1: \def\lae{\mathrel{<\kern-1.0em\lower0.9ex\hbox{$\sim$}}}

2: \pdfoutput=1

3: %

4: % Her X-1 FUSE tomogram paper

5: %

6: % in directory postref.dir

7: %	responding to the referee

8: %	changes incorporated so far:

9: %	1) N\,V changed for example to N\,{\sc v}

10: %	2) Footnote added to title citing use of FUSE

11: %       3) CHIANTI explanations added

12: %	4) More explanation of Fourier Filtered Back Projection

13: %       5) Started a section on interstellar H2 absorption lines

14: %	6) Created Appendices to explain continuum and line models

15: %

16: % version ms_doppler23.tex

17: %	incorporating jraymond's typo corrections and new section

18: %	on Bowen fluorescence

19: %

20: % version ms_doppler22.tex

21: %	finishing touches before sending to co-authors

22: %

23: % version ms_doppler21.tex

24: %	following a recalibration to fix the flux near c3

25: %		that was on a detector boundary,

26: %	also shifted wavelengths to get sharper lines (higher res)

27: %	also tested models to simulate the strong redshifted line:

28: %		gaussian (works, but why?)

29: %		ablation (works with some fiddling)

30: %		nonisotropic x-rays (didn't work)

31: %	downplaying bowen origin of n3 emission at 991

32: %		(given ratios)

33: %	

34: % version ms_doppler20.tex

35: %	now the eclipse fit is with 6 phases, including the anomalous phase

36: %	no separate graph for the anomalous phase

37: % version ms_doppler19.tex 

38: %	after feedback from saku, some from john raymond

39: %	cutting down exposition in the introduction

40: %	showing more sophisticated continuum models

41: %	including a diagram to illustrate the simple simulation

42: %	more discussion on c3 line ratio, both observational

43: %		complications and theoretical analysis

44: %	to include:

45: %	models in which the "anomalously strong" emission lines

46: %	are included in the eclipse fits

47: %

48: % version ms_doppler18.tex

49: %	after sending draft to saku, john raymond

50: %	revisions to make:

51: %		rework section on the simulation, rework the conclusions

52: %		tone down the c3 977A detection--the abs line makes it

53: %			more uncertain--mention fig. 1 plot

54: % for future work:

55: % measure the n3 991 and 4340 fluxes at each orbital phase

56: % get a more accurate ratio by a weighted average

57: % also go through with my idea of fitting the airglow line w/ gaussian

58: %

59: % version ms_doppler17.tex

60: %	finishing revisions of 16.tex

61: %	this is after reorganization of tomography

62: %		notebook (tomography_byline2)

63: %

64: % new version ms_doppler16.tex

65: %	deleting modified tomogram ("error tomogram") sections

66: %	working with continuum fit section

67: %	changing figure order

68: %	back projection replaced with fourier filtered back projection

69: %

70: % new version ms_doppler12.tex

71: % 	at start, fixed abstract, text, to give less emphasis to 

72: %		tomography

73: % new version ms_doppler11.tex

74: % 	main change:

75: %	add section on Bowen fluorescence

76: %	add plot showing bowen trailed spectrogram

77: %

78: %	other changes:

79: %		cut down intro slightly

80: %		move up her x-1 intro, move down tomography intro

81: %

82: % note: total duration is 65-66 Ksec

83: %

84: % to fix in next revision::

85: %		mark km/s on tomograms

86: %		put long term phase in table

87: %		find where I got long term phase from before, mention

88: %		are the other lines anomalous at phi=0.876 also?

89: %		make the axis labels on the o6 eclipse fit plot bigger

90: %		may make sense to show trailed spectrogram first, 

91: %			tomogram 2nd

92: %		measure more exactly the position of the peak in the

93: %			tomogram (I did by eye)

94: %		be clearer about how I've stretched the b&w color scheme

95: %			in the tomograms

96: %		put in references to the table on system parameters

97: %		the caption for figure 2 gets squeezed out (no display)

98: 

99: %\documentstyle[12pt,aaspp4]{article}

100: %\documentclass[preprint2,longabstract]{aastex}

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

102: 

103: %\newcommand{\Msun}{\mbox{ M}_{\odot}}

104: \newcommand{\gae}{\mathrel{>\kern-1.0em\lower0.9ex\hbox{$\sim$}}}

105: \newcommand{\kms}{km~s$^{-1}$}

106: \newcommand{\NV}{\ion{N}{5}\,$\lambda\lambda\,1238.8,\,1242.8$\ }

107: \newcommand{\CIV}{\ion{C}{4}\,$\lambda\lambda\,1548,\,1550$\ } 

108: \newcommand{\SiIV}{\ion{Si}{4}\,$\lambda\lambda\,1393.7,1402.8$\ }

109: 

110: \lefthead{Boroson et al.}

111: \righthead{One Her X-1 Orbit with {\it FUSE}}

112: 

113: \begin{document}

114: 

115: \title{FUSE Observations of a Full Orbit of Hercules X-1: Signatures of 

116: Disk, Star, and Wind$^1$}

117: 

118: \author{Bram S. Boroson}

119: \affil{bram.boroson@gmail.com}

120: 

121: \and

122: 

123: \author{Saeqa Dil Vrtilek and John C. Raymond}

124: \affil{Smithsonian Astrophysical Observatory

125: Mail Stop 83

126: Cambridge, MA 02138

127: saku@head.cfa.harvard.edu, jraymond@cfa.harvard.edu}

128: 

129: \and

130: 

131: \author{Martin Still}

132: \affil{South African Astronomical Observatory

133: PO Box 9

134: Observatory 7935

135: Cape Town, South Africa

136: still@saao.ac.za}

137: 

138: $^1$Based on observations made with the NASA-CNES-CSA Far Ultraviolet 

139: Spectroscopic Explorer. FUSE is operated for NASA by the Johns Hopkins 

140: University under NASA contract NAS5-32985

141: 

142: %***

143: 

144: \begin{abstract}

145: 

146: We observed an entire 1.7 day orbit of the X-ray binary Hercules~X-1 with

147: the Far Ultraviolet Spectroscopic Explorer ({\it FUSE}).  Changes in the

148: O\,{\sc vi}$\lambda\lambda1032,1037$ line profiles through eclipse 

149: ingress and

150: egress indicate a Keplerian accretion disk spinning prograde with the

151: orbit. These observations may show the first double-peaked accretion disk

152: line profile to be seen in the Hercules~X-1 system.  Doppler tomograms of

153: the emission lines show a bright spot offset from the Roche lobe of the

154: companion star HZ~Her, but no obvious signs of the accretion disk. 

155: Simulations show that the bright spot is too far offset from the Roche

156: lobe to result from uneven X-ray heating of its surface. The absence of

157: disk signatures in the tomogram can be reproduced in simulations which 

158: include absorption from a stellar wind. We attempt to diagnose

159: the state of the emitting gas from the C\,{\sc iii}$\lambda977$,

160: C\,{\sc iii}$\lambda1175$, and N\,{\sc III}$\lambda991$ emission lines. The 

161: latter may be enhanced through Bowen fluorescence. 

162: \end{abstract}

163: 

164: \keywords{stars: neutron, X-rays: binaries, ultraviolet: stars, 

165: individuals: Her X-1}

166: 

167: \section{Introduction}

168: 

169: Hercules X-1 (discovered by Tananbaum et al. 1972) is one of the most

170: frequently observed X-ray binary systems. The intermediate mass of the

171: donor star, $\approx 2.2$~M$_\odot$, leads to a wealth of behavior seen in

172: both low mass and high mass systems. Although we suspect the mass flow is

173: primarily through Roche lobe overflow giving rise to an accretion disk, as

174: in the Low Mass X-ray Binaries (LMXB), the variable P~Cygni profiles

175: observed in UV lines suggest either a transient stellar wind or a stellar

176: wind that is photoionized in some regions (Boroson, Kallman, \&\ Vrtilek

177: 2001). There are theoretical grounds as well to expect that winds may

178: arise in this system when X-rays from the neutron star heat the surface of

179: the normal star or accretion disk (Arons 1973; Davidson \&\ Ostriker 1973; 

180: Basko et al. 1977; London, McCray, \&\ Auer 1981; Begelman, McKee, \&\

181: Shields 1983; Begelman \&\ McKee 1983). 

182: 

183: Most LMXBs, including Z-sources and atoll-sources, do not show 

184: persistent pulsations, perhaps because they have neutron 

185: stars with low magnetic fields.  

186: These sources thus lack an essential window into their kinematics. 

187: Light travel-time delays in the 1.24~second Her~X-1 X-ray pulsation period

188: determine that the neutron star is in a nearly circular 1.7~day orbit with

189: semimajor axis a$_{\rm x}=13.86$~seconds. The orbital period is slowly

190: lengthening from mass loss in the system (Deeter et al. 1991). 

191: 

192: Fortuitously, Her~X-1 has high orbital inclination ($>80$ degrees),

193: allowing total X-ray eclipses to constrain the size of the donor star, and

194: the high galactic latitude offers only small reddening, allowing the

195: system to be observed at crucial UV and even EUV wavelengths. 

196: 

197: The system's most mysterious variability is the 35-day X-ray high and low

198: cycle (Giacconi et al. 1973). For 8--11 days (the ``Main-On state'') of

199: this cycle Her~X-1 emits X-rays with a $\sim 10^{37}$ erg~s$^{-1}$ total

200: X-ray luminosity. For a $\sim 4$ day cycle (the ``Short-On state''),

201: halfway through the 35 day cycle, Her~X-1's output is several times lower. 

202: For the remaining portion of the 35 days, the Off state, the observed

203: X-ray flux is only several percent of that during the Main-On state. 

204: 

205: The behavior of the optical flux and X-ray spectrum over the 35-day period

206: shows that the 35-day variability is not isotropic. Instead, the

207: X-ray emission is merely obscured by a warped accretion disk whose shape

208: ``precesses" globally throughout a 35-day period. The optical 

209: emission from the normal star varies with the 1.7~day binary orbit.

210: Its spectral type is variously classified as A through F, because X-rays 

211: from the neutron star heat the surface of the facing side of the Roche 

212: lobe. The spectral type continues to change with the orbital period 

213: throughout the longer 35-day

214: cycle, implying that X-rays continue to heat the surface of HZ~Her.

215: 

216: The transition between the Off and Main-On state is only a 

217: few hours, whereas the Short-On state is entered into more gradually.

218: This indicates that regions of the disk at different radii may obscure 

219: the central source, in which case the disk must precess globally with the 

220: same period. The variation in pulse shape with the long-term period is 

221: sometimes taken to imply that the innermost edge of the accretion disk 

222: precesses with the 35 day period, obscuring local regions on the neutron 

223: star (Scott, Leahy, \&\ Wilson, 2000).

224: 

225: The physical cause of a 35-day global disk precessing disk warp, or a

226: definitive relation between system parameters and disk precession period,

227: remains unknown and is a major goal of investigations of the system.  It

228: has been proposed, but not confirmed, that reprocessed X-ray radiation

229: pressure (Maloney \&\ Begelman 1997; Wijers \&\ Pringle 1999) or an X-ray

230: driven disk wind and corona (Meyer, Meyer-Hofmeister, 1984) could

231: propagate the warped shape in the disk. Similar "long-term periods'' have

232: been observed in other X-ray binaries, including LMC~X-4.  Her~X-1 serves

233: as a prototype of this behavior. 

234: 

235: X-ray dips present another variability that has not been explained 

236: definitively.  Crosa \&\ Boynton (1980) showed that the dips recur

237: every 1.65 days, near, but not at, the 1.62~day beat period between

238: orbital and precessional periods. This was confirmed by long-term X-ray 

239: observations (Scott \&\ Leahy, 1999). The dips are thought to be associated 

240: with the gas stream between the stars, but there is no consensus 

241: explanation for them or their period.

242: 

243: As befits such a frequently observed, enigmatic, yet prototypical source,

244: the methods of spectroscopy have been applied to the system at a wide

245: range of wavelengths. These studies have borne fruit with determinations

246: of the system's elemental abundances and the orbital motion of the normal

247: star (the neutron star's pulse delays indicate only the motion of one

248: component of the binary). 

249: 

250: High resolution spectroscopy in the X-ray range using {\it 

251: XMM} (Jimenez-Garate et al. 2002) shows a

252: multitude of X-ray emission lines presumably from an accretion disk

253: corona. The line ratios indicate that the gas is enhanced from CNO

254: processing from a massive progenitor.  Jimenez-Garate et al. (2005)

255: confirmed the CNO enhancement by observations of more than two dozen

256: emission lines originating in an accretion disk corona. Model predictions

257: of the disk corona's response to illumination by the central X-ray source

258: are in reasonable agreement with the observed fluxes for low and moderate

259: Z elements ( O through S), but the Fe\,{\sc xxv},Fe\,{\sc xxvi} lines are 

260: several times brighter than predicted (Jimenez-Garate et al. 2005). 

261: 

262: From observations of optical absorption lines (Reynolds et al. 1997), and 

263: from optical pulsations that result from the X-rays from the neutron star 

264: periodically striking the surface of the normal star (Middleditch \&\ 

265: Nelson 1976), we know that the normal star has a mass $\approx2.2$~M$_\odot$

266: and the neutron star has a mass 1.5$\pm0.3$~M$_\odot$.

267: The optical spectrum shows absorption lines from HZ~Her, emission at the 

268: Bowen blend near 4640\AA\ (see Schachter, Filippenko, \&\ Kahn 1989 for 

269: a discussion of the formation of these lines), 

270: and emission from He\,{\sc ii}$\lambda4686$. Still et al. (1997) performed 

271: Doppler tomography on these optical emission lines but could not 

272: conclude that they arose in a symmetric accretion disk.

273: 

274: The UV spectrum is crucial for an understanding of the accretion disk and

275: the X-ray illuminated face of HZ~Her. The continuum emission from both the

276: illuminated star and the disk peaks in the UV and the disk contributes a

277: greater fraction than at optical wavelengths (Cheng, Vrtilek, \&\ Raymond

278: 1995).  From models of the variable UV continuum as observed with the

279: International Ultraviolet Explorer ({\it IUE}, Vrtilek \&\ Cheng 1996)

280: showed that a change in the accretion disk precession could explain an

281: anomalously low period of X-ray emission. 

282: 

283: To study global accretion one should observe gas

284: at ionization stages and temperatures resulting from X-ray illumination of

285: the disk, and one would need to resolve Doppler-shifted velocities that

286: correspond to the orbital motion of the neutron star (160 km~s$^{-1}$),

287: motion in the accretion disk (expected to be $\approx300$~km~s$^{-1}$ at

288: the edge of the disk), and the velocities expected in a disk or stellar

289: wind ($\sim1000$~km~s$^{-1}$). 

290: 

291: The strong UV resonance lines from N\,{\sc v}, Si\,{\sc iv}, and C\,{\sc

292: iv}, first seen with {\it IUE} (Howarth \&\ Wilson 1983b), presumably

293: result from photoionization of the accretion disk and HZ~Her (Raymond

294: 1993, Ko \&\ Kallman 1994). These lines are much stronger than the

295: strongest optical high ionization line, He\,{\sc

296: ii}$\lambda4686$, both in absolute flux and equivalent width. 

297: Observations with the Faint Object Spectrograph (FOS) aboard the 

298: Hubble Space Telescope ({\it HST}) showed that these lines are still 

299: present at a few percent of maximum brightness during mid-eclipse when the 

300: disk and heated star should be entirely obscured (Anderson et al. 1994). 

301: The source of this emission may be an expanding wind.

302: 

303: The Goddard High Resolution Spectrograph (GHRS) on {\it HST} first

304: resolved these emission lines at $\approx35$~km~s$^{-1}$ resolution

305: (Boroson et al., 1996) to discern variable broad and narrow emission 

306: components. 

307: 

308: The {\it HST} Space Telescope Imaging Spectrograph (STIS) confirmed

309: that the resonance lines have at least two components (Vrtilek et al.

310: 2001). A broad component arises on the accretion disk while a narrow line

311: component may be associated with HZ~Her. Prior to the STIS

312: observations, the accretion disk had only been observed as it contributed

313: to the continuum light curve or obscured the central X-rays. During

314: eclipse ingress and egress, the broad lines seen with STIS behaved

315: as expected for lines from an accretion disk rotating prograde

316: with the orbital direction.  The blue edge of the line was obscured

317: first in eclipse ingress and appeared first in egress. 

318: 

319: Observations with the Far Ultraviolet Spectroscopic Explorer ({\it FUSE})

320: are complementary to the existing {\it HST} observations. Long-term

321: variability in the system makes it impossible to combine 

322: rigorously 

323: data from different epochs.  Analyzed separately, however, the {\it FUSE}

324: wavelength range of 900-1200\AA\ offers similar advantages and powerful

325: consistency checks to analysis of the {\it HST} bandpass of 1200-1700\AA.

326: Both wavelength ranges have strong resonance lines that respond to X-ray

327: photoionization.  Observations with the Hopkins Ultraviolet Telescope

328: ({\it HUT}) showed that the O\,{\sc vi} doublet has flux comparable to the

329: N\,{\sc v} doublet, the brightest near UV line (Boroson et al. 1997). 

330: 

331: The resonance line doublets offer optical depth information through their

332: doublet ratios, but if the lines are as broad as the doublet separation,

333: it may be impossible to determine the individual contribution of each line

334: where they overlap. The separations of C\,{\sc iv}$\lambda\lambda

335: 1549,1551$, Si\,{\sc iv}$\lambda\lambda\lambda 1393,1403$, N\,{\sc

336: v}$\lambda\lambda 1238.1242$, O\,{\sc vi}$\lambda\lambda 1032,1038$,

337: S\,{\sc vi}$\lambda\lambda 933,944$ correspond to Doppler shifts of 500,

338: 1900, 960, 1650, 3600 km~s$^{-1}$, respectively. The far UV lines O\,{\sc

339: vi} and S\,{\sc vi} compare favorably with the near UV lines (i.e. have

340: greater separation and are less likely to overlap), except for Si\,{\sc

341: iv}, which suffers from confusion with an O\,{\sc iv} blend near 1400\AA. 

342: 

343: For the present observations, we observed an entire 1.7~day binary orbit

344: with {\it FUSE}. A major goal of this program was to apply the method of

345: Doppler tomography (\S7), which has the advantage of

346: diagnosing the accretion flow of different systems without bringing to

347: bear more than a few assumptions. 

348: 

349: Table~1 shows the measured physical parameters of the system and

350: parameters we adopt for our models. 

351: 

352: \section{FUSE Observations}

353: 

354: With {\it FUSE} (the Far Ultraviolet Spectroscopic Explorer) we extend

355: observations in the UV spectral range to 900-1190\AA, a range observed

356: only once before using {\it HUT}, the Hopkins Ultraviolet Telescope aboard

357: the ASTRO-1, carried aboard the Space Shuttle but not placed in orbit

358: (Boroson et al. 1997). While {\it HUT} had a resolution of $\approx3$\AA,

359: {\it FUSE} has a resolution of $\approx0.05$\AA. 

360: 

361: {\it FUSE} is a NASA {\it Origins} mission operated by The Johns Hopkins

362: University. Four aligned telescopes feed two identical far-UV

363: spectrographs. With resolution R$=20000$ FUSE approaches {\it HST} in its

364: utility for our program; the time coverage of the {\it HST} Space

365: Telescope Imaging Spectrograph (STIS) was limited because the

366: detectors were turned off when the spacecraft passed through the South

367: Atlantic Anomaly. The FUSE mission is described in more detail in Moos et

368: al. (2000) and its on-orbit performance is described in Sahnow et.  al.

369: (2000). 

370: 

371: Our {\it FUSE} observations began on June 9, 2001 at 7:47~UT. Table~2

372: shows the log of exposures, each of which is integrated over each {\it

373: FUSE} orbit of the Earth, with gaps when Her~X-1 goes below the horizon.

374: We use the orbital ephemeris of Deeter et al. (1991) to determine the

375: orbital phases of our observation. The exposure times listed in Table~2

376: are in most cases equal to the raw observation time. However, for cases

377: such as observation 13, interrupted by a passage through the South

378: Atlantic Anomaly and not occultation by the Earth, the Exptime listed is

379: in general the minimum good exposure time for any {\it FUSE} detector. 

380: The data were obtained through the LWRS aperature and in TIMETAG mode.

381: 

382: We used the CalFUSE pipeline software version 3.0.8 to extract and

383: calibrate the data from all four {\it FUSE} telescopes.  Below 1100\AA, 

384: where emission features are sharp, we added an

385: offset to each wavelength scale, in intervals of 0.025\AA, so that the

386: absorption and emission features from each detector best agreed. 

387: We tested our wavelength calibration against the interstellar 

388: Si\,{\sc ii}$\lambda1020.6989$ absorption line, which we found to have a mean

389: heliocentric velocity of $-30$~km~s$^{-1}$. The standard deviation of 

390: the centroid of this line from orbit to orbit was $\approx0.03$\AA or 

391: $<10$~km~s$^{-1}$.

392: 

393: 

394: The S/N of the data was $\approx5$ per 0.1\AA\ pixel in the continuum in the

395: region $<1000$\AA\ and $\approx10$ near 1100\AA. The S/N within the 

396: O\,{\sc vi}

397: line had greater variation with orbital phase, as the doublet changed both

398: in shape and in strength. At $\phi=0.75$, the peak S/N within the doublet

399: was $\approx15$ per 0.1\AA\ pixel, while at $\phi=0.5$, the peak S/N was

400: $\approx25$ per pixel. 

401: 

402: For further analysis, we skip the region between 1120--1160\AA\ in the 1B 

403: LiF detector. This region suffers from a systematic decrease in counts 

404: known as the worm..

405: 

406: In Figures~1 and 2, we show average observed FUSE spectra at orbital

407: phases when the emission is dominated by the disk and star, respectively.

408: For the disk-dominated spectrum, we use observation number 20, at

409: $\phi=0.917$.  For the star-dominated spectra, we average observations

410: number 7 through 11 (phases $\phi=0.43--0.60$). 

411: 

412: For the {\it FUSE} spectra at orbital phases dominated by disk emission,

413: we compare with the STIS spectrum (observation root 

414: name O4V401010) during a Short-On state at $\phi=0.904$. 

415: 

416: For the spectrum dominated by star emission, we use the STIS

417: observation with root name O4V452050, observed during a Main-On state

418: approximately 5 months earlier, on January 24, 2001, starting at MJD

419: 51933.779933 and lasting 2620 seconds. The mean orbital phase of this

420: exposure was $\phi=0.455$. This spectrum has not previously been

421: published. 

422: 

423: We compare the spectra with the time-weighted average of 

424: our continuum models (described in \S5). We

425: indicate prominent UV emission lines from the star system.

426: 

427: Geocoronal (airglow) lines produced in the Earth's atmosphere are also 

428: present (see Feldman et al., 2001): H\,{\sc i}\,Lyman-$\gamma$ 973, 

429: O\,{\sc i}\,5 989, H\,{\sc i}\,Lyman-$\beta$+O\,{\sc i}\,4

430: 1026-1027, and N\,{\sc i} 1167.

431: 

432: \section{Interstellar Lines}

433: 

434: %

435: % points to make:

436: %  figure in previous section will ID atomic lines according to Morton

437: %

438: %  compare lyman series with previous determination of lyman alpha

439: %	nh=1e20... I added in another component w/ 1/10 column density

440: %	is this needed? check

441: 

442: Interstellar features in the {\it FUSE} spectra are interesting not only 

443: for the direct information they provide on the interstellar medium (ISM), 

444: but a proper accounting of these features can remove systematic errors 

445: from the analysis of the line and continuum emission from the system itself.

446: 

447: The neutral Hydrogen column density, N$_H$, has previously been determined

448: to be $\approx10^{20}$~cm$^{-2}$ (Boroson et al. 2000) from the wings of

449: the saturated Lyman~$\alpha$ line as observed with the {\it HST} STIS. 

450: The {\it FUSE} bandpass includes further saturated lines in the Lyman

451: series, and these are consistent with N$_H\approx10^{20}$~cm$^{-2}$. 

452: This N$_H$ is also consistent with the E(B-V) value of 0.018 according to 

453: the Bohlin (1975) relation.

454: 

455: From the galactic latitude of 37.52$^\circ$ we should expect a sightline

456: with less H and H$_2$ than typically seen through the galactic disk. 

457: Indeed, while absorption lines from rotational levels $J=0$ through $J=3$

458: are readily identified, they are not saturated.  Thus the level

459: populations should be easy to measure and it should be easy to compensate

460: for the effects of the absorption lines on the spectra. 

461: 

462: In Figure~3, we show a patch of the time-averaged {\it FUSE} spectrum,

463: with the absorption profiles expected from source with a flat spectrum

464: absorbed by columns of (2.8, 7.5, 4.1, 3.4)$\times 10^{14}$ cm$^{-2}$ for

465: absorption from rotational levels $J=0,1,2,3$, respectively.  We generate

466: the H$_2$ profiles from the templates of McCandliss (2003), which are

467: based on Abgrall et al. (1993a,b).  We have assumed a line velocity

468: parameter $b=10$~km~s$^{-1}$ and have convolved the profiles with a

469: Gaussian to simulate the {\it FUSE} Line Spread Function. 

470: 

471: The two lowest rotational energy levels should have populations given by 

472: a Boltzmann factor, taking into account the statistical weights of the 

473: two levels, $g_1/g_0=9$.  The energy difference between the levels is

474: $\Delta E_{01}=170.5$~K.  We then estimate that the temperature of the 

475: intervening H$_2$ gas is $T_{01}=140$~K.  This is hotter than found for 

476: H$_2$ gas from disk stars, consistent with the high galactic latitude

477: of the line of sight (Gillmon, Shull, Tumlinson, \&\ Danforth 2006).

478: 

479: The average molecular fraction, defined by

480: \begin{equation}

481: f_{\rm H2}=\frac{2\mbox{N}(\mbox{H}_2)}

482: {

483: \mbox{N(H I)}

484: +

485: 2\mbox{N(H2)}

486: }

487: \end{equation}

488: is therefore $\gae4\times10^{-5}$ from the observed H$_2$ lines, which 

489: have N(H$_2$)$\approx2\times10^{15}$~cm$^{-2}$.  

490: Ratios of higher rotational levels than $J_1/J_0$ are determined not by 

491: temperature and collisional excitation but by background FUV radiation. 

492: The H2 column density observed for Her~X-1 places it just above the 

493: boundary at which H2 clouds start to become optically thick to the FUV 

494: radiation.  Gillmon et al. (2006) state that $f_{\rm H2}\approx10^{-5}$ 

495: is a typical molecular fraction below this boundary.

496: 

497: Thus the first four levels of rotational excitation of H$_2$ appear to 

498: match observed absorption features and to be caused by H$_2$ clouds that 

499: are not out of the ordinary.

500: 

501: \section{Photometry of Emission Lines}

502: 

503: The lines vary with the binary orbit in a manner similar to that

504: previously observed in the near UV with {\it HST}. The flux peaks 

505: generally near $\phi=0.5$, when the X-ray heated face of HZ~Her

506: points toward the observer, although there may be a dip very close to

507: $\phi=0.5$ as the accretion disk occults the star.

508: 

509: In Figure~4 a through i respectively, we show the photometric variation

510: with orbital phase of the emission lines S\,{\sc vi}$\lambda 933.4$,

511: S\,{\sc vi}$\lambda944.5$, C\,{\sc iii}$\lambda 977$, N\,{\sc

512: iii}$\lambda991$, O\,{\sc vi}$\lambda1031.9$, O\,{\sc vi}$\lambda 1037.6$,

513: S\,{\sc iv}$\lambda 1073$, P\,{\sc v}$\lambda 1128$, and C\,{\sc

514: iii}$\lambda 1176$. We have subtracted the background flux using a model

515: that we present in \S5.  For the N\,{\sc iii} line, we remove the flux of

516: nearby airglow lines. 

517: 

518: In Table~3, we show the cross-correlation coefficients between the

519: measured lightcurves of the spectral lines and the lightcurves of the disk

520: and star contributions to the continuum lightcurve, as determined by our

521: continuum model. 

522: 

523: \section{UV Continuum Fits}

524: 

525: The UV continuum varies, as the optical continuum does, with the 1.7 day

526: orbital period.  The continuum generally peaks near $\phi=0.5$, when the 

527: X-ray heated face of the Roche lobe points towards the viewer, but a 

528: portion of the normal star may be blocked by the accretion disk, and 

529: depending on long-term phase, the actual peak may occur within 

530: $\phi\pm\Delta \phi\approx 0.5\pm 0.1$.  

531: 

532: The optical/UV orbital light curves do not repeat exactly over the 35-day 

533: disk precession cycle. The disk may block X-rays from heating portions of 

534: the star, and as it precesses it projects varying areas into the line of 

535: sight. The precessing disk should have different eclipse light curves 

536: throughout the 35 day cycle.

537: 

538: To fit the UV continuum observed with {\it FUSE}, we assume that X-ray

539: heating of HZ~Her and the accretion disk cause the entirety of continuum

540: emission.  The details of our simulation are similar to those of Vrtilek

541: et al. (1990, especially the appendix), and Howarth \&\ Wilson (1983a)

542: (which was an elaboration on an earlier analysis of Gerend \&\ Boynton,

543: 1976), and binary simulation codes by Wilson \&\ Devinney (1971).  The

544: method is summarized in Appendix~A. 

545: 

546: We fit the spectrum using a reddening $E(B-V)=0.018$ and for the

547: extinction curve, we use Equation~5 in Cardelli, Clayton, \&\ Mathis

548: (1989). 

549: 

550: We also use a more modern determination of the distance to Her~X-1 

551: (Reynolds et al. 1997). This greater distance requires a larger $\dot{M}$ 

552: to reach the same continuum flux. Thus the scale of our $\dot{M}$ values 

553: are large compared with those reported earlier from {\it IUE} and {\it 

554: HST} observations. The {\it IUE} observations found a range of $-\log 

555: \dot{M}=8.30$ to $8.68$ (given changes in the long-term phase during an 

556: anomalous low state), whereas the {\it HST} found $-\log 

557: \dot{M}=8.19\pm0.06$.

558: 

559: When fit our continuum model to the observed {\it FUSE} spectrum from each 

560: {\it FUSE} observation, we find a range of $-\log \dot{M}=8.12$ to 8.38.

561: 

562: For our fits we ignore wavelength regions that contain prominent 

563: emission or absorption lines. We correct for absorption from the 

564: first 4 rotational energy levels of interstellar H$_2$ by multiplying

565: the spectrum by the model presented in \S3.

566: 

567: Table~3 gives the best-fit values of the mass accretion rate, $\dot{M}$, 

568: which we allow to vary as a free parameter for each {\it FUSE} orbit. We 

569: also list the reduced $\chi^2$.

570: 

571: We have used our continuum fits to examine the continuum photometry.  The

572: model allows us to separate the continuum emission

573: into contributions from the heated face of

574: HZ~Her and from the accretion disk. In Figure~5 we show the model flux 

575: near the O\,{\sc vi} lines split into disk and stellar components.

576: 

577: % /Users/bboroson/Programs/xrb.continuum

578: 

579: \section{Eclipse Models}

580: 

581: We made a simple model for emission from a symmetric precessing

582: accretion disk undergoing eclipse by the Roche lobe of its companion. We

583: then allowed free parameters of that model to vary in order to fit the

584: observed O\,{\sc vi} profiles during eclipse ingress and egress.

585: 

586: The disk and Roche lobe geometry are based on the model of Howarth \&\

587: Wilson (1983a), while the formation of the emission lines follows the model

588: of Horne (1995), and is described in Appendix~B. We allow as free parameters 

589: the exponent of a radial

590: power law of optical depth and elements of the Mach turbulence matrix. We

591: allow the normalization of the line flux at each orbital phase to vary as

592: a free parameter.

593: 

594: We calculate the eclipsing edge of the Roche lobe of 

595: HZ~Her assuming corotation with the orbit.

596: 

597: Before fitting the model to the data, we subtracted the mid-eclipse

598: spectrum from all spectra. The origin of the UV mid-eclipse spectra was

599: explored by Anderson et al. (1994) who concluded that it probably does

600: not arise in the accretion disk. The mid-eclipse O\,{\sc vi} lines are 

601: broad, extending from -200 to $+500$~km~s$^{-1}$ heliocentric velocity.

602: 

603: We also apply narrow gaussian absorption lines to our model spectra in

604: order to simulate a possible C\,{\sc ii} interstellar line near 1036\AA, a

605: possible C\,{\sc ii}$*$ line near 1037\AA, and a O\,{\sc i} line near

606: 1039\AA. We also include narrow interstellar absorption near the rest

607: wavelengths of the O\,{\sc vi} doublet. A narrow absorption line can be

608: seen near the blue component of the doublet at $\lambda=1031.9$\AA, but

609: the weaker line at 1037.6\AA\ cannot be seen explicitly. We fix the

610: wavelengths, widths, and optical depths of these lines and do not let them

611: vary in our fit. 

612: 

613: We fix all the parameters describing the disk and orbit to those given in 

614: Howarth \&\ Wilson (1983a), except the outer radius we fix to 

615: $2\times10^{11}$~cm, following Cheng et al. (1995).

616: 

617: The results are shown in Figure~6.

618: 

619: The reduced $\chi^2$ of the fit was 2.8 with 1418 degrees of freedom.

620: 

621: The power law index for the radial dependence on optical depth was 

622: $\alpha=-1.2$, between the $\alpha=-1.5$ result found for the N\,{\sc v} 

623: doublet by similar methods in Boroson et al. (2000) and 

624: the $\approx-0.55$ value expected from 

625: the simulations of Raymond (1993).

626: 

627: Although we based our fits on the model of Horne (1995) which allows for

628: anisotropic turbulence, the fits do not unambiguously settle on

629: particular values of the Mach matrix. Here, for simplicity, we present a 

630: model with isotropic Mach=1 turbulence.

631: 

632: As with the resonance doublets in the near UV, the observed

633: blue component of O\,{\sc vi} near eclipse is stronger than the red 

634: component, 

635: which suggests that much of the line flux is formed where $\tau_0\lae1$. 

636: However, the doublet ratio is uncertain because strong interstellar 

637: C\,{\sc ii} 

638: and O\,{\sc i} absorption lines affect only the red O\,{\sc vi} doublet 

639: component. 

640: Weaker interstellar lines, which we have not modeled, may be present as 

641: well, and may preferentially absorb one or the other O\,{\sc vi} component.

642: 

643: The model lines are double-peaked outside of eclipse, but one peak of the

644: red component is absorbed by interstellar O\,{\sc i} absorption. The observed

645: line profiles are double-peaked at $\phi=0.84$ and double-peaked but 

646: broader at $\phi=0.80$. Observations at $\phi=0.35$ and $\phi=0.39$, when 

647: narrow line emission predominates, show a dip in the flux at the position 

648: of the gap between peaks from the accretion disk line.

649: 

650: The line normalizations we found from our disk eclipse model can be

651: compared with the mass accretion rates inferred from our continuum fits.

652: They do not appear correlated.  At some of the phases investigated, we

653: observe only a small fraction of the accretion disk, and none of the

654: illuminated star, so that systematic errors in the line or continuum

655: models may make accurate estimates of $\dot{M}$ impossible. In particular,

656: we have considered the disk flat when it is probably warped, we have not

657: calculated the ionization structure of the disk, we have ignored possible

658: emission by the accretion stream, and we have ignored the role of winds in

659: this system. 

660: 

661: Chiang (2001) presented models of accretion disk eclipse in Her~X-1 that

662: considered a disk wind. These models were motivated by the lack of

663: double-peaked line profiles in any observations of UV resonance lines.

664: Although the current model predicts double-peaked emission lines when the

665: full disk is visible, the S/N is not strong 

666: enough to rule out a double-peaked structure. In addition, interstellar 

667: absorption lines coincide with three of the four peaks of the O\,{\sc vi} 

668: doublet. 

669: 

670: It is clear that a model of a partially eclipsed

671: accretion disk with parameters considered standard from previous work

672: provides an excellent match to this data set.  There appear to be 

673: too many ways to improve the current fit to justify singling one out. 

674: 

675: \subsection{Anomalous emission at $\phi=0.876$}

676: 

677: The fit at $\phi=0.876$ is particularly poor and we suggest that

678: a process in addition to the partial eclipse of a Keplerian accretion 

679: disk affects the line profile.

680: 

681: Figure~6 shows that even though more of the disk is eclipsed than at 

682: $\phi=0.835$, the emission lines at $\phi=0.876$ are actually stronger. 

683: There also appears to be an additional absorption component near the rest 

684: wavelength of O\,{\sc vi}$\lambda1032$.

685: 

686: The continuum at $\phi=0.876$ has not increased over the continuum at

687: $\phi=0.835$ as much as the line emission has increased.  This is

688: consistent with the earlier result that the X-ray illumination from our

689: line and continuum models at other phases were not correlated. 

690: 

691: % /Users/bboroson/Programs/xrb.continuum

692: 

693: \section{Doppler Tomography}

694: 

695: The Doppler tomography method, developed by Marsh and Horne (Marsh \&\

696: Horne, 1988, Marsh 2005) takes as input an emission line which is

697: broadened by line of sight motion. The line must be observed over a good

698: sampling of the binary orbit. In analogy with medical tomography, a

699: successful Doppler tomogram builds a higher-dimensional view of the object

700: from different observational slices. 

701: 

702: Analysis of 

703: Doppler shifted emission lines is prone to the confusion of different 

704: gas sources which have been projected onto the same observed velocity. 

705: The tomographic method presents a transform of the same data into

706: features separated not only by projected velocity, but also by orbital 

707: variability.

708: 

709: The output tomogram is not a physical image of an accretion disk, but 

710: rather an image in ``projected velocity space.'' A component of the 

711: emission that varies sinusoidally in velocity (but remains constant in 

712: flux) is placed at a point in velocity space 

713: along a circle with a radius equal to the orbital velocity of that 

714: emission component. The phase of the sinusoid determines its position 

715: along that circle.

716: 

717: The method assumes that all of the variation in a spectral line must be 

718: caused by the

719: orbit changing the projection of the motion into the line of sight. (For

720: example, the change in a spectrum must not be caused by a bright spot

721: being obscured by an opaque object.) Even outside of the conditions in

722: which it strictly applies, however, tomographic images can provide a

723: common view of the data and may suggest directions for further analysis. 

724: 

725: Technically, a Doppler tomogram is an inversion of

726: 

727: \begin{equation}

728: \label{eq:tomogram}

729: f(v,\phi)=\int_{-\infty}^{\infty} \int_{-\infty}^{\infty} I(v_x,v_y) 

730: g(v-v_R) d\,v_x d\,v_y

731: \end{equation}

732: 

733: which gives the flux $F(v,\phi)$ at the projected velocity $v$ and the 

734: orbital phase $\phi$ as a summation of the velocity space tomogram 

735: $I(v_x,v_y)$ with a line broadening function $g(V-V_R)$ which is usually 

736: narrow. This integral is an instance of a Radon transform. The line 

737: broadening function for our FUSE spectra 

738: is approximately 0.05\AA and for tomographic analysis we assume it is a 

739: delta function.

740: 

741: The relation between the radial velocity and the two axes of the Doppler 

742: tomogram, $v_x$ and $v_y$ is given by

743: 

744: \begin{equation}

745: \label{eq:doppler}

746: v_R=\gamma - v_x \cos 2 \pi \phi + v_y \sin 2 \pi \phi

747: \end{equation}

748: where $\gamma$ is the systemic velocity of the system.

749: 

750: The formation of the spectrum from the tomogram can be thought of as 

751: follows. The phase $\phi$ determines a direction in the velocity space 

752: plane, and moving along that line, summing the intensity perpendicular 

753: to the line, one ideally forms the spectral line. At $\phi=0.25$, one 

754: views from the right in our figures, at $\phi=0.5$ one views from the 

755: bottom, and at $\phi=0.75$ one views from the left.

756: 

757: In this paper, we determine $I(v_x,v_y)$ from $F(v,\phi)$ by means of

758: Fourier-Filtered Back Projection, which inverts 

759: Equation~\ref{eq:tomogram} through

760: \begin{equation}

761: I(v_x,v_y)=\int_{0}^{1} \bar{f}(\gamma-v_x \cos 2 \pi \phi+v_y \sin 2 \pi 

762: \phi,\phi)  d\,\phi \end{equation}

763: where to calculate $\bar{f}(v,\phi)$ from $f(v,\phi)$, one takes its 

764: Fourier transform, multiplies by a ramp filter and a Wiener (optimal) 

765: filter based on the noise level, and then takes the inverse Fourier 

766: transform.

767: 

768: Although it may seem as if this filtering complicates the method of

769: back-projection, a ramp filter (multiplying the Fourier transform by the

770: frequency) is required in order for back projection to produce a rigorous 

771: inversion of Equation~\ref{eq:tomogram}.

772: Otherwise, back projection would only

773: produce a smeared version of the true tomogram. 

774: 

775: Unfortunately, using a ramp filter also magnifies pixel to pixel noise. We

776: apply a standard Wiener filter that assumes a signal amid a noise

777: component that is constant with frequency. We find the power spectrum of

778: the line spectra is well fit by several components of the form

779: $P(\omega)=P_0 \exp(-k \omega)$ (with $k>0$).  The contribution of the

780: signal is negligible at high frequencies $\omega$, where the power

781: spectrum is fit well by a constant noise level $n$. The action of the

782: Wiener filter is then to multiply the power spectrum by

783: $P(\omega)/(n+P(\omega)$, which is nearly 1 when the signal dominates the

784: noise and negligible at high frequencies where the spectrum is almost

785: entirely noise. We used the same form of $P(\omega)$ for each line at each

786: orbital phase, and made a visual comparison between the actual power

787: spectrum and the function $P(\omega)$.  Because the filter adapts to

788: different noise levels, the amount of smoothing varies from line to line. 

789: For the lines with highest S/N, such as O\,{\sc vi}$\lambda1032$, the

790: turnover frequency for the smoothing filter is on the order of 3 pixels,

791: whereas for weaker lines such as S\,{\sc vi}$\lambda 944$, the smoothing

792: is on the order of 10 pixels. 

793: 

794: We did not apply the popular method of Maximum Entropy regularization,

795: which awards higher probabilities to those tomograms that provide the

796: least information in the sense that they are smoothest. 

797: 

798: \subsection{Applications of Doppler Tomography}

799: 

800: In Figures~7--14 we show Doppler tomograms of the emission lines of

801: S\,{\sc vi} at 933 and 944\AA, N\,{\sc iii}$\lambda 911$, O\,{\sc vi} at

802: 1031\AA\  and 1037\AA, P\,{\sc v}$\lambda 1128$, S\,{\sc iv}$\lambda1073$,

803: and C\,{\sc iii}$\lambda1175$.  The tomogram of C\,{\sc iii}$\lambda977$

804: is not shown as it is contaminated by the presence of saturated

805: absorption. We have interpolated between orbital phases for the

806: integration, which we perform between $\phi=0.2$ and $\phi=0.85$. The

807: grayscale of the tomograms extends linearly from the minimum to the

808: maximum. 

809: 

810: To understand the C\,{\sc iii} multiplet, we consult {\it 

811: CHIANTI}, a software package and atomic database described in

812: Dere et al. 1997 and Landi et al. 2006. The database, 

813: used extensively by stellar and solar astrophysicists, contains energy 

814: levels, wavelengths, radiative transition probabilities, and excitation 

815: data for many ions. The associated software package, written in IDL 

816: (Interactive Data Language), can be used to examine how line ratios vary 

817: with temperature and density, subject to certain limiting assumptions.

818: We find that C\,{\sc iii} contains 

819: emission components at $0, 85, 168, 199, 269$, and 367 km~s${-1}$ 

820: relative to the 1175.59\AA\ component. 

821: 

822: We were able to reduce the artifacts by deconvolving with a

823: triple-gaussian $g(V-V_R)$ in Equation~\ref{eq:tomogram}. We tried

824: Gaussians of equal weight at velocities $(0,+168,+367)$ km~s$^{-1}$

825: (Figure~14) and find there is a single bright spot near the Roche lobe. 

826: 

827: A further problem with the C\,{\sc iii} multiplet is that the model of the

828: continuum in this region predicts an absorption dip. To reduce artifacts,

829: we use the simpler and smoother Kurucz model atmospheres, with 1\AA\

830: resolution for all of the tomograms, instead of the models based on actual

831: stellar spectra observed with {\it FUSE}, which we have computed with

832: 0.1\AA\ resolution and which include some counting statistics noise.

833: 

834: The signal in the tomograms, except as noted above, is concentrated in a

835: peak offset from the Roche lobe of HZ~Her. The traditional signature of an

836: accretion disk is not obviously present. Accretion disks are expected to

837: cause broad line emission that appears as a ring in a tomogram, with

838: diminished flux within the velocity at the edge of the accretion disk, or

839: about 300~km~s$^{-1}$ for Hercules~X-1.  If the disk emission is

840: symmetric, we expect it to be centered on the position of the neutron star

841: in velocity space, $(vx,vy)=(0,-169)$. 

842: 

843: \subsection{Narrow Emission and Broad Absorption Components}

844: 

845: There is a simple interpretation of the tomograms in terms of P~Cygni

846: lines commonly seen in stellar winds from massive stars (and in {\it HST}

847: spectra of Her~X-1, Boroson, Kallman, \&\ Vrtilek 2001). P~Cygni lines, 

848: caused by resonance scattering, are characterized by red-shifted emission 

849: and absorption blue-shifted by velocities common in the 

850: wind. There is a hint of blue-shifted absorption in the trailed 

851: spectrograms at $\phi=0.3-0.6$, although there may be an interstellar 

852: absorption line as well at these wavelengths and the gap between the 

853: peaks of the accretion disk spectrum could also appear to be absorption.

854: 

855: We note that for many tomograms in addition to the bright spot at

856: $(vx,vy)\approx(150,-30)$, there is a dark spot at

857: $(vx,vy)\approx(-100,50)$.  (For O\,{\sc vi}$\lambda1031$, this absorption

858: may have a darker lobe near the rest wavelength from interstellar O\,{\sc

859: vi} absorption.) Near the start of the phase range used by our tomograms,

860: at $\phi\approx0.3$, the observer's line of sight comes from the lower

861: right in Figure~7. As the line of sight passes through both bright and

862: dark spots, the observer at $\phi\approx0.3$ sees a diminished central

863: peak in the spectral lines. 

864: 

865: At $\phi\approx0.5-0.6$, the viewer looks at the system from the bottom

866: of the plot, stacking pixels vertically. The result is red-shifted

867: emission (the bright spot) and blue-shifted absorption (the dark spot). 

868: 

869: Although the blue-shifted absorption at $\phi=0.3-0.6$ may be explained 

870: through the analogy of P~Cygni lines, the red-shifted emission at these 

871: phases does not fit this explanation. In a P~Cygni line, except when a 

872: very hot wind is present (as in a Wolf-Rayet star), all the emission 

873: is caused by spherically symmetric scattering, so that 

874: conservation of photons requires that the forward-scattered flux be 

875: nearly equal to the back-scattered (``absorbed'') flux (some portion of 

876: the forward-scattered emission is blocked by the star itself). 

877: In contrast, any blue-shifted absorption in the Her~X-1 spectrum is only 

878: a small fraction of the emission. Further, P~Cygni emission is only 

879: red-shifted because it overlaps with the absorption on the blue end; the 

880: actual emission spectrum should be symmetric about zero velocity.

881: 

882: %From the weakness of the lines seen at mid-eclipse 

883: %(Anderson et al. 1994), we expect that the wind outside of this cylinder

884: %will contribute $\sim1$~\%\ of the flux at maximum light.

885: 

886: It seems most natural to associate the bright spot with the X-ray heated 

887: face of the normal star, as it follows roughly the same light curve and 

888: appears near the star and far from the disk on the tomograms. Yet 

889: the placement of the spot, or equivalently, the orbital variation with 

890: velocity, cannot be accounted for easily by such a model.

891: 

892: In Figure~15, we show the observed spectrum near O\,{\sc vi} (bold), the

893: spectrum expected if the illuminated surface of HZ~Her emits O\,{\sc vi}

894: Doppler shifted by the local rotational velocity (dashed), and an

895: empirical Gaussian model of the emission lines. The Doppler shifts

896: predicted from the rotation of the surface of HZ~Her are less than

897: observed in the narrow emission lines. 

898: 

899: To model the narrow O\,{\sc vi} emission lines we adapt our model of the

900: continuum emission. This model takes into account the X-ray shadow cast by

901: the disk and the eclipse of portions of the normal star by the disk.  The

902: reprocessed O\,{\sc vi} emission is proportional to the incident X-ray 

903: flux. We

904: also experimented with a model in which the X-ray illumination was

905: non-isotropic, illuminating HZ~Her with greater luminosity at $\phi=0.25$

906: than at $\phi=0.75$. This model still does not account for the observed

907: velocities. 

908: 

909: We also made an empirical model of the lines as Gaussians. The flux in the

910: Gaussian line emission at each phase was proportional to the flux in the

911: stellar component of our continuum model using Kurucz model spectra. The 

912: central velocity in this model varies

913: sinusoidally with phase. We added Gaussian absorption to correspond to

914: features seen in {\it HST} STIS spectra of the N\,{\sc v}$\lambda1240$,

915: Si\,{\sc iv}$\lambda1400$, and C\,{\sc iv}$\lambda1550$ doublets near 

916: $\phi=0.3-0.6$.

917: We set the maximum projected redshift of the emission lines and blueshift

918: of the absorption to occur near $\phi=0.5$, but delayed by $\phi=0.06$,

919: corresponding to a deflection of a wind by the Coriolis effect. The

920: absorption line has maximum covering fraction near $\phi=0.5$ that

921: smoothly varies to 0 at $\phi=0$. (Any wind is probably confined to the 

922: cylinder from the star towards the disk, as the emission lines seen in 

923: mid-eclipse are only $\sim1$\%\ of the peak line strength, as observed 

924: with the {\it HST} FOS by Anderson et al. 1994). Both emission and 

925: absorption components

926: have constant width. Both emission and absorption components vary about a

927: central velocity given by the L1 point. 

928: 

929: When these features are added to the model of the disk spectrum in 

930: \S6, assumed to move with the neutron star's 

931: 169~km~s$^{-1}$ orbit, and random counting statistics noise is added, the 

932: resulting tomogram (Figure~16) resembles the observed tomogram, while 

933: showing only a hint of the presence of an accretion disk. Without the 

934: assumed phase delay of $\Delta \phi=0.06$, the bright spot would appear 

935: rotated to $vy=0$.

936: 

937: \section{Line Ratio Diagnostics}

938: 

939: The {\it FUSE} spectra show emission lines that may serve as diagnostics

940: of density, temperature, or optical depth.

941: 

942: \subsection{C\,{\sc iii} line diagnostics}

943: 

944: The simultaneous measurement of C\,{\sc iii}$\lambda977$ and

945: C\,{\sc iii}$\lambda1176$ can provide us with an important diagnostic. 

946: The C\,{\sc iii} 1176/977 ratio is sensitive to

947: density in the range up to about $n_e=10^{11}$~cm$^{-3}$. At higher

948: densities (up to $n_e\sim10^{15}$), the populations of the ground and

949: metastable levels are entirely determined by collisions.  In that

950: regime, the 1176/977 ratio depends only on temperature and the optical

951: depths in the lines.

952: 

953: We ignored flux from the {\it FUSE} LiF2B detector near the C\,{\sc

954: iii}$\lambda977$ line, even though this detector recorded flux down to

955: exactly 977\AA. The edge of the detector causes the flux measured in this

956: region was significantly lower than the other two detectors, so that

957: including this data would have created an artificial dip at wavelengths

958: $\lambda>977$\AA. 

959: 

960: The fluxes we measure for C\,{\sc iii}$\lambda977$ are lower limits

961: because of a saturated interstellar absorption line at the rest

962: wavelength.  We correct the fluxes for interstellar reddening using

963: E(B-V)=0.018. 

964: 

965: We compare the fluxes of the two lines near $\phi=0.6$, when the fluxes

966: peak. If the C\,{\sc iii}$\lambda977$ line behaves similarly to the

967: O\,{\sc vi} lines, then the bright narrow emission will, near this phase,

968: have the greatest redshift. If so, the line may be redshifted away from

969: the interstellar absorption and the flux measurement may be more accurate

970: near this phase. If the O\,{\sc vi}$\lambda1032$ line were subject to

971: similar interstellar absorption lines as the C\,{\sc iii}$\lambda977$

972: line, the flux at $\phi=0.6$ would be diminished by $\sim$20\%. 

973: 

974: If we measure the raw C\,{\sc iii}$\lambda977$ flux from the two phases

975: surrounding $\phi=0.6$ to be $7\pm1\times10^{-13}$~erg~s$^{-1}$~cm$^{-2}$,

976: then we find a 1175/977 ratio of $0.82\pm0.16$. If we measure the C\,{\sc

977: iii}$\lambda977$ flux at exactly the peak at $\phi=0.6$ and attempt to

978: correct for interstellar absorption lines, we find a ratio of

979: 0.60$\pm0.09$. 

980: 

981: These ratios are at the high end of those predicted by the {\it CHIANTI}

982: software, which uses atomic data from the {\it CHIANTI} database and

983: computes line ratios, given certain simplifying assumptions (for example,

984: the lines are optically thin and the gas is collisionally ionized and not

985: photoionized). If we consider a density of $n_e=10^{13.4}$~cm$^{-3}$, the

986: lower ratio found above would restrict the temperature to

987: $T>6\times10^4$~K. 

988: 

989: The ratio is rendered uncertain observationally by the saturated 

990: interstellar absorption line, and theoretically

991: because of optical depth effects. Raymond (1993) presented models of line

992: emission from an X-ray illuminated accretion disk which took into account

993: optical depth through an escape probability formalism. One version of the

994: model, ``COS", assumed cosmic abundances while the other, ``CNO", assumed

995: that abundances had been altered by CNO processing. The CNO models should

996: be more appropriate for Her~X-1.  In the two cases the $I(1176)/I(977)$

997: ratios were 1.3 and 1.1, respectively. The observed emission at 

998: $\phi=0.6$ may, however, have an origin not on the disk but on the 

999: normal star. 

1000: 

1001: \subsection{Bowen fluorescence and N\,{\sc iii}$\lambda991$ emission}

1002: 

1003: The Bowen Fluorescence process (Schachter et al., 1989) arises because of

1004: the nearly perfect coincidence of the He\,{\sc ii} Ly$\alpha$ and the

1005: O\,{\sc iii} 2p$^2$ -- 2p3d resonance line ($\lambda$304), 

1006: resulting in

1007: O\,{\sc iii} near--UV primary cascades at $\lambda\lambda$3133, 3444 and

1008: secondary cascades at $\lambda$374 back to the ground state.  If

1009: conditions are right, an additional fluorescence occurs, since the

1010: \ion{O}{3}$\lambda$374 line is almost coincident with the two \ion{N}{3}

1011: 2p -- 3d resonance lines, resulting in N III optical primary cascades at

1012: $\lambda$4634, 4641, 4642.  The emission in all of these cascades is

1013: completely dominated by the Bowen process; detection of any of them is a

1014: clear confirmation.  We note that in Her~X-1 the process requires

1015: substantial optical depth ($\sim10^6$) in the \ion{He}{2} Ly$\alpha$ and

1016: \ion{O}{3} $\lambda$374 {\em pumping} lines, because otherwise they will

1017: simply escape without conversion to a Bowen line. 

1018: 

1019: In the Sun, Raymond (1978) discovered \ion{O}{3}$\lambda$304 Bowen

1020: emission.  It was found that measurements of \ion{O}{3}$\lambda$703, a

1021: {\em ground-state} Bowen cascade which, therefore, can also be produced

1022: by collisional excitation, can serve as a density diagnostic, and compared

1023: favorably with other solar estimates of $n_e$.  

1024: 

1025: We may use N\,{\sc iii}$\lambda$990, an analogous ground--state Bowen 

1026: cascade produced as a result of the \ion{N}{3} primary cascades

1027: (``$\lambda$4640''; all other \ion{O}{3} and \ion{N}{3} ground-state

1028: cascades lie below the Lyman limit and hence are unobservable).

1029: 

1030: %\begin{equation}\label{eq:bowen}

1031: %n_e = 1.92 \times 10^{17} [I(\lambda

1032: %990)/I(\lambda 4640) - 0.75(4640/990)] \end{equation} which takes into

1033: %account collisional excitation and de-excitation. 

1034: 

1035: The intensity of N\,{\sc iii}$\lambda990$ resulting from Bowen 

1036: fluorescence is related to the intensity of the Bowen lines at 4640\AA by

1037: \begin{equation}

1038: I_f(991)=(4640/991) B(\mbox{3p2p}) I(4640)

1039: \end{equation}

1040: where $B(\mbox{3p2p})$ is the branching ratio of 2p-2s$^2$p$^2$ 2D versus

1041: 3p-3s. From CHIANTI, we take $B(\mbox{3p2p})=0.55$ and 

1042: $B(\mbox{3d3p})=0.0058$.

1043: 

1044: The remainder of the line intensity is therefore from collisional 

1045: excitation and is given by $I_c(991)=I(991)-I_f(991)$.

1046: 

1047: Taking into account the reddening toward Her~X-1 implied by E(B-V)=0.018,

1048: we find a peak flux in the 991\AA\ line of $8.6\pm0.7\times10^{-13}$

1049: erg~s$^{-1}$~cm$^{-2}$. Non-simultaneous measurements of the optical

1050: $\lambda$4640 line have a peak flux at similar orbital phases of

1051: $7.0\pm0.2 \times 10^{-14}$ erg~s$^{-1}$~cm$^{-2}$ (Still et al. 1997).

1052: The optical flux varies from orbit to orbit, and, moreover, probably

1053: includes contributions from C\,{\sc iii} as well as N\,{\sc iii}.

1054: 

1055: These values imply that 20\% of the N\,{\sc iii}$\lambda991$ line flux is 

1056: the result of the Bowen fluorescence mechanism.

1057: 

1058: We attempt to relate the density to the Bowen flux by

1059: \begin{eqnarray}

1060: \label{eq:bowen}

1061: \frac{I_c(991)}{I_f(991)}  & = & 4\\

1062: \nonumber & = & 

1063:  \frac{n_e N(\mbox{N\,{\sc iii}}) q}{B(\mbox{3d3p}) B(\mbox{3p2p})

1064: N(\mbox{N\,{\sc iii}}) \sigma I(374)}\\

1065: \nonumber & = & 4.9\times10^8 n_e / I(374)

1066: \end{eqnarray}

1067: with $N(\mbox{N\,{\sc iii}})$ the density of N\,{\sc iii} and 

1068: $q=6\times10^{-8}$~s$^{-1}$ 

1069: the excitation rate of the 991\AA\ transition at T=30,000~K.

1070: 

1071: We estimate an upper bound on the O\,{\sc iii}$\lambda374$ intensity (the 

1072: line

1073: may be optically thick) from the O\,{\sc iii}$\lambda3132$ line observed with

1074: the Faint Object Spectrograph on {\it HST} by Anderson et al. (1994). We

1075: account for the other O\,{\sc iii} branches leading to the $\lambda374$ 

1076: line by

1077: using the observed O\,{\sc iii} Bowen spectrum of RR~Tel (Selvelli, Danziger,

1078: \&\ Bonifacio 2007). We then estimate $I(374)\lae 0.032 (D/R)^2$, with

1079: $D=6.6$~kpc the distance to the system and $R$ the radius of the emitting

1080: region. Boroson et al. (2000) found 

1081: $\log n_e=13.4\pm0.2$ 

1082: for the narrow lines from the ratios of lines observed with {\it HST},

1083: and Howarth \&\ Wilson (1983b) found $\log n_e=13.3$ from {\it IUE} 

1084: observations.

1085: Combining with Equation~\ref{eq:bowen}, we find $R\lae 7 \times

1086: 10^{10}$~cm. 

1087: 

1088: This is smaller than the size of the accretion disk 

1089: ($R_{\rm outer}\approx2\times10^{11}$~cm). We note that the tomograms show 

1090: emission from a small region in velocity space.

1091: 

1092: %bram

1093: 

1094: Another test is provided by a comparison with theoretical models of

1095: N\,{\sc iii}$\lambda991$ emission in the absence of the Bowen process. 

1096: Raymond

1097: (1993) found ratios between the N\,{\sc iii} line and C\,{\sc 

1098: iii}$\lambda977$ 

1099: for the COS model of $I(991)/I(977)=0.34$ and for the CNO model of

1100: $I(991)/I(977)=0.53$. The observed ratio of $0.75\pm0.10$

1101: requires enhancement by Bowen fluorescence and provides an excellent 

1102: match to the CNO model if only 70\%\ of the N\,{\sc iii}$\lambda991$ line 

1103: results from collisional excitation.

1104: 

1105: 

1106: \section{Discussion}

1107: 

1108: The far UV spectrum of Hercules~X-1 shows clear evidence for a Keplerian 

1109: accretion disk. A simple model such a disk can fit the broad O\,{\sc vi} 

1110: emission lines near eclipse. The origin of brighter, narrow emission 

1111: lines is still unknown. Doppler tomograms place this emission apart from 

1112: the Roche lobe of HZ~Her.  

1113: 

1114: The bright narrow emission component generally follows the flux expected 

1115: from the illuminated

1116: portion of the normal star. However, it appears brighter than expected at

1117: $\phi=0.2$ and the velocity excursion is also greater than is accounted

1118: for by the rotation of the Roche lobe. A dense portion of a stellar wind

1119: that moves along the surface of the star, or that originates at the L1

1120: point and flows back towards the star, would be red-shifted at $\phi=0.5$

1121: and blue-shifted at $\phi=0.25$ and $\phi=0.75$. Although such a model can

1122: be made to reproduce the gross features of the line variation, there

1123: remain too many possible alternate explanations for this to be convincing.

1124: There could be a source of variable emission, absorption, or

1125: scattering in addition to the normal star, for example, the gas stream or

1126: the surface of the disk. 

1127: 

1128: In the presence of stellar winds, possibly transient or restricted in

1129: solid angle, tomography did not show a clear signature of an accretion

1130: disk in Hercules X-1. 

1131: 

1132: Although we found that a simple model of an accretion disk generally fit the

1133: spectra during eclipse ingress and egress, there was an anomalous

1134: observation in which the O\,{\sc vi} lines brightened as the disk eclipse

1135: progressed. The phase $\phi=0.876$ is in the $\phi=0.8-0.9$ range in which

1136: historically ``pre-eclipse dips'' of X-rays have been observed, and it is

1137: tempting to relate the temporary brightening of the lines to a physical

1138: cause or result of the dips. 

1139: 

1140: Alternately, an anomalous change to the emission lines could result from 

1141: a change in shadowing of the X-ray emission. Boroson et al. (2001) 

1142: concluded that anomalous {\it absorption} within the 

1143: N\,{\sc v}$\lambda\lambda 1238,1242$ doublet changed too rapidly to result 

1144: from material passing over the line of sight. The absorption changed so

1145: rapidly probably because of a rapid change in  

1146: the shadowing of the X-rays which ionize the gas.

1147: 

1148: Future X-ray spectroscopy missions such as {\it Constellation X} may be

1149: able to resolve lines from more highly ionized species than the UV and Far

1150: UV resonance lines (Vrtilek et al. 2004).

1151: If more highly ionized

1152: species track disk material better and are not as prevalent in the narrow

1153: line region, we may be able to make reliable tomograms of the disk in this

1154: or similar systems. Hydrodynamic models of this system may be required to

1155: cut down on the vast parameter space of possible gas flows responsible for

1156: the spectral signatures observed here. 

1157: 

1158: \acknowledgements

1159: 

1160: We would like to thank Jeff Bryant for helping us develop some {\it 

1161: Mathematica} routines for Doppler tomography.  CHIANTI is a collaborative 

1162: project involving the NRL (USA), RAL (UK), MSSL (UK), the Universities of 

1163: Florence (Italy) and Cambridge (UK), and George Mason University (USA).

1164: We would like to thank the anonymous referee for suggestions that 

1165: improved the text and presentation, for pointing us to the S\,{\sc 

1166: vi}$\lambda\lambda 933,944$ lines, and prompting us to analyze the 

1167: interstellar molecular absorption lines.

1168: 

1169: \appendix

1170: 

1171: \section{Details of the Model of the Her X-1 Continuum}

1172: 

1173: With one free parameter, the mass accretion rate $\dot{M}$, we model 

1174: the continuum emission from the accretion disk and X-ray illuminated face 

1175: of the normal star. We do this not by means of radiative transfer 

1176: calculations, but by calculating the heated temperatures of both surfaces, 

1177: and co-adding spectra appropriate for those temperatures, either 

1178: blackbody spectra, models of stellar spectra, or actual spectra of hot 

1179: stars observed with {\it FUSE}.

1180: 

1181: Our model of the far UV continuum emission of Her~X-1 follows closely the 

1182: methods of Vrtilek et al. (1990) and Cheng, Vrtilek, \&\ Raymond (1995).  

1183: We describe the method here in detail, including the minor departures we 

1184: have made to extend the spectral simulation to the far UV. 

1185: 

1186: The model of the accretion disk temperature as a function of radius, $T(r)$, 

1187: includes the effects of both internal heating due to viscous forces and 

1188: heating from X-rays from the neutron star.  

1189: 

1190: When we calculate the disk temperature as a function of radius, $T(r)$, 

1191: we include both the heat  from viscous forces and X-rays from the 

1192: neutron star.  The local energy generated by accretion is given by

1193: \begin{equation}

1194: \sigma T_0(r)^4 = \frac{3 G M \dot{M}}{8 \pi r^3}\left( 

1195: 1-\sqrt{\frac{R}{r}}\right)

1196: \end{equation}

1197: for Stefan-Boltzmann constant $\sigma$, neutron star mass $M$ and radius 

1198: $R$, and disk mass accretion rate $\dot{M}$ (Shakura \&\ Sunyaev 1973).

1199: 

1200: For local energy balance, the energy emitted must equal the energy 

1201: generated plus the X-ray energy absorbed:

1202: \begin{equation}

1203: \sigma T_{\rm eff}^4(r) = \sigma T_0(r) + \frac{a L_x}{4 \pi r}

1204: \frac{\partial (h/r)}{\partial r}

1205: \end{equation}

1206: where we choose the albedo $a=0.5$, and the disk height at radius $r$ is 

1207: $h(r)$.  The X-ray luminosity is related to the gravitational potential 

1208: released by accreting matter:

1209: \begin{equation}

1210: L_{\rm x}=0.5 \frac{G M \dot{M}}{R}

1211: \end{equation}

1212: 

1213: These equations, together with vertical hydrostatic equilibrium, can be 

1214: solved numerically to determine $T(r)$. The surface of the noncollapsed 

1215: star is also assumed to have an albedo $a=0.5$.

1216: 

1217: At each small region in the accretion disk, the temperature $T(r)$ is

1218: compared with the critical disk temperature, $T_{dc}$. If $T(r)>T_{dc}$,

1219: then we assume that the disk region radiates like a blackbody. If

1220: $T(r)<T_{dc}$, then we interpolate through our library of actual stellar

1221: spectra to find a spectrum appropriate for temperature $T(r)$.  From

1222: experience fitting {\it HST} FOS spectra and fitting the Balmer jump, we

1223: fix $T_{dc}=10000$~K so that in practice the disk almost always has

1224: blackbody spectra, except at its edge where it is not illuminated by

1225: X-rays. 

1226: 

1227: The stellar library from 1150 to 7500\AA\ is described in Cannizzo \&\ Kenyon

1228: (1987).  The near UV spectra were obtained with {\it IUE} and are

1229: described in Wu et al. (1982). All the spectra are normalized to have

1230: constant flux in the V bandpass and are then scaled to their absolute

1231: magnitudes using their V-R colors and the Barnes-Evans relation (Barnes,

1232: Evans, \&\ Moffett 1978). 

1233: 

1234: For this paper, we have extended the library of stellar spectra further

1235: into the UV by using {\it FUSE} spectra of main sequence stars between

1236: O\,9.5 and A\,7.  These observations are described briefly in

1237: Table~5.  Stars of temperature greater than $T_{\rm 

1238: sc}=18,900$~K 

1239: are not used in the fits to Her~X-1 and are not included in the table. We 

1240: have interpolated over those wavelengths we

1241: expect are affected by interstellar atomic or molecular absorption, and we

1242: have applied a filter designed to smooth interstellar lines while

1243: retaining stellar features.  We de-reddened each spectrum and scaled the

1244: flux according to the star's visual magnitude V.  We have tested the

1245: method and the use of these stellar spectra by using model stellar spectra

1246: as predicted by Kurucz (1979). Plots of far UV spectra in our library, 

1247: normalized to have a constant V magnitude, are shown in Figure~17.

1248: 

1249: For the purpose of computing the disk shadow, the disk is assumed to have a 

1250: fixed opening angle and a fixed tilt 

1251: from the orbital inclination, although the direction of the disk normal

1252: precesses with the X-ray cycle in a manner described by Gerend \&\ Boynton

1253: (1976) and Howarth \&\ Wilson (1983a).  Because of the finite disk 

1254: opening angle, the outer regions of the disk can

1255: occult the inner regions, and the disk edge itself can radiate, although

1256: it is not illuminated by the central X-rays. 

1257: 

1258: The noncollapsed star is assumed to fill its Roche lobe, and its shape,

1259: visibility, and illumination by the X-ray source are treated according to

1260: methods evolved from Wilson \&\ Devinney (1971) to Howarth \&\ Wilson

1261: (1983a). Following those references, we model the eclipse of the disk using

1262: the Roche potential. We take into account both limb darkening and the

1263: gravity-darkening appropriate for a late-type star. If the

1264: temperature of some region on the surface of the star, after X-ray

1265: illumination, is greater than some critical temperature $T_{sc}$, we

1266: assume the region emits as a blackbody. If the local temperature is less

1267: than $T_{sc}$, we again interpolate through our stellar library to find

1268: the appropriate stellar spectrum with which the spot radiates.  From

1269: earlier experience fitting Her~X-1 UV spectra, we fix $T_{sc}=18900$ (Cheng,

1270: Vrtilek, \&\ Raymond 1995). 

1271: 

1272: The model, as currently realized, does not account for emission or 

1273: shadowing by a gas stream between the noncollapsed star and disk.

1274: 

1275: \section{Details of the Model of the Disk Lines in Eclipse}

1276: 

1277: Accretion disks in theory have emission lines with a double-peaked shape,

1278: with peaks separated by a velocity given approximately by the projection

1279: into the line of sight of the orbital velocity at the edge of the

1280: accretion disk. For Hercules~X-1, with typical neutron star mass, an 

1281: nearly edge-on inclination, and an outer disk radius of $R\approx 

1282: 2\times10^{11}$~cm, we expect the peaks to occur at $\pm 

1283: 300$~km~s$^{-1}$. While greater velocities are found within the outer disk

1284: radius, each ring at a fixed radius emits down to a Doppler velocity of 0

1285: where the gas moves tangential to the line of sight. Regions of high 

1286: Doppler velocity arise from progressively smaller regions in the disk, 

1287: and contribute less to the overall line shape.

1288: 

1289: We have calculated detailed models of emission from an accretion disk and

1290: have fit these to the profiles observed in Her~X-1 during eclipse ingress

1291: and egress. In reality, the disk is likely to be warped, and in future

1292: work we will test individual models of this warped shape. The importance

1293: of the demonstration in this paper is that it shows how easy it is to get

1294: general agreement with the observed line profiles given the standard

1295: picture of the accretion disk. 

1296: 

1297: The method in detail follows that of Horne (1995), which allows for 

1298: nonisotropic turbulence and for Keplerian shear, that is, the local 

1299: dispersion in disk velocity as a result of the Keplerian flow itself. This 

1300: introduces free parameters for the ``Mach matrix" describing the 

1301: turbulent flow, in addition to the power law exponent that we use to 

1302: as a phenomenological description the strength of the emission line as a 

1303: function of radius in the disk. For simplicity, we keep the elements of 

1304: the Mach matrix constant with radius in the disk.

1305: 

1306: Following Horne (1995), the local line profile is given by

1307: \begin{equation}

1308: I_\nu \propto (1-e^{-\tau_\nu})

1309: \end{equation}

1310: where the optical depth $\tau_\nu$ at frequency $\nu$ is given by

1311: \begin{equation}

1312: \tau_\nu=\tau_0 e^{(V-V_0)^2/2\Delta V^2}

1313: \end{equation}

1314: for a given line center optical depth $\tau_0$.

1315: 

1316: The nonisotropic turbulence and shear enter through $\Delta V$, given by

1317: \begin{equation}

1318: \Delta V^2  =  \Delta V^2_{\rm therm} + \Delta V^2_{\rm shear}

1319: 	+ \Delta V^2_{\rm turb}

1320: \end{equation}

1321: or

1322: \begin{eqnarray}

1323: \Delta V^2/C^2_s & = &  \frac{\mu}{\gamma A}+(Q \sin i \tan i \sin

1324: 2 \theta)^2

1325: +\sin^2 i(M_{RR} \cos^2\theta-2 M_{R\theta} \cos \theta \sin \theta\\

1326: \nonumber & + & M_{\theta \theta} \sin^2 \theta)+2 \sin i \cos i (M_{Z R} 

1327: \cos \theta - M_{Z \theta} \sin \theta)+ \cos^2 i M_{ZZ}\\

1328: \end{eqnarray}

1329: %\begin{equation}

1330: %\Delta V^2 =  \Delta V^2_{\rm therm} + \Delta V^2_{\rm shear}

1331: %	+ \Delta V^2_{\rm turb}\\

1332: %+  C^2_s \left[ \frac{\mu}{\gamma A}+(Q \sin i \tan i \sin

1333: %2 \theta)^2

1334: %+\sin^2 i(M_{RR} \cos^2\theta-2 M_{R\theta} \cos \theta \sin \theta 

1335: %+M_{\theta \theta} \sin^2 \theta)\\

1336: %+  2 \sin i \cos i (M_{Z R} \cos \theta - M_{Z \theta} \sin 

1337: %\theta)+ \cos^2 i M_{ZZ}\right]\\

1338: %\end{equation}

1339: where $i$ is the orbital inclination, $\mu$ is the mean molecular weight, 

1340: $\gamma=5/3$, and $A$ is the atomic weight. The shear parameter $Q$ is 

1341: given by

1342: \begin{equation}

1343: Q=\frac{3}{4}\frac{H}{R}\frac{V_{\rm Kep}}{C_s}\frac{\Delta Z}{2 H}

1344: \end{equation}

1345: and, for simplicity, is set to $3/4$, the maximum value for an optically 

1346: thin accretion disk.

1347: 

1348: The Mach matrix is given by the correlations between components of 

1349: turbulence in different cylindrical coordinates:

1350: \begin{equation}

1351: M_{ij}\equiv \frac{\langle \delta V_i \delta V_j \rangle}{C_s^2}

1352: \end{equation}

1353: where $C_s$ is the sound speed.

1354: 

1355:  \clearpage

1356: 

1357: \begin{references}

1358: 

1359: Abgrall, H., Roueff, E., Launay, F., Roncin, J.Y., \&\ Subtil, J.L.

1360: 1993a, A\&AS, 101, 273

1361: 

1362: Abgrall, H., Roueff, E., Launay, F., Roncin, J.Y., \&\ Subtil, J.L.

1363: 1993b, A\&AS, 101, 323

1364: 

1365: Anderson, S.F., Wachter, S., Margon, B., Downes, R.A., Blair, W.P. 

1366: Halpern, J.P. 1994, ApJ, 436, 319

1367: 

1368: Arons, J. 1973, ApJ, 184, 539

1369: 

1370: Barnes, T.G., Evans, D.S., \&\ Moffett, T.J. 1978, MNRAS, 183, 285

1371: 

1372: Basko, M.M., Sunyaev, R.A., Hatchett, S., \&\ McCray, R. 1977, ApJ, 215, 

1373: 276

1374: 

1375: Begelman, M.C., McKee, C.F., \&\ Shields, G.A. 1983, ApJ, 271, 70

1376: 

1377: Begelman, M.C., \&\ McKee, C.F. 1983, ApJ, 271 89

1378: 

1379: Bohlin, R.C. 1975, ApJ, 200, 402

1380: 

1381: Boroson, B., Vrtilek, S.D., McCray, R., Kallman, T., \&\ Nagase, F. 1996,

1382: ApJ, 473, 1079

1383: 

1384: Boroson, B., Blair, W.P., Davidsen, A.R, Vrtilek, S.D., Raymond, J., 

1385: Long, K.S., McCray, R. 1997, ApJ, 473, 1079

1386: 

1387: Boroson, B., Kallman, T.R., \&\ Vrtilek, S.D. 2001, ApJ, 562, 925

1388: 

1389: Boroson, B., Kallman, T., Vrtilek, S.D., Raymond, J., Still, M., 

1390: Bautista, M., \&\ Quaintrell, H. 2000, ApJ, 529, 414

1391: 

1392: Cannizzo, J.K., \&\ Kenyon, S.J. 1987, ApJ, 320, 319

1393: 

1394: Cardelli, J.A., Clayton, G.C., \&\ Mathis, J.S. 1989, ApJ, 345, 245

1395: 

1396: Cheng, F.H., Vrtilek, S.D., \& Raymond, J.C. 1995, ApJ, 452, 825

1397: 

1398: Chiang, J. 2001, ApJ, 549, 537

1399: 

1400: Crosa, L, \&\ Boynton, P.E. 1980, ApJ, 235, 999

1401: 

1402: Davidson, K., \&\ Ostriker, J.P. 1973, ApJ, 179, 585

1403: 

1404: Deeter, J.E., Boynton, P.E., Miyamoto, S., Kitamoto, S., Nagase, F., \&\ 

1405: Kawai, N. 1991, ApJ, 383, 324

1406: 

1407: Dere, K.P., Landi, E., Mason, H.E., Monsignori Fossi, B.C., Young, P.R. 

1408: 1997, A\&AS, 125, 149

1409: 

1410: Feldman, P.D., Sahnow, D.J, Kruk, J.W., Murphy, E.M., Moos, H.W. 2001, 

1411: JGR, 106, 8119

1412: 

1413: Gerend, D. \&\ Boynton, P.E. 1976, ApJ, 209, 562

1414: 

1415: Giacconi, R., Gursky, H., Kellog, E., Levinson, R., Schreier, E., \&\ 

1416: Tananbaum, H. 1973, ApJ, 184, 227

1417: 

1418: Gillmon, K., Shull, J.M., Tumlinson, J., \&\ Danforth, C. 2006, ApJ, 636, 891

1419: 

1420: Horne, K. 1995, A\&A, 297, 273

1421: 

1422: Howarth, I.R., \&\ Wilson, R. 1983a, MNRAS, 202, 347

1423: 

1424: Howarth, I.R., \&\ Wilson, R. 1983b, MNRAS, 204, 1091

1425: 

1426: Jimenez-Garate, M.A., Hailey, C.J., den Herder, J.W., Zane, S., Ramsay, 

1427: G. 2002, ApJ, 578, 391

1428: 

1429: Jimenez-Garate, M.A., Raymond, J.C., Liedahl, D.A., \&\ Hailey, C.J. 

1430: 2005, ApJ, 625, 931

1431: 

1432: Ko, Y.-K., \&\ Kallman, T.R. 1994, ApJ, 431, 273

1433: 

1434: Kurucz, R.L. 1979, ApJS, 40, 1

1435: 

1436: Landi, E., Del Zanna, G., Young, P.R., Dere, K.P., Mason, H.E., Landini, 

1437: M. 2006, ApJSS, 162, 261

1438: 

1439: London, R., McCray, R., \&\ Auer, L.H. 1981, ApJ, 243, 970

1440: 

1441: McCandliss, S.R. 2003, PASP, 115, 651

1442: 

1443: Maloney, P.R., \&\ Begelman, M.C. 1997, ApJ, 491, L43

1444: 

1445: Marsh, T.R., \&\ Horne, K. 1988, MNRAS, 235, 269

1446: 

1447: Meyer, F., \& Meyer-Hofmeister, E. 1984, A\&A, 104, L35

1448: 

1449: Middleditch, J., \&\ Nelson, J. 1976, ApJ, 208, 567

1450: 

1451: Moos, H.W., Cash, W.C., Cowie, L.L., Davidsen, A.F., Dupree, A.K.,

1452: Feldman, P.D., Friedman, S.D., Green, J.C., et al. 2000, ApJ, 

1453: 538, L1

1454: 

1455: Raymond, J.C. 1993, ApJ, 412, 267

1456: 

1457: Raymond, J.C. 1978, ApJ, 224, 259

1458: 

1459: Reynolds, A.P., Quaintrell, H., Still, M.D., Roche, P., Chakrabarty, D., 

1460: \&\ Levine, S.E. 1997, MNRAS, 288, 43

1461: 

1462: Sahnow, D.J., Moos, H.W., Ake, T.B., Andersen, J., Anderson, B.-G.,

1463: Andre, M., Artis, D., Berman, A.F., Blair, W.P., Brownsberger, K.R., et 

1464: al. 2000, ApJ, 538, L7

1465: 

1466: Schachter, J., Filippenko, A.V., \&\ Kahn, S.M.  1989, ApJ,

1467: 340, 1049

1468: 

1469: Scott, D.M., \&\ Leahy, D.A. 1999, ApJ, 510, 974

1470: 

1471: Scott, D.M., Leahy, D.A., \&\ Wilson, R.B. 2000, ApJ, 539, 392

1472: 

1473: Selvelli, P., Danziger, J., \&\ Bonifacio, P. 2007, A\&A, 464, 715

1474: 

1475: Shakura, N.I., \&\ Syunyaev, R.A. 1973, A\&A, 24, 337

1476: 

1477: Still, M.D., Quaintrell, H., Roche, P., \&\ Reynolds, A.P. 1997, MNRAS, 

1478: 292, 52

1479: 

1480: Tananbaum, H., Gursky, H., Kellogg, E.M., Levinson, R., Schreier, E., 

1481: Giacconi, R. 1972, ApJ, 174, L 143

1482: 

1483: Vrtilek, S.D., \&\ Cheng, F.H. 1996, ApJ, 465, 915

1484: 

1485: Vrtilek, S.D., Quaintrell, H., Boroson, B., Still, M., Fiedler, H., 

1486: O'Brien, K., McCray, R. 2001, ApJ, 549, 522

1487: 

1488: Vrtilek, S.D., Quaintrell, H., Boroson, B., Shields, M. 2004, AN, 325, 

1489: 209 

1490: 

1491: Vrtilek et al. 1990. A\&A, 235, 162

1492: 

1493: Wijers, R.A.M.J, \&\ Pringle, J.E. 1999, MNRAS 308, 207

1494: 

1495: Wilson, R.E. \&\ Devinney, E.J. 1971, ApJ, 166, 605

1496: 

1497: Wu, C.-C., Boggess, A., Bohlin, R.C., Holm, A.V., 

1498: Schiffer III, F.H., \&\ Turnrose, B.E. 1982, UVSC Conference, 25

1499: 

1500: \end{references}

1501: 

1502: \clearpage

1503: 

1504: \begin{table}

1505: \begin{tabular}{llll}

1506: Symbol & Adopted value & Meaning & Reference\\

1507: \hline

1508: $q$ & 0.58 & Binary mass ratio & Howarth \&\ Wilson 1983a\\

1509: $M_{\rm ns}$ & 1.4 & Mass of neutron star (M$_\odot$) & \\

1510: $R_{\rm inner}$ & 10$^6$ & Inner radius of accretion disk (cm) & Cheng, 

1511: Vrtilek, \&\ Raymond 1995\\

1512: $R_{\rm outer}$ & 2$\times10^{11}$ & Outer radius of accretion disk (cm) 

1513: & CVR\\

1514: $a$    & 0.5 & Albedo & CVR\\

1515: $D$ & 6600 & Distance to system (pc) & Reynolds et al. 1997\\

1516: E$_{\rm B-V}$ & 0.018 & Reddening parameter & Boroson et al. 2001\\

1517: T$_*$ & 8100 & Polar temperature of normal star (K) & CVR\\

1518: $\Delta \Psi$ & $0.8105$ & Disk precession phase offset & H\&W \\

1519:  $\theta_{\rm d}$ & 4.87 & Disk thickness (degrees) & H\&W \\

1520: $\alpha_{\rm d}$ & 28.72 & Disk tilt from orbital plane (degrees) & H\&W \\

1521: $a_*$ & 6.35$\times 10^{11}$ & Orbital separation, centers of mass (cm) & 

1522: H\&W \\ $i$ & 81.25 & Orbital inclination (degrees) & H\&W \\

1523: \end{tabular}

1524: \caption{Parameters describing the X-ray binary Hercules X-1}

1525: \end{table}

1526: 

1527: \begin{table}

1528: \begin{tabular}{lllcrl}

1529: Root name & MJD (start) & MJD (end) & Orbital Phase & Exptime (s) 

1530: & Comments\\

1531: \hline

1532: B0080101001  &  52069.32464 &  52069.36421 & 0.182562 & 3418 & \\

1533: B0080101002  & 52069.39311 &  52069.43359 & 0.223105 & 3498 & \\

1534: B0080101003   & 52069.46245 &  52069.50297 & 0.263900 & 3501 & \\

1535: B0080101004   & 52069.53186 &  52069.57233 & 0.304712 & 3497 & \\

1536: B0080101005   & 52069.60135 &  52069.64170 & 0.345547 & 3455 & Dip in 

1537: O\,{\sc vi} (double-peak gap?)\\

1538: B0080101006   & 52069.67566 &  52069.71106 & 0.387799 & 2800 & Dip in 

1539: O\,{\sc vi}\\ 

1540: B0080101007   & 52069.74947 &  52069.78043 & 0.429904 & 2675  & \\

1541: B0080101008   & 52069.82233 &  52069.84980 & 0.471734 & 2374 & \\

1542: B0080101009   & 52069.89492 &   52069.91917 & 0.513482 & 2095 & \\

1543: B0080101010  & 52069.96775 &   52069.98854 & 0.555305 & 1796 & \\

1544: B0080101011  & 52070.04000 &   52070.05792 & 0.596954 & 1548 & \\

1545: B0080101012  & 52070.11058 &   52070.12730 & 0.638113 & 1442 & \\

1546: B0080101013  & 52070.15631 &   52070.16500 & 0.662651 & 644 & \\

1547: B0080101014  & 52070.18235 &   52070.19666 & 0.679620 & 1231 & \\

1548: B0080101015  & 52070.22613 &   52070.26604 & 0.712899 & 3447 & \\

1549: B0080101016  & 52070.29521 &   52070.33540 & 0.753612 & 3472 & \\

1550: B0080101017  & 52070.36448 &   52070.40475 & 0.794380 & 3479 & 

1551: O\,{\sc vi}$\lambda1032$ double-peaked\\ 

1552: B0080101018  & 52070.43388 &   52070.47413 & 0.835192 & 3478 & 

1553: O\,{\sc vi}$\lambda1032$ double-peaked\\ 

1554: B0080101019  & 52070.50334 &   52070.54349 & 0.876021 & 3469 & 

1555: Lines anomalously bright\\

1556: B0080101020  & 52070.57267 &   52070.61286 & 0.916809 & 3434 & \\

1557: B0080101021  & 52070.64515 &   52070.68223 & 0.958526 & 3204 & \\

1558: B0080101022  & 52070.71889 &   52070.75160 & 0.000611 & 2811 & Mid-Eclipse\\

1559: B0080101023  & 52070.79200 &   52070.82097 & 0.042515 & 2503 & \\

1560: B0080101024  &   52070.86463 &   52070.88778 & 0.083521 & 2000 & \\

1561: \end{tabular}

1562: \caption{A log of the {\it FUSE} Her X-1 observations.}

1563: \end{table}

1564: 

1565: 

1566: \begin{table}

1567: \begin{tabular}{rrlllll}

1568: Exposure & Orbital Phase & Duration (s) & $\dot{M}$ (Kurucz)  & 

1569: $\chi^2$ (Kurucz) & $\dot{M}$ ({\it FUSE}) & $\chi_{\nu}^2$ ({\it FUSE})\\

1570:          &           &      & ($-\log_{10}$ M$_\odot$) & & ($-\log_{10}$ 

1571: M$_\odot$) \\ \hline

1572: 1  & 0.182 & 3418 & 8.04 & 2.7 & 8.06 & 2.7\\

1573: 2  & 0.223 & 3498 & 8.07 & 2.9 & 8.10 & 3.0\\

1574: 3   & 0.264 & 3501 & 8.09 & 3.1 & 8.19 & 3.2\\

1575: 4   & 0.305 & 3497 & 8.16 & 2.9 & 8.26 & 3.0\\

1576: 5   & 0.346 & 3486 & 8.21 & 2.7 & 8.32 & 2.9\\

1577: 6   & 0.388 & 3059 & 8.24 & 2.5 & 8.33 & 2.7\\

1578: 7   & 0.430 & 2675 & 8.29 & 2.4 & 8.38 & 2.8\\

1579: 8   & 0.472 & 2374 & 8.20 & 2.3 & 8.30 & 2.5\\

1580: 9   & 0.513 & 2095 & 8.07 & 2.6 & 8.18 & 2.7\\

1581: 10  & 0.555 & 1796 & 8.08 & 2.4 & 8.17 & 2.7\\

1582: 11  & 0.597 & 1548 & 8.13 & 2.5 & 8.21 & 2.6\\

1583: 12  & 0.638 & 1444 & 8.20 & 2.2 & 8.27 & 2.3\\

1584: 13  & 0.663 & 751 & 8.22 & 1.2 & 8.30 & 1.3\\

1585: 14  & 0.680 & 1236 & 8.19 & 1.6 & 8.27 & 1.6\\

1586: 15  & 0.713 & 3448 & 8.17 & 2.5 & 8.27 & 2.6\\

1587: 16  & 0.754 & 3473 & 8.20 & 2.5 & 8.22 & 2.6\\

1588: 17  & 0.794 & 3480 & 8.18 & 2.3 & 8.19 & 2.3\\

1589: 18  & 0.835 & 3478 & 8.22 & 2.2 & 8.22 & 2.2\\

1590: 19  & 0.876 & 3469 & 8.17 & 2.2 & 8.18 & 2.2\\

1591: 20  & 0.917 & 3472 & 8.29 & 1.7 & 8.28 & 1.7\\

1592: 21  & 0.958 & 3204 & NA &   NA  &  NA &  NA\\

1593: 22  & 0.001 & 2826 & NA &   NA  &  NA &  NA\\

1594: 23  & 0.042 & 2503 & NA &   NA  &  NA &  NA\\

1595: 24  & 0.084 & 2000 & 8.10 & 1.7 & 8.12 & 1.7\\

1596: \end{tabular}

1597: \caption{Fits to the Her~X-1 continuum as observed with {\it FUSE}.

1598: $\dot{M}$ (Kurucz) and $\chi^2$ (Kurucz) give the mass accretion rate

1599: and goodness of fit parameter for the model fit that uses Kurucz spectra 

1600: while $\dot{M}$ (FUSE) and $\chi^2$ (FUSE) give those parameters for fits 

1601: using a library of {\it FUSE} spectra.} \end{table}

1602: 

1603: \begin{table}

1604: \begin{tabular}{llllll}

1605: Ion & Wavelength (\AA) & $r_{\rm star}$ & $r_{\rm disk}$ & $r_{\rm 

1606: continuum}$ & Comments\\

1607: \hline

1608: S\,{\sc vi} & 933.4 & 0.88 & 0.38 & 0.67 & \\

1609: S\,{\sc vi} & 944.5 & 0.88 & 0.54 & 0.85 & \\

1610: C\,{\sc iii} & 977 & 0.90 & 0.45 & 0.76 & Affected by strong ISM line\\

1611: N\,{\sc iii} & 992 & 0.80 & 0.62 & 0.91 & Near strong airglow lines\\

1612: O\,{\sc vi} & 1031.9 & 0.74 & 0.75 & 0.95 & \\

1613: O\,{\sc vi} & 1037.6 & 0.82 & 0.71 & 0.98 & \\

1614: S\,{\sc iv} & 1073 & 0.75 & 0.63 & 0.88 & \\

1615: P\,{\sc v} & 1128 & 0.75 & 0.58 & 0.86 & \\

1616: C\,{\sc iii} & 1175 & 0.82 & 0.65 & 0.95 \\

1617: \end{tabular}

1618: \caption{Correlation coefficients between the light curves of emission 

1619: lines and continuum components as separated by our model 

1620: (\S5).} 

1621: \end{table} 

1622: 

1623: \begin{table}

1624: \begin{tabular}{crrlrr}

1625: Star & Spectral Type & T$_{\rm eff}$ (K) & FUSE Rootname & E(B-V) & V\\

1626: \hline\\

1627: HD~146813 & B\,1.5 V & 24000 & P1014901 & 0.02 & 9.06\\

1628: HD~74662 & B\,3 V & 18000 &    A1290201 & 0.09 & 8.87 \\

1629: (Interpolated) & B\,4 V & 16500 &  & \\

1630: HD~92288 & B\,6 V & 14000 &    Z9012801 & 0.05 & 7.9\\

1631: (HD~21672,HD~92536) & B\,8 V & 11500 &          (Z9011401,Z9012901) & 

1632: (0.09,0.04) &

1633: 	(6.63,6.32)\\

1634: HD~149630 & B\,9 V & 10800 &   B0910101    & 0.04 & 4.2   \\

1635: (HD~109573,HD~181296) & A\,0 V & 10000 &   (B0910401,P2500101)    & 

1636: (0.0,0.01) & 

1637: 		(5.78,5.03)  \\ 

1638: HD~31647 & A\,1 V & 9300 &     C0380901    & 0.0 & 4.989 \\

1639: HD~115892 & A\,2 V & 9050 &    A0410505    & 0.0 & 2.75 \\

1640: HD~43940 & A\,3 V & 8850 &     C0380101    & 0.0 & 5.87 \\

1641: HD~11636 & A\,5 V & 8500 &     A0410101    & 0.0 & 2.64 \\

1642: (Interpolated) & A\,6 V & 8350 & \\

1643: HD~187642 & A\,7 V & 8050 &    D0990101    & 0.0 & 0.77 \\

1644: \end{tabular}

1645: \caption{FUSE spectra used to form a library of stellar continua versus 

1646: effective temperature in the far UV.}

1647: \end{table}

1648: 

1649: \clearpage

1650: 

1651: \setcounter{figure}{0}

1652: 

1653: %figure 1 is the overall spectrum

1654: 

1655: \begin{figure}

1656: \includegraphics[height=0.4\textheight]{f1a.pdf}

1657: \includegraphics[height=0.4\textheight]{f1b.pdf}

1658: \end{figure}

1659: 

1660: \newpage

1661: \setcounter{figure}{0}

1662: \begin{figure}

1663: \includegraphics[height=0.4\textheight]{f1c.pdf}

1664: \includegraphics[height=0.4\textheight]{f1d.pdf}

1665: \caption{The mean Her X-1 far UV spectrum observed with FUSE ($<1180$\AA) 

1666: and the {\it HST} STIS ($>1150$\AA) near orbital phases when the emission is 

1667: expected to be dominated by the accretion disk. Prominent 

1668: emission lines are indicated by dotted vertical lines. 

1669: We show he average of our continuum models using actual FUSE spectra of hot 

1670: stars together with blackbodies. The component of the model resulting 

1671: from the X-ray heated face of the normal star is indicated by a dotted 

1672: curve.}

1673: \end{figure}

1674: \newpage

1675: 

1676: %figure 2 is the overall spectrum at star-dominated phases

1677: 

1678: \newpage

1679: \setcounter{figure}{1}

1680: \begin{figure}

1681: \includegraphics[height=0.4\textheight]{f2a.pdf}

1682: \includegraphics[height=0.4\textheight]{f2b.pdf}

1683: \end{figure}

1684: \newpage

1685: \setcounter{figure}{1}

1686: \begin{figure}

1687: \includegraphics[height=0.4\textheight]{f2c.pdf}

1688: \includegraphics[height=0.4\textheight]{f2d.pdf}

1689: %spectrum with line ids

1690: \caption{The mean Her X-1 far UV spectrum observed with FUSE ($<1180$\AA) 

1691: and the {\it HST} STIS ($>1150$\AA) near orbital phases when the emission is 

1692: expected to be dominated by the X-ray heated face of the normal star.

1693: The spectrum is plotted as in Figure~1. The lines and continuum are 

1694: brighter at these phases, and the lines are narrower.}

1695: \end{figure}

1696: \newpage

1697: 

1698: \clearpage

1699: 

1700: \begin{figure}

1701: \includegraphics[width=0.9\textwidth]{f3.pdf}

1702: \caption{A sample of the H$_2$ absorption lines in the spectrum of 

1703: Her~X-1, and a model (dotted lines) with flat normalization, of absorption 

1704: from rotational levels $J=0$ through $J=3$.}

1705: \end{figure}

1706: 

1707: \setcounter{figure}{3}

1708: %figure 4a through h are the lightcurves of the lines

1709: \begin{figure}

1710: \includegraphics[height=0.25\textheight]{f4a.png}

1711: \includegraphics[height=0.25\textheight]{f4b.png}

1712: \includegraphics[height=0.25\textheight]{f4c.png}

1713: \includegraphics[height=0.25\textheight]{f4d.png}

1714: \end{figure}

1715: %

1716: \setcounter{figure}{3}

1717: \begin{figure}

1718: \includegraphics[height=0.25\textheight]{f4e.png}

1719: \includegraphics[height=0.25\textheight]{f4f.png}

1720: \includegraphics[height=0.25\textheight]{f4g.png}

1721: \includegraphics[height=0.25\textheight]{f4h.png}

1722: \includegraphics[height=0.25\textheight]{f4i.png}

1723: \caption{Light curves of prominent lines in the FUSE spectrum of 

1724: Hercules~X-1. A model of the continuum flux has been subtracted. Errors 

1725: at the $1\sigma$ level are shown. Emission lines shown: (a) 

1726: S\,{\sc vi}$\lambda933.4$, (b) S\,{\sc vi}$\lambda944.5$, 

1727: (c) C\,{\sc iii}$\lambda 977$, (d) N\,{\sc iii}$\lambda991$, (e) O\,{\sc 

1728: vi}$\lambda1031.9$, 

1729: (f) O\,{\sc vi}$\lambda 1037.6$, (g) S\,{\sc iv}$\lambda1073$, (h) P\,{\sc 

1730: v}$\lambda 1128$, (i) C\,{\sc iii}$\lambda 1176$}

1731: \end{figure}

1732: 

1733: \newpage

1734: 

1735: \setcounter{figure}{4}

1736: \begin{figure}

1737: \epsscale{1.0}

1738: %\plotone{f5.png}

1739: \includegraphics[width=1.0\textwidth]{f5.png}

1740: \caption{Flux versus orbital phase $\phi$ in our model of the continuum 

1741: flux in the 

1742: wavelength range of the O\,{\sc vi} doublet. The continuum is separated 

1743: into disk and stellar components.}

1744: \end{figure}

1745: 

1746: \newpage

1747: 

1748: \setcounter{figure}{5}

1749: %figure 6 is the model of o6

1750: \begin{figure}

1751: %\vspace{-1.5in}

1752: %model of o6

1753: \includegraphics[width=1.25\textwidth]{f6.png}

1754: \caption{Fits to the O\,{\sc vi} lines at phases when the accretion disk is 

1755: partially eclipsed.}

1756: \end{figure}

1757: 

1758: \newpage

1759: 

1760: \clearpage

1761: \setcounter{figure}{6}

1762: \begin{figure}

1763: %\epsscale{0.8}

1764: \includegraphics[width=0.8\textwidth]{f7a.png}

1765: \includegraphics[width=0.8\textwidth]{f7b.png}

1766: %\plotone{f6a.eps}

1767: %\epsscale{1.0}

1768: %\plotone{f6b.eps}

1769: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1770: FUSE spectra of S\,{\sc vi}$\lambda 933$ from $\phi=0.2$ to $\phi=0.85$. 

1771: The grayscale runs from the minimum to the maximum. 

1772: Top: the trailed spectrogram (left) and filtered trailed spectrogram (right).

1773: Middle: the back-projected tomogram (left) and filtered tomogram (right).

1774: Bottom: the trailed spectrogram inverted from the back-projected 

1775: tomogram (left) and from the filtered tomogram (right).}

1776: \end{figure}

1777: 

1778: \clearpage

1779: 

1780: \begin{figure}

1781: %\epsscale{1.0}

1782: \includegraphics[width=0.95\textwidth]{f8a.png}

1783: \includegraphics[width=0.95\textwidth]{f8b.png}

1784: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1785: FUSE spectra of S\,{\sc vi}$\lambda 944$ presented as in Figure~7.}

1786: \end{figure}

1787: 

1788: \begin{figure}

1789: %\epsscale{1.0}

1790: \includegraphics[width=0.95\textwidth]{f9a.png}

1791: \includegraphics[width=0.95\textwidth]{f9b.png}

1792: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1793: FUSE spectra of N\,{\sc iii}$\lambda 991.6$ (Bowen process) presented as in 

1794: Figure~7.} \end{figure}

1795: 

1796: \clearpage

1797: 

1798: \begin{figure}

1799: %\epsscale{1.0}

1800: \includegraphics[width=0.95\textwidth]{f10a.png}

1801: \includegraphics[width=0.95\textwidth]{f10b.png}

1802: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1803: FUSE spectra of O\,{\sc vi}$\lambda 1031.9$ presented as in Figure~7.}

1804: \end{figure}

1805: 

1806: \clearpage

1807: 

1808: \begin{figure}

1809: %\epsscale{1.0}

1810: \includegraphics[width=0.95\textwidth]{f11a.png}

1811: \includegraphics[width=0.95\textwidth]{f11b.png}

1812: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1813: FUSE spectra of O\,{\sc vi}$\lambda 1037.6$ presented as in Figure~7.}

1814: \end{figure}

1815: 

1816: \clearpage

1817: 

1818: \begin{figure}

1819: %\epsscale{1.0}

1820: \includegraphics[width=0.95\textwidth]{f12a.png}

1821: \includegraphics[width=0.95\textwidth]{f12b.png}

1822: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1823: FUSE spectra of S\,{\sc iv}$\lambda 1073$ presented as in Figure~7.}

1824: \end{figure}

1825: 

1826: \begin{figure}

1827: %\epsscale{1.0}

1828: \includegraphics[width=0.95\textwidth]{f13a.png}

1829: \includegraphics[width=0.95\textwidth]{f13b.png}

1830: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1831: FUSE spectra of P\,{\sc v}$\lambda 1128$ presented as in Figure~7.}

1832: \end{figure}

1833: \clearpage

1834: 

1835: \begin{figure}

1836: %\epsscale{1.0}

1837: \includegraphics[width=0.95\textwidth]{f14a.png}

1838: \includegraphics[width=0.95\textwidth]{f14b.png}

1839: \caption{A Fourier-filtered back-projected Doppler tomogram using the 

1840: FUSE spectra of C\,{\sc iii}$\lambda 1176$ presented as in Figure~7.}

1841: \end{figure}

1842: 

1843: \begin{figure}

1844: %\epsscale{1.0}

1845: %\plotone{f15.eps}

1846: \includegraphics[width=1.0\textwidth]{f15.png}

1847: \caption{The O\,{\sc vi}$\lambda1032$ emission line averaged over every 

1848: two {\it FUSE} exposures (thick curve), a model of O\,{\sc vi} emission on 

1849: the surface of the noncollapsed star (dashed), and an empirical model of 

1850: Gaussian emission and absorption models. Both models include flux from a 

1851: model of the accretion disk.}

1852: \end{figure}

1853: 

1854: \begin{figure}

1855: %\epsscale{0.8}

1856: \includegraphics[width=0.85\textwidth]{f16a.png}

1857: \includegraphics[width=0.85\textwidth]{f16b.png}

1858: \caption{A Doppler tomogram of a simple simulation of the 

1859: O\,{\sc vi}$\lambda1032$ line as a sum of moving Gaussians and an accretion 

1860: disk model, with counting statistics included.} 

1861: \end{figure}

1862: 

1863: \begin{figure}

1864: \includegraphics[width=1.0\textwidth]{f17.pdf}

1865: \caption{Sample spectra, with spectral type labelled, in the library of 

1866: {\it FUSE} spectra we use to complement our library of {\it IUE} and 

1867: optical spectra. We use this library in the fitting of the Her~X-1 

1868: continuum. All stars have their V band flux normalized to the 

1869: same value.}

1870: \end{figure}

1871: 

1872: \end{document}

1873: