4. Quality of the parameters and types

A detailed comparison of the morphological types of the whole Coma sample of galaxies with other published studies is presented in Andreon & Davoust (1997). It shows that the main objective of this work and of Paper I, reliable estimates of morphological types, has been reached, since these types are at least as good as the traditional ones, because less subjective, more reproducible and based on images of adequate quality.

The quality of the parameters listed in Tables 4 (click here) and 5 (click here) (magnitudes, effective radii, representative ellipticities, etc.) does not differ from that of the parameters presented in Paper I, because of the close similarity of the data and of the analyses.

Some discrepancies have been found between our values of representative quantities (such as ellipticity or e₄) and published ones, but we stress that they are largely due to differences in the definition of what is a "representative" quantity, whether it is an intensity averaged quantity, or the quantity at the galaxy effective radius, at its maximum or at the extremum, and of what is its value when not just an extremum is present or when we only measure an incomplete range of galaxy radii (i.e. always because of seeing or sky brightness limitations). In Paper I, the comparison of these "representative" quantities shows that the typical errors are of 0.06 on ellipticity (and our ellipticities are systematically larger than the others by 0.05) and of 1.3% on e₄ (and our e₄ are larger than the others by 0.7%). These figures, based on more than 200 comparisons, are also valid for the data presented in this paper.

Aside from errors on sky determination, the effective radii suffer from the existence of two definitions, the radius containing half the light, measured by extrapolating the luminosity growth curve and taking the radius where the integrated magnitude is 0.75 mag fainter than the total one, or the slope of the SuBr profile, measured by the best fit of the SuBr profile with a de Vaucouleurs' law. Adopting the former method, the subjective extrapolation of the growth curve implies a typical error of 0.02 in and for galaxies of range 0.5 to 1.0 in (where is in units of arcsec), or, more precisely, this is the typical scatter between estimates of different observers, all using the same growth curves. Much larger differences have sometimes been found for galaxies whose growth curves differ from the standard ones listed in RC3(de Vaucouleurs et al. 1991), used by us as standards.