data tabulated by Alter, Ruprecht, and Vanysek (1970). Although recent analyses suggest that the selection criteria are violated by a few of the clusters so chosen, the offenders were retained. Since open clusters in general are concentrated within ~100 pc of the galactic plane, the sample of all known clusters is intrinsically biased by the obscuring effect of dust in the Milky Way. Thus, for instance, only along relatively transparent lines of sight are clusters more distant than a few kpc observed optically. Such clear lines of sight are rare in the inner Galaxy (the direction of the Carina spiral arm is exceptional in this respect). Naturally, selection effects present in the population of all known clusters propagate to the OCISM sample. Statistical properties (distances, sizes, masses, distribution in the Galaxy, etc.) of the clusters in the OCISM database were compared to the properties of all known clusters (see, e.g., Lynga 1982) by Leisawitz (1985). In most respects, OCISM clusters are representative of the sample of all well- studied open clusters. However, old open clusters (~100 - 1000 Myr) and nearby clusters (d <~1 kpc) are, by fiat, underrepresented in the database; and clusters in quadrants III and IV of galactic longitude are missing because of the declination limit. SOURCES OF INFORMATION A systematic survey of international astronomical literature was used to locate information about the clusters. Journal articles cited in the bibliographies of Alter, Ruprecht, and Vanysek (1970) and Ruprecht, Balazs, and White (1981), as well as articles cited in the Astronomy and Astrophysics Abstracts, were scanned manually. Most of the articles to which we refer were published more recently than the early 1960s and our survey is essentially complete through late 1987. In its present form, the database contains information from approximately 400 reference sources. Some sources are particularly valuable because they contain self-consistently evaluated descriptions of numerous clusters. Information from the surveys listed in Table 2 is heavily represented in the OCISM database. DATABASE CONTENTS Cluster Identification The OCISM database can facilitate research related to the interstellar matter associated with young star clusters, an interdisciplinary endeavor, by linking the various names by which a cluster and the interstellar gas in its environment are referred. Molecular cloud observers, for example, relate their discoveries to stellar associations or to well-known H II regions, but seldom to Table 2. Open Cluster and Related Interstellar Matter Surveys Reference Principal Contributions (a) ________________________________________________________________________________ Johnson et al. (1961) d, E(B-V), (B-V)t, eSpT Johnson and Svolopoulos (1961) RV(*) Becker (1963) d, E(B-V), AV, eSpT Schmidt (1963) M(*) Hoag and Applequist (1965) d, eSpT Courtes, Cruvellier, and RV(H II) Georgelin (1966) Gordon, Howard, and Westerhout (1968) RV(H I), M(H I) Lindoff (1968) cluster age D'Odorico and Felli (1970) M(dust) Hagen (1970) d, AD, E(B-V), RV(*) Becker and Fenkart (1971) d, AD, LD, E(B-V), AV, eSpT Reddish and Sloan (1971) M(*) van Schewick (1971) cluster proper motion Schwartz (1971) M(H II) Georgelin and coworkers (1970, RV(H II), eSpT 1973, 1975, 1976) Harris (1976) cluster age Fenkart and Binggeli (1979) d, eSpT Moffat and coworkers (1971, 1972, d, AD, E(B-V), eSpT 1973, 1974, 1975, 1979) Israel (1977, 1978, 1980) M(H II) Mermilliod (1981a, b) cluster age Nicolet (1981) d, E(B-V) Blitz, Fich, and Stark (1982) RV(H2) Table 2. (continued) Reference Principal Contributions (a) ________________________________________________________________________________ Janes and Adler (1982) (B-V)t Bruch and Sanders (1983) M(*) Wramdemark (1983) RV(*) Lynga (1981, 1983, 1987) d, AD, LD, E(B-V), AV, age, eSpT, (B-V)t, RV(*), M(*) Hron (1987) RV(*) Leisawitz, Thaddeus, and Bash (1988) RV(H2), M(H2) ________________________________________________________________________________ (a) Abbreviations: d = distance; AD = cluster angular diameter; LD = cluster linear diameter; E(B-V) = reddening; AV = visual extinction; eSpT = cluster member of earliest main sequence spectral type (photometric or spectroscopic type); (B-V)t = color of main sequence turnoff; RV = radial velocity; M = mass; * = pertaining to cluster stars; H II = pertaining to associated ionized gas; H I = pertaining to associated or potentially associated atomic gas; H2 = pertaining to associated or potentially associated molecular gas. the individual visible clusters of stars to which their clouds recently gave birth. Optical observers are vulnerable to being uninformed that the reddening that they measure photometrically sometimes is produced by dust in a known foreground molecular or atomic hydrogen cloud. Observers of interstellar clouds, in turn, may benefit by having constraints placed on the clouds' distances. Essentially all known open clusters have been assigned "OCL" numbers in the Catalog of Star Clusters and Associations (Alter, Ruprecht, and Vanysek 1970) or in its supplement (Ruprecht, Balazs, and White 1981). Clusters with NGC or Messier (M) numbers commonly are referred to by those numbers. Most young clusters are members of OB associations. Many of the clusters in the OCISM database ionize H II regions because they contain O stars. Since the clusters are observed optically, the H II regions are characterized by nebulosity which can be seen in the Palomar Observatory Sky Survey and in the Parker, Gull, and Kirschner (1979) Emission Line Survey of the Milky Way. Most of the nebulae were catalogued by Sharpless (1959) and have "S" numbers. A handful of the H II regions are referred to by their Westerhout (1958) radio source (W) numbers. Spatial Coordinates Galactic and equatorial coordinates of the nominal cluster centers are tabulated. The galactic longitudes and latitudes are in units of degrees. Hours and minutes of right ascension are shown, as are degrees and arcminutes of declination. The equatorial coordinates are for epoch 1950.0. Radial Velocities Approximate coincidence of stellar and interstellar gas radial velocities often is the only clue that an association may exist between a star cluster and a particular parcel of interstellar material. All radial velocities in the OCISM database are in the reference frame of the Local Standard of Rest (LSR). Stellar radial velocities and, occasionally, H II region velocities are published in the heliocentric reference frame. To convert to the LSR reference frame, we assume a standard solar motion of 20 km/s in the direction (lII, bII) = (56.2 deg, +22.8 deg). * Stellar Clusters The velocity dispersion among the stars in a typical open cluster is of the order of 1 km/s (see, e.g., Mathieu 1986) and, in principle, the cluster velocity is the average of the individual stellar velocities. Because of zero- point calibration uncertainties, intrinsically broad spectral lines in early- type stars, contamination of measurements by binary orbital velocities, and the fact that only a limited number of stars in open clusters have been observed with sufficiently high spectral resolution, the radial velocities of most young open clusters are known with an accuracy no better than + or -10 km/s (Hron 1987). Using the rotation curve model of Brand (1986), we calculated for each OCISM cluster a theoretical radial velocity which is the velocity that the cluster, at a known distance, would have if it followed a circular orbit around the galactic center. By comparing the derived radial velocities with measured velocities, when available, we found that the dispersion of the velocity differences is approximately 14 km/s (cf. Brand 1986; Hron 1987). For a large number of clusters, the "circular velocity" is practically as reliable an estimator of cluster radial velocity as any existing observation. This unfortunate state of affairs will persist until modern techniques and instruments are applied to measure the velocities of a large number of young clusters. Lynga (1987) and Hron (1987) incorporated in their catalogs the radial velocities compiled in unpublished work by Wramdemark. Agreement among the velocities credited to these sources therefore is not accidental. * H II Regions Radio and optical recombination lines provide information about the radial velocities of H II regions. Recombination lines are intrinsically broad (a typical FWHM linewidth is ~25 km/s), but the centroid velocities can be measured with an accuracy of the order of 1 km/s in the case of radio lines or a few km/s in the case of optical lines. Differences among reported H II region radial velocities measured using radio recombination lines more often than not can be attributed to peculiar motions of the ionized gas and the fact that different lines of sight are observed (or that the observations are made with beam sizes that differ). For example, significant deviations from the "mean H II region velocity" often occur along lines of sight that contain molecular clouds. Whenever possible, the velocity tabulated in the OCISM database is one that is supposed to characterize the mean H II region velocity. In some cases, several distinct major emission features are reported in an article; these individual features are tabulated here if they are not too numerous. * Molecular Clouds The mean radial velocities of interstellar molecular clouds can be measured with an uncertainty less than 1 km/s. Molecular clouds are discovered and studied most commonly with observations of the spectral line emission from the fundamental rotation transition of carbon monoxide (12CO) molecules. In most cases, 12CO spectral lines are the sources of tabulated "CO cloud" velocities. In some cases, however, 13CO or H2CO line velocities are stored in the database. Different molecular clouds that are potentially associated with the same open cluster may have different radial velocities. Leisawitz, Thaddeus, and Bash (1988) describe and apply systematically criteria to decide if a molecular cloud is "associated" or "potentially associated" with a cluster. All major emission components reported to have been contributed by associated or potentially associated molecular clouds are recorded in the OCISM database. The Blitz, Fich, and Stark (1982) catalog of CO radial velocities toward H II regions also is frequently cited. Proper Motions Absolute proper motions of the clusters are tabulated, in units of arcseconds per century, when available. We tabulate mu(x) and mu(y) where, as is conventional, mu(x) = mu(RA)cos(DEC) and mu(y) = mu(DEC) for the right ascension component of proper motion mu(RA) and the declination component mu(DEC). Distance Among the many factors that might motivate one to study the interstellar matter associated with open clusters is the fact that, by astronomical standards, the distances of these objects are well known. Thus, masses of associated ionized, atomic, and molecular gas components and, for example, infrared luminosities of interstellar clouds can be measured with uncharacteristically small errors. We analyzed the scatter among different authors' estimates of a cluster's distance and determined that the mean dispersion in log(d) for clusters in the OCISM database corresponds to an uncertainty of ~15%; the corresponding uncertainty in gas mass or dust luminosity is ~30%. A large number of researchers have constructed from photometric data dereddened color-magnitude diagrams (CMD) and, by the method of main sequence fitting (i.e., comparison with a "calibrated" CMD), have derived distance moduli. Absolute distance moduli, and cluster distances, generally are obtained by assuming a standard selective extinction ratio (e.g., AV/E(B-V) = 3.0 or 3.1) to determine the extinction. The distances to some open clusters have been derived using the method of spectroscopic parallax. Generally we do not tabulate distances derived in this way because only a few stars in each cluster are spectroscopically observed and, with poor statistics, the method is inferior to main sequence fitting (see, e.g., Hoag and Applequist 1965). Many of the distances entered in the OCISM database are values that were adopted, rather than determined, by the authors cited. For example, Neckel (1967) adopted distances that were measured by Johnson et al. (1961). Angular Diameter Rigorous determination of a star cluster's angular diameter can be made by counting stars and calculating the radial distribution of stellar surface density (see, e.g., Danilov, Matkin, and Pyl'skaya 1985 and references therein). Such measurements suggest that clusters often have high density cores (or nuclei) and relatively low density halos (or coronas). Thus, what is meant by "the cluster angular diameter" is somewhat ambiguous. When an author reports measurements of core and halo diameters, both are entered in the OCISM database. In most cases, cluster angular diameters are measured directly by inspection of a photographic plate or print. The cluster size determined in this manner may be investigator-dependent and may be a function of the limiting magnitude with which the cluster was observed. The print inspection method was employed by Lynga, who estimated the nuclear sizes of nearly all open clusters (Lynga 1981; 1983; 1987). Lynga's measurements are valuable particularly because of their self-consistency. His cluster diameters are contained in the OCISM database. Linear Diameter The linear diameters of well-studied open clusters are known because their distances and angular sizes have been measured. In addition to published estimates of the linear diameters of clusters, the OCISM database includes a value for the size of each cluster which is based on an adopted distance and angular diameter. The distance adopted for this calculation is either the average of two recently reported measurements or the average of all database distance entries if the dispersion among the entries is small. The angular diameters of Lynga were assumed. Age Estimates have been published for the ages of a large number of open clusters based on several methods, most of which are related. Two premises generally are implicit in age determinations: (1) the stars in a cluster form coevally; and (2) the main sequence lifetime of the most massive unevolved cluster member is an estimator of cluster age. There is no concensus among researchers as to the validity of the first assumption (cf. Doom, De Greve, and de Loore 1985; Stahler 1985). For studies of the interaction between clusters and their surrounding interstellar matter, however, this may not be problematic because the derived age will be a measure of the time elapsed since the formation of the most massive cluster stars and these are the stars that can be assumed to dominate the interaction. Only upper limits can be derived for the ages of clusters that are so young (<~5 Myr) that they contain no evolved massive stars. In some cases, however, ages of very young clusters have been estimated by noting that low-mass stars have not yet reached the zero-age main sequence and by comparing photometric data with theoretical pre-main sequence isochrones (see, e.g., Moffat 1972). Not all authors who report information sufficient to calculate a cluster's age also report the age implied by their observation. Typically what is published is the (B-V) color of the main sequence turnoff in the (dereddened) CMD, the corresponding "photometric spectral type," or the spectroscopically determined type of the brightest and bluest star on the cluster's main sequence. The information needed to interpret these parameters as measures of cluster age can be found in numerous articles (e.g., Lindoff 1968; Harris 1976; Janes and Adler 1982; Lynga 1982; Mermilliod and Maeder 1986). Published numerical estimates of the ages of clusters are entered in the OCISM database. Also included in the database are "earliest" main sequence spectral types and main sequence turnoff colors (in magnitudes). Lower case letters are used to designate photometrically determined stellar types and upper case letters are used for true spectral classifications. Masses In principle, one could measure the mass of stars in a cluster and inventory the masses of all interstellar matter components associated with the cluster to determine the "efficiency" with which stars form from dense interstellar clouds. What may well be an insurmountable barrier stands in the way of achieving this objective: to measure star formation efficiency, one must account correctly for all of the interstellar matter that participates in a cluster's formation; obviously, material that has participated (or will participate) in the formation of a significant additional stellar population must not be included in the analysis. At the very least, however, it is possible, and therefore sensible, to assemble the data needed to address this fundamental issue. Considerable care is required to determine what components of molecular and atomic material observed in the direction of a star cluster are indeed associated with the cluster. On the other hand, it is generally unambiguous that ionized gas detected in the direction of a cluster with O star members is associated with the cluster. Fortunately, clear signs of interaction between H II regions and dense interstellar clouds, such as "bright rims" on the cloud surfaces, occasionally enable one to conclude with confidence that a particular molecular cloud is associated with a cluster (see, e.g., Leisawitz, Thaddeus, and Bash 1988). More often, the only hint that a particular atomic or molecular cloud is associated with a cluster is a near-coincidence of stellar and interstellar radial velocities (see, e.g., Gordon, Howard, and Westerhout 1968). Masses of ionized, atomic, and molecular gas components are derived with some assumption made about the distance of the emitting material. Masses reported in the OCISM database are the published masses; no adjustment has been applied to "correct" these values to a common reference distance, even if the distance assumed is discrepant with a cluster's distance. * Stellar The masses of large samples of clusters have been studied by Schmidt (1963), Reddish and Sloan (1971), and Bruch and Sanders (1983). Reddish and Sloan derived masses relative to the mass of Trumpler 1. The cluster masses credited to them in the OCISM database were derived by us assuming that Trumpler 1 has a mass of 63 Mo (Schmidt 1963). Cluster masses quoted by Bruch and Sanders also are based on the Reddish and Sloan relative masses. Bruch and Sanders derived a new conversion factor for absolute mass which turned out to be close to the Schmidt mass estimate for Trumpler 1. Close agreement between the Reddish and Sloan masses and the Bruch and Sanders masses stored in the database is therefore fortuitous. * Atomic Gas Aside from the difficulty associated with determining that a cloud of atomic hydrogen is physically related to an open cluster (see above), there is the added complication that individual H I clouds, seen as features in the 21-cm line profile, generally are difficult to disentangle from "background" emission. Only Gordon, Howard, and Westerhout (1968) and Tovmassian et al. (1973) have published systematic studies of H I emission from the neighborhoods of open clusters (Dr. Bania, of Boston University, presently is observing H I emission from regions around a number of OCISM clusters). * Ionized Gas Published estimates of the mass of gas ionized by a young open cluster that are based on the flux from the entire H II region (or at least the bulk of it) are recorded in the OCISM database. Most of these estimates come from observations of radio continuum emission and thus are not affected by extinction. * Molecular Clouds The dynamical timescale for the interaction of molecular clouds and open clusters was first determined observationally by Bash, Green, and Peters (1977). The interaction has since been investigated in greater detail by Leisawitz, Thaddeus, and Bash (1988). In the Leisawitz et al. CO survey, a distinction was made between catalogued molecular clouds that can be considered with confidence to be "associated" with clusters, clouds that are "potentially associated," and clouds that can be considered with confidence not to be associated. A single entry of data from the Leisawitz et al. survey is made in the OCISM database when the molecular clouds found in the region around a cluster are "potentially," but not conclusively, associated with the cluster; when some of the clouds are considered "associated," two entries are made, the second corresponding to "associated" clouds only and the first corresponding to the "associated" and the "potentially associated" clouds. The tabulated masses are sums of the masses of the clouds in each category. Some regions that were observed by Leisawitz et al. contained no CO emission considered to be even potentially associated with the clusters. Since many H II region neighborhoods have been mapped for CO emission by millimeter wave observers, information about the masses of molecular clouds near very young open clusters is plentiful. A few H II regions in particular have recieved much attention, notably M 17 (NGC 6618) and W 3 (adjacent to IC 1805). A plethora of excellent articles have been neglected, for no reason other than to conserve space in the database, when the information they provide is essentially redundant with data already entered. Generally an entry is not made in the OCISM database for articles in which only a small portion of a molecular cloud is discussed. * Dust In rare instances, a value has been published for the mass of dust associated with an open cluster. D'Odorico and Felli (1970) conducted the only systematic survey of this nature of which we are aware. Although we tabulate the D'Odorico and Felli masses, it is important to note that, unlike the database entries for the masses of interstellar gas components, these dust mass estimates correspond only to dust concentrated in a relatively small region around the stars. A crude but reasonable approximation for the total mass of dust in the extended cluster environment can be obtained by summing the masses of the various associated interstellar gas components and dividing by 100 since the "normal" interstellar gas-to-dust mass ratio is ~100 (see Savage and Mathis 1979 and references therein). Visual Extinction In a majority of cases, values published for the visual extinction of a cluster are derived from a measurement of the (B-V) color excess (reddening) with an assumed value of 3.1 or 3.0 for the ratio AV/E(B-V). Of course, the true distance modulus of a cluster will be estimated incorrectly if the selective extinction ratio assumed to obtain the extinction is not strictly applicable to the line of sight to the cluster. Reddening The reddening of cluster stars by foreground dust is deduced from multi- wavelength photometric studies of the clusters. Even when photometry in a system other than UBV is discussed, it is conventional for an equivalent (B-V) color excess to be provided (see, e.g., Nicolet 1981; Janes and Adler 1982). It is not uncommon for an observer to report a finding of variable reddening toward a young cluster. In other words, the cluster stars show excessive scatter about the nominal main sequence relation in the color-color diagram (e.g., [U-B] shown as a function of [B-V]). Variable extinction is an indication of the presence of a foreground molecular or atomic cloud (which may or may not be physically related to the cluster). What is most often published, and what is stored in the database, is a value for the mean cluster reddening. PROCESSING AND ANALYSIS SOFTWARE To use the FORTRAN programs created to operate on the database, follow these steps: 1) read from tape the source code, the database file, and the reference file; 2) rename the file containing the database "OCDB.DAT" and rename the file containing the references "OCDB.REF"; 3) compile, link and run the source code; The program is menu-driven and should be self-explanatory. You will be prompted for the information needed to execute your commands. The basic processing options are as follows: "change output device and/or format" - user can select output to file or to terminal; complete or limited amounts of output can be requested; "select clusters on which to operate" - all clusters or a specifiable subset of the clusters in the database can be considered; "type list of selected clusters" - list includes cluster coordinates and aliases; "print catalog of cluster information" - in standard output format, this prints the database catalog; a less voluminous output is generated if requested with option (a); "mask out some references" - enables consideration of a subset of database entries selected either according to how currently articles were written or to specifiable article or compilation names (one might wish, for example, to examine database entries from the compilations of Lynga); to generate a complete table of reference codes and corresponding references, send output to a file and select option (h) (see below); ask the program for a more detailed explanation and it will tell you the "masking rules"; "average unmasked data and tabulate" - produces table showing average value of a database parameter (e.g., distance) for each of the selected clusters; table includes standard deviation and number of database entries averaged (only entries that were not masked with option (e) are averaged); "restore database" - necessary after masking either to mask differently or to proceed with no masking; "type complete list of database references" - bibliography of all references cited with corresponding reference codes (132 column output field); codes are used to identify individual articles for "masking." Acknowledgments I am grateful to Dr. Gosta Lynga who, long ago, encouraged me to pursue this compilation. His own Catalog of Open Cluster Data provided great inspiration as well as a wealth of information. I thank Dr. Wayne Warren for agreeing to incorporate the OCISM database in the NSSDC archives and to provide the support needed to distribute the database. The database was updated and prepared for presentation while I was supported first as a National Research Council Resident Research Associate at NASA/GSFC and later by a grant from the NASA Space Astrophysics Data Analysis Program (SADAP proposal R033-87).