8 Conclusions

8.1 New detections and DIB statistics

Figure 1 shows a composite spectrum of all DIBs seen toward the reddened targets in the wavelength range 3906-6812 Å. The detection of 60 new DIBs raised the total number of DIBs to 226 in this line-of-sight, along with 25 possible DIBs awaiting confirmation. A statistical analysis on the DIB population has clarified the presence of three main DIB groups based on their band widths and has illustrated the clustering of DIBs in selected wavelength bands.

8.2 The enhancement of DIBs in BD+63$^{\circ }$ 1964 vs. other reference targets

This survey of diffuse interstellar bands in the optical range has confirmed earlier reports of the unique nature of BD+63$^{\circ }$ 1964 compared to other heavily reddened targets, such as HD 183143 and BD+40$^{\circ }$ 4220 (Ehrenfreund et al. 1997). What mechanism could cause this unusual enhancement in DIB strengths and lead to the identification of many more hitherto undetected DIBs? The Ca II absorption lines at 3933 and 3968 Å, as well as the CH band at 4300 Å seem to indicate at least two clouds of different velocities in the line-of-sight of BD+63$^{\circ }$ 1964. In order to understand the DIB behaviour, models of diffuse clouds have been performed using information on various atomic and molecular species residing in the line-of-sight toward BD+63$^{\circ }$ 1964. The models indicate that the line-of-sight toward this target passes through a cloud edge where ultra-violet radiation may trigger the ionization of DIB carriers (Tuairisg et al. 2000). Further modeling and observational efforts are required to constrain the environmental conditions in this exceptional line-of-sight.

8.3 Future work

Higher signal-to-noise spectra of BD+63$^{\circ }$ 1964 and other reddened targets would undoubtedly confirm many of the possible DIBs listed in Table 3, and possibly lead to the detection of more new DIBs. Important further work would be to complement this work with a comprehensive survey of DIBs in the near infra-red. Observations taken with the INT/MUSICOS spectrograph are currently under analysis. As previously mentioned, the ELODIE spectrograph has limitations for a survey of broader DIBs. Further observations of reddened targets using more suitable spectrographs to detect broad DIBs, are also desirable.

The new survey presented in this paper provides a new basis for comparison with spectroscopic measurements of DIB carrier candidates in the laboratory. However, with the large number of detected DIBs namely $\sim$200 DIBs between 5400 and 6800 Å, we find on average one DIB every 6 Å. Therefore any identification in this range based only on wavelength coincidence could be obtained by pure chance and additional criteria for identification are necessary, such as multiple bands, band widths, DIB strength ratios and correlated variations in different environments (Ehrenfreund & Foing 1995; Herbig 1995).

Follow-up studies of the surveyed DIBs by high resolution spectroscopy and the physical properties of the line-of-sight conditions will further help to reveal the nature of DIB carriers.

Acknowledgements

We thank the staff at OHP for their help during the observations. PE is a recipient of an APART fellowship. SÓT acknowledges support for a research stage at ESA Space Science Department.