4 Results

The raw results of the processing are presented for 22 systems in Table  6, in the same layout as in Paper II. Five values of $\beta-B$ have a formal error less than 0.01, 3 are in the range 0.01-0.02 while the remaining 14 have errors larger than 0.02, a maximum being reached for HIP 55016 (0.084) and HIP 81126 (0.087).

For 16 systems, estimates of $\Delta m$ (and thus of $\beta$) are taken from ground-based sources while the Hipparcos data (ESA 1997) were used for the other systems. The transformation into the Hipparcos photometric system was done for the 10 systems with available spectral types for each component (see Table 5). The procedure is described in Paper II, Sect. 5.2. The details of the mass computation will be found in Paper II.

  

Table 7: Masses of 22 systems. See comments at the end of the table

¹ Correction: "yes" indicates that the $\Delta m$ estimate has been brought into the Hipparcos photometric system, via the spectral types of the components (see Sect. 5.2). "no" means that this conversion could not be done due to the lack of individual spectral types, and "-" means that $\Delta m$ is taken from the Hipparcos catalogue and thus does not need any conversion.
² Difference between the parallax derived from this processing and the Hipparcos catalogue's value (ESA 1997), followed by its standard deviation. Most of the differences are not significant, except for HIP 82817 = Kui75, where the catalogue's value is probably wrong.
³ It is the Hipparcos estimate if "-" stands in the second column, a ground-based one otherwise.
⁴ When unknown, $\sigma$ is taken equal to 0.15.
⁵ Number of available terms for the computation of the standard deviation of the total mass M. N=3 if $\sigma\rm _a$, $\sigma_{\pi}$ and $\sigma\rm _P$ are known. N=2 if $\sigma\rm _P$ is unknown. N=1 if both $\sigma\rm _a$ and $\sigma\rm _P$ are ignored. In the latter case, $\sigma(M)$ is underestimated.

  

Table 8: Masses for 6 systems already studied in the previous paper, reprocessed with the Hipparcos magnitude difference estimate. See comments at the end of the previous table