1: \documentclass[manuscript]{aastex}
2: \usepackage{apj}
3: \shorttitle{A Cepheid is no More}
4: \shortauthors{Macri {\it et al.}}
5: \slugcomment{}
6:
7: \def \ng {NGC$\,$}
8: \def \de {^{\circ}}
9: \def \hr {^{\rm h}}
10: \def \mi {^{\rm m}}
11: \def \se {^{\rm s}}
12: \def \ebv {\langle E(B\!-V)\rangle}
13: \def \evi {\langle E(V\!-I)\rangle}
14: \def \vi {V\!-\!I}
15: \def \vk {V\!-\!K}
16: \def \bv {B\!-\!V}
17: \def \rvi {{\cal R}_{\vi}}
18: \begin{document}
19:
20: \title{A Cepheid is No More: Hubble's Variable 19 in M33}
21:
22: \author{L.M.~Macri, D.D.~Sasselov\altaffilmark{1}\&K.Z.~Stanek\altaffilmark{2}}
23: \affil{Harvard-Smithsonian Center for Astrophysics, 60 Garden St., Cambridge
24: MA 02138, USA} \email{lmacri, sasselov, kstanek@cfa.harvard.edu}
25:
26: \altaffiltext{1}{Alfred P. Sloan Foundation Fellow}
27: \altaffiltext{2}{Hubble Fellow}
28:
29: \begin{abstract}
30:
31: We report on the remarkable evolution in the light curve of a variable star
32: discovered by \citet{hu26} in M33 and classified by him as a Cepheid. Early
33: in the 20th century, the variable, designated as V19, exhibited a 54.7-day
34: period, an intensity-weighted mean $B$ magnitude of $19.59\pm0.23$~mag, and a
35: $B$ amplitude of 1.1~mag. Its position in the P-L plane was consistent with
36: the relation derived by Hubble from a total of 35 variables. Modern
37: observations by the DIRECT project show a dramatic change in the properties of
38: V19: its mean $B$ magnitude has risen to $19.08\pm0.05$~mag and its $B$
39: amplitude has decreased to less than $0.1$~mag. V19 does not appear to be a
40: classical (Population I) Cepheid variable at present, and its nature remains a
41: mystery. It is not clear how frequent such objects are nor how often they
42: could be mistaken for classical Cepheids.
43: \end{abstract}
44:
45: \keywords{Cepheids --- stars: evolution --- galaxies: individual (M33)}
46:
47: \section{Introduction}
48:
49: In his seminal work, ``A spiral nebula as a stellar system, Messier 33,''
50: \citet{hu26} presented the discovery of 35 Cepheid variables in that galaxy,
51: which were used to determine its distance. No other searches for variables in
52: M33 were undertaken for the next fifty years, until the surveys of
53: \citet{vhk75,sc83, kmw87} that discovered a total of 77 new Cepheids and
54: recovered many of Hubble's objects.
55:
56: The DIRECT project \citep{ka98,st98} started in 1996 the first modern CCD-based
57: search for variables in M33, among other targets. Our first two seasons
58: surveyed three $10\arcmin\times 10\arcmin$ fields located in the central area
59: of M33; the variable star content of these fields has been presented in
60: \citet{ma01a} and \citet{st01}. Since these fields had been covered by all
61: previous surveys, we cross-correlated our variable star catalog with published
62: catalogs and successfully identified 57 of the 60 known Cepheids. Two of the
63: three missing variables were short period Cepheids: B08 from \citet{sc83}
64: (P=3.2~d) and Q19113 from \citet{kmw87} (P=13.4~d). These variables are faint
65: and close to our detection threshold; therefore, it was not surprising that
66: they eluded our detection.
67:
68: The third missing variable turned out to be a surprising object, and is the
69: topic of this paper. It is V19 from \citet{hu26}, classified by him as a
70: 54.7-day Cepheid. It is a bright, isolated, and easy to identify star located
71: some $6\arcmin$ north of the nucleus. Hubble's light curve exhibited a
72: variation of $1.1$~mag, which is not present in our data. Additionally, the
73: mean $B$ magnitude of the object increased by $\sim 0.5$~mag since Hubble's
74: original observations.
75:
76: \S2 summarizes the existing photometry for this object; \S3 addresses the
77: current variability, periodicity, and location of this object in the H-R and PL
78: diagrams; and \S4 discusses the possible nature of this object and its known
79: Galactic counterparts.
80:
81: \section{Photometry}
82:
83: \subsection{Data from \citet{hu26}}
84:
85: The discovery of V19 by \citet{hu26} was based on observations obtained at the
86: Mount Wilson 100-inch telescope in 56 epochs between August 1919 and September
87: 1925, with two additional plates from September 1909 and August 1910.
88: Magnitudes were measured using the Argelander method, and were based on local
89: reference stars calibrated through a photometric sequence from the Selected
90: Area 45. Hubble's original magnitudes suffered from zeropoint and scale
91: errors, as pointed out by \citet{sn83} and \citet{cs87}.
92:
93: We calculated a new transformation between Hubble's magnitude scale and the
94: standard system, using our photometry \citep{ma01b} for Hubble's original
95: comparison stars. We identified 121 of these stars in our CCD images, using the
96: finding charts published by \citet{sn83} and performed a least-squares fit
97: between Hubble's photographic magnitudes and our CCD-based $B$ magnitudes.
98: We found
99:
100: \vskip -12pt
101: \begin{equation}
102: m_B = 1.62 (\pm0.04) \left [m_{pg}-18.5 \right ] + 19.68(\pm0.02),
103: \end{equation}
104:
105: \noindent{where $m_B$ is the $B$ magnitude in the standard system and $m_{pg}$
106: is the photographic magnitude as measured by Hubble. The relation has a
107: scatter of 0.18~mag after the rejection of 11 stars which deviated by more
108: than 0.5~mag. Our coefficients agree with those derived by \citet{cs87}. We
109: corrected Hubble's original measurements using the newly derived
110: transformation, and analyzed the data using a suite of programs developed by
111: the DIRECT project \citep{ka98}. We derived a best-fit period of 54.71 days,
112: identical to Hubble's original value, and an intensity-weighted mean $B$
113: magnitude of $19.59\pm0.23$~mag. Figure 1 shows a phased light curve of V19
114: based on Hubble's recalibrated data. The scatter in the light curve is
115: consistent with the {\it r.m.s.} scatter of Equation 1.}
116:
117: \subsection{DIRECT Data}
118:
119: We based our identification of V19 on the finding charts of \citet{hu26},
120: \citet{vhk75} and \citet{hs80}, all of whom mark the same star as the variable.
121: V19 is the DIRECT object D33J013357.1+304512.4, located at R.A.=1h33m57.1s,
122: Dec.=+30d45m12.4s (J2000.0). We ruled out a mis-identification of V19 with a
123: nearby star in Hubble's original finding chart by identifying all variables in
124: the general vicinity ($< 3\arcmin$). This is a generous search area, given the
125: fact that all other variables discovered by Hubble were recovered within a few
126: arcseconds of previously published positions. The nearest variable of any kind
127: is D33J013357.1+304455.2, a faint ($V=21.5$) periodic variable located
128: $17\arcsec$ south of V19. The nearest Cepheid is D33J013354.9+304532.5, with a
129: period of 36 days and located $39\arcsec$ north-west of V19. The nearest
130: Cepheid with a period close to Hubble's original period for V19 is
131: D33J013347.5+304423.2, with a period of 56 days and located $2.5\arcmin$ to the
132: south-west.
133:
134: V19 was observed by the DIRECT project a total of 11, 116 and 60 times in
135: $BVI$, respectively, during the 1996 and 1997 observing seasons. The
136: description of the reduction and analysis of this data set was presented in
137: \citet{ma01b}. Additional observations of another DIRECT field which contains
138: V19 were obtained during the 1999 season; for the purposes of this paper, we
139: performed a preliminary reduction using the standard DIRECT pipeline
140: \citep{ka98} and calibrated the resulting photometry using bright, non-variable
141: stars from the database of \citet{ma01b}. This third year of observations
142: yielded an additional 19, 121 and 35 points in $BVI$, respectively. The mean
143: $BVI$ magnitudes of V19 obtained from the three seasons spanned by our data are
144: $19.08\pm0.05$, $18.21\pm0.05$ and $17.30\pm0.05$~mag, respectively. The light
145: curves exhibit very little variation, of order 0.05~mag peak-to-peak. \S 3
146: contains the analysis of our light curve data\footnote{The light curve of V19
147: and its finding chart can be found at the DIRECT Web page:
148: \url{http://cfa-www.harvard.edu/\~\/kstanek/DIRECT}}.
149:
150: \hfill
151:
152: \subsection{Other photometric data}
153:
154: We searched the literature for additional observations of V19 obtained after
155: 1926 and before 1996. The only time series photometry that we were able to
156: locate was that of \citet{vhk75}, who obtained 67 plates of M33 between January
157: 1966 and November 1974 at the Palomar 48-inch Schmidt, using a variety of plate
158: types and filters. The data for V19 consists of 19 points in $B$ and 41 points
159: in $V$. Unfortunately, the uncertainties quoted by the authors for the
160: individual measurements is 0.7~mag, and there are no data available to check
161: the absolute calibration of their photometric systems.
162:
163: V19 was observed by \citet{hs80} in their survey for blue and red stars in M33,
164: and was designated as R142 in their catalog. We corrected their tabulated
165: photometry according to \citet{wfm90} and obtained $B=19.1$~mag and $B-V =
166: 0.9$~mag, consistent with our own determinations. Unfortunately, the authors
167: did not obtain a spectrum for this star. Additionally, V19 is present in the
168: fields surveyed by \citet{kmw87}, but these authors do not list it in their
169: compilation of variables.
170:
171: \citet{mad85} reported the results of $JHK$ observations of V19 and other M33
172: Cepheids conducted between 1980 and 1983 at a variety of telescopes using a
173: photometer with a $5\arcsec$ diameter aperture. However, the authors expressed
174: concerns about possible contamination by field stars inside their aperture.
175: Therefore, we analyzed $K_s$-band images of V19, obtained by Don McCarthy with
176: the PISCES camera \citep{mcc98} at the refurbished MMT, under $0.6\arcsec$
177: seeing. These images show the presence of another star of similar infrared flux
178: some $3\arcsec$ away. A preliminary calibration of the MMT data, based on
179: isolated 2MASS stars present in the field of view, yields a mean magnitude for
180: V19 of $K_s=16.1\pm0.05$~mag. This, in turn, implies a $\vk$ color of
181: $2.1\pm0.1$~mag, which is somewhat redder than, but consistent with, the one
182: predicted for a 55-day Cepheid, $1.9\pm0.1$~mag \citep[c.f. Equations 5 and 9
183: of][]{ma01c}.
184:
185: Lastly, we retrieved archival HST/WFPC2 frames in which V19 is present (dataset
186: u2c604, proposal \# 0538) from the Space Telescope Archive. The images were
187: obtained in the $F439W$ and $F547M$ filters and show the object (at x=92,y=283
188: on WF4) to be well isolated, with no detectable companions inside our
189: ground-based seeing disk.
190:
191: \section{Analysis}
192:
193: We computed the $J_S$ index \citep{st96} using all 237 $V$ measurements from
194: the DIRECT data, and found a value of $J_S=0.807$, which is above the standard
195: DIRECT threshold of 0.75. However, the overall {\it r.m.s.} scatter is only
196: 0.025~mag, below our usual threshold of 0.04~mag. Therefore, the star does not
197: meet the standard criteria for variability adopted by the DIRECT project.
198:
199: We attempted to find a period from our data, using the standard DIRECT
200: technique of a modified Lafler-Kinman algorithm \citep{st96} and the CLEAN
201: algorithm of \citet{rld87}. No convincing periods were found, which is not
202: surprising given that our frame-to-frame random photometric error (arising from
203: the limitations of fixed-position PSF photometry) is of the order of
204: $0.02-0.03$~mag. We conclude that {\it if} the 54.7-day periodicity seen by
205: Hubble is still present, its amplitude must be less than 0.05~mag. Long-term
206: follow-up observations using larger telescopes would be most useful to detect
207: any small-scale periodicity and/or variability.
208:
209: Figure 2 shows the $B$-band light curve of V19 from 1910 to 2000, based on the
210: data from \citet{hu26}, {vhk75}, and this work. The intensity-weighted mean
211: magnitude shows a dramatic increase, from $19.59\pm0.23$~mag in 1919-1925 to
212: $B=19.08\pm0.05$ in 1996-1999, while the amplitude of pulsation decreases from
213: 1.1~mag to undetectable levels. Figure 3 shows the location of V19 in the H-R
214: diagrams of \citet{ma01b}. We also plot the location of 130 Cepheids with
215: $P>10$~days present in the inner region of M33 \citep{ma01a}. Figure 4 shows
216: the location of V19 in the $BVI$ Cepheid P-L relations defined by the same set
217: of variables, using Hubble's original period of 54.7 days. These figures
218: indicate that the modern optical magnitudes of V19 are somewhat brighter than,
219: but still consistent with, those of 55-day Cepheids in M33.
220:
221: \section{Discussion}
222:
223: The cessation of pulsations, or any significant decrease in amplitude (i.e., by
224: a factor of ten), has never been observed in a classical Cepheid. V19 may be
225: the first case, or it may belong to another class of variable stars. We
226: present here four classes of variable stars whose pulsational behaviour may
227: explain the nature of V19.
228:
229: \vskip 4pt
230:
231: \noindent{{\em Population II Cepheids}: RU Cam, a 22-day Cepheid which in 1965
232: abruptly decreased in amplitude from 1~mag to about 0.1-0.2~mag \citep{df66},
233: was later shown to be a W Virginis variable (Population II Cepheid) that
234: exhibited a highly unstable and modulated light curve thereafter
235: \citep{ks93}. Even before the 1965 event, \citet{pg41} had suggested that RU
236: Cam may be a W Virginis variable. However, RU Cam is of little relevance to
237: our case, because in order for V19 to be a Population II Cepheid, it should be
238: about 2 magnitudes fainter \citep[see][]{al98} than Population I Cepheids of
239: similar period. Furthermore, W Virginis variables are not known to change their
240: mean magnitude, although a few RV Tauri variables (longer-period Population II
241: Cepheids) do that cyclically.}
242:
243: \vskip 4pt
244:
245: \noindent {{\em Peculiar Population I Cepheids}: These objects are extremely
246: rare and entirely confined to much shorter periods. A case in point is the
247: classical Cepheid V473 Lyr of $P=1.49$ days \citep[e.g.,][]{bu86,an98}. It has
248: an amplitude modulation of a factor of 15 which resembles the beating of two
249: closely spaced pulsation modes. This phenomenon is not understood, but it is
250: most likely associated with the interaction of high-order modes. Furthermore,
251: the interval corresponding to low amplitudes in V473 Lyr has a short duration;
252: in general, the phenomenon seems to have very little in common with V19. A more
253: relevant object in this category could be Polaris, now established to be a
254: Population I Cepheid pulsating in the first overtone mode \citep{fc97}. It has
255: a period is 3.97 days, and its amplitude has decreased by a factor of 3 over
256: the past 50 years. Despite earlier reports, this amplitude decrease has
257: stabilized over the past 15 years; the pulsation amplitude is practically
258: constant since 1986 \citep{hc00}. As in the case of V473 Lyr, the amplitude
259: decrease could be due to high-order mode interactions \citep{es01}. However,
260: the pulsation amplitude of Polaris has always been very low ($\leq 0.1$~mag in
261: V), which makes the comparison with V19 difficult. Finally, we should note
262: that there are stars inside the Cepheid instability strip that are not
263: variable, for reasons that are still unknown.}
264:
265: \vskip 4pt
266:
267: \noindent{{\em UU Herculis stars}: These are supergiants with Cepheid-like
268: pulsations that exhibit occasional standstills \citep{sa84}. Pulsations cease
269: abruptly and at mid-cycle (corresponding to the mean magnitude of the star),
270: last a couple of months, and then abruptly start again \citep{sa83}. The UU
271: Herculis stars may alternate between Cepheid-like and RV Tauri-like pulsations
272: (over timescales of several years) with periods of 30 to 90 days, which are
273: always of relatively low amplitude ($\sim$0.3~mag in V). UU Her is the only
274: exception, with amplitudes as high as 0.6~mag in V. Most stars in this class
275: have infrared excesses, low metallicities, and show variable emission in the
276: H$\alpha$ line \citep{kpc97}.}
277:
278: \vskip 5pt
279:
280: \noindent{{\em Luminous Blue Variables}: Given its luminosity (M$_{\rm
281: V}\approx -6.4$~mag), V19 appears to be more closely related to the Luminous
282: Blue Variables (LBVs), which are not always blue. One example is 164 G Sco
283: \citep[M$_{\rm V}=-8.4$, T=10,000K, $R\approx 175 R_{\odot}$,
284: after][]{lba98}. It shows multiperiodic variability ($45-55$ days) of very low
285: amplitude ($\leq 0.1$~mag in V). Occasionally, LBVs experience dramatic
286: outbursts and mass loss episodes which could considerably alter their effective
287: radii and temperatures for extended periods of time. There also exist other
288: luminous supergiants which are not LBVs but exhibit Cepheid-like pulsations,
289: such as V810 Cen \citep{ki98}. However, their pulsations have always been
290: observed to be low ($\leq 0.1$~mag) and multiperiodic.}
291:
292: \vskip 6pt
293:
294: Each of the types described above belongs to a different evolutionary state.
295: The W Virginis (and RV Tauri) variables are most likely low-mass AGB (and
296: tip-of- or post-AGB) stars (hydrogen- and helium-shell burning). The same
297: appears to be true for the majority of UU Hers stars (a related object
298: which "wanders" across the HR diagram is FG Sge). Hence their lower
299: luminosity and common circumstellar infrared excesses. They often show
300: emission in the absorption profile of the H$\alpha$ and other strong lines.
301: There is no evidence that V19 possesses any of the above characteristics and
302: its luminosity is incompatibly higher. The remaining types of variables (LBVs,
303: V810 Cen) are very massive supergiants. At a given pulsation period they are
304: brighter than Population I Cepheids and they are unlikely to be mistaken for
305: Cepheids, due to their low amplitudes and multi-periodic pulsation.
306:
307: In summary, V19 does not belong to any of the above types: it is too luminous
308: to be a low-mass Population II star, and its pulsation was too regular and
309: high-amplitude 75 years ago to be a massive microvariable supergiant.
310: Therefore, V19 appears to be a high-mass Population I star that behaves more
311: like a low-mass Population II RV Tauri variable.
312:
313: Given the bright and isolated nature of V19, and its location in the host
314: galaxy, it is likely that most photographic plates of M33 obtained at 2-m class
315: (or larger) telescopes during the 20th century will yield precise photometry
316: for this object. We encourage the astronomical community to contact us if they
317: are in possession of plate material that might help constrain the past behavior
318: of V19. Future observations, including spectroscopy and long-term, multi-band
319: photometric follow-up, will help resolve the nature of this object. Such
320: resolution is desirable in order to avoid misclassifying objects like V19 as
321: classical Cepheids in distant galaxies.
322:
323: \section{Acknowledgments}
324:
325: %We would like to thank the telescope allocation committees of the FLWO and MDM
326: %Observatories for the generous amounts of telescope time that were devoted to
327: %the DIRECT project.
328: We would like to thank Don McCarthy for the PISCES
329: images of V19 taken at the new MMT. Thanks are due to D. Bersier, N.R. Evans,
330: J. Kaluzny \& B. Paczynski for helpful comments. LMM would also like to thank
331: John Huchra for his support and comments. DDS acknowledges support from the
332: Alfred P. Sloan Foundation and from NSF grant No. AST-9970812. KZS was
333: supported by Hubble Fellowship grant HF-01124.01 from the Space Telescope
334: Science Institute, which is operated by the Association of Universities for
335: Research in Astronomy, Inc., under NASA contract NAS5-26555.
336:
337: \begin{thebibliography}{}
338:
339: \bibitem[Alcock et al.\,(1998)]{al98} Alcock, C. \& MACHO Team,~1998, \aj,
340: 115, 1921
341:
342: \bibitem[Andrievsky et al.\,(1998)]{an98} Andrievsky S.M., Kovtyukh V.V.,
343: Bersier, D., Luck, R.E., Gopka, V.P., Yushchenko, A.V., Usenko I.A.~1998,
344: \aap, 329, 599
345:
346: \bibitem[Burki et al.\,(1986)]{bu86} Burki G., Schmidt E.G., Arellano Ferro
347: A., Fernie J.D., Sasselov, D., Simon N.R., Percy J.R., Szabados L.~1986, \aap,
348: 168, 139
349:
350: \bibitem[Christian \& Schommer\,(1987)]{cs87} Christian, C.A. \& Schommer,
351: R.A.~1987, \aj, 93, 557
352:
353: \bibitem[Demers \& Fernie\,(1966)]{df66} Demers, S., \& Fernie, J.D.~1966,
354: \apj, 144, 440
355:
356: \bibitem[Evans et al.\,(2001)]{es01} Evans, N.R., Sasselov, D.D., \& Short,
357: C.I.~2001, \apj, submitted
358:
359: \bibitem[Feast \& Catchpole\,(1997)]{fc97} Feast, M.W., \& Catchpole,
360: R.M.~1997, \mnras, 286, L1
361:
362: \bibitem[Hatzes \& Cochran\,(2000)]{hc00} Hatzes, A.P., \& Cochran, W.D.~2000,
363: \aj, 120, 979
364:
365: \bibitem[Hubble\,(1926)]{hu26} Hubble, E.~1926, \apj, 63, 236
366:
367: \bibitem[Humphreys \& Sandage\,(1980)]{hs80} Humpreys, R. \& Sandage, A.~1980,
368: \apjs, 44, 319
369:
370: \bibitem[Kaluzny et al.\,(1998)]{ka98} Kaluzny, J., Stanek, K.Z.,
371: Krockenberger, M., Sasselov, D.D., Tonry, J.L. \& Mateo, M.~1998, \aj, 115,
372: 1016
373:
374: \bibitem[Kienzle et al.\,(1998)]{ki98} Kienzle, F., Burki, G., Burnet, M. \&
375: Maynet, G.~1998, \aap, 337, 779
376:
377: \bibitem[Kinman et al.\,(1987)Kinman, Mould \& Wood\,(1987)]{kmw87} Kinman,
378: T., Mould, J.R., and Wood, P.R.~1987, \aj, 93, 833
379:
380: \bibitem[Klochkova et al.,\,(1997)]{kpc97} Klochkova, V.G., Panchuk, V.E. \&
381: Chentsov, E.L.~1997, \aap, 323, 789
382:
383: \bibitem[Kollath \& Szeidl\,(1993)]{ks93} Kollath, Z., \& Szeidl, B.~1993,
384: \aap, 277, 62
385:
386: \bibitem[Lamers et al.\,(1998)]{lba98} Lamers, H., Bastiaanse, M., \& Aerts,
387: C.~1998, in ASP Conf. Ser. 135, eds. P.Bradley \& J.Guzik, p.159
388:
389: \bibitem[Macri et al.\,(2001a)]{ma01a} Macri, L.M., Stanek, K.Z., Sasselov,
390: D.D., Krockenberger, M. \& Kaluzny, J.~2001a, \aj, 121, 861
391:
392: \bibitem[Macri et al.\,(2001b)]{ma01b} Macri, L.M., Stanek, K.Z., Sasselov,
393: D.D., Krockenberger, M. \& Kaluzny, J.~2001b, \aj, 121, 870
394:
395: \bibitem[Macri et al.\,(2001c)]{ma01c} Macri, L.M., Calzetti, D., Freedman,
396: W.L., Gibson, B.K, Graham, J.A., Huchra, J.P., Hughes, S.M.G., Madore, B.F.,
397: Mould, J.R., Persson, S.E. \& Stetson, P.B.~2001c, \apj, 549, xxx
398:
399: \bibitem[Madore et al.\,(1985)]{mad85} Madore, B.F., McAlary, C.W., McLaren,
400: R.A., Welch, D.L., Neugebauer, G. \& Matthews, K.~1985, \apj, 294, 560
401:
402: \bibitem[McCarthy et al.\,(1998)]{mcc98} McCarthy, D.W., Ge, J., Hinz, J.L.,
403: Finn, R.A., Low, F.J., Cheselka, M. \& Salvestrini, K.~1998, \baas, 193, 1109
404:
405: \bibitem[Payne-Gaposchkin\,(1941)]{pg41} Payne-Gaposchkin, C.~1941, Harvard
406: Bull., 915, 10
407:
408: \bibitem[Roberts et al.\,(1987)]{rld87} Roberts, D.H., Lehar, J. \& Dreher,
409: J.W.~(1987), \aj, 93, 968
410:
411: \bibitem[Sandage\,(1983)]{sn83} Sandage, A.R.~1983, \aj, 88, 1108
412:
413: \bibitem[Sandage \& Carlson\,(1983)]{sc83} Sandage, A.R. \& Carlson, G.~1983,
414: \apjl, 267, 25
415:
416: \bibitem[Sasselov\,(1983)]{sa83} Sasselov, D.D.~1983, IVBS \# 2387
417:
418: \bibitem[Sasselov\,(1984)]{sa84} Sasselov, D.D.~1984, \apss, 102, 161
419:
420: \bibitem[Stanek et al.\,(1998)]{st98} Stanek, K.Z., Kaluzny, J., Krockenberger,
421: M., Sasselov, D.D., Tonry, J.L. \& Mateo, M.~1998, \aj, 115, 1894
422:
423: \bibitem[Stanek et al.\,(2001)]{st01} Stanek, K.Z., Macri, L.M., Sasselov,
424: D.D., \& Kaluzny, J.~2001, in preparation
425:
426: \bibitem[Stetson\,(1996)]{st96} Stetson, P.B.~1996, \pasp, 108, 851
427:
428: \bibitem[van den Bergh et al.\,(1975)]{vhk75} van den Bergh, S., Herbst, E. \&
429: Kowal, C.T.~1975, \apjs, 29, 303
430:
431: \bibitem[Wilson et al.\,(1990)]{wfm90} Wilson, C.D., Freedman, W.L. \& Madore,
432: B.F.~1990, \aj, 99, 149
433:
434: \end{thebibliography}
435:
436: \clearpage
437:
438: \begin{figure}
439: \plottwo{f1.eps}{f2.eps}
440: \caption{({\bf left}) Phased light curve of V19, based on the period and magnitudes published in \citet{hu26}. The original photographic magnitudes have been
441: corrected into the standard $B$ band using Equation 1.}
442:
443: \caption{({\bf right}) $B$-band light curve of V19 from 1910 to 1999,
444: based on the data of \citet{hu26}, \citet{vhk75}, and the present
445: work.}
446: \end{figure}
447:
448: \begin{figure}
449: \plottwo{f3.eps}{f4.eps}
450: \caption{({\bf left}) Color-magnitude diagrams of M33 from
451: \citet{ma01b}, showing the location of V19 ($\star$) and $P>10$~d
452: Cepheids ($\bullet$).}
453:
454: \caption{({\bf right)} P-L relations of M33 Cepheids ($\bullet$) from
455: \citet{ma01a}, showing the location of V19 ($\star$).}
456: \end{figure}
457: \end{document}
458: