0708.2220/ms.tex
1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass[apjl]{emulateapj}
3: \journalinfo{To appear in The Astronomical Journal}
4: 
5: \def\kms{\ifmmode{\rm km\thinspace s^{-1}}\else km\thinspace s$^{-1}$\fi}
6: \def\hr{HR~6046}
7: \def\hip{{\it Hipparcos\/}}
8: 
9: \shortauthors{Torres}
10: \shorttitle{\hr}
11: 
12: \begin{document}
13: 
14: \title{HIPPARCOS astrometric orbit and evolutionary status of
15: \hr}
16: 
17: \author{Guillermo Torres}
18: 
19: \affil{Harvard-Smithsonian Center for Astrophysics, 60 Garden St.,
20: Cambridge, MA 02138}
21: 
22: \email{gtorres@cfa.harvard.edu}
23: 
24: \begin{abstract} 
25: 
26: The previously known, 6-yr spectroscopic binary \hr\ has been
27: speculated in the past to contain a compact object as the secondary. A
28: recent study has re-determined the orbit with great accuracy, and
29: shown that the companion is an evolved but otherwise normal star of
30: nearly identical mass as the primary, which is also a giant. The
31: binary motion was detected by the \hip\ mission but was not properly
32: accounted for in the published astrometric solution. Here we use the
33: \hip\ intermediate data in combination with the spectroscopic results
34: to revise that solution and establish the orbital inclination angle
35: for the first time, and with it the absolute masses $M_{\rm A} =
36: 1.38_{-0.03}^{+0.09}$~M$_{\sun}$ and $M_{\rm B} =
37: 1.36_{-0.02}^{+0.07}$~M$_{\sun}$. Aided by other constraints, we
38: investigate the evolutionary status and confirm that the primary star
39: is approaching the tip of the red-giant branch, while the secondary is
40: beginning its first ascent.
41: 	
42: \end{abstract}
43: 
44: \keywords{astrometry --- binaries: general --- binaries: spectroscopic
45: --- methods: data analysis --- stars: fundamental parameters ---
46: stars: individual (\hr)}
47: 
48: \section{Introduction}
49: \label{sec:introduction}
50: 
51: Since the publication of the first preliminary orbits in the 1930's by
52: \cite{Christie:34, Christie:36}, the bright giant star \hr\
53: (HD~145849, HIP~79358, $\alpha = 16^{\rm h} 11^{\rm m} 48\fs05$,
54: $\delta = +36\arcdeg 25\arcmin 30\farcs3$, J2000.0; spectral type
55: \ion{K3}{2}, $V = 5.63$) has been known as a highly eccentric
56: $\sim$6-yr period single-lined spectroscopic binary of particular
57: interest. The relatively large minimum mass inferred early on for the
58: companion ($> 3$~M$_{\sun}$), along with the fact that it was not
59: visible to early observers, led to the speculation that the secondary
60: was a collapsed star \citep{Trimble:69}, more commonly referred to
61: nowadays as a black hole.  Very recently \cite{Scarfe:07} have
62: dispelled this notion by detecting the secondary spectroscopically and
63: showing that it is merely an evolved late-type star, with nothing
64: particularly out of the ordinary in its properties other than the fact
65: that it is faint. On the basis of extensive radial velocity
66: measurements carried out over more than 26 years these authors
67: presented an accurate double-lined orbit for the system yielding
68: minimum masses considerably smaller than previously thought. They also
69: found the components to be nearly identical in mass to within about
70: 1\%, even though they are substantially different in brightness
71: ($\Delta V \approx 3$). They relied on the \hip\ parallax of the
72: system \citep{Perryman:97} to construct a plausible photometric model
73: of the stars (individual magnitudes and colors), but cautioned that
74: the distance could be vitiated by unmodeled photocentric motion.
75: 
76: As described by \cite{Scarfe:07}, \hr\ has never been spatially
77: resolved despite repeated attempts over the years using speckle
78: interferometry and other techniques. However, \cite{Jancart:05} were
79: able to obtain an estimate of the inclination angle by reconsidering
80: the \hip\ Intermediate Astrometric Data in conjunction with the
81: spectroscopic orbit of \cite{Christie:36} to account for the binary
82: motion. With this the absolute masses of the components can in
83: principle be obtained, although the resulting values are somewhat high
84: for late-type giants ($\sim$3.6 M$_{\sun}$) possibly because of the
85: use of outdated spectroscopic elements.
86: 
87: The main motivation for the present work is to revisit the \hip\
88: astrometric solution in the light of the accurate orbit of
89: \cite{Scarfe:07}, and to examine the effect on the published
90: parallax. Additionally, we investigate the evolutionary status of the
91: system with our newly determined masses, aided by current stellar
92: evolution models. We report also a spectroscopic determination of the
93: effective temperature of the primary star that supports the general
94: picture outlined by other constraints in showing the evolved state of
95: the binary.
96: 
97: \section{HIPPARCOS observations and revised astrometric solution}
98: \label{sec:orbit}
99: 
100: Given the 6-yr orbital period of \hr, which is of the same order as
101: the duration of the \hip\ mission (slightly more than 3 yr), it may be
102: expected that the orbital motion would be detectable in the satellite
103: measurements, and that if not properly accounted for, it could bias
104: either the trigonometric parallax, the proper motions, or both.
105: Indeed, proof that a signal of this nature is detectable is given by
106: the fact that the \hip\ team found it necessary to include
107: acceleration terms in the astrometric solution, representing the first
108: derivatives of the proper motions. These terms, $d\mu_{\alpha}^*/dt =
109: -9.08 \pm 1.23$ mas~yr$^{-2}$ and $d\mu_{\delta}/dt = -6.98 \pm 1.53$
110: mas~yr$^{-2}$, are statistically significant. It is unclear why a full
111: orbital model was not applied originally to this system, since the
112: binary nature of the object has been known for a long time. In any
113: case, this has been done more recently by \cite{Jancart:05}, as
114: mentioned earlier.
115: 
116: A total of 78 astrometric measurements were made by \hip\ from 1989
117: December to 1993 February, covering about 53\% of an orbital cycle.
118: Each measurement consisted of a one-dimensional position (`abscissa',
119: $v$) along a great circle representing the scanning direction of the
120: satellite, tied to an absolute frame of reference known as the
121: International Celestial Reference System (ICRS). The measurements
122: available for our analysis are published in the form of `abscissa
123: residuals' ($\Delta v$; see Table~\ref{tab:abscissae}), which are the
124: residuals from the standard five-parameter solution reported in the
125: Catalogue \citep{Perryman:97}.  The five standard parameters are the
126: position ($\alpha_0^*$, $\delta_0$) and proper motion components
127: ($\mu_{\alpha}^*$, $\mu_{\delta}$) of the barycenter at the reference
128: epoch 1991.25\footnote{Following the practice in the \hip\ Catalogue
129: we define $\alpha^* \equiv \alpha \cos\delta$ and $\mu_{\alpha}^*
130: \equiv \mu_{\alpha} \cos\delta$.}, and the trigonometric parallax,
131: $\pi_{\rm t}$. The nominal errors of these measurements have a median
132: value of 2.5 mas.
133: 
134: We incorporated orbital motion into a new astrometric solution
135: following the formalism of \cite{vanLeeuwen:98}, \cite{Pourbaix:00},
136: and \cite{Jancart:05}, including the correlations between measurements
137: from the two independent data reduction consortia that processed the
138: original \hip\ observations \citep[NDAC and FAST;][]{Perryman:97}.
139: Details of this modeling along with another example of an application
140: may also be seen in \cite{Torres:07b}. In the most general case the
141: adjustable quantities of the fit (a total of 12) are the corrections
142: to the five standard \hip\ parameters ($\Delta\alpha^*$,
143: $\Delta\delta$, $\Delta\mu_{\alpha}^*$, $\Delta\mu_{\delta}$, and
144: $\Delta\pi_{\rm t}$), and the seven usual elements of the binary
145: orbit: the period $P$, the semimajor axis of the photocenter $a_{\rm
146: phot}$, the eccentricity $e$, the inclination angle $i$, the longitude
147: of periastron for the primary $\omega_{\rm A}$, the position angle of
148: the ascending node $\Omega$ (equinox J2000.0), and the time of
149: periastron passage $T$. For \hr\ the \hip\ measurements do not
150: constrain $P$, $e$, $\omega_{\rm A}$, or $T$ sufficiently well, so
151: those elements were held fixed at their spectroscopic values as
152: reported by \cite{Scarfe:07}. We solved for the remaining 8 parameters
153: simultaneously using standard non-linear least-squares techniques
154: \citep[][p.\ 650]{Press:92}. The reduced $\chi^2$ of the solution is
155: 0.9895, indicating that the internal errors of the abscissa residuals
156: are realistic.
157: 
158: We list the results in Table~\ref{tab:elements}, where they are
159: compared with those obtained by \cite{Jancart:05}. Significant
160: differences are seen in the semimajor axis and inclination angle,
161: which reflect the spectroscopic constraints used in each case.
162: Jancart's value of $a_{\rm phot}$ is nearly twice as large as ours,
163: and while their inclination angle is substantially smaller than
164: 90\arcdeg, we obtain an orientation that is essentially edge-on. The
165: smaller $i$ value of \cite{Jancart:05} would lead to much larger
166: absolute masses for the stars when combined with the minimum masses of
167: \cite{Scarfe:07}, as noted in \S\ref{sec:introduction}.  There is
168: little doubt that the recent orbital elements by \cite{Scarfe:07}
169: supersede the provisional values of \cite{Christie:36} adopted by
170: Jancart, which suggests our solution should be closer to the
171: truth.\footnote{A preliminary version of the \cite{Scarfe:07} work had
172: been published in 2004 \citep{Scarfe:04}, but was perhaps not yet
173: available to Jancart and collaborators.} An external check on our
174: results is provided by the absolute proper motions we obtain, listed
175: in the bottom section of Table~\ref{tab:elements}. These are different
176: from the values reported in the \hip\ Catalogue, which are likely to
177: be affected also by the unmodeled orbital motion. They agree well,
178: however, with the motions given in the Tycho-2 Catalogue
179: \citep{Hog:00}, as shown in Table~\ref{tab:pm}. The latter are based
180: on the position from the Tycho experiment aboard the \hip\ satellite
181: (epoch $\sim$1991.25) combined with ground base catalog positions
182: going back nearly a century in some cases.  This long baseline tends
183: to average out any orbital motion that has a period of a few years, as
184: in the case of \hr, and therefore yields a more accurate estimate than
185: {\it Hipparcos\/}.  Finally, we note that our revised \hip\ solution
186: did not change the value of the parallax significantly, which was one
187: of the concerns of \cite{Scarfe:07}. This parallax corresponds to a
188: distance of 178 pc.
189: 
190: \begin{figure}
191: \vskip -0.3in
192: \epsscale{1.3}
193: {\hskip -0.2in \plotone{f1.eps}}
194: \vskip -0.3in 
195: \figcaption[]{(a) Abscissa residuals of \hr\ from the \hip\ mission as
196: a function of orbital phase. These are the residuals from the standard
197: 5-parameter solution as published in the Catalogue
198: \citep{Perryman:97}.  The systematic patterns are due to the unmodeled
199: motion of the center of light. (b) $O\!-\!C$ residuals of the \hip\
200: measurements from our new 12-parameter fit that accounts for the
201: orbital motion.\label{fig:hipresid}}
202: \end{figure}
203: 
204: As a way of visualizing the effect of accounting for the orbital
205: motion in the \hip\ solution, we show in Figure~\ref{fig:hipresid}a
206: the abscissa residuals $\Delta v$ resulting from the standard
207: 5-parameter solution as a function of orbital phase. Some systematic
208: patterns are apparent, but they largely disappear after the orbital
209: motion is incorporated into the model. This is shown by the $O\!-\!C$
210: residuals from the 12-parameter fit in Figure~\ref{fig:hipresid}b.
211: 
212: \begin{figure} 
213: \vskip -0.6in
214: \epsscale{1.3} 
215: {\hskip -0.2in \plotone{f2.eps}}
216: %\vskip 0.0in 
217: \figcaption[]{Motion of the photocenter of \hr\ relative to the center
218: of mass of the binary (indicated by the plus sign) as seen by
219: \hip. The one-dimensional abscissa residuals are shown schematically
220: with a filled circle at the predicted location, dotted lines
221: representing the scanning direction of the satellite, and short
222: perpendicular line segments indicating the undetermined location of
223: the measurement on that line (see text). The length of the dotted
224: lines represents the magnitude of the $O\!-\!C$ residual from the
225: computed location.  Three measurements with residuals larger than 5
226: mas were omitted for clarity. Also indicated on the plot is the
227: location of periastron (open circle near the
228: top).\label{fig:hiporbit}}
229: \end{figure}
230: 
231: The projection of the photocentric orbit on the plane of the sky along
232: with a schematic representation the \hip\ measurements is seen in
233: Figure~\ref{fig:hiporbit}, where the axes are parallel to the right
234: ascension and declination directions.  Because these measurements are
235: one-dimensional in nature, their exact location on the plane of the
236: sky cannot be shown graphically. The filled circles represent the
237: predicted location on the computed orbit (see also
238: Figure~\ref{fig:deproject}).  The dotted lines connected to each
239: filled circle indicate the scanning direction of the \hip\ satellite
240: for each measurement, and show which side of the orbit the residual is
241: on. The length of each dotted line represents the magnitude of the
242: $O\!-\!C$ residual.\footnote{The ``$O\!-\!C$ residuals'' are not to be
243: confused with the ``abscissa residuals'' $\Delta v$, which we refer to
244: loosely here as \hip\ ``observations'' or ``measurements''. As
245: indicated earlier, the abscissa residuals are in fact residuals from
246: the standard 5-parameter fit reported in the \hip\ Catalogue, whereas
247: the $O\!-\!C$ residuals (or simply ``residuals'') are the difference
248: between the abscissa residuals and the computed position of the star
249: from a model that incorporates orbital elements.}  The short line
250: segments at the end of and perpendicular to the dotted lines indicate
251: the direction along which the actual observation lies, although the
252: precise location is undetermined. Occasionally more than one
253: measurement was taken along the same scanning direction, in which case
254: two or more short line segments appear on the same dotted lines. The
255: orbit is formally clockwise (retrograde), although the orientation is
256: so close to edge-on ($i = 91\fdg4 \pm 9\fdg7$) that the motion on the
257: plane of the sky may well be direct.
258: 
259: \begin{figure} 
260: \vskip -0.6in
261: \epsscale{1.3} 
262: {\hskip -0.2in \plotone{f3.eps}}
263: %\vskip 0.0in 
264: \figcaption[]{Same as Figure~\ref{fig:hiporbit}, except that the orbit
265: has been de-projected to appear as if it were viewed exactly face-on.
266: The line of nodes is indicated with the dashed line. Motion on the
267: plane of the sky in this figure is direct (counterclockwise).
268: \label{fig:deproject}}
269: \end{figure}
270: 
271: The phase coverage of the \hip\ observations is seen more clearly in
272: Figure~\ref{fig:deproject}, in which we have de-projected the orbit
273: and represented it as if it were viewed with an inclination angle of
274: 0\arcdeg. The measurements happen to cover the periastron passage of
275: 1991.83 (indicated with an open circle), but not apastron.  The path
276: of \hr\ on the plane of the sky as seen by \hip\ is shown in
277: Figure~\ref{fig:hippath}.  The irregular pattern is the result of the
278: combination of annual proper motion (indicated by the arrow),
279: parallactic motion, and orbital motion.
280: 
281: \begin{figure}
282: \vskip -0.6in
283: \epsscale{1.3}
284: {\hskip -0.2in \plotone{f4.eps}}
285: %\vskip 0.0in 
286: \figcaption[]{Path of the center of light of \hr\ on the plane of the
287: sky, along with the \hip\ observations (abscissa residuals)
288: represented as in Figure~\ref{fig:hiporbit} and
289: Figure~\ref{fig:deproject}. The irregular motion is the result of the
290: combined effects of the annual parallax, proper motion, and orbital
291: motion according to the solution described in the text.  The arrow
292: indicates the direction and magnitude of the annual proper
293: motion.\label{fig:hippath}}
294: \end{figure}
295: 
296: The detection of the photocentric motion by \hip\ offers an
297: independent way of estimating the brightness difference between the
298: stars, in the passband of the satellite ($H_p$ band), from the
299: relation between the semimajor axis of the photocentric orbit and that
300: of the relative orbit in angular units \citep[e.g.,][]{vandeKamp:67}:
301: $a_{\rm phot} = a_{\rm rel}(B-\beta)$. Here $a_{\rm rel}$ can be
302: obtained from the spectroscopic minimum masses, the inclination angle,
303: and the parallax through Kepler's Third Law. $B$ represents the mass
304: fraction $M_{\rm B}/(M_{\rm A}+M_{\rm B})$, and $\beta$ is the light
305: fraction given by $L_{\rm B}/(L_{\rm A}+L_{\rm B}) = (1+10^{0.4\Delta
306: H_p})^{-1}$. The magnitude difference is then
307: %
308: \begin{equation}
309: \Delta H_p = 2.5 \log \left[ \left({q\over 1+q}-{a_{\rm phot}\over
310: a_{\rm rel}}\right)^{-1} - 1 \right]~,
311: \end{equation}
312: %
313: in which $q \equiv M_{\rm B}/M_{\rm A}$.  We obtain $\Delta H_p = 2.1
314: \pm 0.6$ mag, somewhat smaller but still broadly consistent with the
315: values in the Johnson $B$ and $V$ bands reported by \cite{Scarfe:07}
316: (see below), considering the difference in the passbands.
317: 
318: \section{Evolutionary status}
319: \label{sec:evolution}
320: 
321: From the analysis of their high-resolution spectra of \hr\
322: \cite{Scarfe:07} estimated the spectral types and luminosity classes
323: of the components to be \ion{K3}{2} and \ion{K0}{4}, and the
324: brightness difference to be $\Delta B \approx 2.5$ mag and $\Delta V
325: \approx 3.0$ mag. They then made use of the \hip\ parallax (which we
326: now know is substantially correct) and the combined system brightness
327: to infer absolute visual magnitudes of $M_V = -0.68$ for the primary
328: and $M_V = 2.32$ for the secondary, ignoring extinction.  $U\!-\!B$
329: and $B\!-\!V$ color indices were computed by making use of standard
330: tabulated values for spectral types and luminosity classes close to
331: what they derived, along with the magnitude differences. The
332: photometry synthesized in this way provides a very good match to the
333: observed $B\!-\!V$ color of the system, and enabled \cite{Scarfe:07}
334: to claim that both stars are evolved, with the primary being near the
335: tip of the giant branch. They also pointed out that evolutionary
336: models by \cite{Pols:98} allow the brightness difference to be as
337: large as observed in \hr\ for component masses that differ by as
338: little as they do in this system ($\sim$1\%), at least for masses in
339: the most probable range for \hr, which they estimated to be 2.0--3.5
340: M$_{\sun}$. This provided a natural explanation for a set of
341: properties that had intrigued earlier investigators.
342: 
343: The minimum masses from the spectroscopic work of \cite{Scarfe:07}
344: combined with our inclination angle give absolute masses for \hr\ of
345: $M_{\rm A} = 1.38_{-0.03}^{+0.09}$~M$_{\sun}$ and $M_{\rm B} =
346: 1.36_{-0.02}^{+0.07}$~M$_{\sun}$ (Table~\ref{tab:elements}), in which
347: the uncertainties are currently dominated by the error in $i$. These
348: masses are considerably smaller than assumed by those authors. We have
349: therefore re-examined the consistency with the evolutionary models by
350: making use of these values along with slightly revised absolute
351: magnitudes for the stars based on our new parallax. Given a distance
352: approaching 200~pc, a small amount of extinction would not be
353: unexpected for the system, and in fact the dust maps of
354: \cite{Schlegel:98} indicate a total reddening in the direction of \hr\
355: of $E(B\!-\!V) \sim 0.03$ mag. We tentatively adopt $E(B\!-\!V) \sim
356: 0.02 \pm 0.01$ mag, which corresponds to $A_V \sim 0.06 \pm 0.03$ mag.
357: With this adjustment to the apparent magnitude $V = 5.63 \pm 0.01$
358: \citep{Haggkvist:87}, and taking into account the magnitude difference
359: from \cite{Scarfe:07} (to which we assign, somewhat arbitrarily, an
360: uncertainty of 0.3 mag), we infer individual magnitudes of $M_V^{\rm
361: A} = -0.62 \pm 0.20$ and $M_V^{\rm B} = 2.38 \pm 0.34$. These are not
362: changed much from the original estimates of \cite{Scarfe:07}.  The
363: $B\!-\!V$ colors proposed by them rely on an external tabulation and
364: are quite sensitive to the luminosity class adopted, so we have
365: preferred to proceed without them here, although they may well be
366: accurate. The models by \cite{Girardi:00} are a convenient choice for
367: a comparison with the observations, since they reach the late
368: evolutionary stages needed for the primary and are tabulated for a
369: wide range of chemical compositions and ages. They also incorporate
370: mass loss due to winds in the giant phase, which turns out to be only
371: a 1--2\% effect for \hr.
372: 
373: \begin{figure}
374: \vskip -0.4in
375: \epsscale{1.4}
376: {\hskip -0.35in \plotone{f5.eps}}
377: \vskip -0.25in
378: \figcaption[]{Model isochrones from \cite{Girardi:00} that match the
379: measured absolute magnitudes (grey areas) and current masses of \hr\
380: within the observational errors, shown for a range of metallicities
381: and ages as labeled. Open circles on each isochrone indicate the
382: expected location of the components, and the plus signs represent the
383: predicted system magnitude and color. The system magnitude and color
384: actually measured are represented by the filled circle and error
385: bar. The model drawn with the heavy line satisfies the additional
386: constraint given by the observed $B\!-\!V$ color of the system.
387: \label{fig:isochrones}}
388: \end{figure}
389: 
390: In Figure~\ref{fig:isochrones} we show several isochrones for
391: different metallicities from this series of models that produce the
392: best simultaneous match to the absolute magnitudes of both components
393: at their current masses (accounting for mass loss). In each case
394: (except for the middle isochrone) the age is the one providing the
395: best agreement among the values tabulated by \cite{Girardi:00}, which
396: come in steps of 0.05 in $\log t$.  The location of the stars on these
397: models is indicated with open circles. The combined color and
398: magnitude predicted for the system in each case is shown by the plus
399: signs, and the filled circle represents the measured values of $M_V$
400: and $B\!-\!V$ for \hr. The system color we have adopted is $B\!-\!V =
401: 1.34 \pm 0.01$ \citep{Haggkvist:87}, which we have then dereddened as
402: indicated above.  Properties of the stars inferred from these models
403: are listed in Table~\ref{tab:models}. If we now impose the additional
404: requirement that the isochrones match the measured system color within
405: its uncertainty, a very good correspondence may be obtained by
406: interpolation for a metallicity of $Z = 0.011 \pm 0.002$
407: (corresponding to [Fe/H] $= -0.24 \pm 0.07$ for the assumed $Z_{\sun}
408: = 0.019$ in these models) and a logarithmic age of $\log t =
409: 9.49_{-0.09}^{+0.04}$ (or $3.1_{-0.6}^{+0.3}$~Gyr). This model is
410: shown with the heavy line in Figure~\ref{fig:isochrones}, and
411: indicates that the primary is indeed approaching the tip of the giant
412: branch, whereas the secondary is beginning its first ascent of the
413: giant branch. We infer from this best-fitting isochrone that the
414: primary star has a radius of $30.9_{-1.1}^{+3.6}$~R$_{\sun}$, a
415: surface gravity of $\log g = 1.60_{-0.08}^{+0.04}$, and an effective
416: temperature ($T_{\rm eff}$) of $4211_{-23}^{+16}$~K, and the secondary
417: has $R = 4.20_{-0.33}^{+0.27}$~R$_{\sun}$, $\log g =
418: 3.33_{-0.04}^{+0.07}$, and $T_{\rm eff} = 5010_{-47}^{+79}$~K. These
419: uncertainties account for all observational errors but exclude
420: possible systematics in the models.  Other properties of the stars for
421: this metallicity are listed in Table~\ref{tab:models}.
422: 
423: \subsection{Additional spectroscopic constraints}
424: 
425: With the goal of providing a check on the temperature of the primary
426: star, we obtained a high-resolution spectrum of \hr\ on UT 5 July 2007
427: using the HIRES instrument \citep{Vogt:94} on the Keck~I telescope.
428: The spectrometer slit was $0\farcs86$, giving a resolving power of
429: $\lambda/\Delta\lambda \approx 55,\!000$. The signal-to-noise ratio of
430: this spectrum is $\sim$120 per pixel. The spectral region
431: $\lambda\lambda$6200--6300~\AA\ was used to measure a number of
432: temperature-sensitive lines mainly of \ion{Fe}{1} and \ion{V}{1}, and
433: also a few of \ion{Ni}{1}, \ion{Sc}{1}, \ion{Si}{1}, and
434: \ion{Fe}{2}. As described by \cite{Gray:91} and others, line-depth
435: ratios (LDRs) of properly selected line pairs are an excellent
436: diagnostic of effective temperature that allow for \emph{relative}
437: measurements with a precision as small as a few K, not only in dwarfs
438: but also in giants \citep{Gray:01}.  The conversion to an
439: \emph{absolute} temperature scale, however, necessarily depends on an
440: external color-temperature calibration since LDR variations are
441: usually compared with corresponding changes in a color index such as
442: $B\!-\!V$, which is an easier quantity to measure than temperature.
443: Therefore, absolute temperatures are much less certain. We measured a
444: total of 26 lines of the primary star\footnote{The secondary is too
445: faint to affect these measurements, and would not influence them even
446: if it were brighter \citep[see][]{Gray:01}.} from Table~1 by
447: \cite{Biazzo:07}, and used their calibrations appropriate for giants
448: for 16 selected line pairs to derive an average temperature from the
449: LDRs of $T_{\rm eff} = 4340 \pm 20$~K (formal error).  The scatter of
450: the individual determinations is only 80~K. The color-temperature
451: relation on which the above calibrations are based is that of
452: \cite{Gray:05}, which combines dwarf and giant temperatures obtained
453: by many different methods. For this work we have preferred to rely on
454: a more sophisticated color-temperature relation such as that by
455: \cite{Ramirez:05}, which accounts not only for luminosity class but
456: also metallicity, and is based on effective temperatures derived
457: homogeneously by the Infrared Flux Method. The conversion from the
458: above LDR-based $T_{\rm eff}$ back to an average color for the primary
459: star was made using the same prescription by \cite{Gray:05}, and gives
460: $B\!-\!V = 1.328 \pm 0.014$. The \cite{Ramirez:05} calibration for an
461: adopted metallicity of [Fe/H] $= -0.24$ (\S\ref{sec:evolution}) then
462: yields $T_{\rm eff} = 4210 \pm 60$~K, in which the error combines
463: photometric uncertainties and the scatter of the calibration. This
464: result is in virtually perfect agreement with the temperature
465: predicted by the \cite{Girardi:00} models, supporting our overall
466: conclusions from the previous section.
467: 
468: \section{Discussion and concluding remarks}
469: 
470: The example of \hr\ is one of a growing number of binaries in which
471: the \hip\ intermediate data have been brought to bear on the orbit of
472: the system, providing complementary information to that afforded by
473: other types of observations \citep[see, e.g.,][]{Pourbaix:04,
474: Fekel:05, Fekel:07, Torres:06, Torres:07a, Torres:07b}. In this case
475: the \hip\ data yield the inclination angle, and allow the absolute
476: masses to be derived for the first time. Additionally they provide an
477: estimate of the brightness difference.
478: 
479: Although the inclination we derive for the orbit is consistent with
480: 90\arcdeg\ and allows for the possibility of eclipses, chances are
481: slim because of the long period compounded by the relatively large
482: uncertainty in $i$. No photometric variability was found by
483: \cite{Percy:93}, who, however, apparently did not observe during the
484: eclipse phases, or by the \hip\ satellite (scatter $\sigma_{H_p} =
485: 0.007$ mag). The latter observations do straddle two of the eclipses,
486: which can be predicted very accurately from the spectroscopic orbit of
487: \cite{Scarfe:07} with errors of only 1.0 days for primary eclipse (HJD
488: $2,\!448,\!652.4$) and 1.4 days for the secondary (HJD
489: $2,\!448,\!419.3$).  If central, the eclipses would last approximately
490: 18 and 22 days, respectively.  Examination of the epoch photometry
491: from \hip\ indicates that both of these eclipses were missed by just a
492: few days. Future eclipses would be expected to occur at Julian dates
493: $2,\!455,\!254.7 \pm 1.2$ and $2,\!457,\!455.5 \pm 1.4$ for the
494: primary, and $2,\!455,\!021.6 \pm 1.4$ and $2,\!457,\!222.4 \pm 1.6$
495: for the secondary. The small separation of only 233 days (10.6\% of a
496: cycle) between the secondary eclipses and the primary events that
497: follow is due to a combination of high eccentricity and a longitude of
498: periastron near zero.
499: 
500: A second implication of the near edge-on orientation of \hr\ is
501: significantly lower component masses than previously assumed, which in
502: turn leads to a linear semimajor axis for the relative orbit of $4.63
503: \pm 0.11$ AU, corresponding to $26.0 \pm 2.4$ mas. The maximum angular
504: separation subtended by the binary is about 43 mas, also considerably
505: smaller than has been suggested in the past \citep[$\sim$70
506: mas;][]{McAlister:76, Halbwachs:81}. About two dozen unsuccessful
507: attempts to detect the secondary have been made over the past 3
508: decades with the speckle technique. Some of them were made at phases
509: in the orbit when the instrumental resolution should have allowed the
510: observers to resolve the pair, suggesting the large disparity in
511: brightness ($\Delta V \approx 3$) as the cause of those
512: non-detections. In principle the expected separations make it a good
513: target for long-baseline interferometry, except that in most cases
514: these instruments observe in the near infrared, where the brightness
515: difference will be even more extreme. From our modeling we expect
516: $\Delta J \sim 3.8$ mag and $\Delta K \sim 4.1$ mag. Still, some of
517: these facilities may have the sensitivity required. They could also
518: resolve the primary itself; the angular diameter should be
519: approximately 1.6 mas, while that of the secondary is only 0.2 mas.
520: 
521: Double-lined spectroscopic binaries with two giant components are a
522: relatively rare occurrence, and usually imply the components must have
523: nearly identical mass.  \hr\ is quite remarkable in that the stars
524: have attained a large brightness difference that is near the maximum
525: for this system, yet the secondary is still visible. Evolution
526: proceeds very rapidly at this stage, and according to the models the
527: difference in brightness may still increase by another half a
528: magnitude or so in $V$ over the next 20~Myr, which represents only
529: $\sim$0.5\% of its present age.
530: 	
531: \acknowledgements 
532: 
533: We are grateful to an anonymous referee for helpful suggestions on the
534: original manuscript.  Partial support for this work from NSF grant
535: AST-0708229 and NASA's MASSIF SIM Key Project (BLF57-04) are
536: acknowledged. This research has made use of the VizieR service
537: \citep{Ochsenbein:00} and of the SIMBAD database, both operated at
538: CDS, Strasbourg, France, as well as of NASA's Astrophysics Data System
539: Abstract Service. Alex Sozzetti is thanked for help in obtaining and
540: reducing the Keck spectrum of \hr.
541: 
542: \begin{thebibliography}{}
543: 
544: \bibitem[Biazzo et al.(2007)]{Biazzo:07}
545:  Biazzo, K., Frasca, A., Catalano, S., \& Marilli, E. 2007, Astr.\
546: Nach., in press
547: 
548: \bibitem[Christie(1934)]{Christie:34}
549:  Christie, W.\ H., 1934, \pasp, 46, 238
550: 
551: \bibitem[Christie(1936)]{Christie:36}
552:  Christie, W.\ H., 1936, \apj, 83, 433
553: 
554: \bibitem[Fekel et al.(2005)]{Fekel:05}
555:  Fekel, F.\ C., Barlow, D.\ J., Scarfe, C.\ D., Jancart, S., \&
556:  Pourbaix, D. 2005, \aj, 129, 1001
557: 
558: \bibitem[Fekel et al.(2007)]{Fekel:07}
559:  Fekel, F.\ C., Williamson, M., \& Pourbaix, D. 2007, \aj, 133, 2431
560: 
561: \bibitem[Girardi et al.(2000)]{Girardi:00}
562:  Girardi, L., Bressan, A., Bertelli, G., \& Chiosi, C.  2000, \aaps,
563: 141, 371
564: 
565: \bibitem[Gray(2005)]{Gray:05}
566:  Gray, D.\ F. 2005, The Observation and Analysis of Stellar
567: Photospheres, 3rd Edition, Cambridge, UK: Cambridge University Press
568: 
569: \bibitem[Gray \& Brown(2001)]{Gray:01}
570:  Gray, D.\ F., \& Brown, K. 2001, \pasp, 113, 723
571: 
572: \bibitem[Gray \& Johanson(1991)]{Gray:91}
573:  Gray, D.\ F., \& Johanson, H.\ L. 1991, \pasp, 103, 439
574: 
575: \bibitem[H\"aggkvist \& Oja(1987)]{Haggkvist:87}
576:  H\"aggkvist, L., \& Oja, T. 1987, \aaps, 68, 259
577: 
578: \bibitem[Halbwachs(1981)]{Halbwachs:81}
579:  Halbwachs, J.-L. 1981, \aaps, 44, 47
580: 
581: \bibitem[H\o g et al.(2000)]{Hog:00}
582:  H\o g, E., Fabricius, C., Makarov, V.\ V., Urban, S., Corbin, T.,
583:  Wycoff, G., Bastian, U., Schwekendiek, P., \& Wicenec, A. 2000, \aap,
584:  355, L27
585: 	
586: \bibitem[Jancart et al.(2005)]{Jancart:05}
587:  Jancart, S., Jorissen, A., Babusiaux, C., \& Pourbaix, D. 2005, \aap,
588:  442, 365
589: 
590: \bibitem[McAlister(1976)]{McAlister:76}
591:  McAlister, H.\ A. 1976, \pasp, 88, 317
592: 
593: \bibitem[Ochsenbein et al.(2000)]{Ochsenbein:00}
594:  Ochsenbein, F., Bauer, P., \& Marcout, J. 2000, \aaps, 143, 23
595: 
596: \bibitem[Percy(1993)]{Percy:93}
597:  Percy, J.\ R. 1993, \pasp, 105, 1422
598: 
599: \bibitem[Perryman et al.(1997)]{Perryman:97}
600:  Perryman, M.\ A.\ C., et al. 1997, The \hip\ and Tycho Catalogues
601:  (ESA SP-1200; Noordwjik: ESA)
602: 
603: \bibitem[Pols et al.(1998)]{Pols:98}
604:  Pols, O.\ R., Schr\"oder, K.-P., Hurley, J.\ R., Tout, C.\ A., \&
605:  Eggleton, P.\ P. 1998, \mnras, 298, 525
606: 
607: \bibitem[Press et al.(1992)]{Press:92}
608:  Press, W.\ H., Teukolsky, S.\ A., Vetterling, W.\ T., \& Flannery,
609: B.\ P. 1992, Numerical Recipes, (2nd.\ ed.; Cambridge: Cambridge
610: Univ.\ Press)
611: 
612: \bibitem[Pourbaix \& Jorissen(2000)]{Pourbaix:00}
613:  Pourbaix, D., \& Jorissen, A. 2000, \aaps, 145, 161
614: 
615: \bibitem[Pourbaix et al.(2004)]{Pourbaix:04}
616:  Pourbaix, D., Jancart, S., \& Jorissen, A. 2004, in Spectroscopically
617:  and Spatially Resolving the Components of the Close Binary Stars, ASP
618:  Conf.\ Ser. 318, eds.\ R.\ W.\ Hilditch, H.\ Hensberge \& K.\
619:  Pavlovski (San Francisco: ASP, p.\ 144
620: 
621: \bibitem[Ram\'\i rez \& Mel\'endez(2005)]{Ramirez:05}
622:  Ram\'\i rez, I., \& Mel\'endez, J. 2005, \apj, 626, 465
623: 
624: \bibitem[Scarfe et al.(2004)]{Scarfe:04}
625:  Scarfe, C.\ D., Griffin, R.\ F., \& Griffin, R.\ E.\ M. 2004, Rev.\
626:  Mexicana Astron.\ Astrof\'\i s, 21, 79
627: 
628: \bibitem[Scarfe et al.(2007)]{Scarfe:07}
629:  Scarfe, C.\ D., Griffin, R.\ F., \& Griffin, R.\ E.\ M. 2007, \mnras,
630:  376, 1671
631: 
632: \bibitem[Schlegel et al.(1998)]{Schlegel:98}
633:  Schlegel, D.\ J., Finkbeiner, D.\ P., \& Davis, M. 1998, \apj, 500, 525
634: 
635: \bibitem[Torres(2006)]{Torres:06}
636:  Torres, G. 2006, \aj, 131, 1022
637: 
638: \bibitem[Torres(2007a)]{Torres:07a}
639:  Torres, G. 2007a, \apj, 654, 1095
640: 
641: \bibitem[Torres(2007b)]{Torres:07b}
642:  Torres, G. 2007b, \aj, 133, 2684
643: 
644: \bibitem[Trimble \& Thorne(1969)]{Trimble:69}
645:  Trimble, V.\ L., \& Thorne, K.\ S. 1969, \apj, 156, 1013
646: 
647: \bibitem[van de Kamp(1967)]{vandeKamp:67}
648:  van de Kamp, P. 1967, Principles of Astrometry (San Francisco: W.\
649:  H.\ Freeman)
650: 
651: \bibitem[van Leeuwen \& Evans(1998)]{vanLeeuwen:98}
652:  van Leeuwen, F., \& Evans, D. W. 1998, \aaps, 130, 157
653: 
654: \bibitem[Vogt et al.(1994)]{Vogt:94}
655:  Vogt, S.\ S.\ et al. 1994, Proc.\ SPIE, 2198, 362
656: 
657: \end{thebibliography}
658: 
659: %\clearpage
660: 
661: \LongTables
662: \begin{deluxetable}{lcccccc}
663: \tabletypesize{\scriptsize}
664: \tablewidth{0pc}
665: \tablecaption{\hip\ Intermediate Astrometric Data for \hr.\label{tab:abscissae}}
666: \tablehead{\colhead{Date} & \colhead{} & \colhead{Orbital} & \colhead{$\Delta v$} & 
667: \colhead{$\sigma_{\Delta v}$} & \colhead{$O\!-\!C$} & \colhead{Consortium} \\
668: \colhead{(HJD$-2,\!400,\!000$)} & \colhead{Year} & \colhead{Phase} & 
669: \colhead{(mas)} & \colhead{(mas)} & \colhead{(mas)} & \colhead{(NDAC/FAST)}}
670: \startdata
671:  47862.0382\dotfill  &  1989.9165  &  0.6830  & \phn$+$3.09  &  2.68  &  $+$2.12  &   N  \\
672:  47899.4398\dotfill  &  1990.0187  &  0.6999  & \phn$-$1.33  &  3.49  &  $-$1.89  &   F  \\
673:  47899.4398\dotfill  &  1990.0189  &  0.7000  & \phn$-$2.65  &  2.55  &  $-$3.21  &   N  \\
674:  47926.5048\dotfill  &  1990.0931  &  0.7123  & \phn$+$2.13  &  1.65  &  $+$1.79  &   F  \\
675:  47926.5048\dotfill  &  1990.0930  &  0.7123  & \phn$+$1.63  &  1.69  &  $+$1.29  &   N  \\
676:  47985.6022\dotfill  &  1990.2552  &  0.7392  & \phn$-$0.49  &  1.76  &  $-$0.33  &   F  \\
677:  47985.6022\dotfill  &  1990.2549  &  0.7392  & \phn$-$2.97  &  1.71  &  $-$2.81  &   N  \\
678:  47999.3721\dotfill  &  1990.2924  &  0.7454  & \phn$-$0.85  &  2.00  &  $-$0.89  &   F  \\
679:  47999.2991\dotfill  &  1990.2924  &  0.7454  & \phn$+$0.61  &  1.47  &  $+$0.57  &   N  \\
680:  48053.5022\dotfill  &  1990.4408  &  0.7700  & \phn$-$0.96  &  1.57  &  $-$0.71  &   F  \\
681:  48053.5022\dotfill  &  1990.4408  &  0.7700  & \phn$-$1.62  &  1.43  &  $-$1.37  &   N  \\
682:  48091.2325\dotfill  &  1990.5437  &  0.7871  & \phn$+$1.17  &  2.71  &  $+$0.93  &   F  \\
683:  48091.2325\dotfill  &  1990.5432  &  0.7870  & \phn$+$2.25  &  2.73  &  $+$2.01  &   N  \\
684:  48115.2294\dotfill  &  1990.6101  &  0.7981  & \phn$+$5.42  &  2.38  &  $+$5.31  &   F  \\
685:  48115.2294\dotfill  &  1990.6098  &  0.7981  & \phn$+$2.05  &  2.39  &  $+$1.94  &   N  \\
686:  48144.5225\dotfill  &  1990.6901  &  0.8114  & \phn$+$1.04  &  1.94  &  $+$0.95  &   F  \\
687:  48144.5225\dotfill  &  1990.6901  &  0.8114  & \phn$+$0.65  &  2.09  &  $+$0.56  &   N  \\
688:  48181.7780\dotfill  &  1990.7918  &  0.8283  & \phn$+$3.07  &  2.74  &  $+$2.72  &   N  \\
689:  48182.2163\dotfill  &  1990.7932  &  0.8285  & \phn$-$1.53  &  1.94  &  $-$1.86  &   F  \\
690:  48182.2163\dotfill  &  1990.7935  &  0.8285  & \phn$-$0.43  &  2.22  &  $-$0.76  &   N  \\
691:  48201.2823\dotfill  &  1990.8454  &  0.8372  & \phn$-$0.61  &  1.93  &  $+$0.28  &   F  \\
692:  48201.2823\dotfill  &  1990.8454  &  0.8372  & \phn$+$2.77  &  3.06  &  $+$3.66  &   N  \\
693:  48243.0304\dotfill  &  1990.9602  &  0.8562  & \phn$-$1.98  &  2.67  &  $-$0.73  &   F  \\
694:  48243.0304\dotfill  &  1990.9598  &  0.8561  & \phn$-$1.02  &  3.05  &  $+$0.23  &   N  \\
695:  48266.5891\dotfill  &  1991.0237  &  0.8667  & \phn$-$2.61  &  2.89  &  $-$1.37  &   F  \\
696:  48266.5525\dotfill  &  1991.0243  &  0.8668  & \phn$-$5.34  &  3.20  &  $-$4.09  &   N  \\
697:  48267.0273\dotfill  &  1991.0245  &  0.8669  & \phn$-$2.19  &  2.50  &  $-$0.98  &   F  \\
698:  48266.9908\dotfill  &  1991.0245  &  0.8669  & \phn$+$0.73  &  2.67  &  $+$1.94  &   N  \\
699:  48294.9690\dotfill  &  1991.1022  &  0.8798  & \phn$-$5.80  &  2.92  &  $-$3.10  &   F  \\
700:  48294.9690\dotfill  &  1991.1019  &  0.8797  & \phn$-$4.27  &  3.35  &  $-$1.58  &   N  \\
701:  48295.4073\dotfill  &  1991.1027  &  0.8799  & \phn$-$1.66  &  2.89  &  $+$1.04  &   F  \\
702:  48295.4073\dotfill  &  1991.1031  &  0.8799  & \phn$-$0.94  &  3.18  &  $+$1.76  &   N  \\
703:  48327.3666\dotfill  &  1991.1911  &  0.8945  & \phn$+$2.25  &  3.32  &  $+$1.51  &   F  \\
704:  48327.3666\dotfill  &  1991.1906  &  0.8944  & \phn$-$0.54  &  3.72  &  $-$1.28  &   N  \\
705:  48327.8050\dotfill  &  1991.1918  &  0.8946  & \phn$+$2.48  &  3.24  &  $+$1.77  &   N  \\
706:  48344.7526\dotfill  &  1991.2382  &  0.9023  & \phn$+$1.05  &  2.43  &  $+$3.80  &   F  \\
707:  48344.6795\dotfill  &  1991.2378  &  0.9023  & \phn$-$2.10  &  5.27  &  $+$0.63  &   N  \\
708:  48384.6744\dotfill  &  1991.3475  &  0.9205  & \phn$+$3.62  &  2.18  &  $-$0.05  &   F  \\
709:  48384.6744\dotfill  &  1991.3475  &  0.9205  & \phn$+$7.30  &  3.27  &  $+$3.63  &   N  \\
710:  48395.7049\dotfill  &  1991.3779  &  0.9255  & \phn$+$2.72  &  1.82  &  $+$1.70  &   F  \\
711:  48395.7049\dotfill  &  1991.3777  &  0.9255  & \phn$+$3.88  &  2.17  &  $+$2.86  &   N  \\
712:  48436.1381\dotfill  &  1991.4882  &  0.9438  & \phn$+$3.63  &  2.22  &  $-$0.93  &   F  \\
713:  48436.1016\dotfill  &  1991.4884  &  0.9439  & \phn$+$3.91  &  2.95  &  $-$0.65  &   N  \\
714:  48455.2042\dotfill  &  1991.5408  &  0.9526  & \phn$+$4.28  &  2.58  &  $-$0.43  &   F  \\
715:  48455.2042\dotfill  &  1991.5406  &  0.9525  & \phn$+$4.37  &  3.14  &  $-$0.34  &   N  \\
716:  48488.4784\dotfill  &  1991.6314  &  0.9676  & \phn$+$4.01  &  2.03  &  $+$2.86  &   F  \\
717:  48488.4419\dotfill  &  1991.6316  &  0.9676  & \phn$+$3.98  &  2.72  &  $+$2.83  &   N  \\
718:  48519.9995\dotfill  &  1991.7182  &  0.9820  & \phn$+$6.03  &  1.68  &  $+$0.35  &   F  \\
719:  48519.9995\dotfill  &  1991.7180  &  0.9820  & \phn$+$6.31  &  2.24  &  $+$0.62  &   N  \\
720:  48543.0468\dotfill  &  1991.7812  &  0.9925  & \phn$-$4.35  &  1.92  &  $-$1.21  &   F  \\
721:  48543.0468\dotfill  &  1991.7811  &  0.9924  & \phn$-$3.45  &  1.85  &  $-$0.31  &   N  \\
722:  48586.1097\dotfill  &  1991.8990  &  0.0120  & \phn$-$2.14  &  2.13  &  $-$0.43  &   F  \\
723:  48586.1097\dotfill  &  1991.8990  &  0.0120  & \phn$+$3.20  &  3.17  &  $+$4.91  &   N  \\
724:  48606.0524\dotfill  &  1991.9536  &  0.0211  & \phn$-$2.08  &  1.91  &  $+$1.98  &   F  \\
725:  48606.0524\dotfill  &  1991.9537  &  0.0211  & \phn$-$0.31  &  2.82  &  $+$3.75  &   N  \\
726:  48640.2398\dotfill  &  1992.0469  &  0.0366  & \phn$-$2.43  &  2.93  &  $-$0.06  &   F  \\
727:  48640.2033\dotfill  &  1992.0471  &  0.0366  & \phn$+$2.16  &  3.14  &  $+$4.53  &   N  \\
728:  48669.0580\dotfill  &  1992.1261  &  0.0497  & \phn$-$1.67  &  1.95  &  $-$1.60  &   F  \\
729:  48669.0580\dotfill  &  1992.1261  &  0.0497  & \phn$-$1.80  &  2.00  &  $-$1.73  &   N  \\
730:  48727.5346\dotfill  &  1992.2865  &  0.0763  & \phn$-$1.45  &  2.97  &  $-$0.21  &   F  \\
731:  48727.5346\dotfill  &  1992.2862  &  0.0763  & \phn$-$3.25  &  3.14  &  $-$2.02  &   N  \\
732:  48728.0459\dotfill  &  1992.2872  &  0.0764  & \phn$+$6.85  &  3.47  &  $+$8.06  &   F  \\
733:  48728.0094\dotfill  &  1992.2875  &  0.0765  & \phn$+$4.19  &  4.20  &  $+$5.40  &   N  \\
734:  48739.9896\dotfill  &  1992.3200  &  0.0819  & \phn$+$0.44  &  1.82  &  $+$0.09  &   F  \\
735:  48739.9896\dotfill  &  1992.3202  &  0.0819  & \phn$+$2.03  &  2.19  &  $+$1.68  &   N  \\
736:  48780.7880\dotfill  &  1992.4323  &  0.1005  & \phn$-$4.94  &  1.99  &  $-$0.37  &   F  \\
737:  48780.7880\dotfill  &  1992.4320  &  0.1005  & \phn$-$3.69  &  1.90  &  $+$0.87  &   N  \\
738:  48781.2263\dotfill  &  1992.4333  &  0.1007  & \phn$-$5.75  &  3.30  &  $-$1.15  &   N  \\
739:  48795.8728\dotfill  &  1992.4736  &  0.1074  & \phn$-$0.22  &  2.09  &  $+$3.70  &   F  \\
740:  48795.8728\dotfill  &  1992.4733  &  0.1073  & \phn$-$3.51  &  3.01  &  $+$0.41  &   N  \\
741:  48832.2517\dotfill  &  1992.5731  &  0.1239  & \phn$+$0.26  &  2.23  &  $+$3.18  &   F  \\
742:  48832.2517\dotfill  &  1992.5725  &  0.1238  & \phn$-$5.10  &  3.03  &  $-$2.18  &   N  \\
743:  48926.3036\dotfill  &  1992.8306  &  0.1666  & \phn$-$0.19  &  3.83  &  $+$1.97  &   F  \\
744:  48926.3036\dotfill  &  1992.8304  &  0.1666  & \phn$-$1.81  &  4.28  &  $+$0.34  &   N  \\
745:  48944.4931\dotfill  &  1992.8802  &  0.1749  & $+$11.27  &  1.89  &  $+$0.50  &   F  \\
746:  48944.4931\dotfill  &  1992.8801  &  0.1748  & $+$11.24  &  2.54  &  $+$0.47  &   N  \\
747:  49025.1768\dotfill  &  1993.1013  &  0.2116  & $+$15.87  &  1.89  &  $+$2.19  &   F  \\
748:  49025.1768\dotfill  &  1993.1011  &  0.2115  & $+$10.99  &  3.22  &  $-$2.69  &   N  \\
749: \noalign{\vskip -6pt}
750: \enddata
751: \end{deluxetable}
752: 
753: %\clearpage
754: 
755: \begin{deluxetable}{lccccc}
756: %\tabletypesize{\scriptsize}
757: \tablewidth{0pc}
758: \tablecaption{Astrometric orbital solutions for \hr.\label{tab:elements}}
759: \tablehead{\colhead{\hfil~~~~~~~~~~~Parameter~~~~~~~~~~~~} & \colhead{\cite{Jancart:05}\tablenotemark{a}} & \colhead{This work\tablenotemark{b}}}
760: \startdata
761: \noalign{\vskip -2pt}
762: \sidehead{Adjusted quantities} \\
763: \noalign{\vskip -8pt}
764: ~~~~$P$ (days)\dotfill                         &  2150 (fixed)                &  2200.77 (fixed) \\
765: ~~~~$a_{\rm phot}$ (mas)\dotfill               &  16.5~$\pm$~2.3\phn          &  9.69~$\pm$~0.85 \\
766: ~~~~$e$\dotfill                                &  0.6 (fixed)                 &  0.6797 (fixed) \\          
767: ~~~~$i$ (deg)\dotfill                          &  46.4~$\pm$~3.5\phn          &  91.4~$\pm$~9.7\phn \\
768: ~~~~$\omega_{\rm A}$ (deg)\dotfill             &  340 (fixed)                 &  9.73 (fixed) \\
769: ~~~~$\Omega_{\rm J2000}$ (deg)\dotfill         &  305.4~$\pm$~6.7\phn\phn     &  50.5~$\pm$~6.6\phn \\
770: ~~~~$T$ (HJD$-$2,400,000)\dotfill              &  50699 (fixed)\tablenotemark{c}               &  50760.4 (fixed) \\
771: ~~~~$\Delta\alpha^*$ (mas)\dotfill             &  \nodata                     &  $+$4.41~$\pm$~0.76\phs \\
772: ~~~~$\Delta\delta$ (mas)\dotfill               &  \nodata                     &  $+$3.77~$\pm$~0.95\phs \\
773: ~~~~$\Delta\mu_{\alpha}^*$ (mas~yr$^{-1}$)\dotfill   &  \nodata               &  $-$3.57~$\pm$~0.91\phs \\
774: ~~~~$\Delta\mu_{\delta}$ (mas~yr$^{-1}$)\dotfill     &  \nodata               &  $-$3.13~$\pm$~1.23\phs \\
775: ~~~~$\Delta\pi_{\rm t}$ (mas)\dotfill          &  \nodata                     &  $+$0.27~$\pm$~0.51\phs \\
776: \sidehead{Derived quantities} \\						                                         
777: \noalign{\vskip -8pt}										                                         
778: ~~~~$\mu_{\alpha}^*$ (mas~yr$^{-1}$)\dotfill   &  \nodata                     &  $-$12.46~$\pm$~0.91\phn\phs \\
779: ~~~~$\mu_{\delta}$ (mas~yr$^{-1}$)\dotfill     &  \nodata                     &  $-$31.13~$\pm$~1.23\phn\phs \\
780: ~~~~$\pi_{\rm t}$ (mas)\dotfill                &  \nodata                     &  5.61~$\pm$~0.51 \\
781: ~~~~$M_{\rm A}$ (M$_{\sun}$)\dotfill           &  \nodata                     &  $1.38_{-0.03}^{+0.09}$ \\
782: ~~~~$M_{\rm B}$ (M$_{\sun}$)\dotfill           &  \nodata                     &  $1.36_{-0.02}^{+0.07}$ \\
783: %\tablebreak
784: \noalign{\vskip -6pt}
785: \enddata
786: 
787: \tablenotetext{a}{$P$, $e$, $\omega_{\rm A}$, and $T$ adopted from
788: \cite{Christie:36}. Other {\it Hipparcos}-related parameters were not
789: reported.}
790: 
791: \tablenotetext{b}{$P$, $e$, $\omega_{\rm A}$, and $T$ adopted from
792: \cite{Scarfe:07}.}
793: 
794: \tablenotetext{c}{Projected forward from the original epoch
795: $2,\!424,\!290$ (HJD) using the more accurate period from the present
796: paper, for easier comparison with our new results.}
797: 
798: \end{deluxetable}
799: 
800: %\clearpage
801: 
802: \begin{deluxetable}{lccc}
803: %\tabletypesize{\scriptsize}
804: \tablewidth{0pt}
805: \tablecaption{Comparison of the proper motion and parallax results for \hr.\label{tab:pm}}
806: \tablehead{\colhead{~~~~~~~~~~Parameter~~~~~~~~~~} & \colhead{\it Hipparcos}  &
807: \colhead{Tycho-2} & \colhead{This paper} }
808: \startdata
809: ~~~$\mu_{\alpha}^*$ (mas yr$^{-1}$)\dotfill  & $-8.89$~$\pm$~0.58\phs      &  $-13.3$~$\pm$~1.4\phn\phs  &  $-12.46$~$\pm$~0.91\phn\phs   \\
810: ~~~$\mu_{\delta}$ (mas yr$^{-1}$)\dotfill    & $-28.00$~$\pm$~0.60\phn\phs &  $-30.8$~$\pm$~1.4\phn\phs  &  $-31.13$~$\pm$~1.23\phn\phs    \\
811: ~~~$\pi_{\rm t}$ (mas)\dotfill               & 5.34~$\pm$~0.60             &  \nodata                    & 5.61~$\pm$~0.51 \\
812: \noalign{\vskip -6pt}
813: \enddata
814: \end{deluxetable}
815: 
816: %\clearpage
817: 
818: \begin{deluxetable}{lccccc}
819: %\tabletypesize{\scriptsize}
820: %\tablecolumns{4}
821: \tablewidth{0pt}
822: \tablecaption{Properties for the components of \hr\ inferred from the models.\label{tab:models}}
823: \tablehead{\colhead{~~~~~~~~~~Property~~~~~~~~~~} & \colhead{$Z = 0.004$}  &
824: \colhead{$Z = 0.008$} & \colhead{$Z = 0.011$} & \colhead{$Z = 0.019$} & 
825: \colhead{$Z = 0.030$} }
826: \startdata
827: 
828: Log age (yr)\dotfill                & 9.40 & 9.45 & 9.49 & 9.55 & 9.60 \\
829: $(B\!-\!V)_{\rm A}$ (mag)\dotfill   & 1.07 & 1.25 & 1.35 & 1.50 & 1.58 \\
830: $(B\!-\!V)_{\rm B}$ (mag)\dotfill   & 0.75 & 0.87 & 0.92 & 1.02 & 1.12 \\
831: $(B\!-\!V)_{\rm tot}$ (mag)\dotfill & 1.05 & 1.22 & 1.32 & 1.47 & 1.54 \\
832: $\Delta J$ (mag)\dotfill            & 3.48 & 3.64 & 3.80 & 3.96 & 4.11 \\
833: $\Delta H$ (mag)\dotfill            & 3.66 & 3.88 & 4.05 & 4.29 & 4.43 \\
834: $\Delta K$ (mag)\dotfill            & 3.68 & 3.90 & 4.08 & 4.33 & 4.54 \\
835: $T_{\rm eff}^{\rm A}$ (K)\dotfill   & 4590 & 4311 & 4211 & 3947 & 3699 \\
836: $T_{\rm eff}^{\rm B}$ (K)\dotfill   & 5312 & 5079 & 5010 & 4819 & 4649 \\
837: $R_{\rm A}$ (R$_{\sun}$)\dotfill    & 23.1 & 29.3 & 30.9 & 42.1 & 56.7 \\
838: $R_{\rm B}$ (R$_{\sun}$)\dotfill    & 3.84 & 4.33 & 4.20 & 4.98 & 5.90 \\
839: \noalign{\vskip -6pt}
840: \enddata 
841: 
842:  \tablecomments{The models used are those of \cite{Girardi:00}, and
843: the observational constraints are the measured masses and absolute
844: magnitudes of the components (see text). The best-fitting model for $Z
845: = 0.011$ satisfies the additional constraint that the combined colors
846: of the components match the observed $B\!-\!V$ color.}
847: 
848: \end{deluxetable}
849: 
850: \end{document}
851: