1995A&A...298..219M

We give a critical analysis of the hypothesis of an accretion disk around young stars. We show that the accretion disk model, which is based mainly on the interpretation of the observed infrared excesses, is often in contradiction with observations. Recent results obtained for some Herbig Ae/Be stars in a serie of papers by Grinin and coauthors, as well as the similarity in the behaviour of Ae/Be stars and T Tauri stars, show that the circumstellar dust near such young objects cannot be entirely confined to thin disks. This conclusion is supported by the correlations between the forbidden line luminosities, the near infrared excess, and the reddening of T Tauri stars from Cabrit et al. (1990ApJ...354..687C). Such correlations clearly indicate that the near infrared excess cannot solely be interpreted as being due to accretion disks. Here, we suggest a new model of the circumstellar environment of young stars, based on the assumption that circumstellar dust is in clouds rather than in a homogeneous, spherical shell. The dusty clouds are optically thick at optical wavelengths and thin in the infrared. Such an envelope produces very little reddening, whereas in the infrared it reveals a significant excess. We estimate that a small amount of dust, distributed in the intercloud circumstellar space, is responsible for the observed reddening. A detailed analysis of the 10µm silicate feature shows that our model can explain the appearance of the feature both in absorption and emission. The absorption feature should be associated with the objects having a flat energy distribution in the infrared. The maximum intensity of the silicate feature indicates that circumstellar disks cannot contribute significantly (at least in some young stars) to the formation of an infrared excess at 10µm.