1: 

2: % For ApJ submission -----------------------------------------

3: %\documentclass[12pt,preprint]{/home/cheongho/LatexStyle/AAS5.02/aastex}

4: %\documentclass[12pt, preprint]{aastex}

5: %\def\baselinestretch{1.3}

6: 

7: \documentclass{emulateapj}

8: \usepackage{apjfonts}

9: 

10: 

11: \lefthead{HAN} \righthead{SOURCE COMPANIONS AND OPTICAL DEPTH}

12: 

13: %==== CUSTOMIZED LATEX MACROS ========================================

14: % Italic bold font

15: \def\dslash{\mathbin{/\mkern-4mu/}}

16: 

17: \newcommand{\rvec}{\mbox{\boldmath $r$}}

18: \newcommand{\xvec}{\mbox{\boldmath $x$}}

19: \newcommand{\yvec}{\mbox{\boldmath $y$}}

20: \newcommand{\zvec}{\mbox{\bf z}}

21: \newcommand{\zetavec}{\mbox{\boldmath $\zeta$}}

22: \newcommand{\svec}{\bold s}

23: 

24: \newcommand{\te}{t_{\rm E}}

25: \newcommand{\re}{r_{\rm E}}

26: \newcommand{\rh}{r_{\rm H}}

27: \newcommand{\retilde}{\tilde{r}_{\rm E}}

28: \newcommand{\thetae}{\theta_{\rm E}}

29: 

30: 

31: % Equation align

32: \def\eqalign#1{\null\,\vcenter{\openup\jot

33:         \ialign{\strut\hfil$\displaystyle{##}$&$

34:         \displaystyle{{}##}$\hfil \crcr#1\crcr}}\,}

35: 

36: %=======================================================================

37: 

38: 

39: \begin{document}

40: \title{Effect of Binary Source Companions on the Microlensing Optical 

41: Depth Determination toward the Galactic Bulge Field}

42: 

43: 

44: \author{Cheongho Han}

45: \affil{Department of Physics, Institute for Basic Science

46: Research, Chungbuk National University, Chongju 361-763, Korea;

47: cheongho@astroph.chungbuk.ac.kr}

48: 

49: %\submitted{Submitted to The Astrophysical Journal}

50: 

51: 

52: 

53: \begin{abstract}

54: 

55: Currently, gravitational microlensing survey experiments toward the 

56: Galactic bulge field utilize two different methods of minimizing 

57: blending effect for the accurate determination of the optical depth 

58: $\tau$.  One is measuring $\tau$ based on clump giant (CG) source 

59: stars and the other is using `Difference Image Analysis (DIA)' 

60: photometry to measure the unblended source flux variation.  Despite 

61: the expectation that the two estimates should be the same assuming 

62: that blending is properly considered, the estimates based on CG stars

63: systematically fall below the DIA results based on all events with 

64: source stars down to the detection limit.  Prompted by the gap, we 

65: investigate the previously unconsidered effect of companion-associated 

66: events on $\tau$ determination.  Although the image of a companion 

67: is blended with that of its primary star and thus not resolved, the 

68: event associated with the companion can be detected if the companion 

69: flux is highly magnified.  Therefore, companions work effectively as 

70: source stars to microlensing and thus neglect of them in the source 

71: star count could result in wrong $\tau$ estimation.  By carrying out 

72: simulations based on the assumption that companions follow the same 

73: luminosity function of primary stars, we estimate that the contribution 

74: of the companion-associated events to the total event rate is

75: $\sim 5f_{\rm bi}\%$ for current surveys and can reach up to 

76: $\sim 6f_{\rm bi}\%$ for future surveys monitoring fainter stars, 

77: where $f_{\rm bi}$ is the binary frequency.  Therefore, we conclude 

78: that the companion-associated events comprise a non-negligible fraction 

79: of all events. However, their contribution to the optical depth is not 

80: large enough to explain the systematic difference between the optical 

81: depth estimates based on the two different methods.

82: 

83: \end{abstract}

84: 

85: \keywords{binaries: general -- gravitational lensing}

86: 

87: 

88: 

89: 

90: \section{Introduction}

91: 

92: The value of the microlensing optical depth toward the Galactic bulge 

93: field has been evolved with the refinement of both experimental 

94: measurements and theoretical predictions.  \citet{paczynski91} and 

95: \citet{griest91} first predicted the optical depth of $\tau\sim 0.4

96: -0.5 \times 10^{-6}$ assuming all events were caused by known disk 

97: stars.  However, the first measurement of the optical depth of 

98: $\tau=3.3 \times 10^{-6}$ reported by OGLE \citep{udalski94} was 

99: significantly higher than the earlier predictions.  This prompted 

100: \citet{kiraga94a} to evaluate the lens contribution of bulge stars 

101: in addition to disk stars.  Soon after, optical depths based on 

102: more refined models of the Galactic bulge were suggested by a number 

103: of authors \citep{kiraga94b, zhao95, han95, metcalf95, zhao96, 

104: bissantz97, peale98, evans02}.  The values of the new predictions  

105: were in the range of $\tau\sim 0.8-2.0\times 10^{-6}$, but these 

106: values still fall systematically below the measurements by OGLE and 

107: MACHO \citep{alcock97}.

108: 

109: 

110: To reconcile the predictions and measurements, various explanations were 

111: suggested.  A set of these explanations suggested new populations of 

112: events.  \citet{mollerach96} pointed out non-negligible contribution 

113: of disk self-lensing events.  \citet{nair99} indicated that disk source 

114: stars behind the bulge has much higher optical depth than bulge stars, 

115: and thus although they are a very small fraction of the stars in the 

116: Baade's window, they can contribute $\sim 5\%-10\%$ of the optical 

117: depth.  \citet{cseresnjes01} pointed out similar contribution of events 

118: associated with source stars belonging to the Sagittarius dwarf galaxy.  

119: \citet{sevenster01} mentioned the possibility of events caused by lenses 

120: in an inner ring located roughly halfway between the observer and bulge.

121: 

122: 

123: Another set of explanations pointed out potential biases in the optical 

124: depth measurements.  One of such biases is the the ``magnification-bias 

125: effect'' \citep{nemiroff94}, pointed out by \citet{alard97} and 

126: \citet{han97}.  The bulge is very crowded field and thus the images of 

127: stars suffer from severe blending.  For efficient processing of images 

128: taken toward such a crowded field, earlier lensing experiments registered 

129: bright resolved stars on a template image and monitored the brightness 

130: variation of only these stars.  For such experiments, although a source 

131: star is fainter than the detection limit imposed by crowding and thus 

132: not registered, it is still possible to detect an event if the star is 

133: located close to the seeing disk of a bright registered star. The 

134: optical depth is measured based on the number of monitored stars, 

135: $N_\star$, by

136: \begin{equation}

137: \tau={\pi\over 2N_\star T} \sum_{i=1}^{N_{\rm tot}}{t_{{\rm E},i} 

138: \over \epsilon(t_{{\rm E},i})},

139: \label{eq1}

140: \end{equation}

141: where $T$ is the total observation time, $N_{\rm tot}$ is the total 

142: number of detected events, $t_{{\rm E},i}$ are the Einstein timescales 

143: of the individual events, and $\epsilon(t_{{\rm E},i})$ is the detection 

144: efficiency of events as a function of $t_{\rm E}$.  Then, if these 

145: faint source stars are not taken into consideration, the number of 

146: source stars effectively monitored by the survey is systematically 

147: underestimated, and thus the optical depth is overestimated.

148: 

149: 

150: Observational efforts to refine the optical depth measurements were 

151: focused also on blending.  One simple but efficient method for the 

152: accurate determination of the optical depth was measuring $\tau$

153: based on events associated with only clump giant (CG) source stars, 

154: because these bright bulge stars are not strongly affected by blending.  

155: Four optical depth measurements based on CG-associated events were 

156: reported by \citet{popowski01}, \citet{afonso03}, \citet{popowski04}, 

157: and \citet{sumi05}.  These values range $\tau\sim 0.9-2.2\times 10^{-6}$, 

158: approaching close to the theoretical predictions.  Another effort of 

159: minimizing the blending effect was the adoption of a new photometry 

160: technique that measures the source star flux variation on images 

161: obtained by subtracting observed images from a seeing-convolved 

162: reference image \citep{tomaney96, alard98, alcock99, wozniak00, alard00, 

163: bond01}.  By suing the DIA method, one can improve the photometry of 

164: events because the measured flux variation is not affected by blending.

165: The optical depth measurements by using this so called ``Difference 

166: Image Analysis'' (DIA) method based on all events with source stars 

167: down to the detection limit were published by \citet{alcock00} and 

168: \citet{sumi03}.  These estimates range $\tau\sim 3.2-3.4 \times 10^{-6}$, 

169: and thus they are systematically greater than the estimates based on CG 

170: stars and the theoretical predictions.

171: 

172: 

173: Prompted by the difference between the optical depth estimates based 

174: on CG stars and the DIA result, we investigate the effect of unresolved 

175: binary companions to source stars on the microlensing optical depth 

176: determination.  Due to the characteristics of the DIA photometry, a 

177: significant fraction of events are associated with faint source stars

178: which could not be identified by the traditional DoPHOT \citep{schechter93}.  

179: Therefore, estimation of the optical depth based on the DIA photometry 

180: requires consideration of faint stars effectively working as source stars 

181: to lensing.  In the previous DIA estimates, this has been done based on 

182: a luminosity function (LF) obtained by using high-resolution instruments 

183: such as the {\it Hubble Space Telescope} ({\it HST}).  However, companions 

184: to source stars cannot be resolved even with the {\it HST}, and thus the 

185: contribution of the companions to the effective number of source stars 

186: has {\it not} been taken into consideration in the previous optical 

187: depth estimations.  Considering that majority of stars have companions 

188: \citep{abt76, abt83, duquennoy91}, the fraction of events associated 

189: with companion stars might not be negligible and could explain a 

190: significant fraction of the gap between the optical depth estimates 

191: based on CG stars and the DIA results.

192: 

193: 

194: 

195: \section{Companion-Associated Events}

196: 

197: Let us think about an event occurred on a companion star whose image is 

198: blended with that of its brighter primary star.  The separation between 

199: the primary and companion stars is, in general, much larger than the 

200: Einstein ring radius of a typical Galactic bulge event caused by a 

201: low-mass star of

202: \begin{equation}

203: r_{\rm E}\sim 1.9\ {\rm AU} \left( {M\over 0.3\ M_\odot}\right)^{1/2}

204: \left( {D_L\over 6\ {\rm kpc}}\right)^{1/2} 

205: \left( 1-{D_L\over D_S}\right)^{1/2},

206: \label{eq2}

207: \end{equation}

208: where $M$ is the mass of the lens, and $D_L$ and $D_S$ are the distances 

209: to the lens and source, respectively.  Then, the primary star, in most 

210: cases, is not participating in the lensing magnification and thus it 

211: simply works as a blending source \citep{dominik98, han98}.  In this 

212: case, the apparent magnification of the event is 

213: \begin{equation}

214: A_{\rm obs}(u) = {F_{\rm p}+A(u)F_{\rm c}\over F_{\rm p}+F_{\rm c}} 

215: = {F_{\rm p}/F_{\rm c}+A(u)\over F_{\rm p}/F_{\rm c} +1},

216: \label{eq3}

217: \end{equation}

218: where $A(u)$ is the true magnification of the event occurred on the 

219: companion star, $u$ is the lens-source (companion star) angular 

220: separation in units of the angular Einstein ring radius $\theta_{\rm E}

221: =r_{\rm E}/D_L$, and $F_{\rm p}$ and $F_{\rm c}$ are the fluxes of the 

222: primary and companion stars, respectively.  The true magnification is 

223: related to the normalized lens-source separation $u$ by

224: \begin{equation} A= {u^2+2 \over u(u^2+4)^{1/2}}.

225: \label{eq4}

226: \end{equation}

227: 

228: 

229: 

230: Although the image of a companion star is blended, the event associated 

231: with the companion star can be detected if its flux is magnified highly 

232: enough to make the apparent magnification of the combined flux of the

233:  primary and companion higher than a threshold value. Due to blending, 

234: the threshold magnification for the event detection becomes higher.  

235: The increase threshold magnification is  

236: \begin{equation}

237: {A'}_{\rm th}=A_{\rm th} \left( 1+{F_{\rm p}\over F_{\rm c}} \right)

238: -{F_{\rm p}\over F_{\rm c}},

239: \label{eq5}

240: \end{equation}

241: where $A_{\rm th}$ is the threshold magnification without blending.  Then, 

242: the threshold lens-source separation (impact parameter) corresponding to 

243: the increased threshold magnification is 

244: \begin{equation}

245: {u'}_{0,{\rm th}} = \left[ {2\over \left( 1-{A'}_{\rm th}^{-2} 

246: \right)^{1/2}}  -2 \right]^{1/2},

247: \label{eq6}

248: \end{equation}

249: implying that as long as the lens approaches the companion star closer 

250: than ${u'}_{0,{\rm th}}$, the event associated with the companion can 

251: be detected.  Therefore, companions can work effectively as source stars 

252: to lensing and thus the events associated with them should be taken into 

253: consideration in the optical depth determination.

254: 

255: 

256: 

257: \section{Companion Contribution to Source Stars}

258: 

259: Then, how much is the contribution of events associated with companion 

260: stars to the total optical depth, $\tau_{\rm c}$, toward the Galactic 

261: bulge field.  For this estimation, we compute the ratio of the number 

262: of events associated with companions, $N_{\rm c}$, to the total number 

263: of events, $N_{\rm tot}$.  The optical depth is related to the mass 

264: distribution along the line of sight toward the observation field, while 

265: the event rate is additionally dependent on the mass function of the lens 

266: matter.  However, since the lens does not discriminate whether the source 

267: star is primary or companion, the optical depth contribution of the 

268: companion-associated events is equivalent to their contribution to the 

269: total event rate, i.e.\ $\tau_{\rm c}/\tau=N_{\rm c}/N_{\rm tot}$.

270: 

271: 

272: 

273: % Figure 1 ------------------------------------------------------

274: \begin{figure}[tb]

275: \epsscale{1.15}

276: \plotone{f1.eps}

277: \caption{\label{fig:one}

278: Luminosity of function of source stars effectively participating in 

279: microlensing (effective luminosity function) toward the Galactic bulge 

280: field.  The dotted and and dashed curves represent the LFs of the 

281: primary and companion stars, respectively, and the solid curve is the 

282: combined LF.  The effective LF of the companion source stars is for 

283: the case of $f_{\rm bi}=1.0$, where $f_{\rm bi}$ is the binary frequency.  

284: The LFs are arbitrarily normalized.  The shaded regions represent the 

285: source stars brightness ranges of the current and future lensing surveys.

286: }\end{figure}

287: 

288: 

289: 

290: 

291: % Table 1 -------------------------------------------------------

292: \begin{deluxetable}{lcc}

293: \tabletypesize{\scriptsize}

294: \tablecaption{Contribution of Companion-Associated Events\label{table1}}

295: \tablewidth{0pt}

296: \tablehead{

297: %\multicolumn{1}{c}{source} &

298: %\multicolumn{3}{c}{detectability} \\

299: \colhead{surveys} &

300: \colhead{source brightness range}  &

301: \colhead{$\tau_{\rm c}/\tau$} 

302: }

303: \startdata

304: \smallskip

305: current-type        & $0.0\lesssim M_V\lesssim 6.0$  & $\sim 0.047f_{\rm bi}$ \\

306: future space-based  & $2.7\lesssim M_V\lesssim 7.5$  & $\sim 0.060f_{\rm bi}$ \\

307: \enddata

308: %\tablecomments{ 

309: %Table comments here.

310: %}

311: \end{deluxetable}

312: 

313: 

314: 

315: 

316: We estimate $\tau_{\rm c}/\tau$ as follows.  First, we assume that 

317: companions follow the same LF as that of primary stars \citep{duquennoy91}.  

318: With this assumption, we assign the brightnesses of the primary and 

319: companion source stars based on the binary-corrected $V$-band LF of 

320: stars in the solar neighborhood listed in \citet{allen00}.\footnote{For 

321: the source star brightness, it might be thought that using the LF of 

322: Galactic bulge stars from {\it HST} observations, e.g.\ \citet{holtzman98},  

323: would be a better choice.  However, this LF is not binary corrected 

324: because companions cannot be resolved even with the {\it HST}.  Since 

325: the purpose of this paper is investigating the effect of the source 

326: companion to the optical depth determination, we use the binary-corrected 

327: LF, although it is based on stars in the solar neighborhood. We note, 

328: however, that there is no significant evidence for the variation of 

329: the field of LF from place to place.} Once each pair of the primary 

330: and companion stars is produced, we then calculate the threshold 

331: magnification ${A'}_{\rm th}$ required for the event detection  and 

332: the corresponding threshold impact parameter ${u'}_{0,{\rm th}}$.  

333: The number of events having magnifications higher than $A'_{th}$ is 

334: directly proportional to $u'_{0,{\rm th}}$ because the distribution of 

335: events is uniform in impact parameter.  Then, the contribution of the 

336: companion-associated events to the total event rate is computed by 

337: \begin{equation}

338: {N_{\rm c}\over N_{\rm tot}} = {N_{\rm c}\over N_{\rm p}+N_{\rm c}},

339: \label{eq7}

340: \end{equation}

341: where 

342: \begin{equation}

343: N_{\rm p}=\sum_{i=1}^{N_{\star,{\rm p}}}u_{0,{\rm th,p}},

344: \label{eq8}

345: \end{equation}

346: \begin{equation}

347: N_{\rm c}=f_{\rm bi}\sum_{i=1}^{N_{\star,{\rm p}}}u_{0,{\rm th,c}},

348: \label{eq9}

349: \end{equation}

350: $N_{\star,{\rm p}}$ is the total number of monitored source primaries that 

351: would have produced events in the absence of binarity, $f_{\rm bi}$ is the 

352: binary frequency, and  $u_{0,{\rm th,p}}$ and $u_{0,{\rm th,c}}$ represent 

353: the threshold impact parameters of the events associated with the primary 

354: and companion, respectively. We note that the event associated with the 

355: primary star is also affected by blending caused by the flux from 

356: the companion star, and thus $u_{0,{\rm th,p}}\leq 1.0$ and 

357: $N_{\rm p}\leq N_{\star,{\rm p}}$.  For the computation of the 

358: threshold impact parameter, we assume that events are detected when 

359: $A_{\rm obs}\geq 3/\sqrt{5}$ adopting the conventional threshold 

360: magnification. We also assume that the combined (primary plus companion) 

361: brightness of the source star monitored by the current surveys is in 

362: the range of $0.0\lesssim M_V \lesssim 6.0$.  With the distance 

363: modulus of $V-M_V\sim 14.5$ and extinction of $A_V\sim 1.0$, this 

364: range corresponds to the apparent magnitude range of 

365: $15.5\lesssim V\lesssim 21.5$.  As the brightness of the monitoring 

366: source star decreases, the relative companion/primary flux ratio 

367: increases, and thus the rate of the companion-associated events 

368: increases.  The future space-based lensing survey using the 

369: {\it Galactic Exoplanet Survey Telescope} ({\it GEST}) mission 

370: proposed by \citet{bennett02} plans to monitor faint main-sequence 

371: stars to optimize the detections of terrestrial planets by minimizing 

372: finite-source effect.  To examine the dependence on the source star 

373: brightness, we also estimate the fraction of the companion-associated 

374: events from this future survey to examine the effect of source star 

375: brightness.  For this estimation, we assume that the range of the 

376: source star brightness is $2.7\lesssim M_V \lesssim 7.5$, which 

377: corresponds to early F to late K-type main-sequence stars.

378: 

379: 

380: In Tabel~\ref{table1}, we present the estimated contribution of the 

381: companion-associated events to the total event rate expected from 

382: the current and future lensing surveys.  In Figure~\ref{fig:one}, 

383: we also present the LFs of the primary and companion source stars 

384: effectively participating in lensing ({\it effective luminosity 

385: function}), where the companion LF is for the case $f_{\rm bi}=1.0$ 

386: to show the possible maximum contribution of the companion-associated 

387: events.  The result says that the contribution of the 

388: companion-associated events is $N_{\rm c}/N_{\rm tot}\sim 5f_{\rm bi}\%$ 

389: for the current lensing surveys and it is slightly higher values of 

390: $\sim 6f_{\rm bi}\%$ for the future lensing survey.  Therefore, 

391: these events do contribute to the total bulge event rate and thus 

392: to the optical depth.  However, the contribution is not large enough 

393: to explain an important fraction of the gap between the optical depth 

394: estimates based on CG stars and the DIA results, implying that exploring 

395: other possible reasons that can explain the gap is needed.

396: 

397: 

398: %Knowing that the contribution of the companion-associated events 

399: %alone cannot explain the the systematic discrepancy between the 

400: %$\tau$ estimates based on the two methods, then, what could be the 

401: %explanation for the difference.  Although not based on a convincing 

402: %evidence, we guess the contamination of the DIA event sample with 

403: %other forms of variables is a prime suspect \citep{belokurov03, 

404: %belokurov04}.  A substantial fraction of events detected by the DIA 

405: %method are associated with very faint stars and thus these events 

406: %suffer from poor photometry.  As a result, the chance of being 

407: %contaminated with non-microlensing events for events in this sample 

408: %would be higher than the CG-associated events, for which precision 

409: %photometry is possible due to the brightness of the source star.

410: 

411: 

412: 

413: \section{Conclusion}

414: 

415: Prompted by the gap between the microlensing optical depth estimates 

416: based on CG stars and the DIA result, we investigated the previously 

417: unconsidered effect of unresolved binary companions to source stars 

418: on the determination of the optical depth toward the Galactic bulge 

419: field.  By carrying out simulations based on the assumption that 

420: companions follow the same LF of primary stars, we assessed the 

421: contribution of the companion-associated events to the total optical 

422: depth would be $\tau_{\rm c}/\tau\sim 5f_{\rm bi}\%$ for current 

423: lensing surveys and can reach up to $\sim 6f_{\rm bi}\%$ for future 

424: surveys monitoring fainter stars.  We, therefore, conclude that events 

425: associated with companion stars comprises a non-negligible fraction of 

426: the total events, but their contribution to the optical depth is not 

427: large enough to explain the systematic difference between the optical 

428: depth estimates based on the two different methods.

429: 

430: 

431: \acknowledgments 

432: This work was supported by the Astrophysical Research Center for the 

433: Structure and Evolution of the Cosmos (ARCSEC") of Korea Science \& 

434: Engineering Foundation (KOSEF) through Science Research Program (SRC) 

435: program.  

436: 

437: 

438: \begin{thebibliography}{99}

439: \frenchspacing

440: \bibitem[Abt(1983)]{abt83}

441: Abt, H.A.\ 1983, \araa, 21, 343

442: 

443: \bibitem[Abt \& Levy(1976)]{abt76}

444: Abt, H.A., \& Levy, S.G.\ 1976, \apjs, 30, 273

445: 

446: \bibitem[Afonso et al.(2003)]{afonso03}

447: Afonso, C., et al.\ 2003, \aap, 404, 145

448: 

449: \bibitem[Alard(1997)]{alard97}

450: Alard, C.\ 1997, \aap, 321, 424

451: 

452: \bibitem[Alard(2000)]{alard00}

453: Alard, C.\ 2002, \aaps, 144, 363

454: 

455: \bibitem[Alard \& Lupton(1998)]{alard98}

456: Alard, C., \& Lupton, R.H.\ 1998, \apj, 503, 325

457: 

458: \bibitem[Alcock et al.(1997)]{alcock97}

459: Alcock, C., et al.\ 1997, \apj, 479, 119

460: 

461: \bibitem[Alcock et al.(1999)]{alcock99}

462: Alcock, C., et al.\ 1999, \apj, 521, 602

463: 

464: \bibitem[Alcock et al.(2000)]{alcock00}

465: Alcock, C., et al.\ 2000, \apj, 541, 734

466: 

467: \bibitem[Allen(2000)]{allen00}

468: Allen, C.W.\ 2000, Astrophysical Quantities, eds.\ A.N. Cox 

469: (The Athlone Press, London), 485

470: 

471: \bibitem[Belokurov, Evans, \& Le Du(2003)]{belokurov03}

472: Belokurov, V., Evans, N.W., \& Le Du, Y.L.\ 2003, \mnras, 341, 1373

473: 

474: \bibitem[Belokurov, Evans, \& Le Du(2004)]{belokurov04}

475: Belokurov, V., Evans, N.W., \& Le Du, Y.L.\ 2004, \mnras, 352, 233

476: 

477: \bibitem[Bennett \& Rhie(2002)]{bennett02}

478: Bennett, D.P., \& Rhie, S.H.\ 2002, \apj, 574, 985 

479: 

480: \bibitem[Bissantz et al.(1997)]{bissantz97}

481: Bissantz, N., Englmaier, P., Binney, J., \& Gerhard, O.\ 1997, 

482: \mnras, 289, 651

483: 

484: \bibitem[Bond et al.(2001)]{bond01}

485: Bond, I.A., et al.\ 2001, \mnras, 327, 868

486: 

487: \bibitem[Cseresnjes \& Alard(2001)]{cseresnjes01}

488: Cseresnjes, P., \& Alard, C.\ 2001, \aap, 369, 778

489: 

490: \bibitem[Dominik(1998)]{dominik98}

491: Dominik, M.\ 1998, \aap, 333, 893

492: 

493: \bibitem[Duquennoy \& Mayor(1991)]{duquennoy91}

494: Duquennoy, A., \& Mayor, M.\ 1991, \aap, 248, 485

495: 

496: \bibitem[Evans \& Belokurov(2002)]{evans02}

497: Evans, N.W., \& Kelokurov, V.\ 2002, \apj, 567, L119

498: 

499: \bibitem[Griest et al.(1991)]{griest91}

500: Griest, K., et al.\ 1991, \apj, 372, L79

501: 

502: %\bibitem[Griest \& Thomas(2005)]{griest05}

503: %Griest, K., \& Thomas, C.L.\ 2005, \mnras, 359, 464

504: 

505: \bibitem[Han(1997)]{han97}

506: Han, C.\ 1997, \apj, 484, 555

507: 

508: \bibitem[Han \& Gould(1995)]{han95}

509: Han, C., \& Gould, A.\ 1995, \apj, 447, 53

510: 

511: \bibitem[Han \& Jeong(1998)]{han98}

512: Han, C., \& Jeong, Y.\ 1998, \mnras, 301, 231

513: 

514: \bibitem[Holtzman et al.(1998)]{holtzman98}

515: Holtzman, J.A., Watson, A.M., Baum, W.A., Grillmair, C.J., Groth, E.J., 

516: Light, R.M., Lynds, R., \& O'Neil, E.J.\ 1998, \aj, 115, 1946

517: 

518: \bibitem[Kiraga(1994)]{kiraga94b}

519: Kiraga, M.\ 1994, Acta Astron., 44, 241

520: 

521: \bibitem[Kiraga \& Paczy\'nski(1994)]{kiraga94a}

522: Kiraga, M., \& Paczy\'nski, B.\ 1994, \apj, L101

523: 

524: \bibitem[Metcalf(1995)]{metcalf95}

525: Metcalf, R.B.\ 1995, \aj, 110, 869

526: 

527: \bibitem[Mollerach \& Roulet(1996)]{mollerach96}

528: Mollerach, S., \& Roulet, E.\ 1996, \apj, 458, L9

529: 

530: \bibitem[Nair \& Miralda-Escud\'e(1999)]{nair99}

531: Nair, V., \& Miralda-Escud\'e, J.\ 1999, \apj, 515, 206

532: 

533: \bibitem[Nemiroff(1994)]{nemiroff94}

534: Nemiroff, R.J.\ 1994, \apj, 435, 682

535: 

536: \bibitem[Paczy\'nski(1991)]{paczynski91}

537: Paczy\'nski, B.\ 1991, \apj, 371, L63

538: 

539: \bibitem[Peale(1998)]{peale98}

540: Peale, S.J.\ 1998, \apj, 509, 177

541: 

542: \bibitem[Popowski et al.(2001)]{popowski01}

543: Popowski, P., et al.\ 2001, ASP Conference Series, Microlensing 2000: 

544: A New Era of Microlensing Astrophysics, eds. J.W. Menzies \& P.\ D. 

545: Sackett (San Francisco: Astronomical Society of the Pacific), 234, 244

546: 

547: \bibitem[Popowski et al.(2004)]{popowski04}

548: Popowski, P., et al.\ 2004, astro-ph/0410319

549: 

550: \bibitem[Schechter, Mateo, \& Saha(1993)]{schechter93}

551: Schechter, P.L., Mateo, M., \& Saha, A.\ 1993, \pasp, 105, 1342

552: 

553: \bibitem[Sevenster \& Kalnajs(2001)]{sevenster01}

554: Sevenster, M.N., \& Kalnajs, A.J.\ 2001, \aj, 122, 885

555: 

556: \bibitem[Sumi et al.(2003)]{sumi03}

557: Sumi, T., et al.\ 2003, \apj, 591, 204

558: 

559: \bibitem[Sumi et al.(2005)]{sumi05}

560: Sumi, T., et al.\ 2005, \apj, submitted (astro-ph/0502363)

561: 

562: \bibitem[Tomaney \& Crotts(1996)]{tomaney96}

563: Tomaney, A.B., \& Crotts, A.P.S.\ 1996, \aj, 112, 2872

564: 

565: \bibitem[Udalski et al.(1994)]{udalski94}

566: Udalski, A., et al.\ 1994, Acta Astron., 44, 165

567: 

568: \bibitem[Wo\'zniak(2000)]{wozniak00}

569: Wo\'zniak, D.R.\ 2000, Acta Astron., 50, 421

570: 

571: \bibitem[Zhao \& Mao(1996)]{zhao96}

572: Zhao, H.-S., \& Mao, S.\ 1996, \mnras, 283, 1197

573: 

574: \bibitem[Zhao, Spergel, \& Rich(1995)]{zhao95}

575: Zhao, H.-S., Spergel, D.N., \& Rich, R.M.\ 1995, \apj, 440, L13

576: 

577: 

578: 

579: 

580: \end{thebibliography}

581: 

582: \end{document}

583: