Abstract

We present stellar photometry in the UBV passbands for the globular cluster M5 ≡ NGC 5904. The observations, taken from short-exposure photographic plates and CCD frames, were obtained in the Ritchey-Cretien (RC) focus of the 2-m telescope of the National Astronomy Observatory ‘Rozhen’. All stars in an annulus with radius 1 ≤ r ≤ 5.5 arcmin were measured. We show that the ultraviolet (UV) colour–magnitude diagrams (CMDs) describe different evolutionary stages in a better manner than the ‘classical’V, B−V diagram. We use HB stars, with known spectroscopic Teff to check the validity of the colour zero-point. A review of all known UV-bright star candidates in M5 is made and some of their parameters are catalogued. Six new stars of this kind are suspected on the basis of their position on the CMD. New assessment of the cluster reddening and metallicity is done using the U−B, B−V diagram. We find that [Fe/H]=−1.38, which confirms the Zinn & West value, contrasting with recent spectroscopic estimates. In an effort to clarify the question of the gap in the blue horizontal branch (BHB) stellar distribution and to investigate some other peculiarities, we use the relatively long-base colour index U−V. A comparison of the observed V, (U−V)0 distribution of horizontal branch (HB) stars with a canonical zero-age horizontal branch (ZAHB) model reveals that the hottest stars rise above the model line. This is similar to the ‘u-jump’ found in the Strömgren photometry. 18 BHB stars with (B−V)0∈[−0.02/0.18] are used to estimate their ultraviolet deficiency. It is shown that low-gravity (log g≤2) Kurucz's atmospheric models fit the observed distribution of these stars along the two-colour diagram well.

Introduction

The photometric works on globular clusters (GCs) very often do not include the U-band observations. The U band, however, is very important, because it provides a comparatively long-base colour index graphic, which removes the degeneracy in horizontal branch (HB) colours for graphic and is also the most sensitive temperature parameter for early-type stars in the UBV system (Buser & Kurucz 1978). This colour index is furthermore useful to obtain stellar atmospheric parameters using theoretical models (e.g. Kurucz 1992). Taking in mind the good resolution of the RC-focus graphic, scale 12.86 arcsec mm−1) of the 2-m telescope of the National Astronomy Observatory ‘Rozhen’, we started a program some years ago in studying the central parts of some globular clusters. Here the results concerning graphic 5904 are presented. In this case, our aim was to produce UBV mag for a large, statistically significant sample of stars at the level of the horizontal branch and above, and to investigate some puzzling colour–magnitude diagram (CMD) features. We draw special attention to stars occupying the blue horizontal branch (BHB). The distribution of these stars poses a number of questions that are still open and need further investigation, as does, for example, the presence of a gap in the BHB, the ultraviolet deficiency of some BHB stars and the real location of the bluest HB stars on the log L, log Teff diagram.

In this paper we present the observational material and reduction procedures in Section 2. The UV-diagrams are analysed in Section 3. The main topics in this section are the presence of a gap, the UV-bright star candidates and the discrepancy between the observed BHB and the zero-age horizontal branch (ZAHB) model. The two-colour diagram features are discussed in Section 4. This includes the observed ultraviolet deficiency of BHB stars and the ultraviolet excess of the red giant branch (RGB) stars. An assessment of the cluster metallicity and reddening is also given. The results obtained are summarized and discussed in Section 5.

Observational material

The present investigation of M5 is based on photographic plates and CCD frames taken of the central part of the cluster. To achieve a better resolution of the stellar images, all the observations were obtained with short-time exposures. The average seeing, measured at full width at half-maximum (FWHM) of the stellar image, is ∼2 arcsec. On each plate (5U, 6B and 5V), nine overlapping regions, covering a distance of up to 5.5 arcmin away from the cluster centre, were scanned with a Joyce–Loebl microdensitometer. The data were smoothed by means of a wavelet transform and subsequently converted into relative intensities using specific calibration curves for each plate (for details see Markov 1994). The digital images were analysed with originally developed reduction procedures. To check the reliability of these procedures part of the observations were processed using daophot (Stetson 1987); a detailed comparison of the results is given in Markov et al. (1997). The efficiency of the wavelet transform in localizing stellar images in crowded fields is described in Borissova et al. (1997). A similar approach had been previously employed by Auriére & Coupinot (1989) to obtain improved photometry for stars in dense globular cluster fields. In order to obtain more precise photometric calibration, supplementary 3V and 3B CCD observations of the M5 core were taken on 1997 March 16 with a SBIG ST-6 camera (pixel graphic. Seeing conditions for the night were stable with graphic. The standard field in M92 of Christian et al. (1985) was observed in order to calibrate the program images.

The transformation of the instrumental magnitudes to the standard U system was performed by means of 24 stars from Arp's (1962) photoelectric standard sequence in the field of M5. This sample covers a wide range of colours: graphic, graphic and graphic. The BHB stars are of special interest in our investigation and separate transformations to the standard system have been employed for them. In this colour region, 7 Arp's photoelectric standard stars and three secondary photometric standard stars, taken from Richer & Fahlman (1987), were used.

The calibration into the B, V standard systems was performed in two steps. First, we used the CCD observations to calibrate BV instrumental magnitudes with respect to the primary standard stars in M92 (Christian et al. 1985). Then, 86 stars, common for photographic and CCD measurements, were used as a local secondary standard to calibrate all other program stars. The mean internal errors are in the range of graphic. The total error including the uncertainties in the photoelectric data is ±0.03 in the stellar magnitudes and ±0.04 in the colour indices.

Our photometry can be compared with earlier works, for example

  • (i)

    The UBV CCD photometry of. For 18 stars in common, the mean differences between our data and those of von Braun et al. (1998) are: graphicgraphic and graphic. Obviously, the two magnitude sets match each other within the error. A systematic difference of about 0.06 mag was found for the U magnitudes of five asymptotic giant branch (AGB) stars (in a sense, von Braun's data are brighter). These stars were not included in the above assessments.

  • (ii)

    The photographic photometry of. The mean difference in the zero point graphic, based on 182 common stars, is 0.01 mag. There seems to be a small colour difference of 0.03 mag for BHB stars (our magnitudes are redder).

  • (iii)

    The CMD mean lines of. Fig. 1 shows the V, graphic diagram on which the Sandquist et al. (1996) fiducial lines are overlaid for comparison.

Figure 1.

V, graphic diagram for stars measured in the annulus with radius graphic.

Figure 1.

V, graphic diagram for stars measured in the annulus with radius graphic.

UV colour–magnitude diagrams

The V, U−V diagram

Fig. 2(a) shows the V, graphic diagram for M5. Two features of this diagram are worth noting: the BHB in this plane is not as steep as that in V, graphic (Fig. 1); the BHB stars form an almost linear sequence with a discontinuity at graphic and graphic. The latter is illustrated in the histogram (Fig. 2a). To test the statistical significance of this BHB feature, the ℓHB co-ordinate deduced from Crocker & Rood (1985), Ferraro, Fusi Pecci & Buonanno (1992) and Dixon et al. (1996) is used. ℓHB is a measure of the star position along the HB in the MV, graphic diagram as marked by its projection on the HB ridge line. The ℓHB parameter grows from the red extreme (RE) to the blue end of the HB. The zero point for ℓHB is set at graphic. The HB ridge line is determined as a non-linear fit to the sample of 160 HB stars supplemented with 68 RR Lyr variables. Their colours are assumed to be randomly distributed inside the instability strip boundaries graphic with a mean magnitude graphic. The scale ratios used are graphic and graphic. A histogram of the ℓHB distribution is shown in Fig. 3 and the cumulative distribution is shown in Fig. 4. The gap is clearly visible at graphic. Following Hawarden (1971) and Newell (1973), we tested the significance of the gap by means of ‘χ2 statistics’ applied to the cumulative distribution, and found that the gap is statistically significant at the 99.93 per cent level. However, Catelan et al. (1998) have recently argued that the Hawarden-Newell technique substantially overestimates the probability that the gap is real. This result, together with the fact that only 80 per cent (228 stars) of all observed HB stars have been involved in our analysis, make the reality of the gap not convincing enough.

Figure 2.

The UV colour–magnitude diagrams. In both diagrams, the place of the suspected gap is indicated by a bar; the UV-bright stars are marked with squares. In panel (a), above the BHB, the histogram of the stellar counts found in the corresponding colour bins is shown.

Figure 2.

The UV colour–magnitude diagrams. In both diagrams, the place of the suspected gap is indicated by a bar; the UV-bright stars are marked with squares. In panel (a), above the BHB, the histogram of the stellar counts found in the corresponding colour bins is shown.

Figure 3.

The ℓHB distribution histogram.

Figure 3.

The ℓHB distribution histogram.

Figure 4.

The ℓHB cumulative distribution function.

Figure 4.

The ℓHB cumulative distribution function.

Fig. 5 shows the theoretical ZAHB [Dorman, Rood & O'Connell (1993), oxygen-enhanced model, graphicgraphicgraphicgraphic overlaid on the observed HB. Squares denote stars with spectroscopically derived Teff (Crocker, Rood & O'Connell 1988). The compatibility of both sets is satisfactory. Fig. 5 shows that although practically all the HB stars lie above the theoretical sequence, the discrepancy becomes marked above graphicgraphic. Grundahl et al. (1999) observed a similar effect (called a ‘u-jump’) in 14 GC (including M5) from u, y photometry. These authors suggested that ‘…the u-jump is a ubiquitous feature, intrinsic to all HB stars hotter than graphic’.

Figure 5.

Comparison between the observed BHB (grey points) and the ZAHB (solid line). Note the general discrepancy between observations and theory as well as the more rapidly rising discrepancy for colours below graphic. The squares mark the positions of M5 BHB stars with spectroscopically measured Teffand log g (Crocker et al. 1988).

Figure 5.

Comparison between the observed BHB (grey points) and the ZAHB (solid line). Note the general discrepancy between observations and theory as well as the more rapidly rising discrepancy for colours below graphic. The squares mark the positions of M5 BHB stars with spectroscopically measured Teffand log g (Crocker et al. 1988).

The U, U−V diagram

We note the following points regarding the U, graphic diagram (Fig. 2b).

  • (i)

    The stars of the upper part of the RGB are as bright as the BHB stars. The U magnitude of the brightest RGB stars increase as the stars become redder.

  • (ii)

    The AGB is clearly distinguishable from the RGB being, in average, more than 0.20 mag brighter. Note that the brightest AGB stars have an astrometric membership probability of graphic per cent (Spassova & Michnevski 1981; Rees 1993). The star on the AGB tip, denoted with a filled square, is ZNG2, a UV star (Zinn, Newell & Gibson 1972).

  • (iii)

    The group of the so-called ‘UV-bright stars’ is clearly separated from the principle sequences. They are indicated by filled squares in Fig. 2. Six of the seven UV-bright stars from Zinn et al. (1972) are measured in our study. Three of them are field stars, and one is a variable 41 (Sawyer Hogg 1973). The star ZNG7 is probably a variable 25. On the frames it appears blended with the RGB star III–140 (BCF). Our U photometry confirms that the cluster member III–37 (Cudworth 1979) is a UV-bright star: although both magnitudes and colours show big dispersion in all studies its position on the classical V, graphic diagram is above the blue HB. Rees (1993) noted as other possible UV-bright stars: III–48 (Arp 1962), B169 and B200 (Barnard 1931). Our results confirm this status only for B169. We suspected four new possible UV-bright stars: I–17, I–44, II–22 and the star numbered 62 in our catalogue. Another two stars were added to the list according to their position on the V, graphic diagram. These two stars, with magnitudes and colours graphic, graphic and graphic, graphic, were detected in the central part of the cluster. Thus, 15 UV-bright stars candidates are found by different authors in M5. Table 1 lists the UV-bright stars, including some that probably belong to the so-called ‘blue nose’ (ƒ ≡ field, graphic member).

Table 1.

Data for UV-bright star candidates detected in GC M5.

Table 1.

Data for UV-bright star candidates detected in GC M5.

The colour–colour diagram

The graphic, graphic relation can be used to determine the foreground reddening towards M5. Our intrinsic colours were derived assuming reddening graphic, with a slope of graphic. For 38 BHB stars with graphic we estimate graphic. The spectroscopically determined Teff and log g for 11 HB stars in M5 (Crocker et al. 1988), also allow us to determine the cluster reddening. The (graphic values for these stars were derived using corresponding relations found in Kurucz (1992). In this way we obtained graphic.

Fig. 6 shows the two-colour diagram for RGB (left-hand panel, crosses) and AGB (right-hand panel, triangles) for stars with V magnitudes above the level of the HB graphic. In the both panels, the unreddened intrinsic two-colour line for Population I normal (luminosity class III) giants (Buser & Kurucz 1992) is drawn (solid line). The two types of stars show the well–known ultraviolet excess graphic. This excess, measured at graphic, is a sensitive indicator for the metal abundance (Wallerstein & Helfer 1966; Sandage 1970). In our case graphic for all stars with graphic were averaged. The mean excess Δ(graphic is graphic. Using the Buser & Kurucz relation between graphic and [M/H] we obtained graphic The quoted error is derived only from the formal error propagation in the photometric and model data. The M5 metallicity assessments found in the literature show a wide range (Briley et al. 1992, their table 3). Our value is in agreement with that found in Zinn & West (1984), which is −1.4, but is inconsistent with the more recent high-resolution spectroscopic measurements carried out by Carreta & Bragaglia (1998), graphicCarreta & Gratton (1997), −0.9, and Sneden et al. (1992), graphic. While we could not offer a certain answer on this topic, we would like to note that Borissova et al. (1999) pointed out a remarkable similarity between the upper red giant branches of M5 and NGC 6229. For this reason only, one might expect both clusters to share the same metallicity. The reliability of the value derived photometrically for NGC 6229 in Borissova et al. (1999), graphic, (based on the Zinn & West scale) is confirmed by Wachter et al. (1998). These authors obtained the global metallicity of the cluster, graphic, from a direct medium-resolution spectroscopy of cluster giants.

Figure 6.

The two-colour diagrams for RGB (left-hand panel) and AGB (right-hand panel) stars. The standard Population I line (Allen 1973, solid line) is overlaid on both diagrams. In the right-hand panel, the mean position of the RGB stars is marked with a dashed line.

Figure 6.

The two-colour diagrams for RGB (left-hand panel) and AGB (right-hand panel) stars. The standard Population I line (Allen 1973, solid line) is overlaid on both diagrams. In the right-hand panel, the mean position of the RGB stars is marked with a dashed line.

The AGB stars (within the same colour range) show a smaller ultraviolet excess in comparison with the RGB stars. This is demonstrated in Fig. 6 (right-hand panel), where the mean position of the RGB stars is marked with a dashed line. The difference in the UV excesses for RGB and AGB stars is obvious. The implied difference in graphic is 0.10 mag at graphic. A smaller UV excess of the AGB stars has been found in practically all globular clusters with available UBV measurements.

Fig. 7 illustrates the HB in the colour–colour plane based on 161 stars (114 BHB and 47 RHB). A dashed line shows the empirical Population I line. A solid line indicates the model ZAHB transformed to graphic, graphic. This two-colour diagram reveals the following features.

Figure 7.

A two-colour diagram for HB stars. The empirical Population I line is marked with a dashed line. ZAHB is also overlaid (bold solid line). Two-colour sequences taken from the Kurucz (1992) atmospheric models for two different values of log g are marked with filled stars and squares.

Figure 7.

A two-colour diagram for HB stars. The empirical Population I line is marked with a dashed line. ZAHB is also overlaid (bold solid line). Two-colour sequences taken from the Kurucz (1992) atmospheric models for two different values of log g are marked with filled stars and squares.

  • (i)

    Stars with graphic fit the empirical Population I relation and ZAHB very well. These are stars occupying the region in the graphic, graphic plane, where differences in g and metal abundance have negligible effect on the observed colours.

  • (ii)

    The cool BHB stars, appearing just to the blue side of the RR Lyrae instability strip, show a small but definite ultraviolet deficiency with respect to the intrinsic two-colour line (Allen 1973). The mean deficiency measured for 18 stars in the colour interval graphic is graphic This effect was first found by Arp (1962) and is based on only five BHB stars with photoelectric measurements. The same effect was also observed in some other clusters: NGC 1904 (Stetson & Harris 1977; Kravtsov et al. 1997), 3201 (Lee 1977), 6723 (Menzies 1974), 5139 (Newell et al. 1969b), 6397 (Newell et al. 1969a) and 1841 (Alcaino et al. 1996), and M 15 (Sandage 1970). To quantify the effect, we reanalysed available UBV data for BHB stars in some of these clusters using the graphic values found in Harris (1996). Table 2 summarizes the results.

    The third column shows the number of stars used. However, extensive photoelectric observations of BHB stars in other globular clusters give somewhat conflicting results. Data for M3, M13 and M92 (Sandage 1970) place such stars essentially on the standard Population I two-colour line. Although the possibility of a small systematic error in our U magnitude cannot be totally neglected, the observed effect in the M5 two-colour diagram may be at least partly real. In our opinion, the wide range of the ultraviolet deficiency among the investigated globular clusters (in many clusters this HB feature is completely missing) points to a possible connection between this observable and the cluster parameters.

    The ultraviolet deficiency of the BHB stars with respect to the canonical ZAHB sequence is smaller (see Fig. 7). Although the scatter of the data is large — mainly as a result of the low quality of the photographic observations — the deviation from the model line is systematic and appears to be real. From Kurucz's (1992) computations it is evident that the graphic colour is a gravity indicator for stars with graphicgraphic. This point is illustrated in Fig. 7, where the graphic and graphic model loci are plotted. It is clearly seen that a satisfactory overlapping of the observed and theoretical sequences is achieved involving low-gravity graphic atmospheric models transformed to the graphic, graphic plane.

    The suggestion of Arp (1962) that ‘… fainter BHB stars, as they get very blue have an increasingly large ultraviolet deficiency relatively to Population I line’ is not confirmed.

  • (iii)

    The RHB stars do not show any ultraviolet deficiency or ultraviolet excess with respect to the standard Population I colour–colour line. It has been known (Sandage 1969, 1970) that the sequence of metal-deficient RHB stars follows an empirical Population I relation. Sandage (1970) gives an explanation in terms of the ‘fortuitous’ cancellation of the effects of a low metal abundance and low surface gravity.

Table 2.

The ultraviolet deficiency for cool BHB stars detected in some globulars.

Table 2.

The ultraviolet deficiency for cool BHB stars detected in some globulars.

Summary and conclusions

UBV photographic and BV CCD observations in the central part graphic of the globular cluster graphic 5904 are presented. The UV colour–magnitude and two-colour diagrams for stars down to graphic are analysed.

The main results of our analysis can be summarized as follows.

  • (i)

    We confirm the presence of a gap in the BHB stellar distribution of M5 that was announced by Brocato, Castellsni & Ripepi (1995) and Drissen & Shara (1998). In the V, graphic diagram, the gap is visible at graphic and graphic. Despite the high statistical significance of this BHB feature, we still consider the answer to the question of its reality uncertain.

  • (ii)

    The ultraviolet excess observed for RG stars, graphic, is used to estimate the metallicity of the cluster. The value obtained in our study, graphic, confirms that derived by Zinn & West (1984), which is −1.4.

  • (iii)

    Six stars are suspected to be new ‘UV-bright’ stars. The UBV data for these stars are summarized in Table 1.

  • (iv)

    In the V, graphic diagram we find that the hottest BHB stars tend to lie above the canonical ZAHB at a temperature threshold of graphic. This resembles the effect observed for 14 globular clusters by Grundahl, Vandenberg & Andersen (1998) and Grundahl et al. (1999, hereafter GCLSA) and called a ‘u-jump’. In our investigation, the BHB sample is considerably richer than those used by GCLSA. However, this result does demonstrate the ability of Johnson's UBV to reproduce an effect observed by means of the Strömgren photometry.

  • (v)

    Ultraviolet deficiency is detected for cool BHB stars. Based on Kurucz's (1992) atmospheric models, we interpret this effect as an indication for lower surface gravity for these stars. Having in mind that ‘u-jump’ demonstrated in GCLSA is intimately connected to low gravities (‘log g-jump’), a conclusion could be drawn that two groups of BHB stars in M5 tend to show surface gravities lower than those predicted by existing ZAHB models.

Finally, we would like to add few remarks concerning the nature of the ultraviolet deficiency and the jump. GCLSA argued that a stellar atmosphere effect — radiative levitation of metals — rather than helium mixing, is the primary cause of the u- and log g-jump phenomenon. One of their arguments against helium mixing is that the size of the jump and its location do not depend on a cluster's individuality. Collecting data for ultraviolet deficiency in clusters with available UBV data, we found that the magnitude of the deficiency varies from cluster to cluster (Table 2) while in some clusters it is not observed at all. These findings, together with the results of GCLSA allow us to suggest that the jump and the deficiency are more likely to have different natures. In addition, taking in view certain evidences for mixing processes in some M5 red giant stars (based on spectral observations referenced in GCLSA), we suppose that the ultraviolet deficiency might be a photometric signature of helium mixing. This draws a possible direction for a future ultraviolet deficiency investigation: to obtain a homogeneous set of UBV data for a large number of GCs and to study the relation of the cluster parameters to the characteristics of the deficiency (if any).

Acknowledgments

The authors are grateful to S. Moehler, F. Grudahl and M. Catelan for helpful information and discussions. We also thank the anonymous referee for their very careful and detailed review of our paper.

This research was supported by the National Science Foundation grant under contract No. F-604/1996 with the Ministry of Education and Sciences.

This research has made use of NASA's Astrophysics Data System Abstract Service.

References

Allen
C. W.
,
1973
,
Astrophysical Quantities
 .
The Athlone Press
,
London
Alcaino
G.
Liller
W.
Alvarado
F.
Kravtsov
V.
Ipatov
A.
Samus
N.
Smirnov
O.
,
1996
,
AJ
 ,
112
,
2020
Arp
H.
,
1962
,
ApJ
 ,
135
,
311
Auriere
M.
Coupinot
G.
,
1989
, in
Grosbol
P. G.
Murtagh
F.
Warmels
R. H.
, eds,
Proc. 1st ESO/ST-ECF Data Analysis Workshop
 .
ESO
,
Garching bei Munchen
, p.
101
Barnard
E. E.
,
1931
,
Publ. Yerkes Obs.
 , Vol.
6
, p.
1
Borissova
J.
Catelan
M.
Spassova
N.
Sweigart
A. V.
,
1997
,
AJ
 ,
113
,
692
Borissova
J.
Catelan
M.
Ferraro
F. R.
Spassova
N.
Buonanno
R.
Iannicola
G.
Richtler
T.
Sweigart
A. V.
,
1999
,
A&A
 ,
343
,
813
Briley
M. M.
Smith
G. H.
Oke
J. B.
Hesser
J. E.
,
1992
,
ApJ
 ,
387
,
612
Brocato
E.
Castellsni
V.
Ripepi
V.
,
1995
,
AJ
 ,
109
,
1670
Buser
R.
Kurucz
R. I.
,
1978
,
A&A
 ,
70
,
555
Buser
R.
Kurucz
R. I.
,
1992
,
A&A
 ,
264
,
557
Buonanno
R.
Corsi
C. E.
Fusi Pecci
F.
,
1981
,
MNRAS
 ,
196
,
435
(BCF)
Carretta
E.
Bragaglia
A.
,
1998
,
A&A
 ,
329
,
937
Carreta
E.
Gratton
R. G.
,
1997
,
A&AS
 ,
121
,
95
Catelan
M.
Borissova
J.
Sweigart
A. V.
Spassova
N.
,
1998
,
ApJ
 ,
494
,
265
Christian
C.
Adams
M.
Barnes
J. V.
Hayes
D. S.
Siegel
M.
Bulcher
H.
Mould
J. R.
,
1985
,
PASP
 ,
97
,
363
Crocker
D. A.
Rood
R. T.
,
1985
, in
Philip
A. G.
ed.,
HB and UV-bright Stars
 ,
L. Davis Press
,
Schenectady, NY
, p.
107
Crocker
D. A.
Rood
R. T.
O'Connell
R. W.
,
1988
,
ApJ
 ,
332
,
236
Cudworth
K. M.
,
1979
,
AJ
 ,
84
,
1866
Dixon
W. V. D.
Davidsen
A. F.
Dorman
B.
Ferguson
H. C.
,
1996
,
AJ
 ,
111
,
1936
Dorman
B.
Rood
R. T.
O'Connell
R. W.
,
1993
,
ApJ
 ,
419
,
596
Drissen
L.
Shara
M. M.
,
1998
,
AJ
 ,
115
,
725
Ferraro
F. R.
Fusi Pecci
F.
Buonanno
R.
,
1992
,
MNRAS
 ,
256
,
376
Grundahl
F.
Vandenberg
D. A.
Andersen
M. I.
,
1998
,
ApJ
 ,
500
,
L179
Grundahl
F.
Catelan
M.
Landsman
W. B.
Stetson
P. B.
Andersen
M. I.
,
1999
,
ApJ
 ,
524
,
242
(GCLSA)
Harris
W. E.
,
1996
,
AJ
 ,
112
,
1489
Hawarden
T. G.
,
1971
,
Observatory
 ,
91
,
78
Kravtsov
V.
Ipatov
A.
Samus
N.
Smirnov
O.
Alcaino
G.
Alvarado
F.
,
1997
,
A&AS
 ,
125
,
1
Kurucz
R. L.
,
1992
,
Proc. IAU Symp. 149, The Stellar Populations of Galaxies
 .
Kluwer
,
Dordrecht
. p.
225
Lee
S.-W.
,
1977
,
A&AS
 ,
28
,
409
Markov
H. S.
,
1994
,
A&A
 ,
291
,
1027
Markov
H.
Valtchev
T.
Borisova
J.
Golev
V.
,
1997
,
A&AS
 ,
122
,
193
Menzies
J.
,
1974
,
MNRAS
 ,
168
,
177
Newell
E. B.
Rodgers
A. W.
Searl
I.
,
1969
a
ApJ
 ,
156
,
597
Newell
E. B.
Rodgers
A. W.
Searl
I.
,
1969
b
ApJ
 ,
158
,
699
Newell
E. B.
,
1973
,
ApJS
 ,
26
,
37
Rees
R. F.
,
1993
,
AJ
 ,
106
,
1524
Richer
H. B.
Fahlman
G. G.
,
1987
,
ApJ
 ,
316
,
189
Sandage
A.
,
1969
,
ApJ
 ,
157
,
515
Sandage
A.
,
1970
,
ApJ
 ,
162
,
841
Sandquist
E. L.
Bolte
M.
Stetson
P. B.
Hesser
J. E.
,
1996
,
ApJ
 ,
470
,
910
Sawyer Hogg
H.
,
1973
,
Publ. David Dunlop Obs.
 ,
3
,
6
Sneden
Ch.
Kraft
R. P.
Prosser
Ch. F.
Langer
G. E.
,
1992
,
AJ
 ,
104
,
2121
Spassova
N. M.
Michnevski
N.
,
1981
,
Rep. Bulgarian Acad. Sci.
 ,
36
,
83
Stetson
P. B.
Harris
W. E.
,
1977
,
AJ
 ,
82
,
954
Stetson
P. B.
,
1987
,
PASP
 ,
99
,
191
Wachter
S.
Wallerstein
G.
Brown
J. A.
Oke
J. B.
,
1998
,
PASP
 ,
110
,
821
Wallerstein
G.
Helfer
H. L.
,
1966
,
AJ
 ,
71
,
350
Von Braun
K.
Chiboucas
K.
Minske
J. K.
Salgado
J. F.
,
1998
,
PASP
 ,
110
,
810
Zinn
R. J.
Newell
E. B.
Gibson
G. B.
,
1972
,
A&A
 ,
18
,
390
Zinn
R. J.
West
M.
,
1984
,
ApJ
 ,
55
,
45