Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://star.arm.ac.uk/~ambn/pep5.ps Äàòà èçìåíåíèÿ: Mon Oct 28 13:15:30 1996 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 20:23:45 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: m 101

PEP5: ASTRONOMICAL PHOTOMETRY INTO THE NEXT MILLENIUM
Mt. Kent Observatory, 7--11 July 1996
Long­term photoelectric and visual monitoring
of selected southern variable stars 1
C. Sterken 2;3
A. Jones 4
A.M. van Genderen 5
M. de Groot 6
3 University of Brussels (VUB), Pleinlaan 2, 1050 Brussels, Belgium
4 Carter Observatory, P.O. Box 2909, Wellington, New Zealand
5 Leiden Observatory, Postbus 9513, 2300RA Leiden, The Netherlands
6 Armagh Observatory, College Hill, Armagh BT61 9DG, Northern Ireland
Abstract
We review the status of photoelectric and visual monitoring of key objects among southern massive stars,
specifically the LBVs/hypergiants HD 6884 (R 40), j Car, AG Car, HDE 326823, HDE 269006 (R 71), and
WR stars HD 5980, WR 40. Data have been obtained over the last two decades in the framework of the Long­
term Photometry of Variables project and through visual monitoring from New Zealand. Unpublished light
curves are presented and novel insights on the light variability of these most enigmatic stars are discussed.
Keywords: LBVs, WR stars, HDE 326823, j Car, AG Car, R 71, R 40, HD 5980, WR 40
1 Introduction
The stars that are discussed here are massive early­type stars such as Luminous Blue Variables (LBVs) and
WR stars. These stars are photometrically variable with a large range of amplitudes (several hundredths of a
magnitude to magnitudes) and on a vast range of time scales (hours, over decades, to centuries). Variability
is partially due to rapid and unsteady mass loss, but several other mechanisms have been proposed to account
for the observed variations, such as binarity and pulsational instability. With respect to the last­mentioned
type of variability, LBVs share this characteristic with all massive evolved stars called the ff Cyg variables.
Figure 1 is a theoretical HR diagram (evolutionary tracks for models with Z = 0:001 from Schaller et al. 1992)
with the location of the stars discussed in this paper. The horizontal boxes around the positions of HD 5980,
AG Car, R 71 and R 40 correspond to the extreme positions observed for these objects. The light variability of
hypergiants and LBVs has been reviewed by Sterken (1989); for a very complete review on LBVs, see Humphreys
& Davidson (1994).
2 The data
2.1 Photoelectric uvby measurements
uvby data were obtained at ESO in the framework of the ``Long­Term Photometry of Variables'' (LTPV) project
which was initiated more than a decade ago (Sterken 1983, 1994). The data were published in four catalogues
by Manfroid et al. (1991, 1994) and by Sterken et al. (1993, 1995), and we refer to these papers (and to Sterken
& Manfroid 1992) for more details on the observing strategy and on the data reduction procedures.
2.2 Photoelectric V BLUW measurements
The V BLUW photometry was made with the 90cm Dutch telescope at ESO equipped with the simultaneous
V BLUW Walraven photometer. A general description of the monitoring campaign (including the observing
strategy and reduction procedures) of luminous and massive stars is given by van Genderen et al. (1985).
1 Based, in part, on observations obtained at the European Southern Observatory at La Silla, Chile
2 Belgian Fund for Scientific Research (NFWO)

2.3 Visual estimates
All new visual estimates presented in this paper were obtained by one of us (AJ) during the last decade from
his home observatory in Nelson, New Zealand (latitude ¸ \Gamma41 ffi ). Part of these data have been incorporated by
F. Bateson in the data bank of the Variable Star Section of the Royal Astronomical Society of New Zealand.
The origins of other visual data are given in the text.
2.4 The compatibility of the different sources of data
Merging of magnitudes and colour indices obtained for the same stars in different photometric systems, or
obtained by different observers at different sites using different equipment, is not a trivial undertaking. For an
idea of the incompatibilities between various versions of the uvby system see, e.g., Manfroid & Sterken 1987.
The major problems are the so­called conformity errors, which arise from the fact that photometric systems
have mutually different passbands, and that there is no way to rigorously evaluate the corrections needed to
properly transform data from one such system to another.
In this paper we only discuss data from ``visual'' bands, i.e. the Str¨omgren y band, the Walraven V band and
the visual ``eye'' band (m V ; for an idea of the large difference in spectral coverage of the uvby and visual bands,
we refer to Fig. 2). In order to bring the Walraven V data to the same scale as the Str¨omgren y magnitudes, the
difference between nearly­simultaneously obtained differential magnitudes was determined, and this correction
was then applied to all the Walraven V data, so that they could be plotted on the same scale. However,
we have not attempted to bring to agreement the visual and photoelectric data, because such an enterprise
involves too many parameters which, by nature, are ill­defined. In particular, the spectral energy distribution
of these emission­line stars combined with the extreme width of the eye's passband makes it impossible to
establish reliable transformation schemes. As will be seen in the following Sections, the strength of the visual
observations lies first of all in their long history of recording, and in the extremely high frequency of observation.
The discussion of the compatibility of data, naturally, brings us to the assessment of the level of accuracy
of our data. Whereas the photoelectric data are to be trusted within \Sigma0: m 003 \Gamma 0: m 005, the visual data should
be acceptable to \Sigma0: m 08 \Gamma 0: m 15. However, we warn that this is true only for data taken at air masses that are
of the same order of magnitude as is the case for the photoelectric data. When larger air masses could not
be avoided (because of the season and the weather, see Fig. 9), the given mean error should not be taken at
face value, and some corrections should be applied. We also wish to stress that switching to a CCD detector
(as is discussed in other contributions to this meeting) will not necessarily represent an automatic increase in
accuracy, see Fig. 3 for an assessment of achievable accuracies. Figure 4 illustrates the problem that CCD­based
colour indices inescapably are based on very different exposure times for the contributing passbands.
3 The observed stars
3.1 HDE326823
HDE 326823 (CD­42 ffi 11834, He3­1330; V = 9:0), an extreme supergiant, belongs to an aggregate of distant OB
supergiants in the Galactic Centre region. Feast et al. (1961) labelled the star as ``peculiar emission object'',
and Stephenson (1974) reports the presence of H and K emission ``somewhat like a nova near maximum''.
Figure 5 shows the photoelectric V and visual light curves of HDE 326823. The photoelectric data reveal
a large­amplitude (\Sigma0: m 15) variation; the average V for the data collected after JD 2448600 is about 0: m 05
brighter than the average V from the data taken before. The visual light curve (10­day averages) displays
a similar behaviour for the data preceding JD 2448600, but from then on, the light curve changes to a wavy
pattern with a characteristic time scale of 300--400 days, and an increase in light output as estimated (m V ) is
also visible. The reason why that onset of brightening was missed by the photoelectric observers was because the
measurements of HDE 326823 had been suspended because the observing time was needed for ``more interesting''
objects. It was the visual observer's alert that made us to observe the star again. The long­term behaviour of
the colours is also very interesting: the b \Gamma y, m 1 and c 1 colour indices display a similar pattern of variability
as y. The range of variability in c 1 is huge: almost 0: m 1. It is clear that the whole system of HDE 326823 has
become bluer---for more details, see Sterken et al. (1995b). The event fits the picture that this H­deficient
N­rich star is on its way towards becoming a WN star, making HDE 326823 a WN­precursor.

3.2 WR 40
WR40 (HD 96548, V ¸ 7:8), the brightest known WN8 star, has been studied by several authors because of
its rather large range of variability (up to 0: m 1 in V ). The star was monitored quite intensively in the LTPV
project (for the impact of the emission line spectrum on the uvby visual passbands, see Fig. 4 of Gosset et al.
1994). Though not necessarily an attractive object for visual monitoring, it has been checked quite regularly
in the hope of discovering flare­like behaviour as was seen in the case of HDE 326823 (Section 3.1). The y light
curve reveals quasi­regular variations on time scales of ¸ 10 to 20 days (see Fig. 6, bottom panel). The average
level of the long­term y light curve (a 6­year overlap with the visual estimates from 1988 on) corresponds very
well with the visual magnitudes (Fig. 6, top), although the onset of the wavy pattern seen after 1994 in the
visual estimates is not present. A frequency analysis of the visual estimates collected after 1993 reveals a very
strong peak in the amplitude spectrum at 0.00263 cycles per day (380 days), exactly the same frequency as
the strongest peak in the spectral window. This is most probably due to colour effects since the star---at a
declination of \Gamma69 ffi ---was observed almost continuously throughout the year; see also the case of j Carinae in
Section 3.4.
3.3 HD 5980
HD 5980 (WN4+O7I, V ¸ 11) is a bright Wolf­Rayet star in the SMC. It is also the only extragalactic Wolf­
Rayet known to show eclipses. The eclipsing character of the star was discovered by Hoffmann et al. (1978)
and the orbital period, P = 19: d 266 was found by Breysacher & Perrier (1980). The system has a strong
eccentricity, viz., e = 0:324 (Breysacher & Perrier 1991). The evidence for the existence of more than two
eclipsing components on the basis of a detailed light curve analysis of the system has also been presented by
Breysacher & Perrier (1991). During eclipses, the width of the optical emission lines decreases by more than a
factor of 2 (Breysacher et al. 1982). The spectacular changes of the emission features registered during the past
14 years suggest that the system has suffered an outburst­like event (Barba & Niemela 1995), also reported for
the UV region by Koenigsberger et al. (1994).
Bateson (1994) reported Jones' discovery of HD 5980 brightening by more than one magnitude in July 1994
(see Fig. 7). Since mid­1995 the star is back at the brightness it had before the 1994 flare. The HD 5980 system
has some very appealing characteristics that makes it an exciting target to observe:
ffl the uniqueness of the object---if indeed both components are W­R stars---in an eclipsing configuration
ffl the high eccentricity of the system
ffl the good knowledge of the distance modulus
ffl the SMC membership, a very interesting aspect from the point of view of chemical composition
ffl the fact that the star has been well­monitored in the past, thus allowing the detection of possible changes
in the orbital period
ffl the low southern declination which allows unique coverage in terms of intensive monitoring (virtually full
nights at slowly­changing air mass)
ffl its brightness brings it well into the observing capabilities of a modest­aperture telescope
ffl though previously known as a low­amplitude variable (except, of course, for its eclipsing character),
HD5890 has proven that regular visual monitoring is essential and very rewarding
3.4 j Carinae
The most enigmatic object among our targets certainly is j Car. Very detailed descriptions of the light variations
of j Car have been published by van Genderen & Th'e (1984) and van Genderen et al. (1994, 1995). In van
Genderen et al. (1995) we presented and analysed two seasons of intense photometric monitoring at a time
resolution that is unprecedented for this star (1993--1994, see top panel in Fig. 8). Due to this large amount of
new data, and also due to the fact that j Car was in a relatively quiescent state (no S Dor phases), a pulsation
period of P = 58: d 6 could be very clearly recognised in the photometric data. Figure 8 illustrates the long­term
aspects of the brightness behaviour of j Car: the straight line is the linear fit (over the same time interval) to

the visual and photoelectric data, yielding an equal slope for both datasets, indicating a steady brightening of
the star from 1988 to 1995 and an apparent decline from 1995 on.
Figure 9 shows the corresponding run of airmass X in function of time: the periodic (annual) appearance
of very high peaks in the air mass (allowed for by the deep southern latitude of the observing site, and by the
fact that the visual observer is much less bound to observing at small hour angles than is the photoelectric
observer). Data collected at such high air masses must be contaminated by the strong colour changes provoked
by the earth's atmosphere, and these effects cannot be eliminated with high accuracy.
3.5 R 40
R40 (HD 6884) is the first detected LBV in the SMC (Szeifert et al. 1993). Together with a steady brightening
over more than half a magnitude in one decade, it changed its spectral type from B8Ia to A3Ia­O. On top of
this brightening, the star displays more or less regular oscillations on a time scale of several months. Figure 10
shows the y--m V correspondance for the period covered by our data. As is the case for j Car, both data sets
reflect the brightening phase in an identical way.
Figure 11 illustrates the microvariability in light and colour for this star. It is very clear that the quasi­linear
brightening is associated with a slow reddening (visible in b \Gamma y, v \Gamma b and u \Gamma v), a behaviour so typical for an
LBV.
3.6 R 71
R71 (HDE 269006, V ¸ 10:6) is another LBV. Figure 12 displays 10­day averages of the visual estimates, and
shows that no dramatic events occurred during the time span covering the measurements, except for a slight
increase in brightness starting late 1990 and ending in mid­1992. Figure 13, on the other hand, illustrates the
very typical LBV behaviour: after a strong maximum that occurred in 1975 (V ¸ 9:9), the star descends to
V ¸ 10:8 \Gamma 11:0 and keeps fluctuating with a period of about 3 years and range ¸ 0: m 15 with superimposed
microvariations of a much shorter time scale.
3.7 AG Car
AG Car (HD 94910) is another most enigmatic LBV of the southern hemisphere. Figure 14 displays the visual
data, the solid curve is a straight line drawn through the LTPV y magnitudes.
Figure 15 is reproduced from van Genderen et al. (1996), and gives the schematic light curve of AG Car
covering 105 years of magnitude mesasurements and estimates. The solid lines cover the observations, the dashed
lines are free­hand inserts that illustrate how the authors think the corresponding time gaps could be covered,
and the dotted line is the lower envelope of the continuous line. The data used for the top panel are almost
exclusively due to photographic monitoring, the curves in the lower part are based on combinations of visual
estimates and photoelectric magnitude measurements. For details on the provenance of these data, we refer to
van Genderen et al. (1996). From 1955 till about 1969 there is a marked ambiguity in the visual estimates:
data collected by de Kock (indicated by dKm V ) and Jones--Bateson (JBa m V )/Bateson & co­workers (Ba et
al mV ) do deviate from each other, and we see no direct way how to find out which curve is the one to be
preferred. Fortunately, this ambiguity does not affect the times of maximum, nor the numbering scheme, since
nearly all peaks are visible in both groups of data.
The numbers 1--38 indicate the peaks---that is, moments where a rising branch of the light curve turns into a
descending branch---which seem to form a regular sequence of cycles. Van Genderen et al. (1996), introducing
a new nomenclature, call these cycles ``normal S Dor (SD) phases''---note that the term ``phase'' does not
mean here a constant added to the argument of a trigonometric function (or a particular point in a measured
sequence), but a stage or interval in time (as used in archeology or history). The lower envelope sketched in
Fig. 15 also shows a wavy pattern on a time scale of about 19 to 27 y. This type of oscillation is coined by van
Genderen et al. (1996) the ``Very­long term S Dor phase (VLT--SD)''.
From 1970 on (after JD 2440000) the photometric history of AG Car is completely and unambiguously
covered by visual estimates and a solid amount of photoelectric data (all photoelectrically observed maxima
are, by the way, also present in the visual light curves). Thus, a unique cycle­count scheme could be established,
and this led to the assignment of the period of about 373 d for the SD phase. This period, in turn, allowed to
tie the preceding times of maximum in a linear ephemeris, which led to a final PSD = 371: d 4 for the normal SD
cycles covering almost 100 years.

The O \Gamma C diagrams for the resulting linear ephemeris are shown in Figs. 16 and 17. The former figure
clearly shows a cyclic pattern on a time scale of ¸ 20 y, as is also seen in the independently derived VLT--SD
cycle covering the same period. The latter figure also suggests a cyclic pattern, though on a time scale that is
about twice as long, again as is seen in the wave length of the underlying VLT--SD phase.
The derivation of the above important findings is based on careful scrutiny of not only old data, but also
visual estimates, and may provoke some scepticism. Therefore, an important test of the linear ephemeris derived
by van Genderen et al. (1996) is provided by the visual data collected after the conclusion of their study. As
Fig. 14 indicates, AG Car has gone through a super­maximum similar to the one observed in 1981--1982, and
it seems to have a structure with at least two discrete maxima, viz. JD2449660 and 2450083, respectively 2
and 3 cycles removed from the latest securely­observed maximum (number 37 in Fig. 14), thus conform the
description given by van Genderen et al. (1996).
4 Conclusions
We have demonstrated that the sytematic monitoring of luminous massive stars is scientifically very rewarding
and yields new insights that were unanticipated even a decade ago. Such measurements need to possess at
the same time high accuracy and high time resolution. At the time of the opening of a new observing
facility like Mt. Kent Observatory, we find it appropriate to stress again that the interpretation of the light
variability of massive stars is first­class science that stands on the shoulders of very long series of photoelectric
measurements---if not more so on long lists of published visual estimates, as was expressed by van Genderen et
al. (1996) in the epilogue of their recent paper on AG Car, j Car and S Dor:
``We like to give tribute to the numerous dedicated (amateur) astronomers of the Variable Star
Section of the Royal Astronomical Society of New Zealand, whose thousands of visual observations
of AG Car and S Dor were of crucial importance for the completion of the light curves when
photoelectric observations were interrupted by large gaps in time. Without those observations
numerous SD phases in the light curve of AG Car would have been missed, making a proper analysis
of the periodicity more difficult, if not impossible.
Many types of peculiar variables, such as the S Dor-- or Luminous Blue Variables, are of crucial
importance for understanding particular phases of stellar evolution. Such better understanding de­
pends crucially on continuous and dedicated monitoring, whether done by professional or by amateur
observers. The possibilities for monitoring campaigns by professionals are declining as a result of
the fact that most major observatories are reducing their number of small telescopes while building
large ones that cannot be used for these purposes. This is a matter of grave concern. Therefore, we
encourage amateur astronomers all over the world to take up and continue observing peculiar
variable stars, even if their equipment is modest, and to make those data available to the sci­
entific community. Establishing more photoelectric telescopes, automatic or observer­operated,
also in the southern hemisphere from where important objects within our Galaxy can be studied, is
of imminent urgency and should have the support of all those who promote galactic astrophysics.''
5 Acknowledgements
C.S. acknowledges financial support from the Belgian Fund for Scientific Research (NFWO). Research at the
Armagh Observatory is grant­aided by the Department of Education for Northern Ireland, and by the UK
PPARC through the provision of the STARLINK network. A.J. thanks the Kingdon­Tomlinson Astronomical
Trust and the Donovan Astronomical Trust for grants of funds for travel. Part of the data discussed in this
paper were collected under observing program ESO 57D­0133. Special thanks are due to D. van der Bij, F.
Robijn, R. van der Heiden, B. van Esch, L. Spijkstra, J. Prein, R. Reijns, J. de Jong, F. Dessing, A. Hollander,
R. van Ojik, J. van Grunsven and G. Fehmers for obtaining the Walraven photometric measurements.
References:
Barba, R., Niemela, V., 1995, IAU Symp. No. 163, p. 254
Bateson, F.M., 1994, IAU Circ. 6102
Breysacher, J., Perrier, C., 1980, AA 90, 207
Breysacher, J., Perrier, C., 1991, IAU Symp. No. 143, 229
Breysacher, J., Moffat, A.F.J.,, Niemela, V.S., 1982, ApJ 257, 116
Feast, M.W., Stoy, R.H., Thackeray, A.D., Wesselink, A.J., 1961, MNRAS 122, 239

van Genderen, A.M., 1979, AAS 38, 381
van Genderen, A.M., Alphenaar, P., van der Bij M.D.P. et al. 1985, AAS 61, 213
van Genderen, A.M., Th'e, P.S., 1984, Sp. Sc. Rev 39, 317
van Genderen, A.M., de Groot, M., Th'e, P.S., 1994, AA 283, 89
van Genderen, A.M., Sterken, C., de Groot, M., Stahl, O., Andersen, J., Andersen, M.I., Caldwell, J.A.R., Casey,
B., Clement, R., Corradi, W.J.B., Cuypers, J., Debehogne, H., Garcia de Maria, J.M., J¨onch­S¨orensen, H., Vaz,
L.P.R., Stefl, S., Suso Lopez, J., Beele, D., Eggenkamp, I.M.M.G., G¨ocking, K.­D., Jorissen, A., de Koff, S., Kuss,
C., Schoenmakers, A., Vink, J., W¨alde, E., 1995, AA 304, 415
van Genderen, A.M., Sterken, C., de Groot, M., 1996, AA (in press)
Gosset, E., Rauw, G., Manfroid, J., Vreux, J.­M., Sterken, C., 1994, in The Impact of Long­Term Monitoring on
Variable­Star Research, NATO ARW, Eds. C. Sterken, M. de Groot, NATO ASI Series C, 436, 101, Kluwer
Academic Publishers
Hoffmann, M., Stift, M.J., Moffat, A.F.J., 1978, PASP 90, 101
Humphreys, R.M., Davidson, K., 1994, PASP 106, 1025
JÜnch­SÜrensen, H., 1994, A&A 292, 363
Koenigsberger, G., Moffat, A.F.J., St Louis, N., Auer, L.H., Drissen, L., Seggewiss, W., 1994, ApJ 436, 301
Manfroid, J., Sterken, C. 1987, AAS 71, 272
Manfroid, J., Sterken, C., Bruch, A., Burger, M., de Groot, M., Duerbeck, H.W., Duemmler, R., Figer, A., Hageman,
T., Hensberge, H., Jorissen, A., Madejsky, R., Mandel, H., Ott, H.­A., Reitermann, A., Schulte­Ladbeck, R.E.,
Stahl, O., Steenman, H., vander Linden, D., Zickgraf, F.­J. 1991, AAS 87, 481
Manfroid, J., Sterken, C., Cunow, B., de Groot, M., Jorissen, A., Kneer, R., Krenzin, R., Kruijswijk, M., Naumann,
M., Niehues, M., Sch¨oneich, W., Sevenster, M., Vos, N. Vogt, N.. 1994, AAS 109, 329
Olsen, E.H.: 1993, A&AS 102, 89
Schaller, G., Schaerer, D., Meynet, G., Maeder, A., 1992, AAS 96, 269
Stephenson, C.B., 1974, ApJ 191, 685
Sterken, C. 1983. The Messenger 33, 10
Sterken, C., 1989, in Physics of Luminous Blue Variables, Davidson, K., Moffat, A.F.J., Lamers, H.J.G.L.M. (eds.),
Kluwer Ac. Publ. , 59
Sterken, C., 1994, in The Impact of Long­Term Monitoring on Variable­Star Research, NATO ARW, Eds. C. Sterken,
M. de Groot, NATO ASI Series C, 436, 1, Kluwer Academic Publishers
Sterken, C., 1996, in IAU Coll. 167, New Developments in Array Technology and Applications, A.G. Davis Philip et al.
(eds.), Kluwer Dordrecht p. 131
Sterken, C., Manfroid, J. 1992. Astronomical Photometry, a Guide. Kluwer Academic Publishers, Dordrecht.
Sterken, C., Manfroid, J., Anton, K., Barzewski, A., Bibo, E., Bruch, A., Burger, M., Duerbeck, H.W., Duemmler, R.,
Heck, A., Hensberge, H., Hiesgen, M., Inklaar, F., Jorissen, A., Juettner, A., Kinkel, U., Liu Zongli, Mekkaden,
M.V., Ng, Y.K., Niarchos, P., P¨uttmann, M., Szeifert, T., Spiller, F., van Dijk, R., Vogt, N., Wanders, I. 1993,
AAS, 102, 79
Sterken, C., Manfroid, J., Beele, D., de Koff, S., Eggenkamp, I.M.M.G., G¨ocking, K., Jorissen, A., Kaufer, A., Kuss,
C., A.P. Schoenmakers, A.P., J.M. Stil, J.M., van Loon, J., Vink, J., Vrielmann, S., W¨alde, E. 1995a, AAS 113,
31
Sterken, C., Stahl, O., Wolf, B., Szeifert, T., Jones, A., 1995b, AA 303, 766
Szeifert, Th., Stahl, O., Wolf, B., Zickgraf, F.­J., Bouchet, P., Klare, G., 1993, AA 280, 508

###
###
###
###
###
###
ORJ#7HII
###
###
###
###
###
###
ORJ#/#/VXQ ##
##
##
##
##
5##
h#&DU
###
+'#####
$*#&DU
+'(#######
:5##
5##
Figure 1: Theoretical HR diagram (evolutionary tracks for models with Z = 0:001 from Schaller et al. 1992)
with the location of the stars discussed in this paper. The horizontal boxes around the positions of HD 5980,
AG Car, R71 and R 40 correspond to the extreme positions observed for these stars.

###
###
###
###
###
###
###
###
###
###
###
1RUPDOLVHG#7UDQVPLVVLRQ#RU#5HVSRQVH
### ### ### ### ### ### ### ### ### ###
ZDYHOHQJWK##QP#
X Y E \
9#
H\H
Figure 2: uvby and V transmission curves, together with the response curve for the eye (night vision). All
curves are normalised to maximum transmission or response.
# # # # ## ## ## ## ## ## ## ## ## ## ## ##
9##PDJ#
####
####
####
####
####
s#PDJ#
b
9
E#\
P #
F #
2OVHQ#######
-RQFK#6RUHQVHQ#######
Figure 3: Standard deviations for uvby (Olsen 1993, photomultiplier at 50cm telescope) and uvbyfi (JÜnch­
SÜrensen 1994, CCD at 1.5­m telescope), from Sterken (1996).
300 400 500 600 700
wavelength (nm)
1
10
100
Exposure
time
Figure 4: Normalised exposure times (in seconds) for CCD frames: filled circles from JÜnch­SÜrensen (1994),
open circles from Surma (1993). Source: Sterken (1996).

##### ##### ##### ##### ##### #####
-'##########
###
###
\
###
###
P 9
#### #### #### #### #### #### #### #### ####
+'(############################################G#DYHUDJHV
Figure 5: y and mV light curves for HDE 326823. The visual estimates are 10­d averages.
##### ##### ##### ##### ##### #####
###
###
P 9 :5###
#### #### #### #### #### #### #### #### ####
##### ##### ##### ##### ##### ##### ##### #####
###
###
\
#### #### #### #### #### #### #### #### #### ####
##### ##### ##### ##### #####
+-'########
###
###
###
\
:5##
Figure 6: Top: visual estimates of WR40 disclosing the brightening events; middle: Str¨omgren y magnitudes;
bottom: high­frequency variations (microvariations) on time scales of 10--20 days.

##### ##### ##### ##### ##### ##### #####
-'##########
###
###
####
####
P 9
#### #### #### #### #### #### #### #### ####
+'#####
#### ### ### ### ### ### ### ###
3KDVH
###
###
###
E
Figure 7: The eclipsing WR star HD 5980. Top: phase diagram of Str¨omgren b measurements based on data
from Breysacher & Perrier (1980), Bottom: raw visual magnitude estimates.
##### ##### ##### ##### ##### #####
-'##########
###
###
###
###
###
P 9
\
#### #### #### #### #### #### #### #### ####
h#&DU
Figure 8: Visual (10­d averages, bottom) and y light curves for j Car (top). The straight lines are fitted to the
data over the same time base.
##### ##### ##### ##### ##### #####
-'##########
###
###
###
###
;
Figure 9: Run of airmass X in function of time for visual estimates of j Car.

##### ##### ##### ##### ##### ##### ##### #####
-'##########
###
####
####
P 9
####
####
9 #### #### #### #### #### #### #### #### #### #### ####
5##
Figure 10: Visual (10­d averages) and y light curves for R 40. The straight lines are fitted to the data over the
same time base.
#### #### ####
+-'########
####
####
####
####
####
####
####
####
\##
####
####
E#\
####
####
Y#E
###
###
###
###
###
X#Y
5###
Figure 11: y; b \Gamma y; v \Gamma b; u \Gamma v light­ and colour curves for R 40.

##### ##### ##### ##### ##### ##### #####
-'##########
####
####
####
P 9#
5### #+'########################################################G#DYHUDJHV#
#### #### #### #### #### #### #### #### ####
Figure 12: 10­day averages of visual estimates of R 71.
#### #### #### #### #### #### #### #### #### #### #### #### #### #### ####
+-'########
####
####
####
####
####
####
\
5### #+'######
#### #### #### #### #### #### #### #### #### ####
Figure 13: R 71: combination of LTPV data and data from van genderen (1979). The line represent the
low­frequency least­squares sine fit (P ¸ 2100 d ).
##### ##### ##### ##### ##### #####
-'##########
#
#
#
P 9#
#### #### #### #### #### #### #### ####
$*#&DULQDH
Figure 14: Visual estimates of AG Car. The continuous line connects the photoelectric y data.

Figure 15: Schematic light curve of AG Car covering more than one century. Source: van Genderen et al.
(1996).
0 2 4 6 8 10 12 14 16 18 20
E
­200
­150
­100
­50
0
50
100
150
200
O­C
(days) 17
18
19 20
21
22
23
24
26
27 28
29
30
31
32
33
34
35
36
37
38
Figure 16: O \Gamma C diagram for SD maxima 17--38 of AG Car for the linear ephemeris constructed with P = 371: d 4,
van Genderen et al. (1996).
­70 ­60 ­50 ­40 ­30 ­20 ­10 0 10 20
E
­200
­150
­100
­50
0
50
100
150
200
O­C
(days) 3
5
6
7
10
12
13
14
15
16
Figure 17: O \Gamma C diagram for all SD maxima of AG Car; van Genderen et al. (1996).