In Figures 10 and 11, the color-magnitude data from Figures 6 and 7, have been moved onto a luminosity--temperature plot using the following steps. First the color was converted to a temperature, using tables from Bessell (1990). Then the bolometric correction, appropriate for these temperatures was applied to the observed V magnitudes. Bolometric magnitude was then compared to the bolometric magnitude of the Sun at 380 pc. Since the reddening found for the stars in Table 13 was small, no reddening correction was attempted. The evolutionary tracks of D'Antona and Mazzitelli (1994) have been overplotted. Note the addition of seven diamonds at the lower right--hand corner of the plot. These diamonds indicate the positions of very low mass, 3 Myr old, objects from the work of Burrows et al. (1993). These models are similar to those mentioned above except that they focus particular attention to absorption of H molecules and Coulomb corrections due to free electrons in cool stellar atmospheres. Consequently these models are expected to be more accurate than other models in the cool regime. The diamonds represent objects of masses from 0.2 M to 0.05 M. From the figure, it is clear that masses determined using the Burrows et al.\ tracks are higher than those with the D'Antona and Mazzitelli tracks. Nonetheless, there are still a significant number of PMS sources cooler than the 0.08 M hydrogen burning limit. These objects are candidate brown dwarfs.
As stated above, no reddening correction was applied to the sources. Since the reddening vector runs in the redward direction along the main sequence, it is possible that these objects are simply highly reddened PMS stars. JHK data will aid in determining the reddening and temperatures of these objects. Spectral line diagnostics of these stars will give a more complete picture of their nature. It has been suggested (Stauffer, private communication) that there may be a dust sheet behind the Orionis cluster, and that the extremely red sources are simply embedded in the cloud. However, this is considered unlikely based on the comparison of Figures 6 and 7. From these figures, one can calculate the mean color of the background populations. The means are R-I = 0.55 and R-I = 0.4 for Orionis stars and stars northwest of the belt respectively. While there is difference in the overall extinction beyond the two groups, it is only about 0.15 in R-I, which translates to an extinction of only 0.5 in V. Global extinction probably does not exceed 0.8 magnitudes at V.
The number of stars near the brown dwarf cutoff may betray the existence of many more. The completeness limit of the diagram is approximately V magnitude 18.5. Note that in Figures 6 and 7, the density of sources in the PMS group decreases only slightly as this limit is approached. This seems to imply the existence of additional stars redward of those in the figure. These sources would either be subject to extreme reddening (A 2) or be below the hydrogen burning limit. In either event, the data do indicate a plethora of extremely low mass objects.
This optical V versus R-I scheme is a very powerful method of detecting very low mass (VLM) objects. The usual method of finding young brown dwarfs is to use deep infrared mapping of star formation regions (cf. Comeron et al.\ 1993, Stauffer et al.\ 1994 and references therein). The advantage of infrared mapping is that very low mass stars are expected to be brightest in these colors. As one moves toward the optical, flux falls off rapidly since cool objects have their flux peak well into the near--IR (although for a sufficiently cool temperature, absorption dominates and objects appear bluer as their effective temperature drops). When one looks at young stars (< 2 Myr in the case of Orionis), there is no discernible break in the luminosity function as one moves below the hydrogen burning limit (Burrows et al.\ 1989). Because of this, young brown dwarfs are intrinsically no more difficult to observe in the optical than low mass stars at the age of Orionis. Although 2 Myr old brown dwarfs have less flux by in optical wavelengths than the IR, it appears that this effect is outweighed by the wide field and low background of optical CCDs with respect to IR arrays. The ability to identify a complete population of PMS stars is very important to our understanding of these objects. Previous samples used to study the binary fraction have been naturally biased toward stars with strong emission lines or X--ray emission. The capacity to identify PMS stars independent of their activity levels will greatly aid us in the pursuit of understanding the nature of these objects.