The occurrence and variability of emission lines is one of the main characteristics of PMS objects, although it has been argued by Thé et al. (1993) that these lines will diminish in strength when arriving on the MS even though the objects are in their pre-ZAMS phase. Not all young objects will, therefore, directly show clear and strong emission lines. We have identified 29 objects with emission features in our sample, Table 4, which is about 50% of the total sample or even 2/3rd if the non-cluster members are not taken into consideration. In the HRDs of Figs. 9 (click here)a-b, the PMS candidates with emission or traces of emission features are indicated. Note the fact that most non-emission and emission line stars are not clearly distributed. Twenty seven emission line objects were also detected spectroscopically in the study of Hillenbrand et al. (1993), or about 35% of their sample of 79 objects.
As stated by Hillenbrand et al. (1993), it is a well known difficulty to calibrate emission lines for a star in a cluster such as NGC 6611. The bright ``background'' nebula will exhibit strong but sharp emission lines. If the star is located in or behind such a nebula it is very challenging to make an appropriatea subtraction of the nebular emission. A good subtraction is only possible on long-slit CCD exposures (Horne 1986) as obtained by us. Another difficulty could be the absorption in the cluster region. If a programme star is located in the cluster, part of the radiation in the hydrogen lines will be absorbed by dark inter-cluster material. It is, therefore, questionable whether the stellar emission is strong enough to be detected. Whether these conjectures are realistic seem to be suggested by the contradicting results of the IDS and CCD data. Eleven objects are detected to have emission lines in one but not in both of the spectra. Conclusions of the PMS nature of a star based on the absence of emission lines, and considering its line variability, should be treated carefully. Consequently, it is difficult to compare detections of emission line objects with those of other samples, such as the ``MS-sample'' as plotted in Fig. 9 (click here)b. Additional spectra are necessary, especially of the more luminous objects, to detect emission features related to the evolutionary process.
The presence of emission lines does not seem to be related to the location in the cluster and the presence of anomalous extinction or an IR-excess, which confirms a similar finding by Hillenbrand et al. (1993).
Note that the spectrum of W339 seems to be of symbiotic origin. Apart from lines indicating the existence of an early type object, some wide absorption bands seem to be present which could indicate the presence of a T Tauri companion. This companion could explain the presence of emission lines other than those of hydrogen.
Figure 10: HR-diagram with the positions of the sample of PMS-candidates of NGC 6611 as taken from Table 7. The emission-line stars are indicated by solid dots. Stars with possible emission by half open dots and the open dots the non emission-line objects. They are compared to: a) The pre-main sequence evolutionary tracks and birthlines, dotted, of Palla & Stahler (1993); b) The post-main sequence evolutionary tracks of Maeder & Meynet (1988) together with the results of Thé et al. (1990) on the MS-candidates of NGC 6611, indicated by stars. Crosses indicate the objects which are not or probably not cluster-members as listed in Table 8. The zero-age main-sequence from Hillenbrand et al. (1993) is also given. The latter does not suite the theoretical results in all the cases