As was described in sect. 6, the nuclear pg-pv color indices have been determined for the majority of the central spiral systems. The nuclear color excesses, as obtained by subtracting the mean colors of Table 3, are listed in Table 7. The color excesses range from -0m.23 to +0m.24
It was found above that the nuclear color excess is negative in those cases in which the spectrum or a direct inspection of plates indicates an unusual activity in the nuclear region. On. account of the possibility of explosive events in such nuclei, it seemed of interest to examine whether the number of physical satellites of a spiral system is in any way related to the nuclear color or to the nuclear color excess.
The upper part of Fig. 6 shows the dependence of Nphys , as listed in Table 7, on the nuclear color index. The plot refers to the spiral systems of classes A - B with known colors, in all 72; the number of physical companions, ranging from -2 to +5, also totals 72. A study of the figure shows, quite conclusively, that there is no indication of any systematic variation in Nphys with CN.
Figure 6. Variation in number of satellites with nuclear color index of central galaxy (upper part), and with nuclear color excess (lower part).
Conditions are somewhat different, when Nphys is plotted against the nuclear color excess CN : a correlation seems to be indicated. About one-third of the spirals, those with color excesses larger than +0m.06 , have a mean number of satellites equal to 0.0; for the remainder the mean number is +1.5. If two spirals with exceptional color excesses are omitted, the coefficient of correlation amounts to -0.31 ± 0.11. Since the coefficient is almost three times the mean error, the result is supposedly significant. The. straight line represents the regression line.
It is naturally tempting to try to interpret this result as indicating that the satellites have been formed from matter, presumably gas, ejected from the nuclear regions of the spiral systems. It should however be pointed out that it is possible to determine the morphological types of satellites with diameters above 2.5 kpc, and that no correlation exists between these types and the nuclear color excess of the central galaxy; for all values of CN there is a mixture of all types from E-So to Ir I. Even without knowing anything about the process by which the gas may be condensed into satellites, we would have expected such a correlation if the suggested interpretation were correct. The material at hand does not seem to offer any other possible explanation.