As mentioned in the introduction we have used new observations as well as older data.
The observed Walraven WULBV photometric data are presented in Table 1. They were collected in August 1985 and August 1986 with a photometer attached to the 90 cm Dutch Light Collector stationed at ESO. These observations were made through a 165 diaphragm, and were calibrated using standard stars published by Lub & Pel (1977) of which the data are updated.
Observations in 
, taken during June 1983, July 1985 and
August 1986, and JHKLM, made in July 1986, are listed in Tables 2 and 3.
The August 1986 observations were all taken at relatively high airmass
and are therefore of low quality.
The equipment used for these observations is described in
Thé et al. (1990).
The data from the literature are taken from:
1) Thé et al. (1990): for a number of selected PMS-candidates the
observations are listed and described in Thé et al. (1990). These
observations include 
as well as JHKLM data.
In this paper different observing 
runs were compared
and no systematic differences between photoelectric data sets were found.
2) Walker (1961): only the photographic observations, taken
on June 1955 with the 100-inch reflector were used. The plates contain
all stars brighter than about 
. Only in some cases did the data
sets appear to be useful with respect to the photoelectric observations.
3) Sagar & Joshi (1979): the observations were carried out between April
1977 and October 1978, with the 104 cm Sampurnanand reflector of the Uttar
Pradesh State Observatory, India, employing standard UBV filters of the
Johnson system.
A minimum of two sets of observations, on different nights, were taken for
each star. The averaged UBV values are taken for our purpose.
4) Chini & Wargau (1990): Johnson UBVRI data were observed with
the 0.5 m telescope on Gamsberg in Namibia and with the
1.23 m telescope of the DSAZ on Calar Alto in Spain from 1978 to 1981.
The JHKL data were taken with the ESO 1 m telescope on La Silla
during August 1987 and with the SAAO 1.9 m telescope in Sutherland during July
1988. Their programme stars were measured at least twice in both the
UBVRI and JHK(L)
Johnson photometric systems. Only the averaged values are given.
5) Hillenbrand et al. (1993): all measurements were made at the Kitt
Peak National Observatory (KPNO). The photometric imaging data, taken in
October 1990 using a 90 cm telescope, consists of
several sets of overlapping UBV optical images covering approximately 1600
arcmin
centered on the cluster. Furthermore mosaicked JHK infrared
images covering roughly the central 700 arcmin
of the cluster
were made with the KPNO 1.3 m.
All observations of our sample stars, as derived from above studies, are listed in Table 2. In this table several data sets are clearly not in agreement with each other, e.g. the UBV data of W 494. Besides observational errors, variability can be the cause. A discussion of useful data for each object will be given in Sects. 4 and 5. A discussion of all errors involved in the observations and the averaging procedures are given by Koulis (1993).
For most of our programme stars a low resolution (172 Å mm
) spectrum
was taken in July 1985 or July 1986 with the Image Dissector Scanner (IDS)
attached to the Boller and
Chivens spectrograph mounted on the ESO 1.52 m telescope. These spectra cover
a spectral range from about 3700 to 7000 Å.
Because the IDS data was incomplete and sometimes not well exposed, we
made low dispersion (2 Å pixel
) spectra for each programme
star with the ESO 1.52 m telescope in June 1992, to which a Boller and
Chivens spectrograph was attached.
The detector is a high resolution Ford Aerospace 2048
2048 CCD,
with a pixel size of 15
15 
m.
The dispersion of 172 Å mm
produces a usable spectral range
from 4200 to 7800 Å.
All the IDS and CCD spectra obtained for our programme stars are given in Koulis (1993). For reasons which will be discussed in Sect. 10, we present only the CCD spectra, see Fig. 1 (click here).