Flamsteed's Star Catalog

Introduction

John Flamsteed (1646–1719) was an English astronomer and first Astronomer Royal, the position of director of the newly built Royal Greenwich Observatory (1676) created by King Charles II. Flamsteed spent some forty years observing and cataloging more than 3000 stars visible from the observatory and studying the motion of the sun, moon, and planets. His lifetime work resulted in the publication of Historia Cœlestis Britannica in three volumes and Atlas Cœlestis. Part of the second and the third volume of Historia Cœlestis, as well as the atlas, were published posthumously under the care of Flamsteed's wife Margaret and his collaborators Joseph Crosthwait and Abraham Sharp.

Under the heading Stellarum Inerrantium Catalogus Britannicus (British Catalogue of Fixed Stars), the third volume of Historia Cœlestis [7] contains the official version of Flamsteed's star catalog with 2,936 entries. This British Catalogue is one of the most influential star catalogues in the history of astronomy and the most accurate of its time.

The British Catalogue groups stars in 55 constellations: 46 of the 48 Ptolemaic constellations (Ara and Corona Australis are omitted entirely, and constellations that extend too far south are only partially covered) and 9 newer constellations (Camelopardalis, Coma Berenices, Canes Venatici, Lacerta, Leo Minor, Lynx, Monoceros, Sextans, Vulpecula) that Flamsteed attributes to Hevelius, although the details are more complicated, as explained on Ian Ridpath's site Star Tales.

The so-called Flamsteed numbers assigned to stars by constellation have their origin in Flamsteed's catalog. As explained in Star Tales, these numbers do not appear explicitly in the official version of Flamsteed's catalog but in the version of it published by the French astronomer Joseph Jérôme de Lalande (1732–1807) in Éphémérides des Mouvements Célestes pour le Méridien de Paris [6]. This version contains 2,935 entries: it omits a suspicious entry without any coordinates in the constellation Lynx.

Flamsteed's star designations are still in use today but it is difficult to find a complete and accurate list of all Flamsteed stars, their Bayer letter assignments, and their designations in modern star catalogs such as the Yale Bright Star Catalogue, the Henry Draper Catalogue, or the Hipparcos Catalogue. Interested in tracing the source of Flamsteed's star designations, I have looked for an electronic version of Flamsteed's catalog on the web, without success. So I have made one myself from digitized versions of Flamsteed's Historia Cœlestis Britannica and Lalande's Éphémérides des Mouvements Célestes (links provided on Star Tales), and published it here in the hope that it may be useful to someone else interested in historical star catalogs.

Catalog data

The file flamsteed_f.dat contains Flamsteed's version of his catalog from Historia Cœlestis [7]. The file flamsteed_l.dat contains Lalande's version of Flamsteed's catalog from Éphémérides des Mouvements Célestes [6]. The common format of these files is described in the ReadMe file, according to the conventions for catalog descriptions of the VizieR service.

File name	Explanation
ReadMe	File descriptions
flamsteed_f.dat	Flamsteed's catalog according to Flamsteed
flamsteed_l.dat	Flamsteed's catalog according to Lalande
flamsteed_diffs.dat	Differences between versions and general remarks

I have copied the data as carefully as I could and triple-checked my work. I am confident that the files are at least very close to the originals but, human error being what it is, I cannot be perfectly sure that no error is left.

According to the title page, the catalog's equinox (and epoch) is 1689 Annum Completum, or January 1st 1690 of the Julian calendar (still in effect in England at the time). This corresponds to Julian Day 2,338,331 or Besselian epoch B1690.0316… or Julian epoch J1690.0369…

In order to verify ecliptic coordinates, I used an axial tilt angle of 23° 29', the value I believe Flamsteed used. Indeed, it minimizes the error between the ecliptic positions given by Flamsteed and the ones I computed from the equatorial ones. Also, as far as I could make out (my Latin being more than rusty), Flamsteed mentions this value and how he measured it as the declination of the Sun on June 10, 1690 (the summer solstice, since England was still using the Julian calendar at the time) on page 115 of the introduction (Prolegomena) to the catalog.

Differences

The file flamsteed_diffs.dat documents all the significant differences between Flamsteed's and Lalande's versions, and other noteworthy features. Its format is more or less free-form and self-explanatory. It is not a pure ASCII file; it contains a few UTF-8 characters (degree symbol and accented letters).

When meaningful, I have indicated by (F) or (L) which of the the Flamsteed or Lalande version I found preferable. In the case of numerical differences, "preferable" refers to the internal consistency of the data for the star (e.g., do the ecliptic coordinates correspond more closely to the equatorial ones), not correctness with respect to modern data. While Lalande has corrected some errors in Flamsteed's catalog, he has introduced others, most of them probably typos.

Older version

The initial version of Flamsteed's catalog that I published on this site consisted of a single file containing both versions and special markers to distinguish the discordant records. After gaining more experience with other star catalogs, I found it preferable to keep different versions in separate files and to avoid default values (inherited from other fields in the same record) and implicit values (inherited from the same field in a previous record). I've also slightly reorganized the colums. Just in case the old version is still needed, it is available, along with its specification, in the following files.

File name	Explanation
ReadMe_old	File description
flamsteed_old.dat	Flamsteed's catalog (both versions)

Corrections

Unfortunately, the catalog contains more errors than it would have if Flamsteed had lived long enough to see it to completion. Flamsteed faced many impediments during his lifetime, including poor health, lack of funding, and a bitter dispute involving Isaac Newton and Edmond Halley about the unauthorized publication of an unfinished version of the catalog.

Flamsteed's life, his methods, and the circumstances surrounding the creation of the catalog are described in great detail in the Account of Flamsteed [5] by the English astronomer Francis Baily (1774–1844). Baily collected Flamsteed's original manuscripts and reviewed Flamsteed's observations and computations in order to explain errors in the catalog and to augment it with stars that Flamsteed had observed but that were omitted from the published version. The Account of Flamsteed contains Baily's corrected and enlarged version of the British Catalogue and detailed descriptions of the corrections.

The following sections, based on Baily's work mentioned above and the second part of Morton Wagman's book Lost Stars [1], discuss some of the issues that affect the catalog. Related data is gathered in the following files.

File name	Explanation
flamsteed_dup.dat	Duplicate entries, stars represented by two catalog entries
flamsteed_nex.dat	Non-existent entries and entries with missing coordinates
flamsteed_pos.dat	Position corrections by Francis Baily

Duplicate entries

There are 22 stars represented by two entries in the catalog. This may be intentional in the cases of the pairs 112 Tau = 23 Aur and 79 Aqr = 24 PsA, which were already duplicated in Ptolemy's catalog. In other cases, the positions or magnitudes given for the two entries are slightly different, which suggests that they result from independent measurements. Baily [5, p. 392] states that the inclusion of duplicates is an oversight of the editors and that Flamsteed intended to avoid it.

The file flamsteed_dup.dat corresponds to Table A in Baily [5, p. 645]. It includes each pair of duplicate entries twice, in reversed orders, and mentions the star's actual constellation. The right ascension of 58 Cam (starred in the file) is in fact not given in the catalog, but Baily [5] in his Note 1151, p. 557 asserts that 58 Cam is the same star as 30 Lyn.

Non-existent entries and entries with missing coordinates

Some entries in Flamsteed's catalog don't seem to correspond to any actual star. There are cases of measurement, computation, or transcription errors, at least one case corresponding to a planet (34 Tauri, Uranus), possible cases of now extinct novae (such as 11 Vulpeculae), and several cases for which no star can be found at the position given by Flamsteed and no definitive explanation has been found.

These cases are summarized in Baily's Tables B, C, D [5, pp. 645–646] and described at length in his Notes [5, pp. 506–643]. About half a century later, C. H. F. Peters [4] re-examined the entries in Baily's Table B in more detail and sometimes came to different conclusions. Wagman [1] summarizes the contributions of these authors, and others when applicable, in the second part of his book.

The file flamsteed_nex.dat corresponds for the most part to Tables B, C, D, E in Baily [5, pp. 645–646]. Its format is described in the ReadMe file. It includes entries with no corresponding star (marked with a *) and the ones with missing coordinates (Table E), along with a hint of explanation or the source of the coordinates supplied by Baily in his revised catalog.

Erroneous coordinates

While reviewing Flamsteed's notes, Baily has corrected the computation and transcription errors that he could find and included the corrected values in his revised catalog. He only intended [5, p. 387] to restore the values that Flamsteed intended to reach, not to override them with more modern results.

The file flamsteed_pos.dat corresponds to Table F in Baily [5, pp. 647–650]. It specifies coordinate increments for stars with erroneous coordinates. For convenience, it also supplies the coordinates assigned by Baily (in his revised catalog) to the stars of Table E and mentioned in flamsteed_nex.dat.

There are a few errors and omissions in Table F, that I've tried to correct. The differences between Table F and flamsteed_pos.dat are documented in the ReadMe file.

Star identifications

Flamsteed's catalog specifies Bayer designations of stars when they exist. Flamsteed also assigned Bayer-style letters of his own to many stars. With a few exceptions, when a Bayer letter corresponds to more than one star in his catalog, Flamsteed assigns an index to the corresponding stars. For instance, 77 and 78 Tauri are marked 1 ad θ and 2 ad θ (1 near θ and 2 near θ) respectively. This feature evolved into the superscript notation used today: 77 and 78 Tauri are now called θ¹ and θ² Tauri.

An important issue for any star catalog is to establish a correspondence between its entries and entries in standard catalogs of the day. Flamsteed's catalog indeed includes cross-references to the catalogs of Ptolemy, Tycho Brahe, Hevelius, and to Bayer's atlas.

Bayer letters

Many of the Bayer designations present in Flamsteed's catalog are now obsolete. On one hand, the letters introduced by Flamsteed have been almost completely discarded. Baily [5, p. 397] thinks that he never intended them to make it to the final version anyway (*). On the other hand, later astronomers have disagreed with some of the associations and attempted to correct them, not always in agreement with each other. Also, the letter assignments in some of the constellations present in Flamsteed's catalog (Argo Navis, Centaurus, Lupus) have been affected by Nicolas-Louis de Lacaille's redesign or relettering of the constellations of the southern hemisphere.

In his revised British Catalogue, Baily has already updated some of the letter assignments. Later astronomers have introduced more changes, some of which have survived and some not. These issues are exhaustively discussed and elucidated in the first part of Morton Wagman's book Lost Stars [1].

(*) For what it's worth, I'm not convinced of this since these letters are used alongside Bayer's to identify stars in versions of other catalogs (Ptolemy's, Ulugh Beg's, Tycho Brahe's and Hevelius's) included in Historia Cœlestis.

Bright Star and Draper numbers

The Yale Bright Star Catalogue is a reference for stars brighter than magnitude 6.5. It identifies stars by HR numbers (for Harvard Revised after the predecessor Harvard Revised Photometry catalog). It includes Flamsteed designations for Flamsteed stars bright enough to be included and whose Flamsteed designation matches the actual constellation of the star. For fainter stars, the HD numbers from the Henry Draper Catalogue are often used.

Morton Wagman's book Lost Stars [1] provides the HR and HD numbers for all existent Flamsteed stars. His earlier article Flamsteed's Missing Stars [3] provides the HR or HD numbers for Flamsteed stars that cannot be found in the Bright Star Catalogue by Flamsteed designation. The article addresses

• nonexistent stars,
• stars with duplicate entries,
• stars absent from the Bright Stars Catalog because they are too faint,
• stars whose Flamsteed number does not appear in the Bright Stars Catalog because the constellation of their Flamsteed designation differs from the constellation to which they now belong.

Also, W. P. Bidelman's article Flamsteed stars not contained in the Yale "Catalogue of Bright Stars" [2] provides the Henry Draper catalog numbers for the Flamsteed stars that do not have a Bright Star number.

Identifications file

The file flamsteed_ident.dat, based on these sources, collects redundant star identifications for all the existent Flamsteed stars: Bayer / Lacaille letters and indexes in use today, Flamsteed designations in the actual constellation of the star, and Bright Star, Draper, and Hipparcos catalog numbers.

File name	Explanation
flamsteed_ident.dat	Star identifications

Its format is described in the ReadMe file. It is subject to automatic validation using the Anafile package of the VizieR service.

Accuracy of the catalog

The file flamsteed_ident.dat also gives the differences between Flamsteed's positions (uncorrected by Lalande or Baily) and the modern positions of the corresponding stars from the Hipparcos catalog (or the SAO catalog if there is no Hipparcos match) adjusted to Flamsteed's epoch and equinox.

The file gives the differences in right ascension (multiplied by the cosine of the declination for normalization), the differences in declination, and the great circle distances, all in minutes of arc. The corresponding histograms below give an idea of the overall accuracy of the catalog (data outside the range of the x axis has been discarded).

Right Ascension	Declination	Distance

In Ptolemy's and Tycho's cases the standard deviations of the longitude and latitude differences have similar values. Here, the standard deviations of the RA and DE differences are quite different (~1.0 and ~0.4 respectively). I would venture to say that the declination accuracy is twice as good as the right ascension accuracy in Flamsteed's catalog.

The peak of the distance histogram occurs at about 0.6 arcmin, vs. 2 arcmin in Tycho's case. Using this as a measure of accuracy, we could say that Flamsteed's catalog is about 3 times more accurate than Tycho's.

References

[1] Morton Wagman, Lost Stars, McDonald & Woodward Publishing Company, Blacksburg, Virginia, 2003.

[2] W. P. Bidelman, Flamsteed stars not contained in the Yale "Catalogue of Bright Stars", Bulletin d'Information du Centre de Données Stellaires, Vol. 38, 1990, p. 13.

[3] Morton Wagman, Flamsteed's Missing Stars, Journal for the History of Astronomy, Vol. 18, No. 3, 1987, p. 209.

[4] Christian Heinrich Friedrich Peters, Flamsteed's Stars "Observed, but not Existing", Memoirs of the National Academy of Sciences, Vol. III, Part 1, 1884, pp. 69–83.

[5] Francis Baily, An account of the Reverend John Flamsteed, the First Astronomer-Royal; compiled from His Own Manuscripts, and Other Authentic Documents, Never Before Published. To which is added his British Catalogue of Stars, Corrected and Enlarged. London, 1835.

[6] Joseph Jérôme de la Lande, Éphémérides des Mouvements Célestes pour le Méridien de Paris, tome Huitième, Paris, 1783.

[7] John Flamsteed, Historia Cœlestis Britannica, Tribus Voluminus contenta. Volumen Tertium, London, 1725.