NGC/IC Positions

Introduction

This web page, and the site you found it in, is partially devoted to debugging J.L.E. Dreyer's New General Catalogue and two Index Catalogues. Though compiled as carefully as possible, the three catalogues are unfortunately replete with hundreds of errors, large and small, almost all from the sources that Dreyer adopted. His own work was meticulous; very few errors in his catalogues are attributable to him. Rather, they are the result of either primitive methods of determining positions on the sky, or of hasty publication of "new" non-stellar objects by the observers.

While twenty-first century astrophysics has moved far beyond these old catalogues, the NGC and IC numbers are still used for larger and brighter non-stellar celestial objects beyond the solar system. Making sure that these numbers are applied to the correct objects is still a reasonable, if no longer urgent, program. And it has provided me with an absorbing hobby for many years.

Cleaning up the NGC/IC entries requires three things: 1) Access to the literature that Dreyer used to assemble the catalogues, 2) a source or sources of accurate positions for the objects included in the catalogues, and 3) a good photographic or digital sky atlas. For the first, I have over the years collected almost all of the original publications that Dreyer consulted. A few objects came in from private communications to Dreyer himself; these letters and lists may still be extant in his archives in England or Ireland, assuming they were kept after his death in 1932. For the second, I use CDS's VizieR service to consult many different catalogues of accurate positions for celestial objects of all kinds (a list of the current ones I use is given below). The third requirement is fulfilled by HEASARC's SkyView; it provides easy access to the optical sky surveys (I occasionally use the non-optical surveys as well). The SDSS DR15 "Navigation Tool" is often useful, as is especially the Pan-STARRS1 Image Access page at STScI. STScI also hosts a Digitized Sky Survey page that can be helpful, too. Finally, I use NED and the SIMBAD mirror at CfA for accessing modern data for individual objects.

This particular file primarily discusses the second requirement -- accurate positions. It is split into six sections:

1. A brief description of the position files.
2. The format of each file with detailed notes about the positions and sources.
3. Selected positions, with notes about precession, astrometric reference systems, and photometry.
4. A history of accurate positions, leading to ...
5. ... a discussion of these files and my work debugging the NGC and IC.
6. Acknowledgments.

What's Here

Files linked through this page are

1. Introduction to the positions' files -- this file.
2. Base J2000.0 positions for the NGC objects . This file is useful as it has *all* of my brief notes about object types, diameters, magnitudes, sources, and other comments for the individual positions.
3. Base J2000.0 positions for the deep sky objects found before Dreyer compiled the NGC, but not included in it. This is identical in layout to base J2000.0 positions file, but has only the historically "missing" objects included.
4. Also in the same format are the Base J2000.0 positions for the IC objects, again with types, occasional comments, cross-identifications, and data.
5. Selected positions for the NGC objects for J2000.0, and for B1950.0.
6. Selected positions for the objects found before Dreyer compiled the NGC, but not included in it, for J2000.0 , and for B1950.0.
7. Selected for the IC objects for J2000.0 , and for B1950.0.
8. The remaining files have the NGC Notes , IC Notes, and Not-NGC Notes notes on missing, misidentified, or otherwise unusual objects. An Introduction to the Notes has background information on the Notes themselves.
9. Finally, the position sources are listed alphabetically with references to the papers or catalogues themselves, comments on the reliability of these so far as I have been able to judge them, and estimates of the standard deviations in the positions listed in each source.

File Format

1. NGC and/or IC number, preceded by "N" or "I". Up to two additional NGC/IC numbers are also noted in this column. Other names are in the Comments column.
   • If there are two or more components to an object that have not already been noted and named in the literature, I use directional suffixes (e.g. "nw" is northwest, "se" is southeast, "m" is middle, etc.). I have added these directional suffixes to ALL the multiple objects here; my goal is to keep nomenclature confusion to a minimum.
   • Note that you can enter an object's position (NOT its name!) into any of the sky survey viewers to find exactly which object carries the NGC/IC number that I've assigned. I usually use HEASARC's SkyView to access the DSS2 red images. This series of plates is reasonably uniform across the sky. In the north, DSS2 blue images are an acceptable alternative. SkyView has many other sky surveys at all wavelengths available, including SDSS. However, for those parts of the sky covered by SDSS or by Pan-STARRS1, color images are available via the SDSS DR14 Navigator Tool and the Pan-STARRS1 Image Access page at STScI.
   • I've added additional suffixes for nearby companions, whether physical or just optical, clearly not seen by the NGC/IC observers (usually just "comp" for "companion," though often with a directional suffix added; the word "component", by the way, is abbreviated "compt" to avoid confusion; and "compact" and "comparison" are always spelled out for the same reason); and for the components of objects that are double or triple stars, again whether optical or physical. Sometimes when the star images are merged on the DSS images, I give only a single position for the pair. I do note, however, that modern surveys such as Gaia, Pan-STARRS, SDSS, 2MASS, and others often resolve the individual components. I've added modern positions, primarily from the four surveys just noted, to the files. My goal has always been to keep confusion to a minimum. Which leads us to ...
   • Commonly-used suffixed numbers (e.g. N0078A, N0078B) are also noted here, though I discourage the use of these numbers -- all the objects have other more appropriate names that we would do better to adopt instead. Most of these suffixed numbers come to us from three sources: Gerard de Vaucouleurs's 1956 survey of southern Shapley-Ames galaxies, Eric Holmberg's 1937 list of double and multiple galaxies, and a list of galaxies in the far northern sky found by Philip Keenan, also published in 1937. Most of these also appear in the first two Reference Catalogues, and in RNGC. The southern ones appear in SGC and ESO. Some of the suffixes also come from UGC, but these are relatively rare, as are additional suffixes scattered through the literature. I've attempted to apply the unsuffixed number to the object seen by the original observer, but there may be mistakes remaining. While I've looked at everything, some objects have had more scrutiny than others.
   • There are usually only three cases where the earlier observers added the alphabetical suffixes that we find so problematic:
     1. The NGC/IC object is composed of two or more "components" (NGC 78 is an example; we are not quite sure which of these its discoverer Pechule actually saw).
     2. The NGC/IC object has nearby "companions" on the sky; these are sometimes actual physical companions, sometimes just optical alignments of objects at different distances (most of these are from Holmberg's 1937 thesis on double and multiple galaxies).
     3. Completely unrelated objects up to a degree or more from the NGC/IC object may bear the NGC/IC number and a suffix simply because the NGC/IC object was the nearest "named" galaxy -- most of these are from de Vaucouleurs's 1956 survey of the southern Shapley-Ames galaxies, from Keenan's 1937 Astrophysical Journal paper, (rarely) from Sulentic and Tifft's 1973 RNGC, or (even more rarely) from Holmberg's 1937 thesis.
     The most bothersome cases arise when there are suffixed objects in two or all three of these categories. NGC 6166 is an example. It is composed of four merging components -- labeled "A" to "D" in a 1961 Astronomical Journal article by Rudolf Minkowski -- and has four companions -- labeled "A" to "D" in RNGC, but "b" to "e" by Holmberg in 1937 (N6166 itself was "a").
   • I've added Question marks and colons in two different situations. First, if I've noticed that the positions from different sources disagree, and I've not yet had a chance to find the source of the problem, I've put question marks on the numbers (I believe that I looked all of these by November 2018, but I may have missed a few). Second, if I have done the digging in the historical literature and am still unable to reliably assign the NGC/IC number to the correct object on the sky, I append colons and question marks to the names to indicate this. These objects have "stories" in the "ngcnotes.all" and "icnotes.all" (the "puzzles [not always!] solved") files -- as of November 2018, I believe that all of these have notes. I am now doing a careful completeness check of my early work; it is possible that I have missed objects needing notes (NGC 7830 is an example of one for which I finally wrote the note in May 2016). If your particular questionable object is not yet covered, drop me a line -- I'll look into it for you. Or you may send me a note about your own research into the problem; in that case, I'll write a note quoting your work if you are agreeable, with full credit to you. Note that the NGC/IC Project web site has most of the historical literature available. ADS also has much of it available in its regular abstract/scanned article holdings. I will eventually be hosting these papers myself; check back now and then.
   • Once in a great while, the original observer gives us more than one position for an object. In these cases, I've used notations like "N2002.1", "N2002.2", and so on for each of the positions. Thankfully, this doesn't happen often.
   • Finally, I've occasionally added descriptive words (e.g. "bar", "comp", "core", "nearby *", "F gal", "knot", "clump", "sup *", often with letters indicating the direction from the main object, e.g. "n", "sw", "ene") in an attempt to specify which part of the object, or which nearby object, the position applies to.
2. RA and Dec for J2000.0.
   • Most of the source catalogues from which I originally worked have used equinox 1950.0 positions. Only within the past quarter century has J2000 become common as the base star catalogues have been constructed in the recently-adopted FK5 and ICRS systems. While I am now working in J2000, I also provide the files in B1950 form for easy reference to the mid-20th century catalogues.
   • Precision given is usually that in the published source, though I am rounding off to 0.01 seconds of time, and 0.1 arcseconds when pulling positions out of e.g. UCAC, 2MASS, GSC3, SDSS, etc. These catalogues carry additional digits primarily as round-off guardians. I do keep additional digits for the Gaia positions -- to 0.001 seconds in RA and 0.01 arcseconds in Dec -- which have nominal errors of 0.6 milliarcseconds or better for stars brighter than about 20th magnitude at the nominal observation epoch 2015.5. At J2000.0, the positions, when corrected for proper motion, have nominal errors of around 15 milliarcseconds. Positions from Pan-STARRS DR1 are nearly as accurate (around 4 milliarcseconds at the observation epoch), so I give them, too, with a precision of 0.001 seconds in RA and 0.01 arcseconds in Dec. The only positions that I give with greater precision are those which define the optical International Celestial Reference System (ICRS), e.g. NGC 3031 = M81.
   • The format is HH MM SS.dddd, +-DD MM SS.ddd (number of places to the right of the decimal points varies, and is only given to more than two places for objects defining the ICRF or for those found in the Gaia and Pan-STARRS catalogues); HH MM SS, +-DD MM.d; HH MM.d, +-DD MM; or HH MM, +-DD.d are used for less accurate positions. Positions published in other formats (e.g. HH MM SS, +-DD MM SS) are rounded, usually in declination.
   • Accuracy is another matter. Since most of the positions are from sources other than my own work, I can only guarantee their accuracy through comparison with positions from yet other sources. This is the main reason that I have always collected more than one position for an object.
   • Of course, I don't see much point of including positions here if they aren't accurate, so I check everything as carefully as I can when I put positions in. The positions are always accurate enough to identify the correct objects -- and as I elaborate below, Brian Skiff and I are making them much more accurate than that when we think it is appropriate (e.g. galaxy nuclei and central stars of planetary nebulae).
   • Some early entries (e.g. from RC2) have been truncated to more realistically reflect accuracy. Lately (after 2008 or so), I have also, as noted above, rounded overly precise positions from the digital and digitized sky surveys to more closely reflect accuracy.
   • I also make mistakes now and then. I fix 'em when I find 'em. If you find one, please let me know so that I can fix it -- if I don't know about it, I can't fix it. Thanks!
3. Position Codes
   • I flag original positions from the historical source papers with a lower case "o", and rejected positions with an asterisk "*", both in this column. The flag denoting Brian Skiff's selected positions is an "s" here, and those positions that I selected have a "c". I'll discuss these selected positions below.
   • I've rejected positions if they differ significantly from the mean of the other positions, or if I know they are incorrect from other considerations, primarily checking the DSS or Pan-STARRS images.
   • I've used the lower case letter "o" to flag positions from the original 19th century lists even when they have accuracies that may approach modern positions. While these may be very accurately measured, the systems of reference stars in use during the late 18th and 19th centuries was not nearly as well-defined as it is now. Recovering modern positions of these reference stars at the earlier epochs is now almost trivial, since we now know proper motions for them, thanks to modern catalogues like Gaia, Hipparcos, Tycho-2, and UCAC.
   • I've also listed the old positions when I've been unable to recover the object. In that case, they always carry a note saying, "Nothing here; nominal pos." The original positions are always included for the objects in the Magellanic Clouds; I needed them for clear identifications in these spectacularly crowded parts of the sky. I will eventually have original positions listed for all the objects.
   • Very precisely specified declinations used to overlap into this column once in a while, though I have cleaned this up. Few positions for extended sources are known with sub-milliarcsecond accuracy, though I have included radio and optical positions for a few galaxies (e.g. NGC 5128) with compact nuclei which are part of the definition of the ICRF which enhanced FK5 in the 1990's -- and, of course, Gaia positions are now available. These, too, are very accurate, so have more precision given.
4. Source Codes. These are decoded in the sources file which also gives my own initial ideas about the accuracy of a source, and other relevant notes about identifications, number of positions given, etc. As of October 2006, all the source codes had entries in the "sources" file, thanks to Francois Oschenbein who listed the missing ones and helped me track them down. I've occasionally found typos, however, so please let me know of any source codes with no reference in the "sources" file.
5. Number of observations taken from the original source. This is usually the number of individual observations going into the final number, but is sometimes difficult to determine (some observers give the number of nights on which observations were made; others give the actual number of observations made per night; some have different numbers for right ascension or declination; etc). If no number is given, n = 1.
6. Types, Comments, Notes, and/or Remarks are entered when needed. The column extends as far as necessary, and currently stretches to at least column 155. I will try to keep it to that length as I will have to fuss with the precession routines (to build B1950.0 files) to push it further.
• Objects other than galaxies have their types noted in the first several columns of this field:

  GC	globular cluster
  OC	open cluster (physical cluster)
  OCR	open cluster remnant (physical cluster)
  OptC	optical cluster (not a physical cluster, if this is known)
  PN	planetary nebula
  BN	bipolar nebula
  DN	diffuse nebula
  EN	emission nebula
  WRN	nebula around a Wolf-Rayet star
  RN	reflection nebula
  HII	HII region
  SNR	supernova remnant
  SA	stellar association, sometimes used interchangeably with SC
  SC	star cloud
  *	single star
  **	double star
  ***	triple star
  Ast	Asterism (more than three stars); usually from sources AH or TDM

  These are used in combination if needed (e.g. N0261 is a "DN+SA"). These types are usually based only on the appearance of the object on the red and blue optical sky survey images; there isn't much astrophysics here, though if I know about a physical classification, I list it (e.g. NGC 1952 = M 1 is an SNR, NGC 6888 is a WRN).
• If the NGC or IC number applies to only part of an object, I either append a "p" to the type (e.g. NGC 650 and NGC 651 are both "PNp"), or attach a note explaining the situation (e.g. NGC 3189 is the southwestern part of NGC 3190, NGC 6960 is part of the Veil Nebula -- an SNR -- and so forth). As I work on these, I will append a "p" to make the situation clear, e.g. "Gp", "SNRp".
• If the object is in another galaxy, that galaxy name is appended to the type (e.g. NGC 152 is a "GC,SMC" -- a cluster in the Small Magellanic Cloud). The types in the Large Magellanic Cloud are no longer from the same sources as the position; I'm trying to use the morphological types derived directly from the sky survey images, unless I know better (e.g. that a particular nebula is an SNR or a reflection nebula). Note that the young "globular" clusters in the LMC are simply called "GC"; they are morphologically indistinguishable from old, red globulars, at least on monochrome images. Many types from ESO catalogue are just the first two letters of the "classification" given there (e.g. "Em" = "Emission nebula"). I've fixed most these in the Clouds as I've noticed them, but it's likely that some, perhaps many, original codes from ESO remain.
• Stars, double stars, and triple stars are almost always noted by asterisks, one per star. Asterisms with more than three stars are usually noted by me "n sts" or "n stars", where "n" is the number of stars. If two stars are widely separated, they are sometimes noted "2 sts" or "2 stars". Sources AH (Archinal and Hynes, "Star Clusters") and TDM (Tom DeMary, private communication) often use the word "Asterism" to flag clumps of stars mistaken for open clusters. These might well be called optical clusters ("OptC") in other sources.
  I've given proper motions for stars if a) the proper motion is large enough to move the resulting J2000.0 position by more than about an arcsecond; b) to show that the object in question is in fact a star and not a galaxy; and c) to show that that some of the multiple stars in the list are in fact either optical or physical multiplets. For almost all the stars with listed proper motions, I've adopted the motions from Gaia EDR3. The format for the proper motion entries is "pm = +123.45, -12.34 mas/yr", with the RA proper motion given first, the Dec motion second. The units are always milliarcseconds per year.
• I am currently entering galaxy types on the de Vaucouleurs revised Hubble-Sandage system. The source (in parentheses) follows the type itself; at the moment, "(VA)" flags types by Ron Buta in "The de Vaucouleurs Atlas of Galaxies", and many RC3 types have been taken from the "Carnegie-Irvine Galaxy Survey" (CGS) by Luis Ho and his collaborators. Others are new classifications by me, sometimes followed by "(HC)", sometimes not. These are generally from the DSS2 images, the SDSS images, or the Pan-STARRS images. As I add more galaxy types, I'll expand the source list here into a short table.
• Comments are usually -- but not always! -- from the original sources; my own notes are usually -- but not always! -- in square brackets when appended to a note not written by me, or a source other than one of my own.
• With a two notable exceptions, all the measurements, particularly for diameters and separations, given in the remarks are in arcminutes unless marked (usually "arcsec" or "deg", or another common abbreviation). The exceptions: 1) position angles are always in degrees measured from north through east, and are flagged "PA"; 2) the "Nominal error" given with sources "CCA" for UGC galaxies and "CK" for planetary nebulae is in arcseconds.
• Many cross-identifications into modern catalogues are in this column, too. I've put in the cross-ids when I've had them in front of me. Unless they are needed to help identify an object, I've generally not had them handy, so many aren't here. NED, LEDA, and SIMBAD are still the best places to find cross-id's (at least for the NGC objects): use one of the positions listed here in a position search. Note that cross-ids in the on-line services, particularly in crowded fields, are not guaranteed to be correct. I speak from experience with NED; I was correcting these up to the day I retired from the NED team in July 2011 -- many more certainly remain to be fixed.
• Some of the cross-ids are simply numbers; prepend the source code to these (e.g. VCC 456, GSC 01234-05678).
• Occasionally, I've included some data here (e.g. galaxy types, diameters, or magnitudes), but these are still unusual. Check Wolfgang's, Steve's, or Malcolm's lists for data. I am beginning (summer 2019) to collect basic data for all the objects. Those data will be in a separate file (or files); see the SGC and SEGC pages for the first steps toward this data-collecting project.
• Abbreviations are generally taken from the NGC, but I've used "sts" for "stars", "deg" for the degree symbol, and a few other non-standard ones. (See the introduction to the Notes files for other common abbreviations.) Since these are intentionally 7-bit ASCII files, I've limited myself to the characters available on a standard US keyboard, at least in the NGC/IC files. As I use and add to the files, I'll try to remember to mention other unusual abbreviations here or in the Notes introduction.

Here are some examples from the position tables with the format shown:

NGC/IC              RA (2000.0)   Dec (2000.0) R So n  Comments, Notes, etc.
  (1)                 (2)           (3)       (4)(5)(6)  (7)
<------------------><------------><----------->|<--><> <-------------------------------------------------------------------------------------------------->
N0015               00 09 02.5    +21 37 27     UZC   
N0015               00 09 02.472  +21 37 28.34  KHJ  1 = NPM1G +21.0004
N0015               00 09 02.48   +21 37 28.6  sUCA2  
N0015               00 09 02.50   +21 37 26.9   CCA    = UGC 00082.  Nominal error =   1.4
N0015               00 09 02.54   +21 37 28.2   GSC  2 = 01184-01268
N0262               00 48 47.1415 +31 57 25.085cICRF   = ICRF 0046+316
N0262 comp e        00 48 52.840  +31 57 30.99 cGai1
N2002.1             05 30 18      -66 53.5     oJH     SC,LMC. "Place of double star ... the chief of a great cluster ..." in the LMC
N2002.1             05 30 21      -66 51.7     cHCe    SC,LMC
N2002.2             05 30 20      -66 52.8     oJH     GC,LMC.  "vB, S, R" "nebula" in the LMC
N2002.2             05 30 20      -66 53.0      ESOB   GC,LMC = ESO 086-SC003
N2006n              05 31 19.9    -66 57 29     HCcd   GC,LMC = [SL63] 538
N2006s              05 31 19      -66 58.3      ESOB   GC,LMC = ESO 086-SC008
N2017               05 39 17.3    -17 50 42    c2MSP 6 6 sts; not a cluster.  Mean position.
N2017               05 39 18      -17 50.8     oJH    
N2017 e *           05 39 22.47   -17 51 18.9  c2MSP   *
N2017 m *           05 39 16.21   -17 50 58.0  c2MSP   *
N2017 n *           05 39 16.96   -17 49 43.0  c2MSP   *
N2017 ne *          05 39 19.40   -17 50 20.0  c2MSP   *
N2017 s *           05 39 20.48   -17 52 02.7  c2MSP   *
N2017 w *           05 39 08.06   -17 49 52.5  c2MSP   *
N2139=I2154         06 01 07.97   -23 40 23.7  cUB10 2
N2398 se comp       07 30 16.52   +24 29 12.4   UCAC  
N2398 nw comp       07 30 13.6    +24 29 26     WS    
N2399               07 29 49.9    -00 12 50     WS     ***
N2399               07 29 50.2    -00 12 51     HCm    ***.  GSC pos for 1, HCo for 2.
N2399               07 29 50.3    -00 12 50    c2MSP 3 ***
N2399               07 29 50.6    -00 12 44    oHS1    ***
N2399 n *           07 29 50.26   -00 12 39.3  c2MSP   *
N2399 s *           07 29 50.49   -00 13 05.9  c2MSP   *
N2399 w *           07 29 49.96   -00 12 47.2  c2MSP   *
N2947=I0547=I2494   09 36 05.82   -12 26 11.7  s2MSP  

Selected Positions

These files with just one entry per object were originally intended to provide a handy source for "sufficiently accurate" positions for all of the NGC/IC objects. Unlike the J2000.0 base files, they have only one entry per object, a "selected" position chosen by Brian Skiff or myself. (Earlier versions of these files also carried a unweighted mean positions calculated from the several acceptable positions in the base files when Brian or I had not yet selected an accurate position. All of these unweighted mean positions have been replaced with selected positions as of May 2016.) As new surveys have appeared, we have updated the selections to reflect either increased accuracy or, if accuracy is already adequate, precision.

As of July 2021, I have entered all the appropriate Gaia EDR3 positions that are within 2-3 arcseconds of the previously selected positions. For over 80% of the objects, the Gaia positions are now the "selected" positions that most accurately place the object on the sky. Given its accuracy -- on the order of a few milliarcseconds -- Gaia EDR3 will probably be the penultimate large survey that I will incorporate in these NGC/IC lists (though the Gaia team has promised us DR4 in 2023 or 2024; stay tuned!). The Vera Rubin 8.4-m telescope is scheduled to see first light sometime in 2022, and should begin routine survey operations in 2023. Hopefully, the image results from the Rubin telescope should give us an online method to evaluate the positions of southern objects just as we now do using the Pan-STARRS image server at STScI for objects north of -30 degrees. In principle, the SkyMapper Southern Sky Survey could be used for checking image character and precise centering, too, but as of July 2021, the SMS pixelation was too coarse for either purpose. I expect this situation to improve in the future, but will also look forward to the Rubin results.

The format of the selected-position files follows the base files closely with the same NGC/IC name I've used there in the first 20 columns. The selected RA follows in column 21 in HH MM SS.dddd format (with varying numbers of digits after the decimal depending on the precision of the data in the base files), the declination sDD MM SS.ddd (same comment) in column 35. A one-letter code in column 51 indicates the source of the selected position: "s" for Skiff and "c" for Corwin. The original observer's position, coded "o", is currently used when the object has a questionable identification or cannot be found. I hope to include all of the original positions in later versions of the base files.

(If you are still using older versions of the files which also contain unweighted mean positions, you will find the number of positions used to find the mean in columns 49 and 50. The calculated standard deviations in arcseconds in RA and Dec are in columns 53 to 58, and columns 59 to 64, respectively. The sources of the selected or mean positions are listed beginning in column 66. The precision to which a mean position is given reflects the precision of the most precise position for the object in the base file.)

Note that precision does not necessarily reflect accuracy, especially with unweighted means. However, I've rejected obvious blunders, and most popular objects had enough accurately-measured positions that overall accuracy in the calculated mean did not suffer much from poor positions. In any case, I have replaced all of these unweighted means with selected positions in current versions of the files, so strongly encourage you to use the latest versions.

Two sets of selected position files are available, identical except for the equinoxes: J2000.0 and B1950.0. The equinox J2000.0 file gives the selected position (or in older versions, the unweighted mean of all the non-rejected positions) taken directly from the base file, while the equinox B1950.0 file has that selected (or unweighted mean) position precessed to B1950.0.

Please note that J2000 positions that I calculated between mid-2001 and 28 October 2005 may have errors of over an arcminute in them due to a bug in GCC 2.96 (the GNU CC and Fortran compilers I was using at the time). Most of the precessed positions are all right, but a few are not (e.g. NGC 1593 where the RA seconds for the incorrect position was listed as 16.333; the correct RA seconds should have been 06.133). Other positions in later batches were affected by compiler errors which produced many precessed positions containing "NaN" ("Not a Number"), indicating computation errors. The base B1950 positions I was using then were not affected by the compiler bug. Subsequent (November 2010 and later) versions of the programs compiled with the GCC compiler that I used on workstations at IPAC (Caltech) correctly precess the positions. Similarly, correctly-precessed positions come from the same programs compiled with GCC 4.8.0 (that I adopted in February 2013) on a Macintosh using OS X 10.11.4. I am currently (July 2021) using OS X 10.11.6 and GCC 6.1.0; this compiler is also free of the bug that caused the problem with the earlier version. My apologies for any problems arising from this.

Some Picky Details

Standard Deviations and Mean Errors

If you are working with files from releases earlier than May 2016, you will find objects with unweighted mean positions still shown. For those, I've listed standard deviations rather than mean errors as the s.d.'s more accurately reflect the uncertainties in the unweighted mean positions. Errors in published positions (see the Sources file) are almost always underestimated, so quoting the standard deviation increases the qualitative perception of the errors to about what they should be in an ideal world. If I ever decide to calculate weighted means (unlikely) -- instead of or in addition to selecting "best" positions -- the mean errors will have more relevance. In the meantime, if you want them, mean errors can be easily calculated by dividing the standard deviations by the square root of the number of positions used. Again, this procedure only applies to older versions of the files.

Equinox J2000.0 and Precession

As of March 2014, I am using J2000.0 for the base files rather than B1950. To find the initial J2000.0 positions, I simply precessed the old B1950.0 base files to J2000.0. Specifically, I did not reenter positions originally published at the equinox of J2000.0. Thus, any errors in the B1950.0 positions will be reflected in the J2000.0 positions that are now the basis for the positions I am currently using. The precession routine, discussed below, represents another potential source of error as well. I emphasize this because many of the older positions have as their native equinox B1950.0 tied to the FK4 list of fundamental astrometric standard stars. Many of the newly-adopted positions have equinox J2000.0 tied to either the FK5 or ICRS list of fundamental standards. The J2000.0 positions were precessed back to B1950.0 for use in the earlier base files. Precessing them forward again to the original equinox of J2000.0 will generally not recover the original positions as published -- approximations in the precession routines could lead to differences of as much as 2-3 tenths of an arcsecond. Simple rounding errors have occured as well. I will be checking all these carefully over the next few upgrade cycles, replacing positions that have been too strongly affected. So far, I have found very few that I have had to fix -- but some probably remain.

The "saving grace" in all of this is that the older B1950.0 positions precessed from that equinox are usually not accurate enough to be affected by the back and forth precession discussed above. Round off errors may still be a problem for small percentage of the objects, but errors introduced by the precession routines are far smaller than the known positional accuracies.

The precession routine that I use now for converting between positions at J2000.0 and B1950.0 adopts the FK4 to FK5 reference frame corrections given in USNO Circular No. 163 (1981, ed. G. H. Kaplan. I gratefully acknowledge the invaluable help of Brent Archinal in working through these detailed accounts of precession, particularly the conversions from FK4 to FK5, and back; also see the discussion of precession in e.g. the "Explanatory Supplement to the Astronomical Almanac", pp. 99-108 [University Science Books, 1992], and the Introduction to FK5). Note that I am specifically not using additional corrections taking into account the areal-dependent distortions between FK4 and FK5, nor have I adopted any magnitude terms, either. In practice, this version of precession agrees almost exactly with that used by the CDS services (VizieR and SIMBAD), and closely approximates the NED routine. Typical differences for test cases between B1950.0 and J2000.0 that I've used as checks are on the order of 0.1 arcseconds or less, though I have occasionally seen differences as large as 0.2-0.3 arcseconds. While I do not know the exact origin of these differences (I suspect they come from the FK4-FK5 areal- dependent distortions noted above), they are small enough to live with given that nearly all current programs require and use J2000.0 positions. (But 2050.0 is less than three decades away!)

FK5 and ICRS/ICRF

Thus, the J2000.0 positions in the base files, if derived from B1950.0 FK4 positions, are expected to be systematically close to, if not exactly on, the FK5 system. (Note again that the differences due to precession are nearly all well within the known accuracies of the original surveys.) I also note that I have not specifically reduced these precessed positions to the current optical International Celestial Reference System (ICRS), as reflected first by the Hipparcos and Tycho-2 positions for over 100,000 bright stars, and now by the Gaia positions for over half a million quasars. (I'll discuss the ICRS more below.) Nevertheless, very few of the older positions have accuracies better than 0.2 arcseconds, far larger than the current systematic uncertainties in the ICRS, so they too can be taken to be consistent with the ICRS as defined by Gaia.

The FK5 optical system itself has been shown to be consistent with ICRS to within the known systematic errors (about 20 milliarcseconds) of the older system. See Ma et al, Astronomical Journal, Vol. 116, p. 516, 1998 and references therein; and the Introduction in Volume 1 of the Hipparcos and Tycho Catalogues, pp. 100-101, for more information about the relationship of FK5 and ICRS as it was defined late in the 20th century.

However, the Gaia astrometry now essentially defines the ICRS, and the Pan-STARRS astrometry is tightly tied to Gaia, so positions from those two catalogues may be taken as currently "definitive" (aside from accidental errors which, unfortunately, do occur. I've flagged those discordant positions whenever I've noticed them). Because over three-quarters of the "selected" positions here come from Gaia and Pan-STARRS, the astrometric system I currently use closely approximates ICRS.

You will occasionally see references to the ICRF3, the third revision of the "International Celestial Reference Frame". This is the basic astrometric reference frame resting on 4536 extragalactic radio sources observed with very long baseline interferometry (VLBI), with special emphasis on 303 "defining sources", including 3C 273B. These are almost all quasars so have, in principle, no detectable proper motion. So, this ought to be a "frozen", non-rotating reference frame against which the motions of nearby objects can be measured. See Charlot et al., A&A 644, A159, 2020 for a more detailed discussion of ICRF3.

The ICRS is the optical realization of the ICRF. The Wikipedia article International Celestial Reference System and Frame has more information and links to additional references on the relationship of the ICRS to the ICRF.

As I noted above, an extragalactic subset of half a million quasars observed with Gaia is now being used as an optical reference frame (see the Gaia DR2 celestial reference frame). While this "Gaia-CRF2" was independently determined, it is consistent with ICRF3 on the 30 microarcsecond level.

The Gaia team has (December 2020) published the first part of their third data release, which I refer to here as "Gaia EDR3" or just "Gae3" as a source code. Details are available in Gaia Early Data Release 3. This data release includes the list of objects so far detected by Gaia, along with their positions, proper motions, parallaxes, magnitudes and some additional data. The full Data Release 3, planned for sometime in 2022, will additionally include radial velocities, spectra and parameters based on spectra, solar system data, variability information, and detailed results for multiple stars. My interest, of course, is primarily in the positional data and proper motions, but there may be information in the full data release that I'll pick up, too.

Gaia DR4 is already being planned, but that is still 2-3 years in the future. I doubt that it will offer any big improvement in the positions on the level of accuracy that interests me for the NGC and IC objects. However, additional objects may find their way into DR4, potentially giving accurate Gaia positions for objects currently without them.

Photometry

Occasionally, I give BVRc (Johnson/Cousins systems) data for stars and galaxies in the notes for the positions (see e.g. NGC 1330). Once in a while, you may also see U and Ic numbers as well. I usually take these from the SDSS with the data converted with the following formulae from Jester et al (AJ 130, 873, 2005):

      V = g - 0.59*(g-r) - 0.01
    U-B = 0.78*(u-g) - 0.88
    B-V = 0.98*(g-r) + 0.22
   V-Rc = 1.09*(r-i) + 0.22
  Rc-Ic = 1.00*(r-i) + 0.21

"u", "g", "r", and "i" are the SDSS magnitudes, and UBV are the standard Johnson/Cousins magnitudes, while Rc and Ic are the Cousins red and near-infrared magnitudes. These formulae are appropriate for all stars with Rc-Ic < 1.15 (B-V < ~2.3), and will also work well for non-emission galaxies (also see the SDSS/UBVRI transformation page).

Note that the SDSS "magnitudes" are actually transformed from the measured fluxes with an inverse hyperbolic sine function. For objects brighter than about 22nd magnitude, these are numerically identical to standard Pogson magnitudes transformed using base 10 logarithms. For fainter objects, the "asinh" magnitudes more accurately reflect the fluxes at low signal-to-noise ratio, and even handle negative fluxes without crashing the transformations. Lupton et al (AJ 118, 1406, 1999) describes this magnitude system in detail. A useful introduction is given on the SDSS web site. Again, for "bright" objects, the SDSS magnitudes are functionally the same as the logarithmic Pogson magnitudes that we've been using for 1.5 centuries.

Brief History of Nonstellar Astrometry

Much of astronomy from the time of William and John Herschel on through the end of the 19th century was devoted to the establishment of an accurate astrometric reference system. The star catalogues that the Herschels used often had systematic errors of several arcseconds, perhaps more. Crude proper motions could be found (and indeed were used by Sir William in 1805 to find the solar motion and apex), but parallaxes were first published only in 1838, after the Herschels' sky surveys were finished.

By the end of the 19th century when photography had supplanted visual observing as the primary means of determining positions of celestial objects, the reference system was well enough developed that it can be usefully accessed today for the derivation of parallaxes and proper motions of stars. The Carte du Ciel/Astrographic Catalogue project mapped the entire sky down to about 11th magnitude, providing an epoch now at least 70-80 years in the past for comparison with current epoch surveys.

Indeed, accurate proper motions were one of the goals of the astrometrists. By the late 19th century, they had a good idea of the large-scale motions of the nearby stars within the galaxy. Lacking, though, was an accepted theory of the nebulae and where they fit in with the stars. No one had found any verifiable parallax or proper motion for any of the nebulae in spite of decades of trying. Large programs of observation of nebulae were mounted at several European and American observtories with the hope of determining proper motions for the objects.

But nebulae are tough to measure accurately. Many are fuzzy, ill-defined patches of pale light compared to the sharper star images. Setting a micrometer wire on the "center" of a nebular image is often a difficult task, though many of the brighter galaxies have sharp nuclei that help make the chore somewhat easier. The numbers of nebulae are small, too, at the brighter magnitudes. While there are something like a thousand nebulae -- most are galaxies, of course -- brighter than 12th magnitude, there are several million stars brighter than that same limit. A twelfth magnitude star is hard enough to measure with a visual micrometer, even with a relatively "large" telescope (say 30- to 50-cm); a nebula of the same brightness with most of its light spread out over an area a few arcminutes in diameter is just that much more difficult.

So, by the turn of the 20th century, only a few thousand nebulae had well-measured positions. Unfortunately, those resulting positions were largely published in observatory annals, essentially lost to the hundreds of researchers engaged in the still-new science of astrophysics. While stellar astrometry carried on, astronomers as a group were slow to start large-scale programs of measurement of nebular images simply to find accurate positions. By the 1920s, they knew that parallaxes and proper motions of the galaxies are essentially unmeasureable: their distances are simply too great.

So, twentieth century galaxy cataloguers essentially started over with positions estimated with accuracies of a couple of arcminutes, often from the NGC or ICs. Who needed better? That accuracy was usually good enough to find the galaxies, and that was pretty much all an astrophysicist needed to point a telescope with a spectrograph or camera attached. Adjustments at the eyepiece could easily center the object for further study.

Use of the galaxy images as a reference frame for astrometry was slow to come, too. While galaxies in principle could define a precise and -- for all practical purposes -- unmoving reference frame, finding the nuclei in the larger images proved to be as difficult as always; on photographic plates, the nuclei are sometimes lost in the overexposed central regions of the galaxies. Plates, however, have a huge dynamic range, and many galaxy nuclei have essentially stellar profiles, so several projects were done using galaxies as reference objects.

However, galaxies are found in large numbers only away from the Galactic plane. The Milky Way is precisely where many of the most interesting classes of stars are located, so the astrometric reference system there had to depend on the stars themselves. If one is forced to use stars over a third of the sky, why not simply do it over the other two thirds as well? That, in fact, is usually what happened until the extragalactic quasi-stellar radio sources were recognized in the 1960s as a potential reference system.

By then, it became obvious that we needed good optical positions for the quasars to compare with the radio positions (see the discussion of the ICRF and the ICRS in the previous section). With the dawning realization that when we observe quasars (or their radio-quiet cousins, QSOs) we are looking at the nuclei of very distant galaxies, interest in determining accurate positions for galaxies once again bloomed.

Now, of course, we use the positions from the modern sky surveys, scanned directly from the sky, or from "second generation" survey plates taken by the big Schmidt telescopes. These positions are strictly on the same system as the stars, so -- assuming that the scanning machines can accurately find the galaxies' nuclei -- we have our accurate positions in hand.

Finally, we need to relate the positions from the surveys to the NGC and IC objects. Automated scanners don't have the NGC in their software as they go over the sky or a plate, so we have to take the time to make the association of the NGC/IC number with the accurate position.

That is what I am doing here.

The NGC/IC Revisions

Overview

The NGC -- and to a lesser extent, the two ICs -- has been used as a source for deep sky objects since it was published. Most of the large, bright optical "non-stellar" objects are in these catalogues. So, they have been subject to extensive work since their publication to determine their accuracy.

The unfortunate summary is "Not very accurate." Dreyer recognized this and published long lists of corrections as appendices to both ICs, as well as in his edition of WH's complete papers. Other early 20th century efforts at correcting the catalogues were made by (for example) Father Johann Hagen at the Vatican Observatory, Karl Reinmuth at Heidelberg, and several Harvard astronomers led by Harlow Shapley. Other lists of corrections came from Lick and Mt. Wilson Observatories in California, Helwan Observatory in Egypt, and several of the European observatories.

Guillaume Bigourdan at the Paris Observatory probably made the biggest contribution, observing about 6600 of the NGC and IC objects from the early 1880s to 1911, some several times over, measuring their positions as accurately as possible with respect to nearby stars. This enormous set of micrometric positions uncovered and corrected hundreds of the NGC's errors, and is still valuable today for helping to identify the objects that the 19th century visual observers saw.

The second half of the 20th century saw the husband-and-wife team of Gerard and Antoinette de Vaucouleurs carry on the vast chore of galaxy cataloguing as a necessary first step toward astrophysical studies of the galaxies. They and their colleagues (myself included) eventually published the three Reference Catalogues of Bright Galaxies where the problem of designation and position errors had to be dealt with for not just the NGC/IC objects, but for many other galaxies as well.

Other specialized catalogues in the visual wavelength region had an easier time of it. There are far fewer open and globular clusters; planetary, emission, and reflection nebulae; dark nebulae; and supernova remnants across the sky than galaxies, so the recent catalogues of these objects are -- at least compared to the galaxy catalogues -- relatively error-free.

Two explicit revisions of the NGC were published before 1990: "The Revised New General Catalogue of Nonstellar Astronomical Objects" by Jack Sulentic and William Tifft; and "NGC 2000.0", edited by Roger W. Sinnott (this second book includes the two ICs as well). Neither was particularly successful at removing errors from the old catalogues, but that was not a major goal in either case.

For that, amateur astronomers have played a major role in recent decades. The NGC/IC Project (of which my particular effort was a part) attempted to collect the most prominent of these into a single, coherent web site, and largely succeeded in that. Unfortunately, the Project itself, while still mostly available online, has become largely static since its latest revisions in 2007. Lately (January 2019), it has become clear that finding a dedicated web programmer to take over administration of the NGC/IC Project is not likely to happen. Steve Gottlieb is still trying to locate such a person, but he is not sanquine about his chances of success. So, it is likely that the NGC/IC Project site will continue as a static set of pages, stuck in 2007. Another major effort mounted by Wolfgang Steinicke led to his book discussing the observers who discovered the NGC objects, "Observing and Cataloguing Nebulae and Star Clusters". This was published in 2010 by Cambridge University Press, and remains the single, most complete reference work on the NGC discoverers in print. Along the way, Wolfgang noted hundreds of the NGC errors, but that was a secondary effort, so his book is not complete in that regard.

Accurate NGC/IC Positions and Corrections

The NGC/IC objects in the southern sky were relatively better covered well before the north thanks to the efforts of one man: Andris Lauberts. His ESO/Uppsala Survey of the southern sky (published in 1982) is a monumental work and was originally the single most valuable source of positions for all deep sky objects larger than an arcminute (and thousands smaller) south of -17.5 degrees. Among the southern objects for which he specifically searched were all the NGC and IC objects.

My own work with the de Vaucouleurs (in the Southern Galaxy Catalogue, SGC), and with Brian Skiff (SEGC, the south-equatorial extension to SGC) supplemented Andris's work, and led to generally correct identifications for the NGC objects -- particularly galaxies -- south of +3 degrees, as well as for the larger IC galaxies, and for many other southern deep sky objects.

Andris went through the Magellanic Clouds, too. In addition, the Clouds have been covered by several observers including Jenni Kay, Mati Morel, and Brian Skiff; and particularly Brent Archinal in his recent book "Star Clusters" with Steven Hynes (the objects in the Clouds are discussed further below). I have not yet had time to reflect all of this work here, though I have made a start. I have, however, done a thorough re-evaluation of the NGC and IC objects in both Magellanic Clouds with the original observations, almost all by John Herschel, at hand.

The northern sky was not as fortunate with respect to the NGC and ICs until 1997 (more below). Many of the objects are too faint to be included in the professional mid-20th-century surveys of brighter galaxies (the CGCG, MCG, and UGC), and -- as I noted above -- the simple existence of an NGC or IC number is not in itself sufficient to excite the interest of a professional astronomer intent on building a career "pushing back the frontiers of science." The same is true of the many so-called "non-existent" star clusters (in Sulentic and Tifft's "Revised New General Catalogue of 1973) discussed by Brent Archinal in his 1993 Webb Society monograph; they are just no longer interesting enough to professional astronomers to merit much study.

The NGC/IC position situation improved dramatically in 1997 when Wolfgang Steinicke self-published his "Revised New General Catalogue and Index Catalogue", both on paper and on the web. A second web edition appeared in mid-1998 and has been followed by more or less yearly updatings since. The version that I've adopted for the position files is the one that Wolfgang posted in 2001. His latest version (as of July 2021) is dated April 2021, and I encourage readers to look to that (or its successors) for a collection of data (magnitudes, diameters, and so forth) beyond the positions and object types that I list so far.

Using the available literature, or the digitized sky surveys (available in the 1990s as RealSky and the Digitized Sky Survey), Wolfgang collected or measured accurate positions for thousands of NGC and IC objects not previously having them. These include the nearly 1900 positions that Wolfgang took from the Lick NPM1G catalogue. These (and about 48,000 fainter) galaxies served as the reference frame for the Lick Northern Proper Motion survey, so their positions are very accurately known: standard deviations are less than 0.3 arcsec, though there is a 0.2 to 0.3 arcsec declination offset with respect to the ICRS. Wolfgang himself has measured thousands more positions using RealSky or RealSky South. There is also a systematic offset of about two arcseconds to the south in the declinations measured using RealSkyView; these are due to a bug in the RealSkyView software (the same bug affects positions derived from DSS with any software originating at Software Bisque in the early 2000s, e.g. "The Sky", "CCDSoft", etc.). I have folded all of these positions, without systematic corrections, into my accurate position files -- they are accurate enough to identify the objects they are meant to.

I've already noted that Wolfgang has published his PhD thesis on the discoverers of the NGC objects as a book, "Observing and Cataloguing Nebulae and Star Clusters". This book gives information about nearly all the early observers, their telescopes, and their observing techniques. It is a valuable complement to the actual lists of deep sky objects which they discovered. Much of Wolfgang's information on the NGC observers is also avialable online at NGC/IC Observers.

Similarly, the publication late in 2003 of "Star Clusters" by Brent Archinal and Steven Hynes has at last given us a nearly-definitive (as of the date of publication) list of star clusters in the Milky Way Galaxy, the LMC, SMC, M31, and the Fornax Dwarf Spheroidal Galaxy. I have typed in all of the positions for the NGC and IC clusters from this marvelous book. These positions come primarily from Brent Archinal himself, or from a massive paper in A&AS, 84, 527, 1990 by Kontizas et al for the LMC. While I had earlier called most of these LMC objects "open clusters" -- with only those with a "compact" note being called "globular" -- it's clear, simply looking at the clusters, that many are indeed actually globulars, at least morphologically. Many others are stellar associations ("star clouds"). I've since reclassified most of the Magellanic Cloud clusters based simply on their appearance, nevertheless noting that several of the LMC "globulars" are quite blue, probably reflecting their formation more recently than the old Galactic globulars.

The next three paragraphs are a detour to document my estimated diameters for the star clusters in the Magellanic Clouds and in the Galaxy as well. Photoelectric magnitudes and colors exist for most of the globular clusters in the Clouds, and similar data are available for many of the open clusters as well. I attempted to gather much of this together in 2018 and 2019; the results, along with my own estimates of major and minor diameters, are given in the base position files. While I have not yet transfered these data to the selected position files ("ngconly.txt" and "iconly.txt"), I will get to that soon.

The open cluster diameters are simply my estimates based on the red DSS2 images. But they are intended to represent an isopleth which encloses 80 to 90 percent of the population of the cluster as seen in the DSS2R images. These diameters must, of course, be treated cautiously as we know that the open clusters are "evaporating" -- as time goes on, more and more stars are ejected from them by the internal dynamics, or pulled from them through interaction with external objects, or even their parent galaxies themselves. This means that the clusters, like galaxies, generally have no "sharp edges". They simply fade away into interstellar space. Many of them are also found in rich star fields; this introduces additional uncertainty to any diameters estimated by eye.

Because I usually had no knowledge of which stars were actual cluster members, the diameters are "naive" -- what I saw in the images is all the information I had. For a more objective look at the clusters, I recommend the recent work by Kharchenko and his colleagues (see e.g. Global Survey of Star Clusters in the Milky Way ... and references therein). They have used proper motions and photometry to determine cluster membership, and have gone on to work out distances, ages, and so on from their data.

Back to positions! I also folded in other large lists of optical positions for galaxies and planetaries, including those from Brian Skiff at Lowell Observatory, Steve Gottlieb in Albany, California, and the extragalactic groups at Bologna, Italy (Vettolani et al), NRAO (Condon et al), and Lyons, France (Paturel et al).

This still left thousands of NGC/IC galaxies, nebulae, and clusters with either no accurate position or only a single position (subject to various kinds of errors including simple typos). For all of these objects, I have measured or estimated positions on the POSS prints, the SERC films, or on DSS images on HEASARC's SkyView web site; or -- ideally -- have extracted positions from one of the astrometric catalogues (primarily GSC until about 2006 or 2007; and Gaia DR1/2/3, Pan-STARRS1, 2MASS, SDSS, UCAC, URAT1, CMC, SMS1, and USNO-B1.0 after 2007).

Next, extending the work I've done to revise the list of 195 bright, large, and/or nearby galaxies in the Astronomical Almanac in the late 1990s and early 2000s, I have made an effort to measure very accurate positions for the few hundred brightest galaxies using the 2-Micron All-Sky Survey (2MASS) images available through IPAC's Infrared Science Archive web site. Though these images are at longer wavelengths (roughly 2 microns, as the survey name indicates) than the optical images I've used in the past, almost all the positions derived from them are for the galaxies' bright nuclei lost to overexposure on optical sky survey images. These nuclei are almost all coincident with the optical or radio nuclei. The exceptions tend to be "interesting" and/or unusual peculiar systems like NGC 520, or very inclined galaxies like NGC 4945 where the optical nuclei are hidden behind dust in the galaxies' disks.

Systematically, the 2MASS coordinate system is defined by ICRS. The positions for point sources measured on the 2MASS images also have quite small standard deviations, on the order of 0.2 arcseconds in both coordinates. The positions in the 2MASS extended source catalog (XSC; online at http://irsa.ipac.caltech.edu/applications/Gator/) and the 2MASS large galaxy atlas (T. Jarrett et al, AJ 125, 525, 2003) are not as good, typically with errors on the order of an arcsecond. Positions in the XSC are occasionally off for large objects near the edge of a 2MASS scan; superposed stars or companion galaxies can also confuse the position. The large galaxy atlas was created specifically to address this and other shortcomings of the automated processing that created the extended and point source catalogues.

More recently (2005 to the present) the Sloan Digital Sky Survey (SDSS) has provided positions and photometry (diameters as well as magnitudes) for half a billion objects, most in the northern sky. The positions are again very accurate (standard deviations on the order of 0.15 arcsec), but only for single, isolated objects with well-defined luminosity profiles. Large, multiple, bright, or otherwise complicated objects have often been broken up by the image-finding algorithm that SDSS uses, so data for the bright objects in the NGC/IC must be approached with considerable caution. For the smaller, fainter objects, however, the SDSS offers data of very high accuracy at optical wavelengths. Unfortunately (my point of view!), SDSS covers only about 30 percent of the sky. Though it continues today, its focus has recently been on deeper and specialized surveys emphasizing spectra. Its web pages of course have more information.

Other surveys that I regularly turn to include the US Naval Observatory surveys (the several UCACs, URAT1, and USNO-B1.0, in particular), the Hipparcos and Tycho catalogues, the Guide Star Catalog (particularly GSC 2.2, GSC 2.3 and GSC-ACT), the Carlsberg Meridian Catalogue, PPM/PPMXL, Pan-STARRS1, and especially the several Gaia data releases. Each has its strengths and weaknesses, but the data from them that I've adopted is not just highly accurate, but appropriate for the object at hand.

Since December 2003, there have been no gaps in the position lists. All of the NGC and IC objects have "accurate" positions, and all have been looked at (as of June 2016) by a "modern" observer with the history of the object in mind. In particular, Wolfgang Steinicke has collected data for the NGC and IC objects, as well as usually considering the observational circumstances of its discovery; Steve Gottlieb is well along with his visual observations of all the NGC objects; Malcolm Thomson has extensive notes on many of the IC objects; Courtney Seligman is building a "Sky Altas" around images and detailed descriptive notes of each of its objects, including all of the NGC and IC objects; and I have looked at images of all the NGC and IC objects, making notes for over 5300 of them (the "Notes" files mentioned above), as well as Brian's and my collection of accurate "selected" positions.

Still to be done is a reconciliation of these several lists. Most of the objects are labeled correctly in all the lists, but some differences remain. There are also differences of opinion among us as to which object(s) (if any!) a few of the numbers apply.

I also still have in progress a careful historical check of the positions that I have taken from other sources: are the objects that the modern positions point at really the ones that were found a century or two ago? While we have made a lot of progress toward answering that question for almost all of the NGC and IC objects, there is still some work that needs to be done. Regard these position files, along with the notes files, as a major step in that direction.

As I mentioned above, the positions have been examined by Brian Skiff and myself to include a single, selected, "best" position for each object. These selected positions are always very good, with errors on the order of at most 0.2-0.3 arcsec -- usually much better -- for stellar objects, or for objects with identifiable nuclei. We originally looked at a list of the accurate positions using VizieR to check the object in the surveys, catalogues, and lists noted above. We chose a position that is close to the mean of all the values available. I prefer positions from the surveys that scanned the skies directly -- Gaia, Pan-STARRS1, 2MASS, UCAC, URAT1, SDSS, CMC, Tycho-2 -- rather than those from photographic plates. Similarly, I prefer any of these "objective" positions to those estimated by me (or any other observer) from a photograph or on-screen image. I almost always reject obvious outliers, though in cases where the positions are more discordant than the usual errors suggest, I have listed them all and have selected one that more or less represents the center of the object. These are fortunately rare.

For the nebulae clearly associated with stars, such as planetary nebulae, reflection nebulae, and some HII regions, I almost always give the positions of the stars themselves as representative of the object. Occasionally, lobes or clumps of nebulosity within the larger object will warrant positions of their own; I list these as separate objects with unique names (based on the NGC number) and positions.

The exceedingly rich globular clusters, as well as the relatively compact clusters in the Magellanic Clouds, usually have more than a single position listed in the objective catalogues, probably for individual stars or close clumps of stars, as the result of incipient resolution. I've taken simple means of the several positions where this leads to a center that seems representative of the object. Lately, researchers needing kinematic or dynamical centers of globular clusters have adopted more sophisticated schemes based on isopleths or isophotes; see, for example, the papers by Goldsbury et al (AJ 140, 1830, 2010) and Noyola and Gebhardt (AJ 132, 447, 2006). I've adopted these positions when they represent the visual star distribution well enough.

For open clusters, I've been guided by the original observer's description and sizes (if given). These come primarily from the Herschels, with John Herschel's positions almost always quoted in the GC and NGC. I've listed these cluster positions in all cases for easy comparison with more modern positions. For the positions that I've determined myself using Goddard's SkyView images of the DSS2, I look, first, at 30 arcminute square fields using the red-plate images, or if those are unsatisfactory in some way -- e.g. two or more fields poorly stitched together, or a cluster larger than 30 arcminutes -- I switch to another color plate or a larger field (usually one degree square). Once I have the entire cluster in the image, I estimate the lengths of the major and minor axes and put the cluster's center at the intersection of the two axes. Thus, with perhaps a little effort, you can see the star distribution that I use to visually represent the cluster.

The problem with an approach like this for the poor, scattered clusters -- particularly those found by John Herschel -- is not being able to properly visualize his view from his description, especially when he does not give apparent diameters. This may have led to some rather fanciful "clusters" on the sky; NGC 6773, 6774, and 6775 are three such objects (see my notes on these). I doubt very much that these are physical clusters of stars, but the detailed studies needed to determine that either have not been done, or I've simply not taken the time to dig them out of the literature. I welcome pointers to papers discussing objects such as these. Of course, information on many of these clusters, published before 2003, will be found in the remarkably complete compendium "Star Clusters" by Brent Archinal and Steven Hynes. I am grateful, too, for extensive correspondence with Brent; he has clarified many objects that I might otherwise still be puzzling over.

Other diffuse objects like bright nebulae have their centers estimated in the same way, that is by assuming they are simple ellipses. They, of course, almost always have irregular shapes, so the approximation as an ellipse can sometimes give misleading positions. I've adjusted these by eye accordingly, usually on the DSS2 red images.

For late-type galaxies without obvious nuclei, I have chosen the center of the bar (if there is one) or of the disk as the position for the galaxy. All these positions, while carrying the lower precision warranted by my estimates, are nevertheless more than accurate enough for unambiguous identification of the intended NGC or IC object. They may not, however, be adequate for physical studies of the object -- caution is advised.

For the missing NGC/IC objects, I simply present the original observer's nominal position from the published monograph or paper. Occasionally, I have access to original observing logs or documents (e.g. the Herschel Archive from the Royal Astronomical Society, or sketches by the Leander-McCormick observers) with more information than eventually appeared in print. The position from the NGC or IC -- rather than the original published position -- is given only if it is relevant to the correct identification of the object, if it figures in a discussion of the adopted position, or if no other position was published prior to the NGC or IC (some observers -- e.g. Barnard -- sent lists of "new" nebulae directly to Dreyer, but never published them).

Finally, I'd better say something about NED, the NASA/IPAC Extragalactic Database, of which I was a team member from early 1991 until my retirement in mid-2011. NED is a database of literature and data about galaxies, aimed at the professional astronomer. Thus, it reflects the professional literature. However, as I note below, I corrected all of the NGC/IC identifications in NED in 2006, though some objects with reevaluations since then have not been updated. Many of NED's NGC/IC identifications have been wrong in the past. But as of March 2006, almost all are correct in that they represented my then-latest considerations of the objects' identifications; relatively few have changed since then.

Nevertheless, if you find a difference, trust my most recent files on this web site, not NED. (If you find a difference that is really annoying, send me a note so that I may ask my colleagues in Pasadena to fix NED. Or simply send NED a note directly.) The NED effort grew out of requests from NED users for the NGC numbers on objects in the Magellanic Clouds. Most of those are star clusters, and had not been previously included in NED. That work also led to the resolution of many more NGC puzzles, primarily in the south where Lauberts, Steinicke, or I had not made positive identifications.

Acknowledgments

Four special notes of thanks are due:

First, to the late Bob Erdmann (1943-2016) for his dozen years of work on the NGC/IC Project's Web site. It is through that work that the Project got its start on the World Wide Web, and that it has received the recognition that it has. Were it not for Bob's dedicated effort, our individual work would still sit in our separate file systems, or in scattered notes in the deep-sky journals known only to a few diehard fans of the historical sky.

Second, to Christopher Watson who picked up the Web site when Bob had to step down as webmaster in early 2009. Chris was able to take over the day-to-day chores of running the site with enviable ease. All of us who have data on the site, and who use it frequently, are grateful for Chris's database and Web experience that made the transition possible at all. Chris had to give up the webmaster's role in 2011 due to the pressure of work and family committments. After Chris stepped aside, the Web site was maintained by Mark Wagner and Steve Gottlieb, with financial support by myself until 2019.

Third, the NGC/IC Project has been kept online since mid-2017 by John Pierce, Mark Wagner, and Steve Gottlieb of the The Astronomy Connection, devoted to San Francisco Bay Area Amateur Astronomy. Their plans include a complete revision and updating of the site, using Wolfgang Steinicke's mature data base as a foundation.

Fourth, as mentioned above, in March of 2008 my long-time colleague and collaborator Brian Skiff at Lowell Observatory began collecting and assessing positions for NGC objects from the "new" astrometric catalogues like 2MASS, UCAC, USNO B-1.0, GSC 2.3, and others. This is fulfilling one of my already-mentioned long-term goals for this work -- letting it be an astrometric reference as well as a guide to the history of 19th century nebular astronomy.

I take great pleasure in thanking several people specifically for their help in this "asymptotic" quest for perfection. First, my colleagues in the NGC/IC Project who have contributed data and copies of the old literature, furnished positions, and suggested identifications, specifically (in no particular order) Brian Skiff, Malcolm Thomson, Wolfgang Steinicke, Steve Gottlieb, Brent Archinal, Chris Watson, Steve Coe, Jenni Kay, Ron Buta, Dave Riddle, and the late Bob Erdmann and Glen Deen. Most of these folks have also spent considerable time writing articles or preparing speeches about the project -- getting the word out and helping others to discover the rewards of historical deep-sky work.

Others who have done special digging in libraries and observatory archives, or who have made contributions of positions, time, or materials include Andris Lauberts, Brian Cuthbertson, Brian Marsden, Daniel Green, Tom DeMary, Harvey MacGillivray, Mati Morel, Jason Adamik, Jim Caplan, Sue French, Don Osterbrock, Leos Ondra, Yann Pothier, Marion Schmitz, Paul Brown, James Bryan, Tony Flanders, Guiseppe Longo, Glen Cozens, Jeffery Corder, Steve Waldee, Courtney Seligman, Jeff Kanipe, Perry Remaklus, Kathrine Haramundanis, Ashraf Shaker, Steven Dick, Brenda Corbin, Angus McDonald, and probably many others whom I must have forgotten -- my apologies to all who have fallen victim to my leaky memory. Please remind me; you deserve credit here, too.

I am also pleased to thank a few others for their direct or indirect contributions to this work:

Laurent Cambresy at CDS first suggested to me that this work might find a place in their holdings (VII/239), and Francois Oschenbein has prepared these lists for their inclusion in the CDS collection. Francois has also helped track down missing and obscure references that I had let go by the wayside over the years.

Gerard and Antoinette de Vaucouleurs got me started on this quest through their support and encouragement beginning in 1965, and ending only with Antoinette's death in 1987 and Gerard's in 1995. Their Reference Catalogues of Bright Galaxies led me directly into the historical byways.

I would also be remiss if I did not acknowledge the extraordinary patience of my wife Kathleen who sacrifices evenings and weekends to the never-ending quest for an accurate representation of the historical deep sky. Morris, Thumper, Angelica, Crabcake, Bunty, Dante, and Nicco have also provided much-needed grounding at various times over the past quarter century, pulling me back to their solid feline world of food and affection, clearly the real objects of earthly delight. Galaxies can sometimes, rightly, become mere phantoms in the sky.

I must also thank all those many musicians, and the audio and electronic engineers, who make the great music of the past thousand-plus years available to us at any time of day or night. We live in a special place and time ("a glorious accident"), and we are far more fortunate than we usually acknowledge or even realize. I'm happy now to explicitly recognize their work, and to paraphrase, "A day without music -- or galaxies! -- is a day without life."

Finally, here are some formal acknowledgments for the several data services without which almost none of this work would have happened. The people behind these services made this project possible, and continue to make it easy.

The Sky Surveys, DSS, and STScI:

The Digitized Sky Surveys were produced at the Space Telescope Science Institute under U.S. Government grant NAG W-2166. The images of these surveys are based on photographic data obtained using the Oschin Schmidt Telescope on Palomar Mountain and the UK Schmidt Telescope at Siding Spring. The plates were processed into the present compressed digital form with the permission of these institutions. [Go to http://archive.stsci.edu/dss/index.html]

The National Geographic Society - Palomar Observatory Sky Atlas (POSS-I) was made by the California Institute of Technology with grants from the National Geographic Society.

The Second Palomar Observatory Sky Survey (POSS-II) was made by the California Institute of Technology with funds from the National Science Foundation, the National Geographic Society, the Sloan Foundation, the Samuel Oschin Foundation, and the Eastman Kodak Corporation.

The Oschin Schmidt Telescope is operated by the California Institute of Technology and Palomar Observatory.

The UK Schmidt Telescope was operated by the Royal Observatory Edinburgh, with funding from the UK Science and Engineering Research Council (later the UK Particle Physics and Astronomy Research Council), until 1988 June, and thereafter by the Anglo-Australian Observatory. The blue plates of the southern Sky Atlas and its Equatorial Extension (together known as the SERC-J), as well as the Equatorial Red (ER), and the Second Epoch [red] Survey (SES) were all taken with the UK Schmidt.

HEASARC's SkyView:

We acknowledge the use of NASA's SkyView facility ( http://skyview.gsfc.nasa.gov) located at NASA Goddard Space Flight Center.

IPAC's Skyview:

This research has made use of IPAC's Skyview Image Display and Analysis Program, developed with support from the National Aeronautics and Space Administration. [Go to https://old.ipac.caltech.edu/skyview/].

NED:

This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. [Go to http://ned.ipac.caltech.edu].

IRSA (2MASS access):

This research has made use of the NASA/IPAC Infrared Science Archive (IRSA) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. [Go to http://irsa.ipac.caltech.edu].

SIMBAD:

This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France. [Go to http://simbad.u-strasbg.fr/].

VizieR:

This research has made use of the VizieR catalogue access tool, CDS, Strasbourg, France. The original description of the VizieR service was published in A&AS 143, 23, 2000. [Go to http://vizier.u-strasbg.fr/viz-bin/VizieR].

LEDA:

We have made use of the LEDA and HyperLEDA databases (http://leda.univ-lyon1.fr).

ADS:

This research has made use of NASA's Astrophysics Data System Bibliographic Services. [Go to http://ui.adsabs.harvard.edu/.]

SDSS:

Funding for SDSS-III has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, and the U.S. Department of Energy Office of Science. The SDSS-III web site is http://www.sdss3.org/. SDSS-III is managed by the Astrophysical Research Consortium for the Participating Institutions of the SDSS-III Collaboration including the University of Arizona, the Brazilian Participation Group, Brookhaven National Laboratory, Carnegie Mellon University, University of Florida, the French Participation Group, the German Participation Group, Harvard University, the Instituto de Astrofisica de Canarias, the Michigan State/Notre Dame/JINA Participation Group, Johns Hopkins University, Lawrence Berkeley National Laboratory, Max Planck Institute for Astrophysics, Max Planck Institute for Extraterrestrial Physics, New Mexico State University, New York University, Ohio State University, Pennsylvania State University, University of Portsmouth, Princeton University, the Spanish Participation Group, University of Tokyo, University of Utah, Vanderbilt University, University of Virginia, University of Washington, and Yale University.

JPL's Solar System Dynamics:

I am pleased to acknowledge JPL's SSD Small-Body Identification tool, and their HORIZONS System ephemeris generator for help with comets and asteroids mentioned in the old observations.

Gaia DR1:

This work has made use of data from the European Space Agency (ESA) mission Gaia, processed by the Gaia Data Processing and Analysis Consortium (DPAC). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

Gaia DR2 and EDR3:

This work has made use of data from the European Space Agency (ESA) mission Gaia ( https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

Pan-STARRS:

The Pan-STARRS1 Surveys were made possible through contributions by the Institute for Astronomy, the University of Hawaii, the Pan-STARRS Project Office, the Max-Planck Society and its participating institutes, the Max Planck Institute for Astronomy, Heidelberg and the Max Planck Institute for Extraterrestrial Physics, Garching, The Johns Hopkins University, Durham University, the University of Edinburgh, the Queen's University Belfast, the Harvard-Smithsonian Center for Astrophysics, the Las Cumbres Observatory Global Telescope Network Incorporated, the National Central University of Taiwan, the Space Telescope Science Institute, and the National Aeronautics and Space Administration under Grant No. NNX08AR22G issued through the Planetary Science Division of the NASA Science Mission Directorate, the National Science Foundation Grant No. AST-1238877, the University of Maryland, Eotvos Lorand University (ELTE), and the Los Alamos National Laboratory. The Pan-STARRS1 Surveys are archived at the Space Telescope Science Institute (STScI) and can be accessed through MAST, the Mikulski Archive for Space Telescopes. Additional support for the Pan-STARRS1 public science archive is provided by the Gordon and Betty Moore Foundation. [Go to http://panstarrs.stsci.edu/ ].

SkyMapper Southern Sky Survey:

The national facility capability for SkyMapper has been funded through ARC LIEF grant LE130100104 from the Australian Research Council, awarded to the University of Sydney, the Australian National University, Swinburne University of Technology, the University of Queensland, the University of Western Australia, the University of Melbourne, Curtin University of Technology, Monash University and the Australian Astronomical Observatory. SkyMapper is owned and operated by The Australian National University's Research School of Astronomy and Astrophysics. The survey data were processed and provided by the SkyMapper Team at ANU. The SkyMapper node of the All-Sky Virtual Observatory (ASVO) is hosted at the National Computational Infrastructure (NCI). Development and support the SkyMapper node of the ASVO has been funded in part by Astronomy Australia Limited (AAL) and the Australian Government through the Commonwealth's Education Investment Fund (EIF) and National Collaborative Research Infrastructure Strategy (NCRIS), particularly the National eResearch Collaboration Tools and Resources (NeCTAR) and the Australian National Data Service Projects (ANDS).