astro-ph0102331/a.tex
1: 
2: %\documentclass{article}
3: %\usepackage{emulateapj,pstricks,apjfonts}
4: \documentstyle[11pt,aasms4]{article}
5: %\usepackage{aasms4}
6: %\documentstyle[11pt,aasms4]{article}
7: 
8: \def\alwaysmath#1{\ifmmode{#1}\else{$#1$}\fi}
9: \def\msun{\alwaysmath{{M}_{\odot}}}
10: \def\teff{\alwaysmath{T_{\rm eff}}}
11: \def\arcsec{\hbox{$^{\prime\prime}$}}
12: \def\etal{{et al.~}}
13: \def\kelvin{{\rm \, K}}
14: \def\kms{\alwaysmath{{{\rm \,km\,sec^{-1}}}}}
15: \def\ersa{\alwaysmath{{\rm \, erg sec^{-1}/\AA}}}
16: \def\ers{\alwaysmath{{\rm \, erg\,sec^{-1}}}}
17: \newcommand\hst{{\sl HST}}
18: \newcommand\HST{{\sl HST}}
19: \newcommand\rosat{{\sl ROSAT}}
20: \newcommand\ngc[1]{NGC\,#1}
21: \setlength{\voffset}{-1.5truecm}
22: 
23: 
24: % SUBMISSION
25: % \lefthead{Paltrinieri, et al.}
26: % \righthead{}
27: % PREPRINT
28: \righthead{Paltrinieri, et al.}
29: \lefthead{VLT OBSERVATIONS OF GLOBULAR CLUSTER NGC\,6712} 
30: 
31: 
32: \slugcomment{Astronomical Journal in Press}
33: \begin{document}
34: 
35: \title
36: {VLT Observations of the Peculiar Globular Cluster NGC\,6712, III: 
37:  The Evolved Stellar Population
38: \altaffilmark{1} 
39: }
40: 
41: \author{
42: Barbara~Paltrinieri\altaffilmark{2}, 
43: Francesco~R.~Ferraro\altaffilmark{3},   
44: Francesco Paresce\altaffilmark{4},  
45: Guido De Marchi\altaffilmark{4,5,6}} 
46:  
47: \altaffiltext{1}{Based on data taken at the VLT, European Southern
48: Observatory, Chile, as part of the ESO observing programme
49: and 63.L-0423(A).}
50: \altaffiltext{2}{Istituto di Astronomia --- Universit\'a La Sapienza,
51: Via G.M. Lancisi 29, I--00161 Roma, Italy;
52: barbara@coma.mporzio.astro.it}
53: \altaffiltext{3}{Osservatorio Astronomico di Bologna, via Ranzani 1,
54: I--40126 Bologna, Italy; ferraro@bo.astro.it}
55: \altaffiltext{4}{European Southern Observatory, Karl Schwarzschild
56: Strasse 2, D--85748 Garching bei M\"unchen, Germany;
57: fparesce@eso.org}
58: \altaffiltext{5}{Space Telescope Science Institute, 3700 San Martin
59: Drive, Baltimore, MD 21218, USA; demarchi@stsci.edu}
60: \altaffiltext{6}{Affiliated to the Astrophysics Division, Space
61: Science Department of ESA}
62: 
63:  
64: \begin{abstract}
65: 
66: We present extensive UBVR photometry of the Galactic globular cluster
67: (GGC) NGC\,6712 obtained with the ESO Very Large Telescope (VLT) which
68: reach down to two magnitudes below the main sequence turn-off and
69: allows us for the first time to determine the age of this cluster. By
70: using the apparent luminosity of the zero age horizontal branch (ZAHB),
71: $V_{ZAHB}=16.32\pm0.05$ and the stellar main sequence (MS) turn--off
72: (TO) magnitude $V_{TO}=19.82\pm0.10$, we obtain $\Delta
73: V_{TO}^{HB}=3.5\pm0.1$ (a value fully compatible with that derived for
74: other clusters) which suggests that, at an age of $\sim 12$ Gyr,
75: NGC\,6712 is coeval with other GGC of similar metallicity.
76: 
77: We derive interstellar reddening by comparing the position and
78: morphology of the red giant branch (RGB) with a wide variety of
79: reference clusters and find $E(B-V)=0.33\pm0.05$, a value significantly
80: lower than had been determined previously. Assuming this value for  the
81: reddening, we  determine a true distance modulus of $(m-M)_0 = 14.55$,
82: corresponding to a distance of $\sim 8$ kpc.
83: 
84: We find a population of 108 candidate blue straggler stars (BSS),
85: surprisingly large when compared with the typical BSS content of other
86: low concentration clusters. Moreover, we detect a very bright blue star
87: in the core of NGC\,6712 that might be a post-AGB star. These results,
88: combined with those already presented in two companion papers, strongly
89: support the hypothesis that NGC\,6712 was, at some early epoch of its
90: history, much more massive and concentrated. The continued interaction
91: with the bulge and the disk of the Galaxy has driven it toward
92: dissolution, and what we now observe is nothing but the {\it remnant}
93: core of a cluster that once was probably one of the most massive  in
94: the Galaxy.
95: 
96: 
97: \end{abstract}
98: 
99: \keywords{
100: globular clusters: individual (NGC\,6712);
101: stars: Population II --- stars: evolution -- 
102: }
103: 
104:  
105: \section{Introduction}
106: 
107: NGC\,6712 is a low-concentration ($c=0.9$, $Log \rho_0 \sim 3$;,
108: Djorgovski \& Meylan 1993), intermediate-high  metallicity ($[Fe/H]=-1.01$
109: Zinn 1985) globular cluster, which has, without doubt, some peculiar
110: characteristics.  First, it is the lowest density GGC containing a low
111: mass X-ray binary (LMXB - namely $1RXS J185304.8-084217$; Voges et al.
112: 1999).  LMXB are thought to form via tidal capture in high density
113: clusters, thus the presence of such an object in the core of a loose
114: cluster is somewhat surprising.  Second, its orbit, as computed by
115: Dauphole et al. (1996), suggests that it is experiencing a severe
116: interaction with the Galaxy due its numerous passages through the disk
117: and bulge.
118: 
119: Within this intriguing scenario, we have undertaken a detailed study of
120: the global stellar population in NGC\,6712, using the ESO {\it Very
121: Large Telescope}. This is the third paper in a series dedicated to this
122: cluster.
123: 
124: 
125: Ferraro et al.2000a (hereafter {\it Paper I}), besides confirming the
126: optical counterpart to the LMXB, discovered the presence of an
127: additional interacting binary candidate (showing strong UV and
128: $H\alpha$ excess) in the core of NGC\,6712, located a few arcsec from
129: the LMXB.
130: 
131: Deep $V$- and $R$-band observations of the lower MS, presented by
132: Andreuzzi et al. 2001 (hereafter {\it Paper II}) confirmed the previous
133: findings by De Marchi et al. (1999) that the mass function of this
134: cluster has been severely depleted of lower mass stars, probably
135: stripped away by the tidal force of the Galaxy.
136: 
137: These results add further support to the scenario that NGC\,6712 is
138: only a pale remnant of a once much more massive cluster.  Takahashi \&
139: Portegies Zwart (2000), using extensive N-body simulations, have
140: recently shown that the inverted mass function observed by De Marchi et
141: al. (1999) is a clear signature of significant mass evaporation since
142: it only appears in clusters which have lost most (up to $\sim 99\%$) of
143: their initial mass.  If this scenario is confirmed, with a current
144: mass of $\sim 10^5  M_{\odot}$ (Pryor \& Meylan 1993) NGC\,6712 might
145: have been one of the most massive clusters ever formed in the Galaxy
146: with $M_{inital} \sim 10^7  M_{\odot}$.
147: 
148: In this paper, we make use of a large data-set obtained with the VLT in
149: order to present and discuss the overall morphology and the main
150: characteristics of the CMD for stars brighter than the MS TO, looking
151: for any additional signature of the stormy past of this cluster.
152: 
153: The plan of the paper is as follows. We present the observations and
154: data analysis Section 2 while the overall properties of the resulting
155: colour--magnitude diagramme (CMD) are described in Section 3.
156: Sections 4, 5, 6, 7 are devoted each to the description of one of the
157: evolutionary sequences (horizontal branch - HB, RGB, blue straggler
158: stars - BSS, and blue objects, respectively).  A discussion of the
159: cluster age and distance is given in Section 8.  Section 9 summarises
160: the main results of this work.
161: 
162: 
163: 
164: 
165: \section{Observations and data reductions}
166: 
167: \subsection{Observations}
168: 
169: A large set of frames covering the central region of NGC\,6712 was
170: obtained in service mode during June 1999 at the {\it ANTU} Very Large
171: Telescope (VLT-UT1) at ESO on Cerro Paranal (Chile) using the FORS1
172: camera equipped with a $2048\times2048$ pixel$^2$ CCD array).  The data
173: were obtained using two different levels of resolution:  {\it (i)} high
174: resolution (HR), with a plate-scale of $0\farcs1/pixel$ and a field of
175: view of $3\farcm4 \times 3\farcm4$; and {\it (ii)} standard  resolution
176: (SR) with a   plate-scale of $0\farcs2/pixel$ and a field of view of
177: $6\farcm8 \times 6\farcm8$.  HR frames were roughly centered on the
178: cluster centre, while the four partially overlapping SR fields (one for
179: each quadrant, Q1, Q2, Q3, Q4, respectively) were roughly centered at
180: $3\farcm5 $ for the cluster centre(see Figure 1).  This configuration
181: allowed us to observe the innermost regions of the cluster with the
182: most appropriate resolution and to cover a large area ($\sim 13\farcm
183: \times 13\farcm$) around the cluster.
184: 
185: The data consist of five 10\,s $B$-, $V$-, $R$-band exposures, five
186: 120\,s $U$-band exposures for each HR and SR field.  An additional
187: 700\,s $H\alpha$ exposure was obtained only in  the HR field.
188: 
189: \subsection{Magnitudes, colour and positions}
190: 
191: All the reductions were carried out using ROMAFOT (Buonanno et al.
192: 1983, Buonanno \& Iannicola 1986), a package specifically developed to
193: perform accurate photometry in crowded fields. Specific subroutines of
194: this package have been used to detect stars in each frame, in
195: particular independent searches have
196:  been performed in the best blue ($U,B$) and in the best red ($V,R$)
197: images, in order to properly optimize the search for blue and red
198: objects.  The {\it blue} and the {\it red} masks built in this way and
199: containing the position of  the stars have been then adapted to each
200: {\it colour-group} image of the same field (the red mask to all the
201: $V$- and $R$-band images and the blue mask for the $U$ and $B$ frames)
202: and the standard PSF-fitting procedure performed.  The output
203: instrumental magnitudes in the same band were then averaged by properly
204: weighing  the
205: photometric quality of individual frames to yield the adopted {\it
206: u,b,v,r} mean instrumental magnitudes.  The  data-sets were then
207: matched  together and a final catalogue with stellar coordinates and
208: the average instrumental magnitude in each filter has been compiled for
209: all the objects identified in each of the FORS1 fields.
210: 
211: Stars in overlapping areas (see Figure 1) were used to transform the X
212: and Y coordinates of each field to a common coordinate system.
213: Multiple measurements of the objects common to adjacent SR fields were
214: averaged, while for objects common to HR and a SR field, the HR
215: magnitudes were taken owing to the higher resolution and quality of
216: these images.  The result of this procedure is a homogeneous set of
217: instrumental magnitudes, colours and positions for all the stars in our
218: frames.
219: 
220: 
221: \subsection{Calibration}
222: 
223: The final instrumental magnitudes were then transformed into the
224: standard Johnson photometric system by using ten photometric standard
225: stars in selected areas PG1528, PG2213, PG2331 (Landolt 1992).
226: Standard aperture photometry was performed on these stars.  The
227: calibration curves linking the instrumental aperture photometry to the
228: Johnson magnitudes are plotted in Figure 2.  A subset of the brightest,
229: unsaturated and most isolated stars was used to link the {\it
230: instrumental}  magnitudes obtained with PSF fitting to those derived
231: with aperture photometry.
232: 
233: $H\alpha$ magnitudes were treated and calibrated as $R$-band images,
234: except for an offset of $\sim 4$ mag that was applied so as to account
235: for the lower filter efficiency.
236: 
237: The calibrated data can be compared   with the photographic $V$- and
238: $B$-band data of Cudworth (1988, hereafter C88):  the residuals in
239: magnitude computed using a sample of $\sim 100$ stars in common with
240: C88 turn out to be very small in both the B and V filters
241: ($V-V_{C88}=0.00$ and $B-B_{C88}=-0.03$) showing an excellent agreement
242: between the photometric zero-points used here and those of C88.
243: 
244: 
245: 
246: \subsection{Photometric errors}
247: 
248: Since we have 5 repeated exposures through each filter, the internal
249: accuracy of our photometry has been estimated from the rms
250: frame-to-frame scatter of the instrumental magnitudes.  We computed this
251: quantity for each individual star according to the formula given in
252: Ferraro et al. (1991).  The rms values obtained for the $U$, $B$, $V$
253: and $R$ frames are plotted in Figure 3 as a function of the mean final
254: calibrated magnitude adopted for each star measured in the HR field.
255: The mean photometric errors are very small ($\sigma <0.05$) over the
256: whole magnitude range covered by our observations in all  filters.
257:  
258: 
259: \section{Colour-Magnitude Diagrammes: overall structure}
260: 
261: As result of the procedure described in the previous section,  a
262: catalogue with magnitudes, colour and positions for a sample of 30,000
263: stars has been obtained in the  region covered by our 
264: observations\footnote{The electronic version of the table 
265: listing magnitudes and positions for the entire sample
266: is available upon request from the second author (e-mail: ferraro@bo.astro.it)}
267: (see Figure 1).  Figure 4 shows the CMD corresponding to stars in  regions
268: located at different distances from the cluster centre.  In particular,
269: in {\it  panel (a)}  all the stars lying in the HR field are plotted
270: while the other three panels ({\it (b),(c)} and {\it (d)},
271: respectively) show the stars located in annuli at increasing distances
272: from the cluster centre.  Here the centre of the cluster has been
273: assumed to be the one defined in {\it Paper I}.
274: 
275: The inspection of the CMD presented in Figure 4 shows that field
276: contamination is particularly severe around NGC\,6712. This  was to be
277: expected, since the cluster is located behind the Scutum stellar cloud
278: which is one of the highest surface brightness regions in the Milky
279: Way.
280: 
281: 
282: The overall characteristics of the CMD in Figure 4 can be summarised as
283: follows:
284: 
285: \begin{itemize}
286: 
287: \item{(1)} The population sampled by the HR field is  dominated by
288: cluster stars:  all sequences in the CMD, namely the RGB, HB, sub-giant
289: branch (SGB), TO region, BSS sequence, are well defined and populated.
290: The field component, however, is also  visible (note the field MS
291: crossing the diagram from $B-V\sim 0.4, V\sim14.$ down to $B-V\sim
292: 1.2, V\sim 20.5$);
293: 
294: \item{(2)} as expected, at increasing distance from the cluster centre,
295: the field contribution rapidly increases with respect to the cluster
296: population;
297: 
298: \item{(3)} although in the region $150''<r<360''$ the field population
299: dominates the overall morphology of the CMD, some features of the
300: cluster population are still visible, such as a few blue HB stars at
301: $18<V<17$ and $(B-V)<0.5$ (see Figure 4{\it b} and 4{\it c}).
302: 
303: \item{(4)} blue HB stars are not visible in the CMD at distances
304: $r>360''$ (see Figure 4 {\it d}).
305: 
306: \end{itemize}
307:  
308:  
309: \subsection{ Field decontamination}
310:   
311: Some features in Figure 4 show that the cluster population is still
312: present in the region $150''<r<360''$, even if  evolutionary sequences
313: are not clearly defined in the CMD.  On the other hand the absence of
314: blue HB stars for $r>360''$ suggests that the cluster population
315: becomes negligible beyond this distance.
316: 
317: This indication is fully confirmed by the investigation of MS stars
318: presented in {\it Paper II}, where we compared the MS cluster
319: population at different distances from the cluster centre with that
320: obtained in a reference field taken at much larger distance ($\sim
321: 42^{\prime}$).  From this comparison   we concluded that the  stellar
322: population at $r>5'$ from the cluster centre is fully consistent with
323: that observed in the reference field, i.e. it is made almost
324: exclusively of field stars.  In Figure 5 we present the projected
325: density distribution ({\it filled circles}) for the evolved population
326: of the cluster. All stars  with $V<20$ plotted in Figure 4 have been
327: counted. The stellar density has been scaled to match the  density
328: profile of MS stars  with $19.5<R<20.5$ (corresponding to $\sim 0.75
329: M_{\odot}$) and $r>2'$ (plotted as {\it large open squares} in Figure 5
330: -- see also Figure 8 in {\it Paper II}).  The figure clearly shows a
331: plateau in the stellar density for $r>4.2'$, which is fully consistent
332: with the density of $ 0.75 M_{\odot}$-mass star measured in the control
333: field at $42'$ from the cluster centre (indicated as the {\it dotted
334: horizontal line} in Figure 5). This confirms that the cluster
335: population becomes indeed negligible for $r>5'$.
336:    
337: On the basis of these considerations, we conservatively assumed that
338: the sample at  $r>360''$ can be regarded as field population (see Figure
339: 4(d)) and used it in order to decontaminate the CMD.To be sure, exact
340: discrimination between cluster and field members is not possible on the
341: basis of the CMD comparison alone. However, in order to study the
342: global characteristics of the stellar population of NGC\,6712, we
343: adopted a statistical approach, following the method described by
344: Mighell et al. (1996) and successfully used by others as well (see for
345: example Bellazzini, Ferraro \& Buonanno, 1999, Testa et al. 1999).  In
346: short, for each star in the {\it contaminated} CMD, the probability of
347: membership is computed by comparing the number of stars lying within a
348: cell (of assigned size) in the {\it contaminated} CMD and in the {\it
349: field} CMD.
350: 
351: The decontamination procedure has been applied to each of the CMD
352: plotted in Figure 4 {\it a, b, c}. Only in the case of the HR field,
353: however, does the  procedure yield acceptable results.  Figure 6 shows
354: the statistically decontaminated CMD of the HR field.  Here the result
355: is quite good as the statistical decontamination successfully removes
356: most of the field stars from the HR CMD and the cluster sequences
357: appear well defined. For this reason, we conservatively decided to
358: limit the analysis of the stellar population of NGC\,6712 to this
359: region.  
360: 
361: 
362: 
363: \subsection{Cluster population: ridge lines}
364: 
365: Figure 5 shows that all the sequences in the CMD are  well defined and
366: populated. Particularly noteworthy is the large population of BSS (see
367: Section 6) which has not been affected by the field decontamination
368: procedure as the field MS appears to be significantly redder than the
369: BSS sequence (see Figure 4{\it a} and 4{\it d}), mostly because of the
370: higher metallicity of the field population.
371: 
372: The mean ridge lines of the main branches (RGB, SGB) in the CMD plotted
373: in Figure 6 were determined following an iterative process. A first
374: rough selection of the candidate RGB stars was performed by eye,
375: removing the HB and part of the AGB stars. Then, the sample was divided
376: into $0.4$ V-mag wide bins and the median colour and magnitude computed
377: in each bin. Stars more distant than $2\sigma$ from the median point
378: were rejected and the median and $\sigma$ recomputed until no more
379: outliers were found, The median points were then interpolated through a
380: spline over an interval of typically 3 magnitudes. The resulting ridge
381: line for the RGB-SGB and TO-MS region is listed in Table 1.
382: 
383:  
384: 
385: \section{The HB morphology and the ZAHB level}
386: 
387: Figure 7 plots HB stars from the decontaminated sample of Figure 6.
388: The bulk of the HB stars populates the red side of the instability
389: strip, as expected since NGC\,6712 is a cluster of intermediate-high
390: metallicity ($[Fe/H]_{Z85}=-1.01$).
391:  
392: Our CCD photometry does not have the proper time-coverage to search for
393: variability, for this reason we identified the RR-Lyrae variables 
394: lying in the FORS1-HR field of view  
395: from previous studies.
396: The 7 known RR Lyrae variables (namely V1, V5, V6, V12, V18, V19, V20)
397: identified in  our field are plotted as large filled triangles
398: in Figure 7. Their position in the CMD clearly shows that we observed
399: these variables at random phases of their pulsation cycle. Since our
400: observations were not homogeneously spread over the pulsation period,
401: we are unable to derive meaningful average magnitudes and colours for
402: these objects.
403: 
404: In order to more quantitatively describe the morphology of the HB, we
405: computed the parameter $HBM=(B-R)/(B+V+R)$ (Lee et al. 1990, but see
406: Fusi Pecci et al. 1992, Buonanno et al. 1997), where $B$ and $R$ are
407: the number of stars bluer and redder than the RR Lyrae gap,
408: respectively, and $V$ is the number of RR Lyrae variables.  Using our
409: sample, this parameter takes on the value
410: $HBM=(23-127)/(23+7+127)=-0.66\pm0.12$, a figure typical of clusters of
411: intermediate-high metallicity which confirms that  the HB morphology of
412: NGC\,6712 closely resembles that of NGC\,6171 (Ferraro et al. 1991), a
413: cluster with similar metallicity.
414: 
415: To provide a characterisation of the extreme blue HB population with
416: respect to the red one, we also computed  Buonanno's (1993) index
417: $P_{HB}=(B2-R)/(B+V+R)$, where $B2$ is the number of the blue HB stars
418: bluer than $(B-V)_0=-0.02$. This parameter is, of course,
419: reddening-dependent: assuming $E(B-V)=0.33$, as explained in Section
420: 5.2,we obtain $P_{HB}=-0.78\pm0.14$.
421: 
422: 
423: Ferraro et al. (1999a, hereafter F99) presented a new procedure for the
424: determination of the level of the ZAHB in the observed (V,B-V)-CMD.
425: This procedure uses synthetic HB models   to reproduce the observed HB
426: morphology. Once a proper agreement between the observed and the
427: synthetic HB has been obtained, the value of $V_{ZAHB}$ can be derived
428: from  the  model which has been used to construct the synthetic HB.
429: 
430: F99 derived $V_{ZAHB}=16.32\pm0.05$ for NGC\,6712 adopting the C88
431: photometry. As shown in Section 2.3,  the C88 photometry nicely agrees
432: with that presented in this paper. Thus, in the following, we adopt
433: this  value  for the ZAHB level. The ZAHB is shown as a solid line in
434: Figure 6; as expected, it is consistent with  the lower boundary of the
435: magnitude distribution of the red HB clump.
436:  
437:   
438: \section{The RGB: metallicity and reddening}
439: 
440: Once properly calibrated, the shape and position of the RGB 
441: in the CMD can be used to derive photometric estimates 
442: of the cluster metallicity.
443: 
444: \subsection{Metallicity scales}
445: 
446: The most widely used metallicity scale for GGC is that proposed by Zinn
447: and collaborators in the 1980s (Zinn \& West 1984, Zinn 1985) and
448: essentially based on integrated light indexes and low resolution
449: spectra. There is now a growing body of reasons (see Carretta \&
450: Gratton 1997 - herafter CG97 and Rutledge et al. 1997) suggesting a
451: revision of the {\it classical} Zinn scale, according to the results of
452: the latest high-resolution spectral data which have the advantage of
453: measuring directly Iron absorption lines.  In the following, we adopt
454: the CG97 metallicity scale ($[Fe/H]_{CG97}$).
455: 
456: Moreover, there is much observational evidence that {$\alpha$}-elements
457: (such as Si and Ca) are often enhanced with respect to Iron for
458: Population II stars and thus the knowledge of the Iron abundance alone
459: in not sufficient to correctly evaluate the {\it global} metal content
460: in each cluster.  For this reason, the {\it global} metal index
461: ($[M/H]$) (including also the   {$\alpha$}-element enhancement) has
462: been defined (see Salaris, Chieffi \& Straniero, 1993). For a more
463: detailed discussion of this topic, we refer the reader to F99 and
464: Ferraro et al. 2000b, and the references therein.
465:  
466: Here, we just note that NGC\,6712 turns out to have, according to Table
467: 2 in F99, the following estimated metallicity in different scales:
468: $[Fe/H]_{Z85}=-1.01$, $[Fe/H]_{CG97}=-0.88$, and $[M/H]=-0.71$.
469: 
470: 
471: \subsection{RGB parameters: metallicity and reddening}
472:  
473: A complete set of RGB parameters and metallicity  indicators in the
474: classical $(V,B-V)$ and in the IR plane has been recently presented by
475: F99 and Ferraro et al. (2000b), respectively.  These papers also
476: reported on an independent calibration of RGB-parameters in terms of
477: the cluster metallicity (both in the CG97  and {\it global} scales).
478: In particular, F99 presented a system of equations (see Table 4 in F99)
479: which can be used to simultaneously derive an estimate of metal
480: abundance (in terms of $[Fe/H]_{CG97}$ and $[M/H]$) and reddening from
481: the morphology and location of the RGB in the $(V,B-V)$ CMD.  This
482: system of equations is the equivalent to the so-called SRM method
483: defined by Sarajedini (1994) in the $(V,V-I)$ plane.
484: 
485: For NGC\,6712 we used the mean ridge line listed in Table 1 to measure
486: three RGB-parameters:  $(B-V)_g$, defined as the observed RGB colour at
487: the HB level; the two RGB slopes $S_{2.0}$ and $S_{2.5}$ defined as the
488: slope of the line connecting the intersection of the RGB and HB with
489: the points along the RGB located, respectively, 2.0 and 2.5 mag
490: brighter than the HB.  Assuming the ZAHB level obtained in Section 4
491: and the mean ridge line listed in Table 1, we obtained a first set of
492: RGB parameters, namely $(B-V)_g=1.26$, $S_{2.0}=4.69$ and
493: $S_{2.5}=4.20$ (see F99 for the definition of these parameters).  With
494: these values as {\it "input"}, the method soon  achieved  convergence
495: on the following values: $E(B-V)=0.33\pm0.05$,
496: $[Fe/H]_{CG97}=-0.80\pm0.20$ and $[M/H]=-0.62\pm0.20$.  The errors
497: associated with the measurements  are conservative estimates of the
498: global uncertanties, formal errors being much smaller than those
499: assumed.
500: 
501: The photometric estimate of the cluster metallicity derived with the
502: technique above fully agrees with previous estimates and confirms that
503: NGC\,6712 is an intermediate-high metallicity cluster.
504:   
505: On the other hand, the value of the reddening obtained with this
506: procedure appears significantly lower than that listed in the
507: literature:  $E(B-V)=0.48$, Webbink (1986) and $E(B-V)=0.46$, Harris
508: (1996).  Past reddening determinations for this cluster, however, are
509: quite uncertain:
510: 0.48 (Sandage \& Smith, 1966),
511: 0.38 (Kron \& Guetter, 1976),
512: 0.40 (Alcaino, 1977),
513: 0.33 (Martins \& Harvel 1981),
514: 0.42 (Cudworth 1988).  Moreover, Janulis \& Smriglio (1992) confirmed the
515: existence of differential reddening in the region of NGC\,6712.
516:  
517: 
518: \section{Blue Stragglers}
519: 
520: BSS were first observed in the 1950's in the outer region of the GGC M3
521: (Sandage 1953), while the first sample of BSS in the cores of dense GGC
522: were discovered with HST (Paresce et al. , 1991) in 47 Tuc.  Since
523: then, systematic searches for these objects in GGC cores have been
524: performed from the ground and from  space.  The main result of this
525: extensive search is that they have been found in all the GGC properly
526: observed, thus suggesting that BSS are a normal component of the GGC
527: population (see the catalogue by Sarajedini 1993, and Ferraro,
528: Bellazzini \& Fusi Pecci 1995).
529:  
530: BSS are thought to form via two main mechanisms: (1) the merger of two
531: stars in a primordial binary system and (2)  stellar collisions (Bailyn
532: 1995, Bailyn \& Pinsonneault 1995).  Many authors (Bailyn 1992, Fusi
533: Pecci et al. 1992, Ferraro, Bellazzini \& Fusi Pecci 1995) suggested
534: that BSS in loose clusters might be produced from the coalescence of
535: primordial binaries, whereas in high density clusters BSS might arise
536: from stellar collisions (depending on the survival-destruction rate of
537: primordial binaries).  In particular, collisional BSS can be used as a
538: diagnostic of the dynamical evolution of GGC (see, for example, the
539: exceptionally large population of BSS recently found in M80 by Ferraro
540: et al. 1999b).
541: 
542: From the analysis of the CMD plotted in Figure 4 and Figure 6,
543: one sees that NGC\,6712 has a quite large population of BSS.
544: As already said, this population is not affected by field contamination,
545: since the  BSS sequence is significantly bluer than the Field MS.
546: 
547: 
548: On the basis of their location in the $(V,B-V)$ CMD we identified a sample of 
549: 108 BSS. 
550: The identification numbers, the V,B,R and U magnitudes and the X and Y
551: coodinates (in pixels [1 px$=0".1$]  with respect to the adopted
552: cluster center) for the candidate BSS are listed in Table 2. 
553: Figure 8 shows zoomed CMDs where the BSS candidates are plotted as
554: large filled circles. As can be seen, all the candidates selected
555: in the  $(V, B-V)$ CMD lie also in the BSS region in the $(U, U-V)$ CMD. 
556: Moreover, this latter CMD suggests that a few additional objects might be
557: considered BSS candidates. We conservatively decided, however, 
558: to limit our analysis to those stars which are
559: BSS candidates in both CMDs. In particular, we
560: intended to avoid regions
561: of the CMD which can be populated by {\it artificial} stars
562: resulting from blending effects (see Ferraro et al. 1991, 1992).
563: 
564: It is worth noticing that five objects slightly redder and brighter
565: that the BSS sequence (plotted as stars in Figure 8) seem to delineate
566: a track for post-BSS evolution (E-BSS evolved BSS).
567:   According to their position in the
568: CMD, they could be BSS which have left the MS to evolve toward the RGB
569: (BSS in SGB phase).  Candidate BSS in such an evolutionary phase have
570: been observed in other clusters (see for example the CMD of NGC5053 by
571: Nemec \& Cohen 1989 or the CMD of the external region of M3, Buonanno
572: et al. 1994).  We did not, however, include these stars in our sample
573: as they lie in a region of the CMD where field contamination starts to
574: be severe. Since we performed only a statistical decontamination of the
575: observed sample, cluster membership of stars plotted in Figure 6 cannot
576: be guaranteed, expecially in highly contaminated regions of the CMD.
577: The selected E-BSS candidate are, nevertheless, worth further
578: investigation in order to better assess their membership and true
579: nature and for this reason they are also listed in Table 2.
580: 
581: Even with the conservative selection criterion adopted above, the
582: number of BSS  found in the central regions of NGC\,6712 is
583: surprisingly large.  Although in absolute terms, NGC\,6712 has much
584: fewer BSS than M80 (Ferraro et al. 1999b), it shows a population
585: comparable with that of NGC\,6388 and NGC2808 (see Sosin et al. 1998)
586: which are much more massive (a factor $\sim 16$) and concentrated (a
587: factor  $\sim 100-500$) than NGC\,6712.
588: 
589: In order to quantitatively compare the BSS population across GGC, its
590: fraction with respect to the total has to be taken into account.  In
591: doing so, Ferraro et al. (1997, 1999b) defined the specific fraction of
592: BSS $F_{HB}^{BSS}=N_{BSS}/N_{HB}$ as the number of BSS normalised to
593: the number of HB stars observed in the same cluster region. In
594: NGC\,6712 the specific frequency calculated in this was is $\sim 0.69$
595: similar to that obtained in M3 (Ferraro et al. 1999b) which is again
596: more massive ( $M_V=-8.75$) and concentrated ($c=2.0$ ) than
597: NGC\,6712.
598:   
599: If one limits the comparison to clusters with similar structure (for
600: example with comparable concentration  $c\sim1$ and integrated
601: magnitude), this population appears indeed large. Note that NGC5897
602: ($M_V=-7.27$ and $c=0.95$), for example, has only 34 BSS (Ferraro, Fusi
603: Pecci \& Buonanno 1992) and $F_{HB}^{BSS}=0.5$.
604:  
605: Generally, BSS are more centrally concentrated with respect to the
606: other stars in the cluster, although there are notable exceptions to
607: this general rule, such as M\,3 (Ferraro et al. 1993, 1997) and M\,13
608: (Paltrinieri et al. 1998).  In NGC\,6712 we  compared the radial
609: distribution of BSS  with respect to RGB and SGB objects (in the range
610: $17.5<V<20$). The cumulative radial distributions are plotted, as a
611: function of the projected distance ($r$) from the cluster centre, in
612: Figure 9, where one sees that BSS (solid line) are more centrally
613: concentrated than the RGB+SGB sample (dashed line).  The
614: Kolmogorov-Smirnov test shows that the probability of drawing the two
615: populations from the same distribution is only $0.1\%$.  The two
616: distribution are, thus,  different at a high level of significance ($>3
617: \sigma$).
618:  
619:  
620: \section{Blue Objects}
621: 
622: \subsection{A post-AGB bright object, or two?}
623: 
624: Figure 6 shows that a few blue objects are present in the very central
625: regions of NGC\,6712.  The three faint blue objects ($V>19$ and
626: $B-V<0.5$), namely star $\#10261$, $\#9774$ and $\#8916$, in our
627: catalogue, have been discussed in {\it Paper I}.  In particular, star
628: $\#9774$ is star $S$ of Anderson et al. (1993) and, on the basis of its
629: position, is the best candidate to the optical counterpart to the
630: luminous X-ray source detected in this cluster. Star $\#10261$ is a
631: relatively bright UV star, located a few arcsec away from the X-ray
632: source, and it is the only object in the HR-field showing a significant
633: $ H\alpha$ emission (see next Section). For these reasons, in {\it
634: Paper I} we suggested  it is an additional  promising  interacting binary
635: candidate.
636: 
637: From the analysis of Figure 6, it is also evident that the bluest
638: object in the CMD is a bright star (namely $\# 9620$) roughly located
639: at  the HB level, $V=16.67$ but significantly bluer ($B-V=0.03$) than
640: the bluest HB  stars ($B-V \sim 0.3$)
641: 
642: Figure 10 shows  the $(U,U-B)$ CMD of all the stars plotted in Figure
643: 6, and star $\# 9620$ still shows as the most UV-bright object in the
644: field.  The three faint UV stars discussed in {\it Paper I} are plotted
645: as open triangles.
646: 
647: The position of the UV-bright star in the CMD closely resembles that of
648: the UV-bright post-AGB star found in M3 (vZ1128, see Buonanno et al.
649: 1994). Such objects are indeed very rare in GGC: only a few post-AGB
650: stars have been found in GGC due to their short evolutionary lifetime
651: of $\sim 10^5$ yr (only 0.5 Post-AGB stars are expected to be found in
652: a typical $\sim 10^5 L_{\odot}$ cluster).
653: 
654: To assess whether the position of star $\# 9620$ is compatible with
655: that of a post-AGB star, we performed a qualitative comparison with
656: theoretical models.  A $t\sim 12.5$ Gyr (see Section 8.2) isochrone
657: with appropriate metallicity ($Z=0.004$) from Bertelli et al. 1994 has
658: been over-plotted in the CMD of Figure 10, after having been shifted to
659: fit the main loci, so as to mark the location in this diagramme of the
660: post-AGB track and the subsequent cooling sequence.  The location of
661: star $\# 9620$ in the CMD nicely agrees with that predicted by the
662: models.
663:  
664: Note that star $\# 9620$ is located only $\sim 19''$  away from the
665: cluster centre. Although, in principle, we cannot exclude that this
666: object is a field star, it is worth  noting that no similar  UV-bright
667: objects have been found in the sample that we assumed as FIELD (see
668: Figure 4d).  Moreover, another similar UV-bright object (lying in the
669: same region of the CMD) has been found in the field labelled as SR-Q1
670: in Figure 1, at $4.4'$ from the cluster centre (see Figure 4c).  These
671: objects deserve a more detailed spectroscopic analysis, as proving that
672: both are cluster members would add further support to the scenario that
673: NGC\,6712 was much more massive in the past.
674:  
675: 
676: \subsection{The $(H\alpha, H\alpha-R)$ CMD}
677: 
678: Some stars in GGC are expected to show $H\alpha$-emission: a few bright
679: red giants (see Cacciari \& Freeman 1983) or rare objects possibly
680: connected to planetary nebulae (see Jacoby et al. 1997) or some exotic
681: objects connected to interacting binaries (as cataclysmic variable and
682: LMXB; see Cool et al. 1995).  According to Cool et al. 1995, the
683: $(H\alpha, H\alpha-R)$ CMD can  efficiently serve the purpose of
684: revealing $H\alpha$-excess objects.  Indeed, in this diagramme stars
685: with $H\alpha$ in absorption  will have $(H\alpha-R)>0$ while stars
686: with $H\alpha$ in emission  will feature $(H\alpha-R)<0$.  Typical cool
687: giants and SGB stars in a cluster will have a weak $H\alpha$ absorption
688: line ($EW(H\alpha)=1 A$) and, for this reason, they will have
689: $(H\alpha-R)\sim 0$.
690: 
691: Figure 11 shows the $(H\alpha, H\alpha-R)$ CMD for stars in the HR
692: field.  The brightest RGB stars are saturated in the long $H\alpha$
693: exposure, so they are not plotted in that figure.  As expected, the
694: bulk of the cluster population, constitued as it is by relatively cool
695: stars, defines a sequence lying in the CMD at $H\alpha-R\sim 0$. The
696: $H\alpha$-excess star discussed in {\it Paper I} is marked by its
697: identification number in our catalogue.  Interestingly enough, on the
698: right of the main sequence  there are a few stars which seem to define
699: a secondary sequence at $(H\alpha-R)\sim 0.15$ almost parallel to the
700: main one.  According to their position in this diagramme, they should
701: have $H\alpha$ in absorption.  These stars are the hottest HB stars and
702: BSS.  In particular, blue HB stars (with $(B-V)<0.5$ see Figure 6)
703: which are plotted as large filled circles define  a very narrow
704: sequence, well defined and well separated from the main one.  The
705: bright UV-object (star $\#9620$) discussed in the previous section is
706: plotted in Figure 11 as an open square. According to its position it
707: turns out to have an $H\alpha$ absorption line of intensity
708: intermediate between that of giants and blue HB stars.
709:  
710: Figure 11 shows that the $(H\alpha, H\alpha-R)$ CMD can be used as an
711: independent method to select blue HB stars, and can provide a direct
712: measure of the strength of the $H\alpha$ absorption line in these stars
713: (see Cool et al. 1995).
714: 
715: 
716: \section{Turn-Off point and age}
717:  
718: \subsection{Comparison with NGC\,6171: the relative age of NGC\,6712}
719:  
720: An accurate determination of  relative ages of GGC provides fundamental
721: clues to the formation time-scale of the Galactic halo (see Buonanno,
722: Corsi \& Fusi Pecci 1989, Vandenberg, Stetson \& Bolte 1996, Rosenberg
723: et al. 1999 -hereafter R99- and reference therein).
724: 
725: Since the SGB-TO region in the CMD plotted in Figure 6 is very well
726: defined and populated, we can use it to locate  the MS-TO, which is, as
727: well known, a good age indicator.
728:  
729: As mentioned above, the whole CMD of NGC\,6712 closely resembles that
730: of NGC\,6171, a cluster with similar 
731: metallicity. In Figure 12 the mean ridge lines of NGC\,6171 (solid
732: lines) from Ferraro et al. 1991, have been shifted (by $\Delta V=0.65$
733: and $\Delta (B-V)=0.0$) to match the CMD of NGC\,6712.  The ZAHB level
734: for NGC\,6171 has been assumed to be $V_{ZAHB}=15.70\pm0.10$ according
735: to  Table 2 of F99). Figure\,12 shows that the mean loci of NGC\,6121
736: agree quite nicely with the CMD of NGC\,6712.  Small residual
737: differences in colour between the mean ridge line and the CMD in the
738: extreme red region and at the TO region are  probably due to small
739: residual uncertainties in the colour equation affecting the relatively
740: old photometry of NGC\,6171.
741: 
742: Note that the lack of any colour shift corresponds exactly  to what is
743: expected for these two clusters since they have formally the same
744: reddening (see Ferraro et al. 1991 for NGC\,6171).
745: 
746: To obtain a quantitative estimate of the relative age of the two
747: clusters we used the so-called {\it vertical method}:  this is  the
748: most widely used relative-age indicator, based on the TO luminosity
749: with respect to the HB level (Buonanno, Corsi \& Fusi Pecci 1989).
750:  
751: From Table 1, the TO level in NGC\,6712 is located at $V_{TO}=19.82\pm
752: 0.10$. Using this value and the $V_{ZAHB}$  computed in Section 4, we
753: derived $\Delta V_{TO}^{HB}=3.50\pm 0.15$ mag.  For NGC\,6171, assuming
754: $V_{TO}^{NGC\,6171}=19.20\pm0.10$ (Ferraro et al. 1991) and
755: $V_{ZAHB}^{NGC\,6171}=15.70\pm0.10$ (from F99), we found $\Delta
756: V_{TO}^{HB}=3.50\pm 0.15$ mag.
757:  
758: From  this comparison, we can conclude that NGC\,6712 is nearly coeval
759: with NGC\,6171.  On the other hand, NGC\,6171 has been found (R99) to
760: be coeval with  other clusters belonging to the same metallicity class
761: (NGC288, NGC\,6218, NGC\,6362, NGC\,6723).  Thus, we can safely
762: conclude that NGC\,6712 does not show any significant difference in age
763: with respect to the bulk of the intermediate metal-rich population of
764: GGC.
765:  
766:  
767: \subsection{Cluster distance and absolute age}
768: 
769: The evaluation of absolute ages requires the knowledge (or the implicit
770: assumption) of the distance modulus for the cluster. Different loci in
771: the CMD have been assumed in the past  as standard candles (HB - see
772: F99, MS - see Gratton et al. 1997, WD - see Renzini et al. 1996) to
773: determine cluster distances. F99 used the HB and, in particular, the
774: ZAHB as a reference candle and derived a new homogeneous distance scale
775: for a sample of 64 GGC. Here we use the same procedure to derive the
776: distance to NGC\,6712.
777: 
778: The Straniero Chieffi \& Limongi (1997) theoretical HB models were used
779: to derive the absolute magnitude of the ZAHB as a function of the
780: metallicity (see eq. 4 in F99). From this relation and assuming
781: $[Fe/H]_{CG97}=-0.8$ (section 5.2), we get $M_V^{ZAHB}=0.75$. Assuming
782: $V_{ZAHB}=16.32$ (see Section 4), we obtain an apparent distance
783: modulus $(m-M)_V=15.57$ and finally (considering $E(B-V)=0.33$, see
784: Section 5.2) a true modulus of $(m-M)_0=14.55$, which corresponds to a
785: distance of $\sim 8.1$ Kpc.
786: 
787: This value turns out to be significantly larger than that obtained by
788: C88 ($6.5$ Kpc). The origin of such a large discrepancy is due to many
789: differing assumptions between C88 and this paper.  First of all, the
790: reddening adopted here ($E(B-V)=0.33$) is quite different from that
791: used by  C88 ($E(B-V)=0.42$).  Second, the assumption of the absolute
792: level of the HB is different:  C88 used $M_V^{HB}=0.85$ and we used
793: $M_V^{HB}=0.75$ and finally there is a small difference (0.07 mag) in
794: the observed value of the HB level (C88 used $V_{HB}=16.25$).
795: 
796: In addition, C88 adopted the relation $A_V=3.2E(B-V)$ instead of the
797: canonical $R_V=3.1$. This (or even higher) value of $R_V$ can be
798: reasonable for NGC\,6712 since it is in a highly contaminated region.
799: Although the different assumption for $R_V$ between us and C88 produces
800: only a small effect($\sim 0.04$ mag), the impact of a different $R$
801: coefficient on the determination of the true distance modulus of
802: NGC\,6712 should be kept in mind.  At the moment, however, the main
803: source of uncertainty in the distance error budget  is  the reddening:
804: the assumed error  ($\pm 0.05$), in fact, produces an uncertainty of
805: $\sim 0.15$ mag in the distance modulus and an error of $\sim 6\%$
806: ($\sim 0.5$ Kpc) in the true distance.  Taking into account this and
807: the uncertainties on the other parameters concurring to the
808: determination of the distance, we finally conclude that the  global
809: uncertainty cannot be less than $10\%$ and assume $d=8\pm1$ Kpc as a
810: final conservative estimate of the distance to NGC\,6712.
811: 
812: From the distance modulus obtained above and $V_{TO}=19.82\pm0.10$
813: (from Table 1), we get $M_V^{TO}=4.25\pm0.2$.  With this value and the
814: metallicity estimate ($[Fe/H]_{CG97}=-0.8$) obtained in Section 5.2, we
815: can derive the age for NGC\,6712 using theoretical relations linking
816: these observables to the cluster age.  In doing so, we used the
817: relations recently obtained by R99 (see their Appendix A), for three
818: sets of theoretical models:  SCL97, Cassisi et al. (1998) and
819: VandenBerg et al. (2000) and obtained $t=12.6$,  $12.9$ and $11.5$ Gyr,
820: for SCL97, Cassisi et al. (1998) and VandenBerg et al. (2000),
821: respectively.  Taking into account that the global uncertainty of the
822: method can hardly be better than $15\%$, we conservatively assume
823: $t=12\pm2$ Gyr as the absolute age of NGC\,6712.
824: 
825:   
826: \section{Summary and Conclusions}
827: 
828: In this paper we have presented extensive, high resolution, $UBVR$
829: photometry of the galactic globular cluster NGC\,6712, obtained with
830: the ESO-Very Large Telescope.  This data set gives us the opportunity
831: to investigate the evolved stellar population of the cluster, yielding
832: new constraints on  its metallicity, age and  structural parameters.
833: In particular we find the following results:
834: 
835: \begin{itemize}
836:  
837: \item{\it Metallicity and Reddening:} The location and morphology of
838: the RGB  in the (V,B-V)-CMD, compared with a large set of reference
839: clusters, fully confirm that NGC\,6712 is an intermediate metal-rich
840: cluster. Using the method described in F99 we get $[M/H]=0.62 \pm 0.2$,
841: $[Fe/H]_{CG97}=-0.80\pm0.2$.  The method also yields an estimate of the
842: reddening which  turns out to be $E(B-V)=0.33\pm0.05$, a value
843: significantly lower than previously thought.
844:        
845: 
846: \item{\it Distance:} Using the theoretical level of the ZAHB (from
847: SCL97) as a reference candle, we derived $(m-M)_V=15.57$ and a true
848: distance modulus $(m-M)_0=14.55$ which corresponds to a distance of
849: about $\sim 8$ Kpc. This value is significantly higher than that
850: obtained by C88 (who found 6.5 Kpc), mainly because of the new reddening
851: determination.
852:  
853: \item{\it Age:} Coupling the apparent luminosity of the  ZAHB,
854: $V_{ZAHB}=16.32\pm0.05$ with $V_{TO}=19.82\pm0.10$, we get $\Delta
855: V_{TO}^{HB}=3.5\pm0.1$, a value fully compatible with that obtained in
856: other clusters.  This suggests that NGC\,6712 is nearly coeval, at
857: least, with other clusters with similar metallicity.  An average
858: absolute age of $t=12\pm 2 Gyr$ has been derived from comparison with
859: the theoretical models.
860:   
861: \item{\it BSS:} The  cluster reveals a large population of BSS: we
862: selected a sample of 108 BSS from $(V,B-V)$ and $(U,U-V)$ CMD. This
863: number is surprisingly large when compared with the BSS  population of
864: clusters with similar mass and central concentration.
865:  
866: \item{\it Post-AGB star:} We discovered the presence of a bright
867: UV-bright object in the core, whose position in the CMD closely matches
868: that of a star evolving in the post-AGB phase. A second objects with
869: similar characteristics has been located further out in the cluster at
870: $\sim 4\farcm4$ from the centre.
871:  
872: \end{itemize} 
873:  
874: 
875: The overall scenario emerging when we combine some of the results
876: listed above with those already presented in the two previous papers of
877: this series ({\it Paper I} and {\it II}), is quite exciting.
878:  
879: NGC\,6712 is experiencing a severe interaction with the disk and the
880: bulge of the Galaxy, as suggested by its orbit (Dauphole et al.
881: 1996).  The existence (presented in De Marchi et al. 1999, and
882: confirmed by {\it Paper II}) of an {\it inverted} mass function due to
883: a severe depletion of low mass stars in the MS luminosity function,
884: probably stripped away by the tidal force of the Galaxy, fully supports
885: this scenario.
886:  
887: Takahashi \& Portegies Zwart (2000), using N-body simulations,
888: interpreted this rare feature as a clear signature of severe mass loss
889: and suggested that NGC\,6712 might be only the remnant "core"  of a
890: once much more massive cluster, that  has lost most (up to $\sim 99
891: \%$) of its initial mass through the interaction with the Galaxy.  Thus
892: the hypothesis  that NGC\,6712 might have been much more massive (and,
893: possibly, much more concentrated) in the past, gained  support from a
894: theoretical point of view.
895: 
896: Now, this scenario finds additional support from a variety of
897: observational facts.  The presence of a LMXB and our recent discovery
898: ({\it Paper I}) of an additional promising interacting binary candidate
899: located a few arcsec away from the optical counterpart to the LMXB,
900: suggest that strong stellar interactions might have  occurred at some
901: remote stage of the cluster evolution.  Moreover, the large BSS
902: population discovered in this paper adds additional support to the fact
903: that star collisions might have occurred in the past,  generating most
904: (or part) of the observed BSS. At that time, NGC\,6712 probably was a
905: massive and concentrated cluster and collisional BSS (and other exotic
906: objects such as interacting binaries) formed copiously via dynamical
907: collisions. Then, these stars have migrated towards the centre, because
908: of mass segregation, where we now see them.  The continued action of
909: tidal stripping and disk shocking has removed most of the cluster mass,
910: driving it towards dissolution.  What we now observe is nothing but
911: the  remnant core of a disrupting cluster and its population of
912: peculiar objects, which are otherwise totally unexpected for its actual
913: mass.
914: 
915: If this scenario is further confirmed, our observations would have
916: served the purpose of reconstructing the stormy history of NGC\,6712
917: and possibly, for the very first time, to characterize the effects of
918: the disk shocking on the evolution of  stars in a globular cluster.
919: 
920:  
921: 
922:  
923: \acknowledgments
924: 
925: This research was partially supported by the {\it Agenzia Spaziale
926: Italiana} (ASI).  The financial support of  the {\it Ministero della
927: Universit\`a e della Ricerca Scientifica e Tecnologica} (MURST) to the
928: project {\it Stellar Dynamics and Stellar Evolution in Globular
929: Clusters} is kindly acknowledged.  F. R. F. gratefully acknowledges the
930: hospitality of the {\it Visitor Program} during his stay at ESO, when
931: most of this work has been carried out.  The data presented in this
932: paper were obtained as part of an ESO Service Mode programme.
933: 
934:  
935:  
936: \begin{references}
937: 
938: \reference{} Alcaino, G., 1977, PASP, 89, 491
939: 
940: \reference{}Anderson, S. F., Margon, B., Deutsch, E.,W., Downes, R.A.,
941:  1993, AJ, 106, 1049
942: 
943: \reference{} Andreuzzi, G., et al. , 2001, A\&A, in press {\it (Paper II)}
944: 
945: \reference{} Bailyn, C.D., 1992, ApJ, 392, 519
946: 
947: \reference{} Bailyn, C.D., 1995, ARA\&A, 33, 133
948: 
949: \reference{} Bailyn, C.D., Pinsonneault, M.H., 1995, ApJ, 439, 705
950: 
951: \reference{} Bellazzini, M., Ferraro, F.R., Buonanno, R., 1999,
952:  MNRAS, 307, 619
953: 
954: \reference{} Bertelli, G., Bressan, A., Chiosi, C., Fagotto, F., Nasi, E.
955: 1994,
956: A\&AS, 106, 275
957: 
958:  
959: \reference{} Buonanno, R., Buscema, G., Corsi, C.E., Ferraro, I.,
960:   Iannicola, G. 1983, A\&A, 126, 278
961: 
962: \reference{} Buonanno, R., Corsi, C. E., Fusi Pecci, F. 1989, A\&A, 216, 80
963: 
964: 
965: \reference{} Buonanno, R. $\&$ Iannicola, G. 1989, PASP, 101, 294
966: 
967: \reference{} Buonanno, R., 1993, in ASP  Conf. Ser. 48, The globular
968: cluster-Galaxy connection, ed. G.H. Smith \& J.P. Brodie,
969: (San Francisco: ASP), 131
970: 
971: \reference{} Buonanno, R., Corsi, C.E., Buzzoni, A., Cacciari, C., Ferraro,
972: F.R.,
973:   Fusi Pecci, F. 1994, A\&A, 290, 69 (PH94)
974: 
975: \reference{} Buonanno, R., Corsi, C., Bellazzini, M., Ferraro, F. R., Fusi
976: Pecci,
977: F. 1997, AJ, 113, 706
978: 
979: 
980: \reference{} Cacciari, C., Freeman, K. C. 1983, ApJ, 268, 185
981: 
982: 
983: \reference{} Carretta, E., Gratton, R. G. 1997, A\&AS, 121, 95
984: 
985: \reference{} Cassisi, S., Castellani, V., degl'Innocenti, S., Weiss, A. 1998,
986: 
987: A\&AS, 129, 267
988: 
989: \reference{} Cool, A. M., Grindlay, J. E., Cohn, H. N., Lugger, P. M.,
990: Slavin, S.
991: D. 1995, ApJ, 439, 695
992: 
993: \reference{} Cudworth, K.M. 1988, AJ, 96, 105 (C88)
994: 
995: \reference{} Dauphole, B., Geffert, M., Colin, J.,
996:  Ducourant, C., Odenkirchen, M.,
997:  Tucholke, H.-J., 1996, A\&A, 313, 119
998: 
999: \reference{}De Marchi, G., Leibundgut, B., Paresce, F., Pulone, L., 1999,
1000: A\&A, 343, 9
1001: 
1002: \reference{} Djorgovski, S. G. 1988,
1003: in { The Harlow-Shapley Symposium on Globular Cluster Systems in Galaxies},
1004: ed. J. E. Grindlay  \& A. G. D. Philip
1005: (Dordrecht: Kluwer),  333
1006: 
1007: 
1008: \reference{} Djorgovski, S., Meylan, G.,
1009: 1993, in Structure and Dynamics of Globular Clusters, ed.
1010: S. G. Djorgovski \& G. Meylan (ASP: San Francisco), 325
1011: 
1012: \reference{}Ferraro, F.R., Clementini, G., Fusi Pecci, F., Buonanno, R.,
1013: 1991,
1014:    MNRAS, 252, 357
1015: 
1016: 
1017: \reference{} Ferraro, F.R., Fusi Pecci, F., Buonanno, R., 1992,
1018:    MNRAS, 256, 376
1019: 
1020: \reference{} Ferraro, F.R., Clementini, G., Fusi Pecci, F., Sortino, R.,
1021:    Buonanno, R., 1992, MNRAS, 256, 391
1022: 
1023: \reference{} Ferraro, F.R., Fusi Pecci, F., Cacciari, C., Corsi, C.E.,
1024: Buonanno, R., Fahlman, G.G., Richer, H.B. 1993, AJ, 106, 2324
1025: 
1026: 
1027: \reference{} Ferraro, F.R., Bellazzini, M., Fusi Pecci, F., 1995, A\&A, 294,
1028: 80
1029: 
1030: \reference{} Ferraro, F.R., Paltrinieri, B., Fusi Pecci, F.,Cacciari, C.,
1031:  Dorman, B., Rood, R.T., Buonanno, R., Corsi, C.E., Burgarella, D., Laget,
1032: M.,
1033:  1997, A\&A, 324, 915
1034: 
1035: \reference{}Ferraro, F.R., Messineo, M., Fusi Pecci, F., De Palo, M.A.,
1036: Straniero, O., Chieffi, A., Limongi, M., 1999a, AJ, 118, 1738 (F99)
1037: 
1038: \reference{}Ferraro, F.R., Paltrinieri, B., Rood, R.T., Dorman, B.,
1039: 1999b, ApJ, 522, 983
1040: 
1041: \reference{} Ferraro, F. R., Paltrinieri, B., Paresce, F., De Marchi, G.
1042: 2000a,
1043: ApJ, 542, L29 {\it (Paper I)}
1044: 
1045: \reference{} Ferraro, F. R.,Montegriffo, P., Origlia, L., Fusi Pecci, F.,
1046: 2000b, AJ, 119, 1282
1047: 
1048: \reference{} Fusi Pecci, F., Ferraro, F.R. , Corsi, C.E., Cacciari, C.,
1049:   Buonanno, R. 1992, AJ, 104, 1831
1050: 
1051: 
1052: \reference{} Gratton, R. G., Fusi Pecci, F., Carretta, E., Clementini, G.,
1053: Corsi,
1054: C. E., Lattanzi, M. 1997, ApJ 491, 749
1055: 
1056: \reference{}Harris, W.E., 1996, AJ, 112, 1487
1057: 
1058: \reference{}Kron, G.E., Guetter, H.H., 1976, AJ, 81, 817
1059: 
1060: \reference{} Jacoby, G. H., Morse, J. A., Fullton, L. K., Kwitter, K. B.,
1061: Henry,
1062: R. B. C. 1997, AJ, 114, 2611
1063: 
1064: \reference{} Janulis, R., Smriglio, F., 1992, Baltic Astr., 1, 430
1065: 
1066: \reference{} King, I. R. 1966, AJ, 71, 64
1067: 
1068: 
1069: \reference{}Landolt, A.U., 1992, AJ, 104, 340
1070: 
1071: \reference{} Lee, Y. W., Demarque, P., Zinn, R. 1990, ApJ, 350, 155
1072: 
1073: \reference{}Martins, D.H., Harvel, C.A., 1981, ApJ, 250, 135
1074: 
1075: \reference{} Mighell, K. J., Rich, R. M., Shara, M., Fall, S. M. 1996,
1076: AJ, 111, 2314
1077: 
1078: \reference{} Nemec, J. M., Cohen, J. G. 1989, ApJ, 336, 780
1079: 
1080: \reference{} Paresce, F., Meylan, G., Shara, M., Baxter, D., Greenfield, P.,
1081: et al. , 1991, Nature, 352, 297
1082: 
1083: \reference{} Paltrinieri, B., Ferraro, F.R., Fusi Pecci, F., Rood, R.T.,
1084: Dorman, B., 1998, in Ultraviolet Astrophysics Beyond the IUE Final Archive
1085: ed. B. Harris, Noordwijk,  The Netherlands, p.565
1086: 
1087: \reference{} Pryor, C., Meylan, G.,
1088: 1993, in Structure and Dynamics of Globular Clusters, ed.
1089: S. G. Djorgovski \& G. Meylan (ASP: San Francisco), 357
1090: 
1091: 
1092: \reference{} Renzini, A., Bragaglia, A., Ferraro, F. R., Gilmozzi, R.,
1093: Ortolani,
1094: S., Holberg, J. B., Liebert, J., Wesemael, F., Bohlin, R. C. 1996, ApJL,
1095: 465, 23
1096: 
1097: \reference{}Rosenberg, A., Saviane, I, Piotto, G., Aparicio, A. 1999, AJ,
1098: 118,
1099: 2306 (R99)
1100: 
1101: \reference{} Rutledge, G. A., Hesser, J. E., Stetson, P. B., Mateo, M.,
1102: Simard,
1103: L., Bolte, M., Friel, E. D., Copin, Y. 1997, PASP, 109, 883
1104: 
1105: \reference{} Sandage, A. R., 1953, AJ, 58, 61
1106: 
1107: \reference{} Sandage, A. R., Smith, A., 1966, ApJ, 144, 866
1108: 
1109: \reference{} Salaris, M., Chieffi, A., Straniero, O. 1993, ApJ, 414, 580
1110: 
1111: \reference{} Sarajedini, A., 1993 in Blue Straggles, ed. R.A. Saffer, ASP
1112: Conf
1113: Series, 53, 14
1114: 
1115: \reference{} Sarajedini, A. 1994, PASP, 106, 205
1116: 
1117: \reference{}  Sosin, C., Piotto, G., Djorgovski, S. G., King, I.R., Rich, R.
1118: M.,
1119: Dorman, B., Liebert, J., Renzini, A. 1998, in Advances in Stellar Evolution
1120: ed.
1121: R.T.Rood \& A. Renzini (Cambridge: Cambridge U. Press), 92
1122: 
1123: \reference{} Straniero, O., Chieffi, A., Limongi, M. 1997, ApJ, 490, 425
1124: 
1125: \reference{} Takahashi, K., Portegies Zwart, S.F., 2000, ApJ, 535, 759
1126: 
1127: \reference{} Testa, V., Ferraro, F. R., Chieffi, A., Straniero, O., Limongi,
1128: M.,
1129: Fusi Pecci, F. 1999, AJ, 118, 2839
1130: 
1131: \reference{} Trager, S.C., Djorgovski, S.G., King, I.R. 1993
1132:  in { ASP Conf. Ser. 50, Structure and Dynamics of Globular Clusters, }
1133: ed. Djorgovski S. G. \& Meylan, G.,
1134: (San Francisco: ASP), 347
1135: 
1136: \reference{} Vandenberg, D. A., Stetson, P. B., Bolte, M. 1996, ARA\&A 34,
1137: 461
1138: 
1139: \reference{} Vandenberg, D. A., Swenson, F. J., Rogers, F. J., Iglesias, C.
1140: A., Alexander, D. R., 2000, ApJ, 532, 430
1141: 
1142: \reference{} Voges, W., et al.  1999, A\&A, 349, 389
1143: 
1144: \reference{} Webbink, R. F. 1986, Highlights of astronomy. Vol. 7 -Proceeding
1145:  of the Nineteenth IAU General Assembly, Delhi, India. Dordrecht, D. Reidel
1146: Publishing Co., 185
1147: 
1148: \reference{} Zinn, R. 1985, ApJ, 293, 424
1149: 
1150: 
1151: \reference{} Zinn, R., West, M. J. 1984, ApJS, 55, 45
1152: 
1153:  
1154: \end{references}
1155:  
1156: 
1157: \clearpage
1158: 
1159: \begin{deluxetable}{cccccccc}
1160: \tablewidth{\textwidth}
1161: \label{}
1162: \tablecaption{RGB-SGB Mean ridge line}
1163: \tablehead{
1164: \colhead{B-V} &
1165: \colhead{V} & 
1166:  &
1167: \colhead{B-V} &
1168: \colhead{V} & 
1169: &
1170: \colhead{B-V}&
1171: \colhead{V}
1172: }
1173: \startdata
1174:      2.110 &    13.510 &  &   1.214 &    16.700 & &    0.932 &     19.370 \nl
1175:     2.048 &    13.540 &   &  1.195 &    16.900 &  &  0.910 &     19.420 \nl
1176:     1.978 &    13.570 &   &  1.176 &    17.100 &  &   0.888 &     19.460 \nl
1177:     1.909 &    13.700 &   &  1.155 &    17.300 &  &   0.875 &     19.490 \nl
1178:     1.824 &    13.900 &   &  1.138 &    17.500 &  &   0.861 &     19.530 \nl
1179:     1.761 &    14.100 &   &  1.127 &    17.700 &  &   0.845 &     19.580 \nl
1180:     1.691 &    14.300 &   &  1.115 &    17.900 &  &   0.832 &     19.650 \nl
1181:     1.640 &    14.500 &   &  1.103 &    18.100 &  &   0.826 &     19.730 \nl
1182:     1.580 &    14.700 &   &  1.096 &    18.300 &  &   0.822 &     19.820 \nl
1183:     1.521 &    14.900 &   &  1.085 &    18.500 &  &   0.824 &     19.900 \nl
1184:     1.478 &    15.100 &   &  1.079 &    18.700 &  &   0.826 &     19.990 \nl
1185:     1.435 &    15.300 &   &  1.061 &    18.930 &  &   0.829 &     20.090 \nl
1186:     1.392 &    15.500 &   &  1.048 &    19.110 &  &   0.835 &     20.200 \nl
1187:     1.356 &    15.700 &   &  1.023 &    19.220 &  &   0.842 &     20.300 \nl
1188:     1.325 &    15.900 &   &  1.008 &    19.250 &  &   0.856 &     20.460 \nl
1189:     1.288 &    16.100 &   &  0.995 &    19.280 &  &   0.868 &     20.580 \nl
1190:     1.261 &    16.300 &   &  0.973 &    19.310 & &    0.894 &     20.810 \nl
1191:     1.240 &    16.500 &   &  0.953 &    19.330 & &    0.917 &     21.020 \nl
1192: \enddata
1193: \end{deluxetable}
1194:  
1195: 
1196: \clearpage
1197: 
1198: \begin{deluxetable}{ccccccccc}
1199: \tablewidth{\textwidth}
1200: \label{}
1201: \tablecaption{BSS candidates in NGC6712}
1202: \tablehead{
1203: &
1204: \colhead{\#} &
1205: \colhead{ID} &
1206: \colhead{V} & 
1207: \colhead{B} &
1208: \colhead{R} & 
1209: \colhead{U} &
1210: \colhead{X} & 
1211: \colhead{Y}  
1212:  }
1213: \startdata
1214:   &   BSS  1 &   111 & 19.45 & 20.18 & 18.95 & 20.46 & -400.973 & -998.374 \nl
1215:   &   BSS  2 &   144 & 19.36 & 20.10 & 18.85 & 20.20 &  804.297 & -980.577 \nl
1216:   &   BSS  3 &   262 & 18.56 & 19.27 & 18.08 & 19.58 &  728.953 & -916.481 \nl
1217:   &   BSS  4 &   540 & 17.75 & 18.33 & 17.35 & 18.65 &  792.174 & -791.430 \nl
1218:   &   BSS  5 &   670 & 18.02 & 18.68 & 17.57 & 19.05 & -322.614 & -743.712 \nl
1219:   &   BSS  6 &  1155 & 18.66 & 19.31 & 18.20 & 19.56 & -196.280 & -593.166 \nl
1220:   &   BSS  7 &  1177 & 18.57 & 19.24 & 18.11 & 19.48 & -117.403 & -598.961 \nl
1221:   &   BSS  8 &  1242 & 18.79 & 19.36 & 18.39 & 19.71 &  -45.187 & -580.513 \nl
1222:   &   BSS  9 &  1254 & 18.51 & 19.25 & 17.99 & 19.32 &  532.380 & -579.290 \nl
1223:   &   BSS 10 &  1456 & 18.93 & 19.62 & 18.47 & 19.84 &   20.270 & -526.844 \nl
1224:   &   BSS 11 &  1504 & 18.52 & 19.11 & 18.12 & 19.52 &  303.193 & -523.704 \nl
1225:   &   BSS 12 &  1568 & 18.36 & 18.93 & 17.99 & 19.24 & -398.783 & -481.762 \nl
1226:   &   BSS 13 &  1571 & 19.23 & 19.96 & 18.76 & 20.15 & -426.108 & -486.179 \nl
1227:   &   BSS 14 &  1742 & 18.76 & 19.43 & 18.28 & 19.65 &  -76.106 & -473.851 \nl
1228:   &   BSS 15 &  1850 & 18.58 & 19.35 & 18.06 & 19.63 & -638.202 & -449.283 \nl
1229:   &   BSS 16 &  1903 & 19.12 & 19.77 & 18.64 & 19.95 &  112.571 & -438.082 \nl
1230:   &   BSS 17 &  1906 & 18.77 & 19.54 & 18.22 & 19.76 &  120.261 & -427.270 \nl
1231:   &   BSS 18 &  1920 & 19.48 & 20.14 & 19.03 & 20.44 & -190.172 & -436.096 \nl
1232:   &   BSS 19 &  1930 & 19.00 & 19.66 & 18.54 & 19.92 & -888.934 & -431.771 \nl
1233:   &   BSS 20 &  2042 & 18.81 & 19.57 & 18.32 & 19.79 &  331.887 & -415.190 \nl
1234:  &   BSS 21 &  2428 & 19.39 & 20.14 & 18.88 & 20.33 &  528.692 & -351.818 \nl
1235:   &   BSS 22 &  2544 & 19.09 & 19.78 & 18.62 & 20.02 &  170.663 & -335.835 \nl
1236:   &   BSS 23 &  2546 & 19.36 & 20.05 & 18.91 & 20.23 &  162.471 & -335.651 \nl
1237:   &   BSS 24 &  2547 & 19.08 & 19.71 & 18.65 & 19.93 &  152.264 & -330.614 \nl
1238:   &   BSS 25 &  2623 & 18.79 & 19.49 & 18.30 & 19.67 & -330.428 & -300.632 \nl
1239:   &   BSS 26 &  2749 & 19.23 & 19.93 & 18.76 & 20.11 & -305.596 & -305.598 \nl
1240:   &   BSS 27 &  2965 & 19.25 & 19.96 & 18.78 & 20.10 & -293.869 & -268.150 \nl
1241:   &   BSS 28 &  2978 & 18.56 & 19.19 & 18.14 & 19.48 &  -45.357 & -269.560 \nl
1242:   &   BSS 29 &  3171 & 19.10 & 19.82 & 18.61 & 20.01 &  193.588 & -237.935 \nl
1243:   &   BSS 30 &  3253 & 18.72 & 19.30 & 18.31 & 19.56 &  244.559 & -231.358 \nl
1244:   &   BSS 31 &  3434 & 18.60 & 19.21 & 18.19 & 19.46 & -799.539 & -201.215 \nl
1245:   &   BSS 32 &  3501 & 18.79 & 19.47 & 18.33 & 19.68 & -805.423 & -188.600 \nl
1246:   &   BSS 33 &  3568 & 17.62 & 18.23 & 17.21 & 18.62 &  166.820 & -167.683 \nl
1247:   &   BSS 34 &  3684 & 18.69 & 19.49 & 18.14 & 19.67 &  208.239 & -157.790 \nl
1248:   &   BSS 35 &  3814 & 18.60 & 19.36 & 18.08 & 19.57 &  -59.101 & -139.850 \nl
1249:   &   BSS 36 &  4087 & 19.38 & 20.15 & 18.84 & 20.23 & -319.318 & -101.848 \nl
1250:   &   BSS 37 &  4157 & 17.68 & 18.13 & 17.36 & 18.56 & -790.025 &  -91.930 \nl
1251:   &   BSS 38 &  4239 & 19.39 & 20.13 & 18.89 & 20.29 & -492.933 &  -77.139 \nl
1252:   &   BSS 39 &  4329 & 19.29 & 20.07 & 18.80 & 20.22 &  543.722 &  -30.792 \nl
1253:   &   BSS 40 &  4332 & 18.32 & 19.05 & 17.82 & 19.23 &  633.899 &  -62.242 \nl
1254:   &   BSS 41 &  4337 & 18.61 & 19.17 & 18.22 & 19.46 &  739.292 &  -51.804 \nl
1255:   &   BSS 42 &  4350 & 19.32 & 19.97 & 18.88 & 20.11 & -764.300 &  -56.442 \nl
1256:   &   BSS 43 &  4431 & 17.95 & 18.40 & 17.61 & 18.75 &  186.752 &  -46.434 \nl
1257:   &   BSS 44 &  4463 & 18.87 & 19.49 & 18.45 & 19.69 & -411.709 &  -32.642 \nl
1258:   &   BSS 45 &  4469 & 18.63 & 19.23 & 18.22 & 19.49 &  295.844 &  -36.940 \nl
1259:   &   BSS 46 &  4470 & 19.28 & 19.90 & 18.84 & 20.08 &  287.043 &  -32.371 \nl
1260:   &   BSS 47 &  4528 & 18.87 & 19.60 & 18.36 & 19.82 &  -70.825 &  -32.440 \nl
1261:   &   BSS 48 &  4572 & 19.27 & 20.05 & 18.74 & 20.34 & -150.785 &  -25.653 \nl
1262:   &   BSS 49 &  4723 & 19.50 & 20.23 & 18.97 & 20.39 & -325.664 &   -1.551 \nl
1263:   &   BSS 50 &  4845 & 18.58 & 19.34 & 18.07 & 19.75 &  -82.528 &   19.827 \nl
1264:   &   BSS 51 &  5111 & 19.03 & 19.81 & 18.49 & 19.99 & -835.504 &   58.053 \nl
1265:   &   BSS 52 &  5181 & 18.39 & 18.99 & 17.99 & 19.32 &  451.931 &   80.364 \nl
1266:   &   BSS 53 &  5198 & 18.61 & 19.34 & 18.10 & 19.61 &  -56.833 &   69.893 \nl
1267:   &   BSS 54 &  5364 & 19.28 & 19.95 & 18.83 & 20.20 & -198.830 &  108.252 \nl
1268:   &   BSS 55 &  5514 & 19.01 & 19.82 & 18.47 & 20.03 & -336.323 &  121.265 \nl
1269:   &   BSS 56 &  5631 & 18.32 & 19.00 & 17.85 & 19.29 & -404.621 &  146.077 \nl
1270:   &   BSS 57 &  5712 & 19.13 & 19.80 & 18.65 & 19.97 &  329.344 &  150.904 \nl
1271:   &   BSS 58 &  5816 & 18.52 & 19.13 & 18.09 & 19.43 &  172.368 &  179.946 \nl
1272:   &   BSS 59 &  5938 & 17.85 & 18.46 & 17.43 & 18.90 &   50.804 &  191.088 \nl
1273:   &   BSS 60 &  6003 & 18.04 & 18.67 & 17.60 & 18.89 &  889.243 &  202.474 \nl
1274:   &   BSS 61 &  6014 & 17.58 & 18.07 & 17.24 & 18.50 &  621.351 &  212.900 \nl
1275:   &   BSS 62 &  6243 & 18.32 & 19.05 & 17.95 & 19.42 &  324.106 &  250.420 \nl
1276:   &   BSS 63 &  6244 & 19.08 & 19.77 & 18.63 & 19.95 &  317.493 &  259.019 \nl
1277:   &   BSS 64 &  6433 & 18.91 & 19.73 & 18.36 & 19.97 &   38.265 &  280.186 \nl
1278:   &   BSS 65 &  6556 & 19.19 & 19.97 & 18.65 & 20.11 & 1065.295 &  282.922 \nl
1279:   &   BSS 66 &  6594 & 18.65 & 19.29 & 18.21 & 19.48 & -947.905 &  292.308 \nl
1280:   &   BSS 67 &  6671 & 18.87 & 19.54 & 18.41 & 19.79 &  149.435 &  311.900 \nl
1281:   &   BSS 68 &  6699 & 18.96 & 19.77 & 18.42 & 19.96 & -267.150 &  322.899 \nl
1282:   &   BSS 69 &  6701 & 18.99 & 19.78 & 18.44 & 19.99 & -282.224 &  317.300 \nl
1283:   &   BSS 70 &  6797 & 18.98 & 19.74 & 18.52 & 19.96 &  114.881 &  326.295 \nl
1284:   &   BSS 71 &  6817 & 19.20 & 19.95 & 18.67 & 20.15 &  472.171 &  323.492 \nl
1285:   &   BSS 72 &  7298 & 18.76 & 19.42 & 18.31 & 19.66 & -325.091 &  413.188 \nl
1286:   &   BSS 73 &  7305 & 18.68 & 19.30 & 18.25 & 19.51 &  259.097 &  416.515 \nl
1287:   &   BSS 74 &  7506 & 18.43 & 19.16 & 17.92 & 19.41 &   40.085 &  454.635 \nl
1288:   &   BSS 75 &  7563 & 19.00 & 19.74 & 18.50 & 19.90 & -571.343 &  465.771 \nl
1289:   &   BSS 76 &  7963 & 19.08 & 19.86 & 18.54 & 20.01 &  470.976 &  541.285 \nl
1290:   &   BSS 77 &  8036 & 18.04 & 18.64 & 17.62 & 18.87 &  428.476 &  568.233 \nl
1291:   &   BSS 78 &  8193 & 18.19 & 18.81 & 17.74 & 19.07 &  306.234 &  601.330 \nl
1292:   &   BSS 79 &  8918 & 18.20 & 18.89 & 17.73 & 19.10 &  186.325 &  809.361 \nl
1293:   &   BSS 80 &  9113 & 19.04 & 19.82 & 18.50 & 19.99 &  479.085 &  879.267 \nl
1294:   &   BSS 81 &  9115 & 19.23 & 19.88 & 18.78 & 20.07 & -528.758 &  889.671 \nl
1295:   &   BSS 82 &  9294 & 19.24 & 20.02 & 18.71 & 20.09 &  288.298 &  954.655 \nl
1296:   &   BSS 83 &  9297 & 19.24 & 19.99 & 18.73 & 20.30 &  564.001 &  953.698 \nl
1297:   &   BSS 84 &  9317 & 19.03 & 19.67 & 18.58 & 19.85 &   -1.994 &  964.133 \nl
1298:   &   BSS 85 &  9589 & 19.15 & 19.87 & 18.66 & 20.08 & -316.575 & -115.017 \nl
1299:   &   BSS 86 &  9599 & 18.92 & 19.67 & 18.47 & 19.88 &   17.417 & -119.848 \nl
1300:   &   BSS 87 &  9618 & 18.34 & 19.05 & 17.84 & 19.30 &  -91.522 & -110.438 \nl
1301:   &   BSS 88 &  9688 & 18.03 & 18.63 & 17.60 & 18.96 &   11.660 &  -63.064 \nl
1302:   &   BSS 89 &  9723 & 19.23 & 19.92 & 18.73 & 20.15 & -261.537 &  -51.061 \nl
1303:   &   BSS 90 &  9740 & 18.28 & 19.00 & 17.80 & 19.32 & -120.423 &  -35.079 \nl
1304:   &   BSS 91 &  9763 & 18.31 & 18.96 & 17.86 & 19.23 &  -18.104 &  -27.268 \nl
1305:   &   BSS 92 &  9765 & 19.16 & 19.92 & 18.67 & 20.11 &  -19.759 &  -21.688 \nl
1306:   &   BSS 93 &  9797 & 18.18 & 18.71 & 17.83 & 19.25 & -104.733 &  -15.695 \nl
1307:   &   BSS 94 &  9798 & 18.59 & 19.29 & 18.12 & 19.59 & -115.067 &  -13.541 \nl
1308:   &   BSS 95 &  9909 & 18.46 & 19.10 & 18.02 & 19.40 & -185.532 &   66.507 \nl
1309:   &   BSS 96 &  9955 & 17.59 & 18.27 & 17.12 & 18.61 &   40.322 &   84.100 \nl
1310:   &   BSS 97 &  9961 & 19.13 & 19.89 & 18.59 & 20.13 &  177.678 &   71.732 \nl
1311:   &   BSS 98 & 10002 & 18.69 & 19.36 & 18.25 & 19.65 &   15.869 &  106.062 \nl
1312:   &   BSS 99 & 10014 & 18.31 & 18.86 & 17.93 & 19.22 &   77.432 &  115.310 \nl
1313:   &   BSS100 & 10046 & 18.77 & 19.57 & 18.29 & 19.83 & -233.953 &  123.758 \nl
1314:   &   BSS101 & 10053 & 19.22 & 19.92 & 18.72 & 20.14 & -228.244 &  132.875 \nl
1315:   &   BSS102 & 10092 & 17.48 & 18.12 & 17.04 & 18.60 &   16.717 &  167.770 \nl
1316:   &   BSS103 & 10159 & 18.58 & 19.33 & 18.13 & 19.58 &  108.702 &  202.048 \nl
1317:   &   BSS104 & 10226 & 18.40 & 19.04 & 18.05 & 19.05 & -194.530 & -178.189 \nl
1318:   &   BSS105 & 10277 & 19.14 & 19.83 & 18.66 & 20.07 &  -14.713 &  -80.536 \nl
1319:   &   BSS106 & 10286 & 19.14 & 19.87 & 18.67 & 20.13 & -220.658 &  -49.144 \nl
1320:   &   BSS107 & 10365 & 19.02 & 19.76 & 18.29 & 20.04 &  -46.893 &  -75.749 \nl
1321:   &   BSS108 & 10376 & 18.50 & 19.14 & 18.04 & 19.46 &   75.401 &  -18.944 \nl
1322:    &         &        &       &        &    &       &     & \nl
1323:   &   E-BSS  1 &   816 & 17.31 & 18.16 & 16.77 & 18.46 &    6.936 & -685.088 \nl
1324:   &   E-BSS  2 &  1692 & 17.33 & 18.18 & 16.82 & 18.60 &  477.852 & -483.990 \nl
1325:   &   E-BSS  3 &  2533 & 17.14 & 17.92 & 16.64 & 18.39 & -142.159 & -313.275 \nl
1326:   &   E-BSS  4 &  3437 & 17.19 & 17.94 & 16.68 & 18.37 &  768.774 & -200.574 \nl
1327:   &   E-BSS  5 &  3442 & 17.27 & 18.11 & 16.75 & 18.51 & -840.575 & -188.349 \nl
1328: \enddata
1329: \end{deluxetable}
1330:  
1331: 
1332: 
1333:  
1334: 
1335: \clearpage
1336: \centerline { FIGURE CAPTION}
1337: 
1338: 
1339: Fig. 1 ~~~Location of the five FORS1 fields observed in
1340: NGC\,6712.  Four fields were observed at Standard Resolution (SR):
1341: Q1,Q2,Q3,Q4 and one at High Resolution (HR), respectively.  The HR
1342: field is  roughly centered on the cluster centre.  The cluster centre
1343: is located at the origin of the coordinates.
1344: 
1345: \smallskip
1346: 
1347: Fig. 2 ~~~Calibration curves for the U, B, V, R filters.
1348: 
1349: 
1350: \smallskip
1351:  
1352: 
1353: Fig. 3 ~~~Internal photometric errors for the HR field. The
1354: rms values of the frame-to-frame scatter are plotted versus the U, B,
1355: V, R magnitude of each star.  
1356: 
1357: \smallskip
1358:  
1359: 
1360: Fig. 4 ~~~The $V$, $B-V$ colour magnitude diagramme for the
1361: whole sample of stars observed in NGC\,6712. In  Panel (a) only stars
1362: lieing in the FORS-HR field are plotted, while Panel (b),(c) and (d)
1363: show stars in annuli at increasing distance from the cluster centre.
1364: 
1365: \smallskip
1366: 
1367: 
1368: Fig. 5 ~~~The observed radial density profile for the
1369: evolved stellar population ({\it filled circles}). All stars plotted in
1370: Figure 4 with  $V<20$ have been counted. The stellar density has been
1371: scaled to match the density profile of unevolved MS stars
1372: with $19.5<R<20.5$ (corresponding to $\sim 0.75 M_{\odot}$) for $r>2'$
1373: ({\it large open squares}) see also Figure 8 in {\it Paper II}).  The
1374: dotted horizontal line represents the density of $ 0.75 M_{\odot}$
1375: star measured in the control field at $42'$ from the cluster centre.
1376: The distance $r=5'$ from the cluster centre is indicated by  a vertical
1377: arrow.
1378:  
1379: \smallskip
1380:  
1381: Fig. 6 ~~~The $V$, $B-V$ colour magnitude diagramme for
1382: stars observed in the HR field after the statistical decontamination
1383: from field stars (see text)
1384:  
1385:  
1386: \smallskip
1387: 
1388: Fig. 7 ~~~HB stars in NGC\,6712. The solid heavy line is
1389: the $V_{ZAHB}$ level, determined by F99. Seven RR Lyrae variables are
1390: plotted as large filled triangles.
1391: 
1392:  
1393: \smallskip
1394: 
1395: Fig. 8 ~~~Blue Stragglers Star candidates {\it (large
1396: filled circles)} in NGC\,6712.  The zoomed CMD in the BSS region in the
1397: $(V,B-V)$ {\it (panel a)} and $(U,U-V)$ {\it (panel b)}, plane.
1398: The 5 objects plotted as {\it stars} are suspected to be evolved (E-BSS)
1399: {\it i.e.} BSS in the SGB, which have left the MS to evolve toward the RGB.
1400:  
1401: \smallskip
1402: 
1403: Fig. 9 ~~~Cumulative radial distribution ($\phi$) for BSS
1404: (heavy solid line) compared to the RGB-SGB (dashed line) as a function
1405: of their projected distance ($r$) from the cluster centre in arcmin.
1406: 
1407: \smallskip
1408:  
1409: Fig. 10 ~~~$U,U-B$ CMD for NGC\,6712 from FORS1-HR images.
1410: The bright blue object $\#9620$ is plotted as a {\it large filled
1411: triangle}.  The three faint UV stars (plotted as {\it open triangles})
1412: are discussed in {\it Paper I}. An isochrone from Bertelli et al.
1413: (1994) is also plotted for reference.  As can be seen, the bright blue
1414: object is located along the post-AGB cooling sequence.
1415: 
1416: \smallskip
1417:  
1418: Fig. 11 ~~~$(H\alpha, H\alpha-R)$ CMD for NGC\,6712 from
1419: FORS1-HR images. The $H\alpha$ excess star  $\#10261$ (plotted as a
1420: {\it large filled triangle}) is the star discussed in {\it Paper I}.
1421: Blue HB stars (with $B-V<0.5$) are plotted as large filled circles, BSS
1422: as selected in Section 6 are plotted as large asterisks.  The small
1423: open square is the bright blue object discussed in Section 7.1.
1424: 
1425: \smallskip
1426:  
1427:  
1428: Fig. 12 ~~~The mean ridge line of NGC\,6171 shifted
1429: [$\Delta V=0.62, \Delta (B-V)=0.0$] to match the CMD of NGC\,6712.
1430: 
1431:  
1432: 
1433: \end{document}
1434:  
1435: 
1436:  
1437: 
1438: 
1439: