1: 

2: \documentclass[preprint2]{aastex}

3: 

4: 

5: 

6: 

7: 

8: \newcommand{\kepler}{{\it Kepler}}

9: \newcommand{\tpf}{{\it TPF}}

10: \newcommand{\hst}{{\it HST}}

11: \newcommand{\cmpss}{cm~s$^{-2}$}

12: \newcommand{\mps}{m~s$^{-1}$}

13: \newcommand{\kps}{km~s$^{-1}$}

14: \newcommand{\mpsini}{$M_p{\rm sin}i$}

15: \newcommand{\Msun}{${\rm M_\odot}$}

16: \newcommand{\Rsun}{${\rm R_\odot}$}

17: \newcommand{\Mjup}{${\rm M_J}$}

18: \newcommand{\xonb}{XO-3b}

19: \newcommand{\xon}{XO-3}

20: \newcommand{\Rjup}{${\rm R_J}$}

21: \newcommand{\vMs}{1.41}		

22: \newcommand{\eMs}{0.08}

23: \newcommand{\vRs}{2.13}		

24: \newcommand{\eRs}{0.21}

25: \newcommand{\sptype}{F5V}

26: \newcommand{\vrvK}{1471}		

27: \newcommand{\ervK}{48}

28: \newcommand{\vvsini}{18.54}

29: \newcommand{\evsini}{0.17}

30: \newcommand{\vfeh}{-0.177}

31: \newcommand{\efeh}{0.027}

32: \newcommand{\vDs}{260}		

33: \newcommand{\eDs}{23}

34: \newcommand{\vjd}{2454025.3967}	

35: \newcommand{\ejd}{0.0038}	

36: \newcommand{\vap}{0.0476}	

37: \newcommand{\eap}{0.0005}

38: \newcommand{\vperiod}{3.1915426}	

39: \newcommand{\eperiod}{0.00014}

40: \newcommand{\vecc}{0.260}

41: \newcommand{\eecc}{0.017}

42: \newcommand{\vMp}{13.02}		

43: \newcommand{\eMp}{0.64}	

44: \newcommand{\vMptot}{13.25}

45: \newcommand{\eMptot}{0.64}

46: \newcommand{\vRp}{1.95}		

47: \newcommand{\eRp}{0.16}

48: \newcommand{\vincl}{79.32}	

49: \newcommand{\eincl}{1.36}

50: \newcommand{\valq}{0.0275}		

51: \newcommand{\errq}{0.0005}

52: 

53: \slugcomment{Submitted for publication in the Astrophysical Journal}

54: 

55: 

56: \begin{document}

57: 

58: \title{XO-3b: A Massive Planet in an Eccentric Orbit Transiting an \sptype\ Star}

59: 

60: \author{

61: Christopher~M.~Johns--Krull\altaffilmark{1,2},

62: Peter~R.~McCullough\altaffilmark{3},

63: Christopher~J.~Burke\altaffilmark{3},

64: Jeff~A.~Valenti\altaffilmark{3},

65: K.~A.~Janes\altaffilmark{4}, 

66: J.~N.~Heasley\altaffilmark{5},

67: L.~Prato\altaffilmark{6},

68: R.~Bissinger\altaffilmark{7},

69: M.~Fleenor\altaffilmark{8},

70: C.~N.~Foote\altaffilmark{9},

71: E.~Garcia--Melendo\altaffilmark{10},

72: B.~L.~Gary\altaffilmark{11},

73: P.~J.~Howell\altaffilmark{4},

74: F.~Mallia\altaffilmark{12},

75: G.~Masi\altaffilmark{13},

76: T.~Vanmunster\altaffilmark{14}

77: }

78: 

79: \email{cmj@rice.edu}

80: 

81: \altaffiltext{1}{Dept. of Physics and Astronomy, Rice University, 6100 Main Street, MS-108, Houston, TX 77005}

82: \altaffiltext{2}{Visiting Astronomer, McDonald Observatory, which is operated by the University of Texas at Austin.}

83: \altaffiltext{3}{Space Telescope Science Institute, 3700 San Martin Dr., Baltimore MD 21218}

84: \altaffiltext{4}{Boston University, Astronomy Dept., 725 Commonwealth Ave.,

85: Boston, MA 02215}

86: \altaffiltext{5}{University of Hawaii, Inst. for Astronomy, 2680 Woodlawn Dr., Honolulu, HI 96822}

87: \altaffiltext{6}{Lowell Observatory, 1400 West Mars Hill Road, Flagstaff, AZ,

88: 86001}

89: \altaffiltext{7}{Racoon Run Observatory, 1142 Mataro Court, Pleasanton, CA 94566}

90: \altaffiltext{8}{Volunteer Observatory, 10305 Mantooth Lane, Knoxville, TN 37932}

91: \altaffiltext{9}{Vermillion Cliffs Observatory, 4175 E. Red Cliffs Drive, 

92: Kanab, UT 84741}

93: \altaffiltext{10}{Esteve Duran Observatory, El Montanya, Seva, 08553 Seva, Barcelona, Spain}

94: \altaffiltext{11}{Hereford Arizona Observatory, 5320 E. Calle Manzana, Hereford, AZ 85615}

95: \altaffiltext{12}{Campo Catino Astronomical Observatory, P.O. BOX 03016,

96: Guarcino (FR) Italy}

97: \altaffiltext{13}{Bellatrix Observatory, Via Madonna de Loco, 47 03023 

98: Ceccano (FR) Italy}

99: \altaffiltext{14}{CBA Belgium Observatory, Walhostraat 1A, B-3401 Landen, Belgium}

100: 

101: \begin{abstract}

102: We report the discovery of a massive (\mpsini $= \vMp \pm \eMp$ \Mjup;

103: total mass $\vMptot \pm \eMptot$\ \Mjup),

104: large ($\vRp \pm \eRp$\ \Rjup) planet in

105: a transiting, eccentric orbit ($e = \vecc \pm \eecc$) around a 10$^{\rm th}$

106: magnitude \sptype\ star in the constellation Camelopardalis.  We designate the 

107: planet \xonb, and the star \xon, also known as GSC 03727-01064.  

108: The orbital period of \xonb\ is $\vperiod \pm \eperiod$ days. 

109: \xon\ lacks a trigonometric distance; we estimate its 

110: distance to be \vDs$\pm$\eDs\ pc.

111: The radius of \xon\ is \vRs$\pm$\eRs\ \Rsun, its mass is \vMs$\pm$\eMs\ \Msun,

112: its $v$sin$i = \vvsini \pm \evsini$ \kps, and its metallicity is 

113: [Fe/H] $= \vfeh \pm

114: \efeh$.  This system is unusual for a number of reasons.  \xonb\ is one of 

115: the most massive planets discovered around any star for which the orbital 

116: period is less than 10 days. The mass is near the deuterium burning limit of

117: 13 \Mjup, which is a proposed boundary between planets and brown dwarfs. 

118: Although Burrows et al. (2001) propose that formation in a disk or formation

119: in the interstellar medium in a manner similar to stars is a more logical way

120: to differentiate planets and brown dwarfs, our current observations are not 

121: adequate to address this distinction.  \xonb\ is also unusual in that its 

122: eccentricity is large given its relatively short orbital period.

123: Both the planetary radius and the inclination

124: are functions of the spectroscopically determined stellar radius.

125: Analysis of the transit light curve of \xonb\ suggests that the

126: spectroscopically derived parameters may be over estimated.  Though

127: relatively noisy, the light

128: curves favor a smaller radius in order to better match the steepness of

129: the ingress and egress.  The light curve fits imply a planetary

130: radius of $1.25 \pm 0.15$ \Rjup, which would correspond

131: to a mass of $12.03 \pm 0.46$ \Mjup.  A precise trigonometric parallax

132: measurement or a very accurate light curve is needed to resolve the 

133: uncertainty in the planetary mass and radius.

134: \end{abstract}

135: 

136: \keywords{binaries: eclipsing -- planetary systems -- stars: individual

137: (GSC 03727-01064) -- techniques: photometric -- techniques: radial velocities}

138: 

139: \section{Introduction}

140: 

141: There are now over 200 extrasolar planets known (http://exoplanets.eu/),

142: and most of these have been discovered using the radial velocity technique.

143: As a result, most of these systems yield the minimum mass of the planet

144: (\mpsini), its orbital semi-major axis, and properties of the central

145: star.  The data to date have dramatically changed our appreciation of

146: the diversity of planetary systems that can form, and have particularly

147: focussed the community's attention on the role that planetary migration

148: plays in the planet formation process (see review by Papaloizou et al. 2007).

149: As the precision and duration of radial velocity surveys increases, lower

150: mass planets and planets in longer orbits continue to be found.  As a

151: result, it has been suggested that the fraction of stars hosting

152: planets may be as high as 50\% (see Udry et al. 2007).  Therefore, it

153: appears that the planet formation process is relatively efficient.

154: 

155: The generally accepted model of giant planet formation is that of core

156: nucleated accretion (e.g., Pollack et al. 1996; Bodenheimer et al. 2000;

157: Hubickyj et al. 2005) in which gas giant planets first form a several

158: ($\sim 10$) earth mass core of solids via non-elastic collisions in a disk

159: followed by the runaway accretion of gas from the disk once the mass of 

160: the core is sufficient to exert a strong gravitational influence.  Such a

161: model naturally predicts a positive correlation between the metallicity of 

162: the disk (and parent star) and the ease with which planets can form: there 

163: are more building blocks to form the solid core in higher metallicity cases.

164: The expected correlation has been suspected for some time (e.g., Gonzalez

165: 1997; Santos et al. 2003), and has now been demonstrated in an unbiased way

166: (e.g., Santos et al. 2004; Fischer \& Valenti 2005) and has been taken as

167: evidence in support of the core accretion model of planet formation.  The

168: competing model, gravitational instability in a 

169: massive circumstellar disk (e.g., Boss 1997, 2000; Mayer et al. 2002) does 

170: not predict such a relationship.  Livio and Pringle (2003) showed

171: that lower disk metallicity reduces the efficiency of Type II migration

172: which may account for some of the increased

173: probability for finding a planet around more metal rich stars.  There is

174: some observational evidence suggesting a higher frequency for planets at

175: smaller separations around metal-rich stars 

176: which may support such a metallicity--migration

177: relationship (Jones 2004; Sozzetti 2004).  As a result, the 

178: planet--metallicity correlation may (at least in part) be due to a 

179: metallicity--migration relationship and may not simply signify a 

180: metallicity--formation relationship.

181: 

182: Additional observations of extrasolar planets are necessary to distinguish

183: between the competing modes of planet formation and the conditions which

184: affect planet migration scenarios.  For example, the core accretion model

185: requires several Myr to form Jovian planets in a disk (Inaba et al. 2003),

186: while gravitational instabilities could potentially form these planets around

187: very young stars.  Jovian planets are suspected around a few $\sim 1$ Myr

188: young stars (e.g., CoKu Tau/4 -- Forrest et al. 2004; GQ Lup -- Neuh\"auser

189: et al. 2005); however, clear detection and firm mass determinations from

190: techniques such as radial velocity variations remain elusive on such

191: young stars.  Transiting extrasolar planets offer the opportunity to determine

192: both the mass and radius of the planet, and hence the planet's density.

193: Such planets allow us to better constrain the variety of extrasolar planet

194: properties and offer another potential way to distinguish between planet 

195: formation scenarios.  For example, the high

196: core mass derived for the transiting Saturnian mass planet in orbit around 

197: HD 149026 (Sato et al. 2005) has been interpreted as strong support for 

198: the core accretion model.  

199: There are now $\sim 20$ known transiting extrasolar planets 

200: (http://exoplanets.eu/).  Several of these planets have unexpectedly large

201: radii (see Figure 3 of Bakos et al. 2007 and discussion therein), perhaps

202: suggesting additional heating of these planets other than that from 

203: irradiation by the star they orbit.  Additional effort is needed to

204: understand the structure of extrasolar planets and their origins.

205: 

206: Here, we report the discovery of the third transiting planet from the 

207: XO Project.

208: The XO Project aims to find planets transiting

209: stars sufficiently bright to enable interesting follow up

210: studies (McCullough et al. 2005).  \footnote{This paper includes data taken

211: on the Haleakala summit maintained by the University of Hawaii,

212: the Lowell Observatory, the Hobby-Eberly Telescope, 

213: the McDonald Observatory of The University of Texas

214: at Austin, and five backyard observatories.}

215: This planet presents some very interesting properties:  it

216: has a mass of \mpsini $=\vMp \pm \eMp$ \Mjup\ and is in a short period (3.2 d), 

217: eccentric orbit ($e = \vecc \pm \eecc$) around the apparently single \sptype\

218: star GSC 03727-01064.  \xonb\ has a radius of $\vRp \pm \eRp$ \Rjup, making

219: it substantially larger in both radius and mass than most of the other reported transiting

220: planets.

221: In \S 2 we discuss the observations leading to the discovery of \xonb.

222: In \S 3 we present our analysis of these observations to determine the

223: stellar and planetary properties.  In \S 4 we give a discussion of \xonb\

224: in the context of other extrasolar planets and current planet formation

225: models, and \S 5 summarizes our conclusions.

226: 

227: \section{Observations}

228: 

229: \subsection{XO Project Photometry}

230: 

231: McCullough et al. (2005) describe the instrumentation, operation,

232: analysis, and preliminary results of the XO Project.  In summary, the XO

233: observatory monitored tens of thousands of bright ($V<12$) stars twice

234: every ten minutes on clear nights for more than 2 months per season

235: of visibility for each particular star, over the period September 2003

236: to September 2005. From our analysis of more than 3000 observations per

237: star, we identified \xon\ (Figure \ref{fig:allsky}) as one of dozens of stars

238: with light curves suggestive of a transiting planet. 

239: With the XO cameras on Haleakala, we observed four transits or 

240: segments of transits of \xon\ 

241: in 2003 and three in 2004, on Julian dates 2452934, 2452937, 2452998, 

242: 2453001, 2453301, 2453320, and 2453352.

243: Table~\ref{table:lc} provides a sample of the photometry for \xon\

244: from the XO cameras.  The full table is available in the online

245: edition.  

246: From the survey photometry of \xon\ (Figure \ref{fig:xolc}),

247: which has a nominal standard deviation of 0.8\% or 8 mmag

248: per observation, we determined a preliminary light

249: curve and ephemeris, which we used to schedule

250: observations of higher quality with other telescopes,

251: as described in the next subsections. 

252: 

253: \subsection{Additional Photometry}

254: 

255: As outlined by McCullough and Burke (2006), once an interesting candidate

256: is detedcted in the XO photometry, the candidate is released to an Extended

257: Team (E.T.) of sophisticated amateur astronomers for additional observation.

258: In 2006 September through November and 2007 January through March, transit

259: events of \xon\

260: were observed by members of the E.T. from a total of 4 backyard 

261: observatories. 

262: Table~\ref{table:lc} also

263: provides E.T. photometry for \xon.  For the E.T. light

264: curves, the median differential magnitude out of transit provides the

265: flux normalization and the standard deviation out of transit provides

266: the uncertainty in the measurements.

267: A fifth E.T. observatory was used to obtain all-sky

268: photometry for \xon. These observatories are equipped with telescopes of 

269: aperture $\sim 0.3$ m.  The telescopes, equipped with CCD detectors,

270: are suitable for obtaining light curves sufficient to confirm the nature

271: of the transit and obtain good timing information.  A network of such

272: telescopes is well suited to observe candidates with known positions and 

273: ephemerides.  Here, we use two such telescopes in North America and two

274: in Europe.  Unlike the case of XO-1b (McCullough et al. 2006) and XO-2b

275: (Burke et al. 2007), we were not able to schedule observing time on a 

276: larger ($\sim 1$-m) telescope to obtain additional photometry -- the

277: E.T. photometry is the best we have obtained to date on \xonb.

278: 

279: We estimate all-sky photometric $B, V, R_{\rm C}$, and $I_{\rm C}$ magnitudes

280: for \xon\ and several nearby reference stars (Figure \ref{fig:allsky} and 

281: Table \ref{table:allsky}) calibrated using a total of 6 Landolt areas 

282: (Landolt 1992).  Using the Hereford Arizona Observatory (E.T. member

283: BLG) 0.36-meter telescope on photometric nights 2006 

284: October 27 and 2006 November 5, we measured the fluxes of 48 and 18 Landolt

285: stars, respectively at an air mass similar to that for \xon\ and established

286: the zero points of the instrumental magnitudes and transformation equations 

287: for the color corrections for each filter and the CCD. 

288: The derived magnitudes for \xon\ differed on the two dates by 0.03 magnitudes

289: or less in all 4 colors. 

290: The $B, V, R_{\rm C}$, and $I_{\rm C}$ absolute photometric accuracies

291: are 0.03, 0.03, 0.03, and 0.05 mag r.m.s.,

292: including both the formal error and an estimated systematic error. 

293: The Tycho magnitudes for \xon\ listed in 

294: Table \ref{table:star} transform (via Table 2 of Bessel 2000)

295: to Johnson $V = 9.86$, i.e. 0.06 mag (2-$\sigma$) fainter than our estimate.

296: 

297: \subsection{Spectroscopy}

298: 

299: In order to measure the orbital elements of the system, and in particular

300: the mass ratio, as well as to determine the characteristics of the host star,

301: we obtained spectra of \xon\ with two-dimensional cross-dispersed echelle 

302: spectrographs (Tull et al. 1995; Tull 1998) at the coude focus of the

303: 2.7-m Harlan J. Smith (HJS) telescope and via a fiber optic cable on the 11-m

304: Hobby-Eberly Telescope (HET).  Both telescopes are located at McDonald 

305: Observatory.  The HJS spectra were obtained in a traditionally scheduled

306: manner, while the HET spectra were obtained in queue scheduled mode.

307: An iodine gas cell is used on the HET to provide the wavelength reference for

308: velocity determination.  While an iodine cell is available for HJS telescope,

309: it is not a facility instrument and the observations of \xon\ collected

310: here were done as a part of a radial velocity survey of young stars (Huerta

311: 2007; Huerta et al. 2007) for which very high (few m s$^{-1}$) velocity

312: precision is not required.  Wavelength and resulting velocity calibration

313: is accomplished by taking thorium-argon reference lamp spectra before and

314: after each observation of \xon\ at the HJS telescope.  At the HJS we 

315: obtained one spectrum per night, and at the HET 

316: we obtained or two spectra per night.  At both telescopes, the 

317: spectral resolution was $R \equiv \lambda/\Delta\lambda 

318: \approx 60000$, and data were obtained on a total of

319: 21 nights.  The two-dimensional echelle

320: spectra were reduced using IDL procedures described in Hinkle et al. (2000)

321: which include bias subtraction, flat fielding using a quartz lamp spectrum,

322: and optimal extraction of the data.  Table \ref{table:rvobs} gives a log of

323: the spectral observations.

324: 

325: \section{Analysis}

326: 

327: \subsection{Ephemeris}

328: 

329: We follow the same procedure used by Burke et al. (2007) for XO-2b to

330: refine the ephemeris of \xonb.  

331: We adopt the transit midpoint calculated for the event of 2006 October 16

332: as our ephemeris zeropoint: 2454025.3967$\pm$0.0021.  All measured

333: transit midpoints used to refine the ephemeris are reported in Table

334: \ref{table:ttime}.

335: To determine the orbital period, we minimize the $\chi^{2}$ difference between 

336: the observed transit times and a constant-period ephemeris model.

337: Based upon 3 transit events observed either in multiple passbands by the 

338: same observer, or observed by different observers,

339: we estimate the $1-\sigma$ uncertainty in the time of the center of an

340: individual transit to be 5 minutes, which we adopt as the

341: uncertainty for each of the transits observed in a single bandpass

342: by only one observer.

343: The best-fit period is \vperiod$\pm$\eperiod days.

344: 

345: \subsection{Radial Velocity Measurements}

346: 

347: Using spectra obtained at the HET, we measured \xon's radial velocities with

348: respect to the topocentric frame using iodine absorption lines superposed on

349: the spectra of \xon.  We modeled the extracted spectra using a template

350: stellar spectrum and the absorption spectrum of the HET iodine gas 

351: cell (Cochran 2000).  For the analysis of XO-1 (McCullough et al. 2006) and

352: XO-2 (Burke et al. 2007), a high resolution spectrum of the Sun and the 

353: Earth's atmosphere (Wallace et al. 1998) was used for the template stellar

354: spectrum.  We followed this same procedure initially for \xon; however,

355: the resulting radial velocity uncertainties were substantially higher than

356: we achieved for XO-1 and XO-2.  We suspect that the higher $v$sin$i$ of 

357: \xon\ compared to these other two stars contributes to this increased

358: uncertainty, but we were also concerned that the higher temperature of

359: \xon\ relative to these stars (and the Sun) resulted in spectral differences

360: large enough to increase the uncertainty further.  Therefore, we repeated

361: the HET radial velocity determinations using a high resolution 

362: ($\lambda/\delta\lambda \sim 60000$) spectrum of the F5V star HD 30652 

363: from the SPOCS sample of stars (Valenti \& Fischer 2005, hereafter VF05) 

364: as the template

365: stellar spectrum.  The effective temperature of HD 30652 is 6424 K

366: (VF05), within 5 K of our derived $T_{eff}$ for \xon\

367: (see below), and $v$sin$i = 16.8$ km s$^{-1}$ which is close to the value

368: of 18.54 we find for \xon\ below.  The spectrum of HD 30652 is very similar

369: in appearance to that of \xon.  

370: 

371: Using an IDL\footnote{IDL is a software product of ITT Visual Information Solutions.}

372: implementation of Nelder and Mead's (1965) 

373: downhill simplex $\chi^2$ minimization algorithm, ``Amoeba,'' we adjusted

374: parameters of our model spectrum to fit the observations.

375: The model includes convolution of our model spectra with

376: a best-fitting Voigt profile to approximate the (slightly non-Gaussian)

377: line-spread function of the instrument.  The free parameters of our model

378: are a continuum normalization factor, the radial velocity of the star,

379: the radial velocity of the

380: iodine lines (which represent instrumental deviations

381: from their expected zero velocity with respect to the observatory),

382: and an exponent (optical depth scale factor) that scales the depths of

383: the lines as an arbitrary method of adjusting the spectrum of HD 30652

384: to even more closely match that of \xon.  Due to the iodine absorption, 

385: we could not

386: estimate the continuum level by interpolating between local maxima

387: in the spectrum, so instead we solved for the continuum iteratively,

388: as required to improve the fit between our model and the observations.

389: In the manner described above, for each $\sim$15 \AA\ section of each

390: individual spectrum within the region of the recorded spectrum with

391: significant iodine absorption, $5100 - 5700$ \AA,

392: we estimated the radial velocity of the star.  From the approximately

393: normal distribution of the resulting radial velocity estimates for each

394: epoch, we

395: calculated the mean radial velocity and its uncertainty.

396: The 1-$\sigma$ internal errors of the radial velocity measurements from

397: the HET spectra range from $\sim 120$ to $\sim 200$ \mps\ per epoch.  These

398: uncertainties are an order of magnitude larger than those achieved on 

399: XO-1 ($\sim 15$ \mps) and XO-2 ($\sim 20$ \mps).

400: We transformed our measured radial velocities to the barycentric frame

401: of the solar system and subtracted the mean radial velocity and report

402: these values in Table \ref{table:rvobs}.  These velocity measurements are

403: shown in Figure \ref{fig:rvfit}.

404: 

405: The HJS radial velocity measurement technique used here is described in 

406: detail in Huerta (2007) and Huerta et al. (2007).  We summarize it here.

407: The radial velocity shift of each observed spectrum is determined by a

408: cross correlation analysis of the observed spectrum with respect to a

409: reference spectrum.  To avoid complications which might result from a

410: spectral type mismatch, we use one particular HJS epoch (2454137.8215)

411: of \xon\ as the reference spectrum.  Again, wavelength calibration is

412: done by averaging the wavelength solution from Thorium-Argon reference

413: lamps taken before and after each stellar observation.  A total of 15

414: spectral orders which contain numerous, strong stellar lines and no

415: detectable telluric absorption lines are used in the cross correlation

416: analysis.  The radial velocity shift from the 15 orders are averaged

417: to get the final radial velocity shift for each observation, and the 

418: standard deviation of this mean is computed and adopted as the 

419: uncertainty of the cross correlation analysis.  Barycentric radial

420: velocity corrections are then applied to the observations to determine

421: accurate relative velocities for all the HJS spectra of \xon.

422: 

423: In addition to the uncertainty found above from the order-to-order

424: scatter in the HJS radial velocity measurements, there is an uncertainty

425: associated with the fact that our wavelength calibration is based on

426: spectra (Thorium-Argon) that are not observed simultaneously with the

427: stellar spectrum as is the case for iodine cell observations.  In order

428: to evaluate this uncertainty, a total of 6 stars from the Lick Planet

429: Search sample (Nidever et al. 2002; Butler et al. 1996; Cumming et al. 1999)

430: which are known to be stable at the 5 -- 20 \mps\

431: level are observed each night and analyzed in exactly the same way

432: as we treat the spectra of \xon.  The internal uncertainty in the radial

433: velocity shift measurements for these standard stars based on the measured

434: order-to-order scatter is typically $< 20$ \mps.  For the 2007 February

435: observing run (observations 3-10 at the HJS), the time series of radial 

436: velocity measurements for these standard stars show an average standard 

437: deviation of 123 \mps, with all but one of these stars showing a measured 

438: standard deviation less than 126 \mps.  Therefore, we adopt the average

439: value of 123 \mps\ as the uncertainty due to the non-simultaneous nature of 

440: the HJS wavelength calibration for the 2007 February HJS observations of

441: \xon\ and add it in quadrature to the uncertainty from the order-to-order

442: scatter for each observation.  For the 2006 October observations,

443: the radial velocity standard star data gives an uncertainty of 132 \mps\

444: due to the non-simultaneous nature of the target and wavelength calibration

445: spectra.  In addition, Huerta (2007) and Huerta et al. (2007) show that

446: there are systematic offsets of the order of 100 \mps\ in the radial

447: velocity standard star measurements from one observing run to the next, 

448: and speculate that this is caused by replacing the spectrometer slit plug

449: between observing runs (the slit plug is not moved during the observing

450: runs).  Using the radial velocity standard star data, we estimate that there

451: is a systematic shift of $140 \pm 33$ \mps\ between the 2006 October and 

452: 2007 February observations.  We correct our 2006 October radial velocity

453: measurements of \xon\ by this amount and add this associated uncertainty

454: in quadrature with the other radial velocity uncertainties described

455: above.  The measured radial velocity shifts and the total uncertainty are 

456: reported in Table \ref{table:rvobs} and shown in Figure \ref{fig:rvfit} phased

457: to the ephemeris known from the transits.  The uncertainty for the radial 

458: velocity ($= 0$ \mps) of the reference epoch (2454137.8215) is set equal to 

459: the lowest uncertainty determined for the other HJS relative radial velocity

460: measurements.

461: 

462: An eccentricity approximately equal to zero is expected theoretically

463: for hot Jupiters in $\sim 3$ day orbits (Bodenheimer et al. 2001) and

464: was our expectation for \xonb.

465: Therefore, we expected sinusoidal radial velocity variations for \xon\ with

466: a phasing consistent with the transit observations.  The first two observations

467: obtained with the HJS, taken at phase $\sim 0.64$ and $\sim 0.95$ are

468: inconsistent with this expectation, forcing us to consider eccentric

469: orbits.  Many additional data points are required to adequately constrain an

470: eccentric orbit, so we performed intense observing of \xon\ with both the

471: HJS and the HET in late 2006 and early 2007.  When fitting the orbit, 

472: in addition to the radial velocity data, we also use the time of mid

473: transit as a constraint in the fitting.  Additionally, since the HJS and

474: HET data are on different relative velocity scales, we treat as a free

475: parameter the offset between these two scales.  In the orbit fitting,

476: we keep as fixed the orbital period and phases determined from the transit

477: ephemeris, and treat as free parameters the center of mass velocity of the

478: system, the eccentricity, the velocity amplitude $K$ for \xon, the longitude

479: of periastron, the phase of periastron passage, and the offset between 

480: the HJS and HET radial velocity measurements.  We use the nonlinear least 

481: squares technique of Marquardt (see Bevington \& Robinson 1992) to find the 

482: best fit parameters for the orbit, which are given in Table \ref{table:planet}.

483: Figure \ref{fig:rvfit} shows the radial velocity data along with the

484: orbital solution.  In this plot, the HET data are shifted (by adding

485: 1192.52 \mps\ to the HET radial velocities) to match the 

486: HJS spectra and the center of mass velocity (1002.68 \mps) of all points is 

487: subtracted

488: for display purposes, since our center of mass velocity is based largely on

489: only relative data.  (As a result of these velocity shifts, the HJS velocities

490: in Table \ref{table:rvobs} have had 1002.68 \mps\ subtracted from them for 

491: display in Figure \ref{fig:rvfit}, and the HET velocities in Table 

492: \ref{table:rvobs} have had 189.84 \mps\ added to them for display in Figure

493: \ref{fig:rvfit}.)  Uncertainties in the orbital fit parameters are derived

494: by Mont\'e Carlo simulation of the data: for 1000 simulations we construct

495: fake radial velocity data using the orbital fit and applying Gaussian

496: random noise at a level commensurate with the observed data, and then fit this

497: model data using the same procedure outlined above.  Because the orbital

498: fit produces a minimum reduced $\chi^2 = 0.53$, we believe our radial velocity

499: uncertainty estimates may be a little too large.  This can be seen in

500: Table \ref{table:rvobs} where we report the observed minus calculated

501: ($O-C$) velocities of the fit.  These values are typically less than our

502: derived uncertainties for the radial velocity measurements.  The standard

503: deviation of the HJS $O-C$ values is 122 \mps\ compared to a mean 

504: derived uncertainty of 156 \mps.  For the HET data, the standard deviation of

505: the $O-C$ values is 90 \mps compared to mean derived uncertainty of 169

506: \mps (though this is dominated by a single observation), indicating that the 

507: derived uncertainties for both data sets are too large.  Reducing these 

508: uncertainties would

509: reduce the uncertainties in the fit parameters, but we leave them as given,

510: since we can not justify any specific reduction in our measured radial

511: velocity uncertainties. From the fit, we find the velocity 

512: semi-amplitude of the orbit of \xon\ is K = \vrvK$\pm$\ervK\ \mps.

513: 

514: As described below, the $v$sin$i$ of \xon\ is relatively large ($18.5 \pm 0.2$

515: \kps), which is substantially larger than most stars around which planets have

516: been discovered (HAT-P-2b is another example of a planet orbiting a star which

517: has a relatively large $v$sin$i = 19.8 \pm 1.6$ \kps, Bakos et al. 2007).  

518: The experience of Mandushev et al. (2005) and the similarity between the

519: primary in that system with \xon\ gives concern that \xon\ could be a 

520: hierachical triple system in which two lower mass stars comprise a short

521: period eclipsing binary star which are themselves orbiting a more massive, 

522: significantly brighter primary in a much longer period orbit.  In the case 

523: of Mandushev et al. (2005), the short period binary stars

524: display tens of \kps\ velocity shifts, but the strong, broad lines of the

525: system almost hide this signal.  The apparent velocity shift of the system

526: primary is then the result of weaker lines moving around the (stationary) lines

527: of the primary, producing line profile distortions which are misinterpreted 

528: as radial velocity variability of the primary.

529: 

530: Line bisector analysis is a standard technique to look for evidence that line

531: distortions are producing false radial velocity variations (e.g., Hatzes et

532: al. 1997; Martinez Fiorenzano et al. 2005), and have been used successfully 

533: to identify false radial velocity variations caused by star spots (e.g.,

534: Queloz et al. 2001; Bouvier et al. 2007; Huerta et al. 2007) and 

535: hierarchical triple systems 

536: (e.g., Santos et al.  2002; Mandushev et al. 2005).  To improve signal-to-noise

537: in the bisector analysis, it is common practice to compute the bisector of 

538: the spectrum cross correlation function (Queloz et al. 2001; Santos et al. 

539: 2002; Mandushev et al. 2005).  Here, we calculate the line bisector of the 

540: total cross correlation function computed from all 15 orders in the HJS 

541: spectra used in the determination of the radial velocities.  We then compute

542: the bisector span, which is the difference of the line bisector at two

543: reference levels.  Here we define the bisector span, $S_B$, as the value

544: of the bisector at an absolute cross correlation level of 0.15 minus the 

545: bisector at an absolute cross correlation level of 0.75.  These span values

546: are given in Table 

547: \ref{table:rvobs}.  Because the spectral regions of the HET data 

548: which contain strong stellar lines also contain numerous iodine absorption 

549: lines, we restrict ourselves to the HJS data for the analysis of line

550: bisectors.  Previous studies which used bisector analysis to reveal a false

551: positive in planet searches find a clear correlation between the measured

552: radial velocity and the bisector span (Queloz et al. 2001; Santos et al.

553: 2002; Mandushev et al. 2005; Bouvier et al. 2007; Huerta et al. 2007).

554: In these cases, 

555: whether due to spots (e.g., Queloz et al. 2001; Bouvier et al. 2007; Huerta

556: et al. 2007) or due to a hirearchical triple system (Santos et al. 2002; 

557: Mandushev et al. 2005), the full range measured in $S_B$ is approximately 

558: the same as the full range of the measured radial velocities.  

559: 

560: On the other hand, Torres et al. (2004) use the lack of a correlation between

561: $S_B$ and the radial velocity to help confirm a planet transiting OGLE-TR-56b.

562: In addition, for OGLE-TR-56b the full range in $S_B$ is less than one third 

563: the full range in the radial velocities, which Torres et al. (2004) argue is 

564: additional evidence for the planetary nature of this object.  

565: Figure \ref{fig:bisect} shows the bisector span,

566: $S_B$, plotted versus the measured radial velocity for the HJS spectra of

567: \xon.  Also given in the figure is the value of the linear correlation

568: coefficient and its associated false alarm probability (Bevington \&

569: Robinson 1992) for these data.  In addition to there being no 

570: significant correlation

571: between $S_B$ and the radial velocity, the full range of $S_B$ is very

572: small compared to the full range in the radial velocity.  

573: Figure \ref{fig:bisect} and Table \ref{table:rvobs} show that

574: the full range in $S_B$ in \xon\ is less than one

575: tenth the full range in the radial velocity variations for \xon.

576: 

577: As a last check that the photometric dimmings and radial velocity variations

578: of \xon\ are due to a planetary mass companion and not the result of a

579: hierarchical triple system with later type stars present, we looked for an

580: infrared (IR) excess in \xon.  O'Donovan et al. (2006) used a redder than 

581: expected $V-K$ measurement as one piece of evidence to reject the planetary

582: hypothesis for the F star GSC 03885-00829.  Using the photometry reported in

583: Table \ref{table:star}, we estimate the color $V-K_{\rm s} = 1.01 \pm 0.06$ 

584: for \xon.  Based on the effective temperature and gravity found below for 

585: \xon, we assign a spectral type of \sptype\ based on the calibrations presented

586: in Cox (2000).  In the color system of Bessell and Brett (1988),

587: an F5V star such as XO-3 is expected to have an intrinsic 

588: $(V - K_{\rm s})_\circ = 1.10$. For comparison to the 2MASS colors reported in

589: Table \ref{table:star}, we use the color transformation relations found

590: in the {\it Explanatory Supplement to the 2MASS All Sky Data Release and 

591: Extended Mission Products} by Cutri et al. 

592: (http://www.ipac.caltech.edu/2mass/releases/allsky/doc/), and find that we

593: must add 0.039 mag to the Bessel \& Brett value in 

594: order to estimate $(V - K_{\rm s})_\circ$ = 1.14 on the 2MASS system. Thus, 

595: the observed $V-K_{\rm s} = 1.01$ color of XO-3 is 0.13 mag bluer 

596: ($\sim2$-$\sigma$) than its expected color.

597: The slight bluing may be the result of the

598: relatively low metallicity we derive (see below), but there certainly is

599: no evidence for a near IR excess suggesting a hierarchical triple system.

600: The bisector analysis and the $V-K_{\rm s}$ color of \xon\ give us confidence

601: that both the photometric dimming and radial velocity variations we measure 

602: in this star are indeed due to a planetary mass companion.

603: 

604: 

605: \subsection{Spectroscopically-Derived Stellar Properties and Planetary Mass}

606: 

607: We used the software package SME (Valenti \& Piskunov 1996) to fit each

608: of the 10 spectra of \xon\ from the HJS telescope with synthetic spectra.

609: We used

610: the methodology of VF05, including their minor

611: corrections to match the Sun and remove abundance trends with temperature

612: (negligible in this case). Because of gaps between echelle orders,

613: the HJS spectra are missing the wavelength intervals 6000--6123

614: \AA, which was included in the Valenti \& Fischer

615: analysis. These wavelength intervals are also missing from our extracted

616: HET spectra because the relevant echelle orders span the gap between

617: the two detectors.

618: 

619: We averaged our SME results for the 10 HJS spectra, obtaining

620: the parameter values in Table \ref{table:sme}. 

621: Each value in the last column of the table, labeled ``Precision'' because

622: systematic uncertainties are not included,

623: is the standard deviation of the 10 measurements divided

624: by $\sqrt{9}$ to yield the formal uncertainty in the mean.

625: The median value of

626: each derived parameter (not given) differs from the mean by less than the

627: uncertainty in the mean. The final row in the table gives [Si/Fe], which

628: VF05 used as a proxy for alpha-element enrichment,

629: when interpolating isochrones.

630: Figure \ref{fig:mcd} shows \xon's spectrum in the region of the \ion{Mg}{1}

631: B triplet, which is the dominant spectroscopic constraint on gravity. These

632: three Mg lines also have a significant impact on the global [M/H]

633: parameter, which is used to scale solar abundances for all elements

634: other than Na, Si, Ti, Fe, and Ni.

635: 

636: Following the methodology of Fischer \& Valenti (2005), we interpolated

637: Yonsei-Yale (Y$^2$) isochrones (Demarque et al. 2004) to determine

638: probability distribution functions for the mass, radius, gravity, and age

639: of \xon. The trigonometric parallax of \xon\ is unknown, so we initially

640: assumed distances of 240, 260, and 280 pc with an adopted uncertainty of

641: 10 pc in each case (which affects the width of the resulting distribution

642: functions).  In order to perform the isochrone analysis, we need a $V$ magnitude

643: corrected for extinction for \xon.  We measured $V = 9.80\pm0.03$

644: and $B = 10.25\pm0.03$ for \xon\ (Table \ref{table:star}).  These

645: observations give a measured $B-V = 0.45\pm0.04$ for \xon.

646: VandenBerg \& Clem (2003) determined empirical color-temperature relations

647: for stars including variations with gravity.  We downloaded their high

648: temperature table (mentioned in footnote 1 of their paper) and 

649: interpolated in the table to the effective temperature, surface gravity, 

650: and metallicity of \xon\ determined above.  Doing so gives a predicted

651: $(B-V)_\circ = 0.440$ for \xon.  As discussed below, the lightcurve

652: analysis and the derived properties of \xonb\ favor a larger stellar

653: gravity for \xon.  So, we also interpolate the predicted colors to a

654: gravity of log$g = 4.2$ (a little more than 3$\sigma$ higher than 

655: the spectroscopically derived value).  Doing so results in a predicted 

656: $(B-V)_\circ = 0.446$.  These two predicted values are within

657: 0.01 and 0.001 mag of the measured $B-V$ of \xon, with an

658: uncertainty of 0.04 in the measured color.  We therefore find no

659: compelling evidence for significant reddening to \xon\ and we adopt

660: $V = 9.80\pm0.03$ as the extinction corrected magnitude of the star.

661: 

662: We used our spectroscopic effective temperature, spectroscopic gravity,

663: and an assumed distance to derive a bolometric correction by interpolating

664: the ``high temperature'' table from VandenBerg \& Clem (2003).

665: We combined the bolometric

666: correction with the observed V-band magnitude to determine stellar

667: luminosity. Then we used the stellar luminosity and our spectroscopic

668: effective temperature, iron abundance, and alpha-element enrichment to

669: interpolate the Y$^2$ isochrones to produce the probability distribution 

670: functions in Figure \ref{fig:stellarparam}. 

671: 

672: The parameters we derive place \xon\ above the zero-age main sequence.  As 

673: a result, reasonable fits to the data are possible for pre-main sequence 

674: evolutionary tracks.  However, if \xon\ were indeed a pre-main sequence star,

675: we would expect strong \ion{Li}{1} absorption at 6708 \AA.  For example, in 

676: their survey of Pleiades F stars, Boesgaard et al. (1988) show that the

677: \ion{Li}{1} line at 6708 \AA\ is approximately as strong as the nearby 

678: \ion{Ca}{1} line at 6718 \AA\ is stars of similar spectral type to \xon.  We 

679: show in Figure \ref{fig:licheck} the Li line region in \xon\ produced by 

680: averaging the spectra obtained at the HET (this region is free of iodine 

681: absorption).  The position of the \ion{Li}{1} line is marked and the nearby 

682: \ion{Ca}{1} line is clearly visible.  The Li line is not detectable, 

683: indicating that \xon\ is substantially older than the Pleiades, therefore we

684: conclude \xon\ is not a pre-main sequence star.

685: 

686: The most probable distance, mass, radius, and age for \xon\ are 260 pc, 

687: 1.41 M$_\odot$, 2.13 R$_\odot$, and 2.69 Gyr respectively, for our best fit

688: gravity of log$g = 3.95$.  Combined with our measured semi-amplitude

689: K = \vrvK$\pm$\ervK\ \mps, this stellar mass implies a mass of 

690: $M_p$sin$i = $\vMp$\pm$\eMp\ \Mjup\ for the planet \xonb.

691: If we assume that \xon\ is at 240 pc instead, then the most probable 

692: mass, radius, age, and gravity are 1.36 M$_\odot$, 1.95 R$_\odot$, 2.78 Gyr,

693: and log$g = 4.00$, still placing the star well above the main sequence.

694: As discussed below, one of the unusual properties of \xonb\ is the relatively

695: large radius we derive for the planet.  The planetary radius is approximately

696: proportional to the inferred stellar radius.  The planetary radius could be

697: smaller if we could justify a smaller stellar radius; however, doing so begins 

698: to bring the inferred gravity in conflict with the spectroscopically derived

699: value of log$g = 3.95\pm0.062$.  Taking a 3$\sigma$ upper limit, the largest

700: allowed value for the gravity is log$g = 4.14$, corresponding to a stellar

701: mass of 1.27 M$_\odot$ and a stellar radius of 1.62 R$_\odot$ for an inferred

702: distance of 200 pc.  Together with the measured velocity semi-amplitude, 

703: these values imply a planetary mass of $M_p$sin$i = 12.14\pm0.40$ \Mjup\

704: where the uncertainty here is just that resulting from the radial velocity fit.

705: Clearly, a trigonometric parallax measurement for 

706: \xon\ will be of great value in better establishing both the stellar and 

707: planetary parameters.

708: 

709: \subsection{Light Curve Modeling and the Planetary Radius}

710: 

711: For additional photometric

712: analysis and lightcurve fitting, all the E.T. photometric data shown in

713: Figure \ref{fig:etlc} were averaged into 7.4 minute bins.

714: Binning permitted

715: outlier rejection and empirical estimation of the noise by the scatter

716: of the individual observations, which is helpful in cases such as this

717: in which residual calibration errors can be significant.

718: Differences in the depth and the shape

719: of the transit light curve in different photometric bands are both expected and

720: observed to be

721: substantially smaller than the photometric uncertainties in the E.T.

722: lightcurves, so we combined the light curve from all photometric bands

723: to improve the signal to noise ratio.  The

724: binning begins with phasing the light curves

725: using the refined ephemeris determined in \S3.1.

726: The final binned light curve consists of a robust

727: average of E.~T.\ photometric measurements in bins 7.4 minutes in duration.

728: This robust average is determined by first calculating

729: the median and median absolute deviation for each bin and rejecting points

730: in the bin that deviate by more than $4\sigma$ from the median.  Next,

731: the mean and standard deviation are calculated for each bin and points

732: which are more than $4\sigma$ from the mean are rejected.  This rejection

733: based on the mean and standard deviation is repeated one more time before

734: a final (``robust") mean is calculated.

735: The final binned light curve is supplied in Table \ref{table:lc} and

736: is shown in Figure \ref{fig:lcurvefit} along

737: with the fits described below.  

738: The uncertainty in each binned light curve data point is the standard 

739: deviation of (surviving) measurements in the bin divided by the square root 

740: of the number of measurements in that bin.  The uncertainties in all bins are

741: then multiplied by a correction factor as described below (the uncertainties

742: in Table \ref{table:lc} do not have this correction factor applied).

743: 

744: We modeled the mean transit light curve using the analytic transit model

745: of Mandel and Agol (2002).  For a planet with an

746: eccentric orbit, the transit model has ten parameters: transit

747: midpoint ($t_{o}$), orbital period, stellar mass ($M_{\star}$),

748: stellar radius ($R_{\star}$), planet radius ($R_{\rm p}$), inclination

749: ($i$), quadratic limb darkening law coefficients ($u_{1}$ and

750: $u_{2}$), eccentricity ($e$), and longitude of perihelion ($\omega$).

751: Throughout the following analysis, the orbital period remains fixed at

752: the period found in \S3.1.  The

753: radial velocity data provide the best estimates for $e=0.26$ and

754: $\omega=-15.4$ that remain fixed during the $\chi^{2}$ minimization.

755: For each fit presented below, we fix the stellar parameters to specific

756: values based on the spectral synthesis and isochrone analysis presented

757: above.

758: We interpolate quadratic limb darkening coefficients from Claret (2000) for 

759: the Cousins R photometric bandpass ($u_1 = 0.220$; 

760: $u_2 = 0.380$) based on the gravity, effective temperature, and 

761: metallicity ([M/H]) determined from the spectroscopic analysis

762: (Table \ref{table:sme}).

763: The remaining parameters,

764: $t_{o}$, $R_{\rm p}$, and $i$, are allowed to vary and are solved for

765: using the ``Amoeba" algorithm mentioned earlier.

766: For input into the Mandel and Agol (2002) transit model, we determine the

767: projected separation between star and planet, $\delta$, in the case of

768: nonzero eccentricity following the procedure outlined by

769: Hilditch (2001).  For a given $i$, $\omega$, and $e$, we determine the

770: true anomaly, $\theta{\rm min}$, that minimizes $\delta$ (Equation~4.9

771: of Hilditch 2001).  The resulting $\theta{\rm min}$ corresponds to the

772: mean anomaly at transit midpoint providing the zeropoint to convert

773: observed times to $\theta$ via Kepler's equation.

774: 

775: We then initially fit the light curve assuming the

776: stellar mass (1.41 \Msun) and radius (2.13 \Rsun) derived from the

777: spectral fits as described above.  The free parameters of the light curve

778: fit are then the radius of the planet and the inclination of the orbit

779: (and the time of mid-transit).  The resulting fit is shown

780: in the top panel of Figure \ref{fig:lcurvefit}.  We then repeated the light

781: curve fits using the extremes in the stellar mass (1.33 and 1.90 \Msun)

782: and radius (1.49 and 2.36 \Rsun) which result from the $1\sigma$ uncertainties

783: in these parameters derived above.  Together, this then gives us estimates

784: of the planetary radius ($R_p = \vRp \pm \eRp$ \Rjup) and orbital inclination

785: ($i = 79.^\circ32 \pm 1.^\circ36$) which are also given in Table 

786: \ref{table:planet}.  The total mass of the planet is then

787: $\vMptot \pm \eMptot$ \Mjup.  Examination of the figure shows that the

788: model does not fit all aspects of the lightcurve: the model ingress and

789: egress appear noticeably longer than the observed ingress and egress. 

790: 

791: In order to better model the shape of the ingress and egress while still

792: maintaining the total duration of the transit, the radius of the star and

793: the planet both need to be reduced and the inclination must increase so that

794: the total chord length the planet traverses accross the face of the star

795: stays approximately the same.  Figure \ref{fig:stellarparam} shows that

796: from the isochrone analysis, the stellar mass and radius both decrease

797: as the distance is assumed to be smaller.  We have

798: performed the isochrone analysis every 10 pc, so we computed model

799: light curves, stepping down in distance 10 pc at a time from our favored

800: distance of 260 pc.  For each new distance, we adopt the corresponding stellar

801: mass and radius resulting from the isochrone analysis and compute $\chi^2$

802: for the fit to the light curve.  The minimum in $\chi^2$ occurs between 190

803: and 200 pc.  Fitting a parabola to the lowest 4 $\chi^2$ values, gives a

804: best fit to the light curve for a distance of 185 pc which gives 

805: $M_* = 1.24$ \Msun, $R_* = 1.48$ \Rsun, and log$g = 4.19$.  This fit is

806: shown in the lower panel of Figure \ref{fig:lcurvefit}.  As mentioned

807: earlier, the the best $\chi^2$ is quite large ($\sim 129$) for the

808: 63 degrees of freedom in our fit, indicating that the photometric 

809: uncertainties have been underestimated.  We attempt to correct this by

810: determing the multiplicative factor required to give a $\chi^2 = 63.0$

811: (a reduced $\chi^2 = 1.0$).  The uncertainties shown in Figure 

812: \ref{fig:lcurvefit} have been multiplied by this factor.  Having done

813: this, we can then assign as the uncertainty in $R_P$ and $i$

814: the difference between the best fit parameters and those we obtain

815: when $\Delta\chi^2 = 1.0$.  Doing so,

816: we find that the transit lightcurve is best fit for $R_p = 1.25 \pm 0.15$

817: \Rjup\ and $i = 83.^\circ32 \pm 1.^\circ26$.  The total mass of the planet

818: is then $12.03 \pm 0.46$ \Mjup.  This value for the planetary

819: radius is considerably lower than the value derived above from the default

820: stellar parameters.  As discussed below, we regard this as a lower limit

821: to the true planetary radius; however, we note that the ambiguity 

822: described here can be greatly diminished by obtaining very accurate

823: photometry of a transit of \xon\ and/or by obtaining a precise trigonometric

824: parallax measurement.

825: 

826: \section{Discussion}

827: 

828: \subsection{Comparison to other Transiting Planets\label{sec:cf}}

829: 

830: The $\sim 13$ \Mjup\ planet \xonb\ is unusual in many respects compared to

831: the sample of known extra-solar planets.  The mass of \xonb\ is quite large 

832: compared to most of the known extra-solar planets.  Zucker and Mazeh (2002) 

833: noted that the most massive short period planets are all found in multiple

834: star systems.  Udry et al. (2003) emphasized the general lack of massive 

835: planets on short period orbits, particularly for planets orbiting single 

836: stars, and interpreted these results in terms of planetary migration 

837: scenarios.  The star \xon\ is not known to be a binary, though

838: it is relatively unstudied.  While it is in the Tycho-2 catalog

839: (H{\o}g et al. 2000), \xon\ does not have a significant proper motion measurement

840: from the {\it Hipparcos} mission.  It is not known if any of the nearby

841: stars seen in Figure \ref{fig:allsky} are common proper motion companions

842: or not.  Udry et al. (2003) point out that among planets orbiting single

843: stars known at that time, there were no planets more massive than 2 \Mjup\

844: with periods less than 100 d (see also Eggenberger et al. 2004).  Using 

845: data for 218 extra-solar planets as compiled on the Geneva Extrasolar Planet 

846: website (http://www.exoplanets.eu) updated as of 27 May 2007, this general 

847: lack of short period, very massive planets is still apparent.  There are only 3

848: other planets (HD 162020b, HD 17156b, and HAT-P-2b)

849: orbiting apparently single stars with periods less than 30 days

850: and masses larger than 2.5 \Mjup, and one of those (HD 162020b) is suspected

851: to actually be a much more massive brown dwarf (Udry et al. 2002).  There is

852: only 1 planet other than \xonb\ with a mass larger than 2.5 \Mjup\ and a

853: period less than 4 days (HD 120136b, M$_2$sin$i = 4.14$ \Mjup, Butler et al.

854: 1997), and this star is known to be in a stellar binary system (Hale 1994).

855: 

856: The eccentricity of \xonb\ is also quite rare relative to other known

857: extra-solar planets, given the short orbital period of \xonb.  The 

858: eccentricity of the shortest period extra-solar planets are all quite low

859: ($e < 0.05$) and most are consistent with zero eccentricity (Halbwachs et

860: al. 2005).  This is generally believed to be due to either tidal 

861: circularization (e.g., Wu 2003; Ivanov \& Papaloizou 2007) or the result of 

862: smooth orbital migration in a disk which is not thought to produce significant

863: eccentricity (Murray et al. 1998).  The eccentricity of \xonb\ is larger than

864: all other planets with periods less than that of \xonb.  

865: We discuss the

866: eccentricity of \xonb\ further in Section \ref{sec:tidal}.  While the mass, period, and 

867: eccentricity of \xonb\ make it quite rare among extra-solar planets, it

868: is not totally unique.  The recently discovered HAT-P-2b (Bakos et al. 2007)

869: is very similar in many respects with a period of 5.6 d, $M_p = 9.04$ \Mjup,

870: and $e = 0.52$.  

871: Also, the hot Neptune GJ 436b has a period of 2.6 days and e = 0.15

872: (Butler et al 2004; Deming et al 2007). Apparently the assumption,

873: commonly-held prior to 2007, that short-period planets should have

874: circular orbits was incorrect.

875: 

876: It is now well established that host star metallicity correlates with the

877: likelihood of finding a Jupiter mass companion orbiting the star in the

878: sense that higher metallicity stars are more likely to have planets (Santos

879: et al. 2004; Fischer \& Valenti 2005).  The metallicity of \xon\ ([M/H] 

880: = -0.20) is relatively low, and the above mentioned studies show that planets

881: are found around only $\sim 3$\% of stars with similar metallicity.

882: There are only two other very 

883: massive planets ($M_p > 5$ \Mjup) known around stars (HD 111232, HD 114762)

884: with metallicities lower than that of \xon.  The correlation between host

885: star metallicity and the probability of finding a planet is often taken as

886: support for the core accretion model (e.g., Pollack et al. 1996; Bodenheimer 

887: et al. 2000; Hubickyj et al. 2005) of planet formation.  Under that assumption

888: though, it may be surprising that a few very massive planets have also been

889: found around low metallicity stars.   It has also been suggested that 

890: host star metallicity decreases on average as the planet mass increases 

891: which appears inconsistent with the core accretion model (Ribas \&

892: Miralda--Escud\'e 2007).  On the other hand, gravitational instabilities in 

893: a disk are not expected to produce a metallicity-planet correlation 

894: (Boss 2002).  Taking this notion further, Ribas and Miralda--Escud\'e

895: (2007) have suggested that the sample of extra-solar planets consists of

896: objects formed via two different paths: core accretion in a disk and 

897: fragmentation of a pre-stellar cloud.  The high mass, relatively large

898: eccentricity, and low metallicity of the \xonb\ system would then ``fit''

899: the notion of Ribas and Miralda--Escud\'e of a planet

900: formed from the collapse of a pre-stellar cloud, but of course the

901: observations of \xonb\ do not prove (or disprove) either formation scenario.

902: 

903: \subsection{How Big is \xonb?}

904: 

905: Potentially, one of the most unusual and interesting properties of \xonb\ is

906: the large radius of the planet.  At an age of 2-3 Gyr, giant planets and

907: low mass brown dwarfs are expected to have radii very close to that of Jupiter

908: (Burrows et al. 2001; Baraffe et al. 2003).  Using the tables in Baraffe et 

909: al. (2003), the radius of a 13 \Mjup\ planet is expected to be 1.03 \Rjup\ at

910: 1 Gyr, and 0.97 \Rjup\ at 5 Gyr.  Including heating from the central star can

911: increase the expected radius for so-called ``hot Jupiters" (e.g., Bodenheimer 

912: et al. 2003).  Fortney et al. (2007) compute planetary structure models up to 

913: 11.3 \Mjup\ for a range of orbital separations, including the insolation

914: produced by absorption of solar radiation.  At an age of 300 Myr or older,

915: all the models computed have radii less than 1.3 \Rjup.  In all cases

916: analyzed by Fortney et al., their models

917: predict a substantially smaller radius than the $\vRp \pm \eRp$ \Rjup\ 

918: determined based upon the values of the stellar mass and radius inferred 

919: from the spectroscopic and isochrone analysis of \xon.  Such a large

920: discrepancy between the observed and predicted planetary radius is 

921: reminiscent of TReS-4 which has a radius of $1.674 \pm 0.094$ \Rjup\

922: (Mandushev et al. 2007).  However, the radius of \xonb\

923: is quite uncertain, and the light curve itself favors a smaller planetary 

924: radius of $R_p = 1.25 \pm 0.15$ \Rjup.  Because \xonb\ is in an eccentric 

925: orbit, for comparison with the models of Fortney et al. (2007)

926: we compute the average orbital

927: separation between \xonb\ and its star and then adjust this separation to

928: take into account the difference in luminosity between

929: \xon\ and the Sun.  The smaller planetary radius for \xonb\ is associated

930: with a smaller

931: stellar radius for \xon, so for this exercise, we take $R_* = 1.57$ \Rsun.

932: We thus compute an effective distance of 0.024 AU to use for \xonb\ when

933: using the tables of Fortney et al. (2007).  Interpolating in these tables,

934: the predicted radius for \xonb\ is 1.18 \Rjup\ at 1 Gyr and 1.11 \Rjup\ at

935: 4.5 Gyr.  Both values are lower than the smaller radius inferred for \xonb,

936: but only by about $1\sigma$.  A more precise radius for \xonb\ would be

937: a very interesting 

938: benchmark for the theory of extra-solar planets.

939: 

940: The actual radius of \xonb\ can be better constrained by a true parallax

941: measurement and more precise photometric observations (ideally in multiple

942: colors) of additional transits.  The constraint from this work that favors the larger

943: radius for \xonb\ is the stellar gravity inferred from the spectroscopic

944: analysis.  A number of investigators have pointed out that stellar gravity

945: determinations are notoriously difficult and the relative error in this

946: parameter is often substantially larger than almost all other measured

947: properties of transiting extra-solar planetary systems.  Many of these

948: investigators have elected to use the transit light curve fits in combination

949: with stellar isochrone models to estimate the stellar and planetary radius

950: and hence log$g$ for the star (e.g., Sozzetti et al. 2004, 2007; O'Donovan 

951: et al.  2006; Bakos et al. 2007; Mandushev et al. 2007).  The photometry

952: of \xonb\ is not as precise as that used in those studies, so

953: we do not emphasize 

954: the photometrically-derived parameters as those investigators have. However, we can

955: examine the spectroscopically determined log$g$ to estimate by how much,

956: and in what sense, it is in error.  

957: 

958: The above studies typically use spectroscopically

959: determined effective temperatures with isochrone models to determine the

960: stellar gravity, implicitly assuming the gravities determined from isochrone

961: analyses are more accurate than the spectroscopic values.  

962: We can compare the gravities determined from the Y$^2$ isochrones and our

963: spectroscopic analysis

964: for stars of similar spectral type to \xon.  The sample

965: of stars studied by VF05 all have accurate 

966: {\it Hipparcos} parallax measurements, allowing them to use the isochrones 

967: with the spectroscopically determined $T_{eff}$ to determine the stellar

968: gravity.  This gravity can then be compared to the gravity VF05

969: derive from the spectra alone, using the same techniques

970: used in this paper.  Figure \ref{fig:loggcheck} shows the result of

971: this comparison for 79 stars with $T_{eff} > 6200$ K from VF05.

972: Recall $T_{eff} = 6429 \pm 50$ for \xon.  There is considerable

973: scatter in this figure; however, in the immediate vicinity of the 

974: spectroscopic gravity determined for \xon, the gravities determined from

975: the isochrones are lower.  This is the opposite sense to what is suggested

976: by the transit light curve of XO-3.  As a result, there is no clear indication in

977: this sample of stars that the spectroscopically determined gravity for 

978: \xon\ is biased low.  When making this comparison, it is appropriate to

979: consider the accuracy of the gravities predicted by the Y$^2$ isochrones.

980: Hillenbrand and White (2004) report excellent agreement (better that 3\%)

981: between dynamically determined stellar masses from eclipsing binaries and 

982: masses determined from the Y$^2$ isochrones for main sequence stars more 

983: massive than $\sim 0.6$ \Msun; therefore, we find no reason to doubt the

984: Y$^2$ isochrones appropriate for \xon.  We also note that results

985: from transit light curve analyses are subject to various assumptions and

986: potential biases that may exist in the models used for the fit.  For example,

987: Aufdenberg et al. (2005) confirm the prediction of Allende Prieto et al.

988: (2002) that 1D atmosphere models, including the ATLAS models used by 

989: Claret (2000) to determine limb darkening coefficients, predict too much

990: limb darkening at optical wavelengths, depending on the treatment of 

991: convection and convective overshoot in the model. In fitting light curves,

992: such a bias will

993: favor models with smaller impact parameters and smaller stellar

994: radii.  Therefore, until better data

995: can be obtained for \xon\ (a precise parallax measurement and/or more

996: precise transit photometry), the stellar

997: and hence planetary radius will remain significantly uncertain.  At this point,

998: taking the photometric lightcurve and the spectroscopic gravity into 

999: account, we estimate the radius of \xonb\ to be $1.10 < R_p < 2.11$ \Rjup\

1000: including $1\sigma$ uncertainties on both limits.  The lower limit is almost

1001: certainly too low however, given the results of Aufdenberg et al. (2005)

1002: and Allende Prieto et al. (2002) on limb darkening showing

1003: that the values we have used here are likely overestimated.

1004: 

1005: \subsection{Tidal Circularization and the Eccentricity of \xonb}

1006: \label{sec:tidal}

1007: 

1008: As mentioned in Section \ref{sec:cf}, an interesting aspect of the orbit of \xonb\ is its 

1009: significant eccentricity, $e = \vecc \pm \eecc$, because

1010: most planets with orbital periods

1011: similar to \xonb\ are believed to have already been tidally circularized

1012: (Halbwachs et al. 2005).  

1013: Independent of exactly

1014: how \xonb\ arrived at its current orbit, it is interesting to consider

1015: how long it can remain in such an orbit under the influence of tidal

1016: circularization.  

1017: The question of tidal circularization of close

1018: extra-solar planets has been studied by several investigators (see 

1019: Adams \& Laughlin 2006 and references therein).  The equations governing the

1020: tidal circularization of extra-solar planets depend on very uncertain

1021: planetary quality factors, $Q_p$ (for example, see discussion in Gu et al.

1022: 2003) and in some cases equally uncertain stellar quality factors, $Q_*$,

1023: if considering tides raised on the star by the planet.  As an 

1024: illustration of the difficulty in estimating quality factors, Mathieu

1025: (1994) point out that theoretically determined values of $Q_*$ imply a

1026: slow rate for close binary stars to circularize which is contradicted by

1027: observations in stellar clusters of various ages.  As a result, many

1028: investigators try to use the observations (of binary stars and extra-solar

1029: planets) to empirically estimate the quality factors.  For Jupiter

1030: mass extra-solar planets, $Q_p \sim 10^5 - 10^6$ (e.g., Gu et al. 2003;

1031: Adams \& Laughlin 2006).  

1032: 

1033: Circularization timescales depend linearly on

1034: $Q_p$, so we assume $Q_p = 10^6$ to get predictions at the long end of what can

1035: currently be estimated.  Using equation (18) of Gu et al. (2003) and

1036: equation (3) of Adams \& Laughlin (2006) to estimate the circularization

1037: time for \xonb, we find timescales of 0.29 Gyr and 0.33 Gyr, respectively,

1038: using values corresponding to the large radius (1.9 \Rjup) for \xonb.

1039: The circularization timescale in both studies depends on the reciprocal of

1040: the planetary radius

1041: to the fifth power, so the circularization timescale grows to 2.92 Gyr and 3.32

1042: Gyr, respectively, using values corresponding to the small radius (1.25

1043: \Rjup) for \xonb.  Clearly, there is considerable uncertainty in the value

1044: of the circularization time; however, because circumstellar disks appear to

1045: be lost after 10-20 Myr (e.g., Haisch et al. 2001; Mamajek et al. 2004), it 

1046: appears that \xonb\ perhaps should have circularized by now if it originally

1047: arrived at its current location while the circumstellar disk was still in 

1048: place (i.e. through migration).  If instead, \xonb\ has been scattered in

1049: to its current location (e.g., Ford \& Rasio 2007) more recently, there may

1050: not have been enough time for circularization to occur, particularly for

1051: the parameters corresponding to the smaller radius for \xonb.  A third

1052: possibility is that additional, undetected planets orbiting \xon\

1053: maintain the eccentricity of \xonb\ (e.g., Adams \& Laughlin 2006).  It is

1054: interesting to note though that for planets in short period, eccentric

1055: orbits such as \xonb, the tides raised in the planet can deposit

1056: substantial energy into the planet (e.g., Gu et al. 2003; Adams \& 

1057: Laughlin 2006) which can inflate it to radii coresponding to

1058: the larger value we measure (Gu et al. 2003).  It therefore appears

1059: \xonb\ can again serve as a very interesting benchmark for studies of tidal

1060: circularization and tidal heating once a more accurate radius can be

1061: established for the planet.

1062: 

1063: \section{Summary}

1064: 

1065: \xonb\ is a massive planet or low mass brown dwarf in a short period ($3.2$

1066: d), eccentric ($e=0.26$) orbit, around a somewhat evolved F5 star.  There is 

1067: relatively large uncertainty in the planetary parameters owing to the unknown 

1068: distance to \xon.  The mass of \xonb\ is 11.57 -- 13.97 \Mjup\ and the radius

1069: is 1.10 -- 2.11 \Rjup.  The larger mass and radius are favored by our

1070: spectroscopic analysis of \xon, and the smaller values are favored by

1071: the light curve analysis.  A precise trigonometric parallax measurement

1072: or more accurate photometric light curve date are needed to distinguish 

1073: between these values.

1074: 

1075: \acknowledgments

1076: 

1077: We wish to thank the anonymous referee for many useful comments which

1078: improved the manuscript.

1079: The University of Hawaii staff have made the operation on Maui possible; we

1080: thank especially Bill Giebink, Les Hieda, 

1081: Jake Kamibayashi,

1082: Jeff Kuhn, Haosheng Lin, Mike Maberry,

1083: Daniel O'Gara, 

1084: Joey Perreira, Kaila Rhoden, and the director of the IFA, Rolf-Peter Kudritzki.

1085: 

1086: This research has made use of a Beowulf cluster constructed by Frank

1087: Summers; the SIMBAD database, operated at CDS, Strasbourg, France;

1088: data products from the Two Micron All Sky Survey (2MASS) and

1089: the Digitized Sky Survey (DSS);

1090: source code for transit light-curves (Mandel \& Agol 2002);

1091: and community access to the 

1092: Hobby-Eberly Telescope (HET), which is a joint project of the University

1093: of Texas at Austin, the Pennsylvania State University, Stanford

1094: University, Ludwig-Maximilians-Universit\"{a}t M\"{u}nchen, and

1095: Georg-August-Universit\"{a}t G\"{o}ttingen.  The HET is named in honor

1096: of its principal benefactors, William P. Hobby and Robert E. Eberly.

1097: We thank the HET night-time and day-time support staff.

1098: We also wish to thank the support staff at McDonald Observatory, and to

1099: especially thank M. Huerta for his assistance observing at the 2.7 m HJS

1100: Telescope.

1101: 

1102: XO is funded primarily by NASA Origins grant 

1103: NNG06GG92G and the Director's Discretionary Fund of STScI.

1104: C.~M. Johns--Krull and L. Prato wish to ackowledge partial support from 

1105: NASA Origins of Solar Systems grant 05-SSO05-86.

1106: We thank Brian Skiff and Josh Winn for noting that a preprint had an error in the coordinates and there was a sign error in our radial velocity fitting algorithm, respectively; both were corrected prior to publication.

1107: 

1108: \begin{thebibliography}

1109: 

1110: \bibitem[Adams \& Laughlin(2006)]{2006ApJ...649.1004A} Adams, F.~C., \& 

1111: Laughlin, G.\ 2006, \apj, 649, 1004

1112: 

1113: \bibitem[Allende Prieto et al.(2002)]{2002ApJ...567..544A} Allende Prieto, 

1114: C., Asplund, M., L{\'o}pez, R.~J.~G., \& Lambert, D.~L.\ 2002, \apj, 567, 

1115: 544 

1116: 

1117: \bibitem[Aufdenberg et al.(2005)]{2005ApJ...633..424A} Aufdenberg, J.~P., 

1118: Ludwig, H.-G., \& Kervella, P.\ 2005, \apj, 633, 424

1119: 

1120: \bibitem[Bakos et al.(2007)]{2007ApJ...670..826B} Bakos, G.~{\'A}., et al.\ 

1121: 2007, \apj, 670, 826

1122: 

1123: \bibitem[Baraffe et al.(2003)]{2003A&A...402..701B} Baraffe, I., Chabrier, 

1124: G., Barman, T.~S., Allard, F., \& Hauschildt, P.~H.\ 2003, \aap, 402, 701

1125: 

1126: \bibitem[Bessell(2000)]{2000PASP..112..961B} Bessell, M.~S.\ 2000, \pasp, 

1127: 112, 961 

1128: 

1129: \bibitem[Bessell \& Brett(1988)]{1988PASP..100.1134B} Bessell, M.~S., \& 

1130: Brett, J.~M.\ 1988, \pasp, 100, 1134

1131: 

1132: \bibitem[Bevington \& Robinson (1992)]{Bev} Bevington, P.\ R. \& Robinson, 

1133: D.\ K. 1992, Data Reduction and Error Analysis for the Physical Sciences, 

1134: (New York: McGraw Hill)

1135: 

1136: \bibitem[Bodenheimer et al.(2000)]{2000Icar..143....2B} Bodenheimer, P., 

1137: Hubickyj, O., \& Lissauer, J.~J.\ 2000, Icarus, 143, 2

1138: 

1139: \bibitem[Bodenheimer et al.(2001)]{2001ApJ...548..466B} Bodenheimer, P., 

1140: Lin, D.~N.~C., \& Mardling, R.~A.\ 2001, \apj, 548, 466 

1141: 

1142: \bibitem[Bodenheimer et al.(2003)]{2003ApJ...592..555B} Bodenheimer, P.,

1143: Laughlin, G., \& Lin, D.~N.~C.\ 2003, \apj, 592, 555

1144: 

1145: \bibitem[Boesgaard et al.(1988)]{1988ApJ...327...389B} Boesgaard, A.~M.,

1146: Budge, K.~G., \& Ramsay, M.~E.\ 1988, \apj, 327, 389

1147: 

1148: \bibitem[Boss(1997)]{1997Sci...276.1836B} Boss, A.~P.\ 1997, Science, 276, 

1149: 1836

1150: 

1151: \bibitem[Boss(2000)]{2000ApJ...536L.101B} Boss, A.~P.\ 2000, \apjl, 536, 

1152: L101

1153: 

1154: \bibitem[Boss(2002)]{2002ApJ...567L.149B} Boss, A.~P.\ 2002, \apjl, 567, 

1155: L149 

1156: 

1157: \bibitem[Bouvier et al.(2007)]{2007A&A...463.1017B} Bouvier, J., et al.\ 

1158: 2007, \aap, 463, 1017

1159: 

1160: \bibitem[Burke et al.(2007)]{2007arXiv0705.0003B} Burke, C.~J., et al.\ 

1161: 2007, ArXiv e-prints, 705, arXiv:0705.0003

1162: 

1163: \bibitem[Burrows et al.(2001)]{2001RvMP...73..719B} Burrows, A., Hubbard, 

1164: W.~B., Lunine, J.~I., \& Liebert, J.\ 2001, Reviews of Modern Physics, 73, 

1165: 719

1166: 

1167: \bibitem[Butler et al.(1997)]{1997ApJ...474L.115B} Butler, R.~P., Marcy, 

1168: G.~W., Williams, E., Hauser, H., \& Shirts, P.\ 1997, \apjl, 474, L115

1169: 

1170: \bibitem[Butler et al.(1996)]{1996PASP..108..500B} Butler, R.~P., Marcy, 

1171: G.~W., Williams, E., McCarthy, C., Dosanjh, P., \& Vogt, S.~S.\ 1996, 

1172: \pasp, 108, 500

1173: 

1174: \bibitem[Butler et al.(2004)]{2004ApJ...617..580B} Butler, R.~P., Vogt, 

1175: S.~S., Marcy, G.~W., Fischer, D.~A., Wright, J.~T., Henry, G.~W., Laughlin, 

1176: G., \& Lissauer, J.~J.\ 2004, \apj, 617, 580

1177: 

1178: \bibitem[Claret(2000)]{2000A&A...363.1081C} Claret, A.\ 2000, \aap, 363, 

1179: 1081 

1180: 

1181: \bibitem{Cochran2000} Cochran, W. 2000, FTS spectrum of I2 Cell HRS3 at 69.9 C.,

1182: ftp://nsokp.nso.edu/FTS\_cdrom/FTS50/001023R0.004

1183: 

1184: \bibitem[Cox(2000)]{2000asqu.book.....C} Cox, A.~N.\ 2000, Allen's 

1185: astrophysical quantities, 4th ed.~Publisher: New York: AIP Press; Springer, 

1186: 2000.~Editedy by Arthur N.~Cox.~ ISBN: 0387987460

1187: 

1188: \bibitem[Cumming et al.(1999)]{1999ApJ...526..890C} Cumming, A., Marcy, 

1189: G.~W., \& Butler, R.~P.\ 1999, \apj, 526, 890

1190: 

1191: \bibitem[Demarque et al.(2004)]{2004ApJS..155..667D} Demarque, P., Woo, 

1192: J.-H., Kim, Y.-C., \& Yi, S.~K.\ 2004, \apjs, 155, 667 

1193: 

1194: \bibitem[Deming et al.(2007)]{2007ApJ...667L.199D} Deming, D., Harrington, 

1195: J., Laughlin, G., Seager, S., Navarro, S.~B., Bowman, W.~C., \& Horning, 

1196: K.\ 2007, \apjl, 667, L199

1197: 

1198: \bibitem[Eggenberger et al.(2004)]{2004A&A...417..353E} Eggenberger, A., 

1199: Udry, S., \& Mayor, M.\ 2004, \aap, 417, 353

1200: 

1201: \bibitem[Fischer \& Valenti(2005)]{2005ApJ...622.1102F} Fischer, D.~A., \& 

1202: Valenti, J.\ 2005, \apj, 622, 1102 

1203: 

1204: \bibitem[Ford \& Rasio(2007)]{2007astro.ph..3163F} Ford, E.~B., \& Rasio, 

1205: F.~A.\ 2007, ArXiv Astrophysics e-prints, arXiv:astro-ph/0703163

1206: 

1207: \bibitem[Forrest et al.(2004)]{2004ApJS..154..443F} Forrest, W.~J., et al.\ 

1208: 2004, \apjs, 154, 443

1209: 

1210: \bibitem[Fortney et al.(2007)]{2007ApJ...659.1661F} Fortney, J.~J., Marley, 

1211: M.~S., \& Barnes, J.~W.\ 2007, \apj, 659, 1661

1212: 

1213: \bibitem[Gonzalez(1997)]{1997MNRAS.285..403G} Gonzalez, G.\ 1997, \mnras, 

1214: 285, 403

1215: 

1216: \bibitem[Gu et al.(2003)]{2003ApJ...588..509G} Gu, P.-G., Lin, D.~N.~C., \& 

1217: Bodenheimer, P.~H.\ 2003, \apj, 588, 509

1218: 

1219: \bibitem[Haisch et al.(2001)]{2001ApJ...553L.153H} Haisch, K.~E., Jr., 

1220: Lada, E.~A., \& Lada, C.~J.\ 2001, \apjl, 553, L153

1221: 

1222: \bibitem[Halbwachs et al.(2005)]{2005A&A...431.1129H} Halbwachs, J.~L., 

1223: Mayor, M., \& Udry, S.\ 2005, \aap, 431, 1129

1224: 

1225: \bibitem[Hale(1994)]{hale94} Hale, A. 1994, AJ, 107, 306

1226: 

1227: \bibitem[Hatzes et al.(1997)]{1997ApJ...478..374H} Hatzes, A.~P., Cochran, 

1228: W.~D., \& Johns-Krull, C.~M.\ 1997, \apj, 478, 374

1229: 

1230: \bibitem[H{\o}g et al.(2000)]{2000A&A...355L..27H} H{\o}g, E., et al.\ 

1231: 2000, \aap, 355, L27 

1232: 

1233: \bibitem[Hilditch(2001)]{2001icbs.book.....H} Hilditch, R.~W.\ 2001, An 

1234: Introduction to Close Binary Stars, by R.~W.~Hilditch, pp.~392.~ISBN 

1235: 0521241065.~Cambridge, UK: Cambridge University Press

1236: 

1237: \bibitem[Hillenbrand \& White(2004)]{2004ApJ...604..741H} Hillenbrand, 

1238: L.~A., \& White, R.~J.\ 2004, \apj, 604, 741

1239: 

1240: \bibitem[Hinkle et al.(2000)]{2000SPIE.4008..720H} Hinkle, K.~H., Joyce, 

1241: R.~R., Sharp, N., \& Valenti, J.~A.\ 2000, \procspie, 4008, 720 

1242: 

1243: \bibitem[Hubickyj et al.(2005)]{2005Icar..179..415H} Hubickyj, O., 

1244: Bodenheimer, P., \& Lissauer, J.~J.\ 2005, Icarus, 179, 415

1245: 

1246: \bibitem[Huerta (2007)]{2007PhDThesis} Huerta, M.\ 2007, PhD Thesis, Rice

1247: University

1248: 

1249: \bibitem[Huerta et al.(2007)]{2007ApJ...sub} Huerta, M., Johns--Krull, C.M.,

1250: Hartigan, P., Prato, L.A., \& Jaffe, D.\ 2007, \apj, submitted

1251: 

1252: \bibitem[Inaba et al.(2003)]{2003Icar..166...46I} Inaba, S., Wetherill, 

1253: G.~W., \& Ikoma, M.\ 2003, Icarus, 166, 46

1254: 

1255: \bibitem[Ivanov \& Papaloizou(2007)]{2007MNRAS.376..682I} Ivanov, P.~B., \& 

1256: Papaloizou, J.~C.~B.\ 2007, \mnras, 376, 682

1257: 

1258: \bibitem[Jones(2004)]{2004AIPC..713...17J} Jones, H.~R.~A.\ 2004, AIP 

1259: Conf.~Proc.~713: The Search for Other Worlds, 713, 17

1260: 

1261: \bibitem[Landolt(1992)]{1992AJ....104..340L} Landolt, A.~U.\ 1992, \aj, 

1262: 104, 340 

1263:  

1264: \bibitem[Livio \& Pringle(2003)]{2003MNRAS.346L..42L} Livio, M., \& 

1265: Pringle, J.~E.\ 2003, \mnras, 346, L42

1266: 

1267: \bibitem[Mamajek et al.(2004)]{2004ApJ...612..496M} Mamajek, E.~E., Meyer, 

1268: M.~R., Hinz, P.~M., Hoffmann, W.~F., Cohen, M., \& Hora, J.~L.\ 2004, \apj, 

1269: 612, 496

1270: 

1271: \bibitem[Mandel \& Agol(2002)]{2002ApJ...580L.171M} Mandel, K., \& Agol, 

1272: E.\ 2002, \apjl, 580, L171 

1273:  

1274: \bibitem[Mandushev et al.(2005)]{2005ApJ...621.1061M} Mandushev, G., et 

1275: al.\ 2005, \apj, 621, 1061 

1276: 

1277: \bibitem[Mandushev et al.(2007)]{2007arXiv0708.0834M} Mandushev, G., et 

1278: al.\ 2007, ArXiv e-prints, 708, arXiv:0708.0834

1279: 

1280: \bibitem[Mart\'inez et al.(2005)]{2005A&A...442..775} Mart\'inez Fiorenzano, 

1281: A.~F., Gratton, R.~G., Desidera, S., Cosentino, R., \& Endl, M. 2005,

1282: A\&A, 442, 775

1283: 

1284: \bibitem[Mathieu(1994)]{1994ARAA...32...465} Mathieu, R.D. 1994, ARA\&A, 32,

1285: 465

1286: 

1287: \bibitem[Mayer et al.(2002)]{2002Sci...298.1756M} Mayer, L., Quinn, T., 

1288: Wadsley, J., \& Stadel, J.\ 2002, Science, 298, 1756

1289: 

1290: \bibitem[McCullough \& Burke(2007)]{2007ASPC..366...70M} McCullough, P.~R., 

1291: \& Burke, C.~J.\ 2007, Transiting Extrapolar Planets Workshop, 366, 70

1292: 

1293: \bibitem[McCullough et al.(2005)]{2005PASP..117..783M} McCullough, P.~R., 

1294: Stys, J.~E., Valenti, J.~A., Fleming, S.~W., Janes, K.~A., \& Heasley, 

1295: J.~N.\ 2005, \pasp, 117, 783 

1296: 

1297: \bibitem[McCullough et al.(2006)]{MCC06} McCullough, P.~R., 

1298: et al.\ 2006, \apj, 648, 1228

1299: 

1300: \bibitem[Murray et al.(1998)]{1998Sci...279...69M} Murray, N., Hansen, B., 

1301: Holman, M., \& Tremaine, S.\ 1998, Science, 279, 69

1302: 

1303: \bibitem[Neuh{\"a}user et al.(2005)]{2005A&A...435L..13N} Neuh{\"a}user, 

1304: R., Guenther, E.~W., Wuchterl, G., Mugrauer, M., Bedalov, A., \& 

1305: Hauschildt, P.~H.\ 2005, \aap, 435, L13

1306: 

1307: \bibitem[Nidever et al.(2002)]{2002ApJS..141..503N} Nidever, D.~L., Marcy, 

1308: G.~W., Butler, R.~P., Fischer, D.~A., \& Vogt, S.~S.\ 2002, \apjs, 141, 503

1309: 

1310: \bibitem[O'Donovan et al.(2006)]{2006...ApJ...651L.61} O'Donovan, F.~T., et al.

1311: 2006, \apjl, 651, L61

1312: 

1313: \bibitem[O'Donovan et al.(2006)]{2006ApJ...644.1237O} O'Donovan, F.~T., et 

1314: al.\ 2006, \apj, 644, 1237

1315: 

1316: \bibitem[Papaloizou et al.(2007)]{2007prpl.conf..655P} Papaloizou, 

1317: J.~C.~B., Nelson, R.~P., Kley, W., Masset, F.~S., \& Artymowicz, P.\ 2007, 

1318: Protostars and Planets V, 655

1319: 

1320: \bibitem[Pollack et al.(1996)]{1996Icar..124...62P} Pollack, J.~B., 

1321: Hubickyj, O., Bodenheimer, P., Lissauer, J.~J., Podolak, M., \& Greenzweig, 

1322: Y.\ 1996, Icarus, 124, 62

1323: 

1324: \bibitem[Queloz et al.(2001)]{2001A&A...379..279} Queloz, D., Henry, G.~W.,

1325: Sivan, J.~P., Baliunas, S.~L., Beuzit, J.~L., Donahue, R.~A., Mayor, M., 

1326: Naef, D., Perrier, C., \& Udry, S. 2001, A\&A, 379, 279

1327: 

1328: \bibitem[Ribas \& Miralda-Escud{\'e}(2007)]{2007A&A...464..779R} Ribas, I., 

1329: \& Miralda-Escud{\'e}, J.\ 2007, \aap, 464, 779

1330: 

1331: \bibitem[Santos et al.(2002)]{2002A&A...392..215S} Santos, N.~C., et al.\ 

1332: 2002, \aap, 392, 215

1333: 

1334: \bibitem[Santos et al.(2003)]{2003csss...12..148S} Santos, N.~C., 

1335: Israelian, G., \& Mayor, M.\ 2003, The Future of Cool-Star Astrophysics: 

1336: 12th Cambridge Workshop on Cool Stars , Stellar Systems, and the Sun (2001 

1337: July 30 - August 3), eds.~A.~Brown, G.M.~Harper, and T.R.~Ayres, 

1338: (University of Colorado), 2003, p.~148-157., 12, 148

1339: 

1340: \bibitem[Santos et al.(2004)]{2004A&A...415.1153S} Santos, N.~C., 

1341: Israelian, G., \& Mayor, M.\ 2004, \aap, 415, 1153

1342: 

1343: \bibitem[Sato et al.(2005)]{2005ApJ...633..465S} Sato, B., et al.\ 2005, 

1344: \apj, 633, 465 

1345: 

1346: \bibitem[Sozzetti(2004)]{2004MNRAS.354.1194S} Sozzetti, A.\ 2004, \mnras, 

1347: 354, 1194

1348: 

1349: \bibitem[Sozzetti et al.(2004)]{2004ApJ...616L..167} Sozzetti et al. 2004,

1350: \apjl, 616, L167

1351: 

1352: \bibitem[Sozzetti et al.(2007)]{2007ApJ...664.1190S} Sozzetti, A., Torres, 

1353: G., Charbonneau, D., Latham, D.~W., Holman, M.~J., Winn, J.~N., Laird, 

1354: J.~B., \& O'Donovan, F.~T.\ 2007, \apj, 664, 1190

1355: 

1356: \bibitem[Torres et al.(2004)]{2004ApJ...609.1071T} Torres, G., Konacki, M., 

1357: Sasselov, D.~D., \& Jha, S.\ 2004, \apj, 609, 1071

1358: 

1359: \bibitem[Tull(1998)]{1998SPIE.3355..387T} Tull, R.~G.\ 1998, \procspie, 

1360: 3355, 387 

1361: 

1362: \bibitem[Tull et al.(1995)]{1995PASP..107..251T} Tull, R.~G., MacQueen, 

1363: P.~J., Sneden, C., \& Lambert, D.~L.\ 1995, \pasp, 107, 251 

1364: 

1365: \bibitem[Udry et al.(2007)]{2007prpl.conf..685U} Udry, S., Fischer, D., \& 

1366: Queloz, D.\ 2007, Protostars and Planets V, 685

1367: 

1368: \bibitem[Udry et al.(2002)]{2002A&A...390..267U} Udry, S., Mayor, M., Naef, 

1369: D., Pepe, F., Queloz, D., Santos, N.~C., \& Burnet, M.\ 2002, \aap, 390, 

1370: 267

1371: 

1372: \bibitem[Udry et al.(2003)]{2003A&A...407..369U} Udry, S., Mayor, M., \& 

1373: Santos, N.~C.\ 2003, \aap, 407, 369

1374: 

1375: \bibitem[Valenti \& Fischer(2005)]{2005ApJS..159..141V} Valenti, J.~A., \& 

1376: Fischer, D.~A.\ 2005, \apjs, 159, 141 (VF05)

1377: 

1378: \bibitem[Valenti \& Piskunov(1996)]{1996A&AS..118..595V} Valenti, J.~A., \& 

1379: Piskunov, N.\ 1996, \aaps, 118, 595 

1380: 

1381: \bibitem[VandenBerg \& Clem(2003)]{2003AJ....126..778V} VandenBerg, D.~A., 

1382: \& Clem, J.~L.\ 2003, \aj, 126, 778 

1383: 

1384: \bibitem[Wallace et al.(1998)]{1998assp.book.....W} Wallace, L., Hinkle, 

1385: K., \& Livingston, W.\ 1998, An atlas of the spectrum of the solar 

1386: photosphere from 13,500 to 28,000 cm-1 (3570 to 7405 A), Publisher: Tucson, 

1387: AZ: National Optical Astronomy Observatories

1388: 

1389: \bibitem[Wu(2003)]{2003ASPC..294..213W} Wu, Y.\ 2003, Scientific Frontiers 

1390: in Research on Extrasolar Planets, D. Deming \& S. Seager (eds.), ASP Conf.

1391: Series, 294, 213

1392: 

1393: \bibitem[Zucker \& Mazeh(2002)]{2002ApJ...568L.113Z} Zucker, S., \& Mazeh, 

1394: T.\ 2002, \apjl, 568, L113

1395: 

1396: \end{thebibliography}

1397: 

1398: \clearpage

1399: 

1400: \begin{deluxetable}{cccccc}

1401: \tabletypesize{\small}

1402: \tablewidth{0pt}

1403: \tablecaption{{\rm XO Survey \& E.T.\ Light Curve Data}\tablenotemark{a}}

1404: \startdata

1405: \hline

1406: \hline

1407: Heliocentric Julian Date & Light Curve & Uncertainty & Filter & N\tablenotemark{b} & Observatory \\

1408:                          &  [mag]      & (1-$\sigma$) [mag] & & & \\

1409: \hline

1410: 2452932.10913 & -0.0017 & 0.0027  &   W\tablenotemark{c} &  1 &  XO \\

1411: 2452932.10938 & -0.0032 & 0.0026  &   W\tablenotemark{c} &  1 &  XO \\

1412: 2452932.11597 &  0.0024 & 0.0026  &   W\tablenotemark{c} &  1 &  XO \\

1413: 2452932.11621 &  0.0021 & 0.0025  &   W\tablenotemark{c} &  1 &  XO \\

1414: 2452932.12305 & -0.0016 & 0.0026  &   W\tablenotemark{c} &  1 &  XO \\

1415: \hline

1416: \enddata

1417: \tablenotetext{a}{The complete version of this table is in the electronic edition of

1418: the Journal.  The printed edition contains only a sample.}

1419: \tablenotetext{b}{Average of N measurements}

1420: \tablenotetext{c}{The filters used in the XO project telescopes are

1421: described in McCullough et al. (2005).  The bandpass is essentially

1422: flat from 4000 -- 7000 \AA.}

1423: \label{table:lc}

1424: \end{deluxetable}

1425: 

1426: \begin{deluxetable}{ccccc}

1427: \tabletypesize{\small}

1428: \tablewidth{0pt}

1429: \tablecaption{{\rm All-sky Photometric Magnitudes}}

1430: \startdata

1431: \hline

1432: \hline

1433: Star\tablenotemark{a} & B     & V     & ${\rm R_C}$ & ${\rm I_C}$ 	\\

1434: \hline

1435: \xon\ 	       & 10.25 &  9.80 &  9.54       &  9.28 \\

1436:    1           & 11.37 & 10.82 & 10.50       & 10.18 \\

1437:    2           & 12.38 & 11.62 & 11.17       & 10.76 \\

1438:    3           & 14.81 & 13.07 & 12.17       & 11.30 \\

1439:    4           & 10.13 &  9.03 &  8.43       &  7.88 \\

1440:    5           & 12.67 & 11.87 & 11.43       & 10.92 \\

1441:    6           &  9.26 &  8.78 &  8.50       &  8.21 \\

1442:    7           & 15.52 & 13.46 & 12.41       & 11.50 \\

1443:    8           & 15.34 & 13.84 & 12.59       & 11.51 \\

1444: \enddata

1445: \tablenotetext{a}{Stars are identified in Figure \ref{fig:allsky}.}

1446: \label{table:allsky}

1447: \end{deluxetable}

1448: 

1449: \begin{deluxetable}{lcl}

1450: \tabletypesize{\small}

1451: \tablewidth{0pt}

1452: \tablecaption{{\rm The Star \xon}}

1453: \startdata

1454: \hline

1455: \hline

1456: Parameter & Value & Reference\\

1457: \hline

1458: RA (J2000.0) & $ 04^h21^m52^s.71 $ & a,b \\

1459: Dec (J2000.0) & +57\arcdeg49\arcmin01\arcsec.9 & a,b \\

1460: $V, V_T, V_{calc}$ & 9.80$\pm$0.03,$9.904\pm0.027$,$9.86\pm0.027$ & c,b,d\\

1461: $B-V, B_T-V_T, (B-V)_{calc}$ & 0.45$\pm$0.04,$0.451\pm0.040$,$0.383\pm0.034$ & c,b,d\\

1462: $V-R_C$ & 0.26$\pm$0.04 & c\\

1463: $R_C-I_C$ & 0.26$\pm$0.06,& c\\

1464: $J$   & $9.013\pm0.029$ & e\\

1465: $J-H$ & $0.168\pm0.034$ & e\\

1466: $H-K_{\rm s}$ & $0.054\pm0.026$ & e\\

1467: Spectral Type & \sptype & c \\

1468: d & $\vDs \pm \eDs$ pc & c \\

1469: $(\mu_\alpha,\mu_\delta)$ & ${\rm (-0.6 \pm 2.7, 1.6 \pm 2.6)~mas~yr^{-1}}$ & b \\

1470: GSC 	& 03727-01064 & a \\

1471: \enddata

1472: \tablerefs{\\

1473: a) SIMBAD\\

1474: b) Tycho-2 Catalogue, H{\o}g et al (2000) \\

1475: c) this work \\

1476: d) Calculated from Tycho-2 measurements using standard transformations\\

1477: e) 2MASS, Skrutskie e al. (2006)

1478: }

1479: \label{table:star}

1480: \end{deluxetable}

1481:                                                                                             

1482: \begin{deluxetable}{cccccc}

1483: \tabletypesize{\small}

1484: \tablewidth{0pt}

1485: \tablecaption{{\rm Radial Velocity Shifts}}

1486: \startdata

1487: \hline

1488: \hline

1489: Heliocentric& Radial Velocity &  Uncertainty & $O-C$ & Bisector & \\

1490: Julian Date &  Shift [\mps] &  (1 $\sigma$) [\mps]  & [\mps] & Span [\mps] & Telescope \\

1491: \hline

1492:     2454037.0112 &     1265 &    209 &  107 & -31 & HJS \\

1493:     2454038.0098 &     2001 &    216 &  174 &   4 & HJS \\

1494:     2454137.8215 &        0 &    134 &  -92 &   0 & HJS \\

1495:     2454138.8124 &      216 &    134 & -112 & 170 & HJS \\

1496:     2454139.8128 &     2625 &    139 & -117 & 133 & HJS \\

1497:     2454140.8131 &      267 &    139 &  -47 & 203 & HJS \\

1498:     2454141.8047 &       77 &    149 &   13 & 112 & HJS \\

1499:     2454142.7967 &     2761 &    139 &   -1 &  68 & HJS \\

1500:     2454143.7999 &      881 &    143 &  242 &  42 & HJS \\

1501:     2454144.7967 &      -35 &    161 &   34 & 152 & HJS \\

1502: \\

1503:     2454005.8587 &     1454 &    129 &  103 & \nodata & HET \\

1504:     2454006.8597 &     -976 &    122 &   16 & \nodata & HET \\

1505:     2454113.5842 &      -71 &    184 &   44 & \nodata & HET \\

1506:     2454121.7100 &     -981 &    188 &  -43 & \nodata & HET \\

1507:     2454121.7190 &     -965 &    204 &  -16 & \nodata & HET \\

1508:     2454122.5635 &    -1152 &    182 &   52 & \nodata & HET \\

1509:     2454127.7046 &     -240 &    195 &   32 & \nodata & HET \\

1510:     2454128.7048 &    -1271 &    183 &   21 & \nodata & HET \\

1511:     2454158.6220 &     1191 &    144 &  -32 & \nodata & HET \\

1512:     2454159.6091 &     -278 &    132 &  -31 & \nodata & HET \\

1513:     2454162.6193 &      -30 &    194 & -248 & \nodata & HET \\

1514: \enddata

1515: \label{table:rvobs}

1516: \end{deluxetable}

1517: 

1518: \begin{deluxetable}{lccc}

1519: \tabletypesize{\small}

1520: \tablewidth{0pt}

1521: \tablecaption{{\rm Transit Timing Measurements}}

1522: \startdata

1523: \hline

1524: \hline

1525:   &  &  & Transit Midpoint  \\

1526:  UT Date & Observer & Filter & (HJD - 2450000.0)  \\

1527: \hline

1528: 2003 Dec 24 & XO & W\tablenotemark{a} & 2997.72729 \\

1529: 2004 Oct 22 & XO & W\tablenotemark{a} & 3300.90479 \\

1530: 2004 Nov 10 & XO & W\tablenotemark{a} & 3320.05713 \\

1531: 2004 Dec 12 & XO & W\tablenotemark{a} & 3351.99341 \\

1532: 2006 Sep 21 & CF & R & 3999.86011 \\

1533: 2006 Sep 21 & MF & R & 3999.85913 \\

1534: 2006 Oct 16 & TV & R & 4025.39673 \\

1535: 2006 Oct 20 & EM & V & 4028.58984 \\

1536: 2006 Oct 23 & CF & B & 4031.77759 \\

1537: 2006 Oct 23 & CF & R & 4031.77954 \\

1538: 2006 Oct 23 & CF & V & 4031.78052 \\

1539: 2006 Nov 08 & CF & B & 4047.73022 \\

1540: 2006 Nov 08 & CF & R & 4047.73804 \\

1541: 2006 Nov 08 & CF & V & 4047.73340 \\

1542: 2006 Nov 24 & EM & I & 4063.69019 \\

1543: 2006 Nov 24 & MF & I & 4063.69897 \\

1544: 2006 Nov 27 & MF & I & 4066.88525 \\

1545: 2007 Jan 26 & EM & I & 4127.53076 \\

1546: 2007 Feb 15 & CF & V & 4146.67725 \\

1547: 2007 Mar 19 & MF & R & 4178.58936 \\

1548: \enddata

1549: \tablenotetext{a}{The filters used in the XO project telescopes are

1550: described in McCullough et al. (2005).  The bandpass is essentially

1551: flat from 4000 -- 7000 \AA.}

1552: \label{table:ttime}

1553: \end{deluxetable}

1554: 

1555: \begin{deluxetable}{lcl}

1556: \tabletypesize{\small}

1557: \tablewidth{0pt}

1558: \tablecaption{{\rm Orbital Solution and Planetary Parameters \xonb}}

1559: \startdata

1560: \hline

1561: \hline

1562: Parameter & Value & Notes\\

1563: \hline

1564: $P $ 		& \vperiod$\pm$\eperiod\  d                     & \\

1565: $t_c $	 	& \vjd$\pm$\ejd\          HJD                   & \\

1566: $e$             & \vecc$\pm$\eecc                               & \\

1567: $\omega$        & $-15.4\pm6.6$          degrees               & \\

1568: $T_\circ$       & 2454024.7278$\pm$0.0570  HJD                  & \\

1569: $K $ 		& \vrvK$\pm\ervK$\        \mps                  & \\

1570: $M_{\rm p}{\rm sin}i$   & \vMp$\pm$\eMp                         & b\\

1571: $R_{\rm p}/R_{\rm s} $ 	& (0.92$\pm$0.04) $\times$ \Rjup/\Rsun 	& a\\

1572: $i $ 			& \vincl$\pm$\eincl\ deg		& b,c\\

1573: $a $ 			& \vap$\pm$\eap\ A.U. 			& b \\

1574: $M_{\rm p} $ 		& \vMptot$\pm$\eMptot\ \Mjup		 	& b,d\\

1575: $R_{\rm p} $ 		& \vRp$\pm$\eRp\ \Rjup			& a,b,c\\

1576: \enddata

1577: \tablecomments{\\

1578: a) \Rjup\ = 71492 km, i.e. the equatorial radius of Jupiter\\

1579: b) for $M_*$ = \vMs$\pm$\eMs\ \Msun \\

1580: c) for $R_*$ = \vRs$\pm$\eRs\ \Rsun \\

1581: d) \Mjup\ = 1.8988e27 kg 

1582: }

1583: \label{table:planet}

1584: \end{deluxetable}

1585: 

1586: 

1587: \begin{deluxetable}{ccc}

1588: \tabletypesize{\small}

1589: \tablewidth{0pt}

1590: \tablecaption{{\rm Results of the SME Analysis}}

1591: \startdata

1592: \hline

1593: \hline

1594:           &      &  Precision \\

1595: Parameter & Mean &  (1 $\sigma$)	\\

1596: \hline

1597: $T_{eff} [K]$	&6429 	&   50		\\

1598: log$g$ [\cmpss]	&3.95  	&   0.062	\\

1599: V~sin~$i$ [\kps] &18.54 &   0.17	\\

1600: \ [M/H]     	&-0.204	&   0.023	\\

1601: \ [Na/H]	&-0.346	&   0.046	\\

1602: \ [Si/H]	&-0.171	&   0.017	\\

1603: \ [Ti/H]	&-0.262	&   0.049	\\

1604: \ [Fe/H]	&-0.177 &   0.027	\\

1605: \ [Ni/H]	&-0.220	&   0.033	\\

1606: \ [Si/Fe]   	&0.006	&   0.032	\\

1607: \enddata

1608: \label{table:sme}

1609: \end{deluxetable}

1610: 

1611: \begin{deluxetable}{lccc}

1612: \tabletypesize{\small}

1613: \tablewidth{0pt}

1614: \tablecaption{{\rm Spectroscopically Derived Stellar parameters}}

1615: \startdata

1616: \hline

1617: \hline

1618: Parameter	&	@ 240 pc&	@ 260 pc&	@ 280 pc\\

1619: \hline

1620:    		&	1.33	&	1.36	&	1.39	\\

1621: Mass [\Msun]   	&	1.36	&	1.41	&	1.43	\\

1622:     		&	1.39	&	1.44	&	1.48	\\

1623:     		&	    	&	    	&	    	\\

1624:     		&	1.92	&	2.08	&	2.22	\\

1625: Radius [\Rsun] 	&	1.95	&	2.13	&	2.27	\\

1626:     		&	1.99	&	2.17	&	2.34	\\

1627:     		&	    	&	    	&	    	\\

1628:     		&	3.98	&	3.93	&	3.87	\\

1629: Log(g) [\cmpss] &	4.00	&	3.95	&	3.89	\\

1630:     		&	4.02	&	3.97	&	3.91	\\

1631:     		&	    	&	    	&	    	\\

1632:     		&	2.66	&	2.53	&	2.41	\\

1633: Age [Gyr]     	&	2.78	&	2.69	&	2.68	\\

1634:     		&	3.06	&	2.83	&	2.96	\\

1635: \enddata

1636: \tablecomments{

1637: For each parameter, the middle row is the maximum likelihood value, and the

1638: values in the rows above and below span the 68\% likelihood of the probability

1639: distributions (cf. Figure \protect{\ref{fig:stellarparam}}). The three columns correspond to three assumed

1640: distances for \xon.}

1641: \label{table:stellarparam}

1642: \end{deluxetable}

1643: 

1644: \clearpage

1645: 

1646: \begin{figure}

1647: \plotone{f1.eps}

1648: \caption{\xon\ is centered, indicated by the two hash marks.

1649: Stars from Table \ref{table:allsky} are circled. 

1650: North is up; East to the left. The DSS image, digitized from a

1651: POSSII-F plate with a IIIaF emulsion and an RG610 filter,

1652: subtends 15\arcmin\ of declination.

1653: \label{fig:allsky}}

1654: \end{figure}

1655: 

1656: \begin{figure}

1657: \plotone{f2.eps}

1658: \caption{A total of 2969 individual observations of \xon\ by the two XO cameras

1659: over two seasons 2004 and 2005 are shown wrapped and phased according

1660: to the transit ephemeris and averaged

1661: in 0.01-day bins (line).  The top panel is the full lightcurve, and

1662: the bottom panel shows the region around phase 0 (the primary transit,

1663: marked with a vertical dashed line in the top panel)

1664: enlarged.

1665: From these data we identified the star

1666: as a candidate for more refined photometry with other telescopes at

1667: epochs of expected transits (Figure \ref{fig:etlc}).

1668: \label{fig:xolc}}

1669: \end{figure}

1670: 

1671: \begin{figure}

1672: \plotone{f3.ps}

1673: \caption{Time series photometry of \xon\ from 2005 -- 2007, with dates, 

1674: observers, and

1675: filters indicated. The observations have been averaged in 0.006-day bins.

1676: The figure is in color in the electronic edition.

1677: \label{fig:etlc}}

1678: \end{figure}

1679: 

1680: \clearpage

1681: \thispagestyle{empty}

1682: \setlength{\voffset}{-10mm}

1683: \begin{figure}

1684: \epsscale{0.75}

1685: \plotone{f4.eps}\\[5mm]

1686: \caption{Top: The measured radial velocity data are shown along with the best fit

1687: orbit.  The filled circles are based on data from the 2.7 m Harlan J. Smith

1688: Telescope, and the filled squares come from data taken with the 11-m HET.

1689: The radial velocity curve of \xon\ traces out an eccentric orbit

1690: with a velocity amplitude K = \vrvK$\pm$\ervK\ \mps, implying \xonb's mass is

1691: \mpsini $= \vMp\pm\eMp$\ \Mjup. The period and phase used to fold 

1692: the measured radial 

1693: velocities are fixed at values determined by the photometric transits.  The 

1694: measured center of mass velocity with respect to the solar system's barycenter

1695: has been subtracted (its value is arbitrary since it is largely based on the

1696: intrinsically relative radial velocity measurements from the HJS).

1697: Middle: A plot of the radial velocity residuals, again phase folded with

1698: the ephemeris determined from the photometric transits.

1699: Bottom:  The radial velocity residuals are shown again, this time as a 

1700: function of time, showing no long term trend.

1701: \label{fig:rvfit}}

1702: \end{figure}

1703: \clearpage

1704: \setlength{\voffset}{0mm}

1705: 

1706: \begin{figure}

1707: \plotone{f5.eps}

1708: \caption{The bisector span is plotted as a function of the measured

1709: radial velocity for all the HJS spectra of \xon.  There is no correlation

1710: of the span with radial velocity and the full amplitude of the span

1711: variations is less than a tenth of the full amplitude of the radial

1712: velocity variations.

1713: \label{fig:bisect}}

1714: \end{figure}

1715: 

1716: 

1717: \begin{figure}

1718: \epsscale{0.8}

1719: \plotone{f6.eps}

1720: \caption{

1721: The mean spectrum of \xon\ as observed (black

1722: histogram) and modeled with SME (curve, colored blue in the electronic edition)

1723: in the region of the Mg b triplet.

1724: Labels note the elements responsible for the indicated spectral lines.

1725: Intermittent line segments (tan) beneath the horizontal

1726: axis indicate wavelength intervals used to constrain the

1727: spectroscopic parameters. Very short and intermittent line segments (black)

1728: immediately above

1729: the spectrum indicate wavelength intervals used to constrain the

1730: continuum fit.  A color version of this figure appears in the electronic

1731: version of the paper.

1732: \label{fig:mcd}}

1733: \end{figure}

1734: 

1735: \begin{figure}

1736: \plotone{f7.eps}

1737: \caption{The stellar mass, radius, gravity, and age which result from the

1738: isochrone analysis.  The solid line shows the derived values as a function

1739: of the assumed distance, and the gray region shows the 68\% confidence

1740: limit on these parameters.  Dashed lines are drawn at 260 pc corresponding

1741: to the spectroscopically determined log$g = 3.95$.

1742: \label{fig:stellarparam}}

1743: \end{figure}

1744: 

1745: \begin{figure}

1746: \plotone{f8.eps}

1747: \caption{The spectrum of \xon\ in the wavelength region of the \ion{Li}{1}

1748: doublet at 6708 \AA.  The position of the Li feature is marked.  The Li

1749: feature is expected to be approximately as strong as the \ion{Ca}{1} line

1750: at 6718 \AA\ in Pleiades age stars of similar spectral type, and even

1751: stronger at younger ages.  Therefore, we conclude \xon\ is not a 

1752: pre-main sequence star, but instead is likely on the post-main sequence.

1753: \label{fig:licheck}}

1754: \end{figure}

1755: 

1756: \begin{figure}

1757: \plotone{f9.eps}

1758: \caption{The combined E.T. transit light curve of \xon\ and the best fit

1759: transit light curve models computed under different assumptions.  The upper

1760: panel (a) fixes the stellar mass and radius to the values determined from

1761: the combined spectroscopic and isochrone analysis.  The lower panel (b) is

1762: determined by minimizing $\chi^2$ in the light curve fit by letting the

1763: stellar mass and radius vary according to the isochrone results for different

1764: assumed distances.  This best fit produces a stellar gravity (log$g$) more 

1765: than $3\sigma$ greater than the measured value from the spectroscopic

1766: analysis.

1767: \label{fig:lcurvefit}}

1768: \end{figure}

1769: 

1770: \begin{figure}

1771: \plotone{f10.eps}

1772: \caption{Comparison of spectroscopically derived gravities to those derived

1773: from isochrone analysis for the sample of stars from VF05

1774: which have effective temperatures similar to \xon.  Stars with

1775: [M/H]$< 0.04$ are shown as blue squares and those with [M/H]$> 0.04$ are

1776: shown with red circles.  

1777: The green dashed line in Figure 10 is the line of equality.

1778: The color version of the figure appears in the

1779: electronic version of the paper.

1780: \label{fig:loggcheck}}

1781: \end{figure}

1782: 

1783: 

1784: \end{document}

1785: