Next Contents Previous


4.5.1. Presence of rings. Galaxies that possess external rings such as NGC 2859 [RSB(r)0+, Atlas, p. 42] are denoted by R preceding the designation of the basic type (SB in this case). The difficulty in designating rings in SB galaxies is that it is often very hard to distinguish a true ring from a form in which two spiral arms are tightly coiled and nearly touch after each has made a turn of 180°. Examples where it is certain that confusion would exist on small-scale plates between this form and true rings include NGC 3185 (Atlas, p. 43; note especially the description given), 2217 (Atlas, p. 43), and 3081 (Atlas, p. 11; note the description). More easily distinguished cases are 3504 (Atlas, p. 46), and 4750 (Atlas, p. 21). None of these galaxies have true rings. They are classed as (R)1 systems by de Vaucouleurs to denote ``pseudo rings.''

True rings do occur in such galaxies as NGC 2859 (Atlas, p. 42), but it is not certain that the rings are attached to the main body of the parent galaxy. Examples of definitely attached partial rings outside the regular spiral pattern include the faintest outer structure in NGC 2685 (Atlas, p. 7 insert), NGC 4736 (Atlas, p. 16), NGC 4457 (Atlas, p. 9), NGC 3368 (Atlas, p. 12), NGC 1068 (Atlas, p. 16), and NGC 5101 (Atlas, p. 42). True rings appear to occur predominantly in early-type systems, although cases such as NGC 4736 [(R)SA(r)ab, de Vaucouleurs 1963b; or Sb, Hubble-Sandage: Atlas 1961] exist. This case is particularly interesting because partial resolution of the ring into stars (B appeq 22 or MB appeq - 8) is seen on 200-inch plates.

4.5.2. Increasingly detailed notation. In his objections to Hubble's simplified system, Reynolds (1927a, b) pointed out that spiral arms differ in character. Some systems have ``massive'' arms, such as M33 (Atlas, p. 36), NGC 4567 (Atlas, p. 13), and M51 (Atlas, p. 26), while others have thin, delicate, filamentary arms, such as NGC 2841 (Atlas, p. 14), NGC 488 (Atlas, p. 15), NGC 628 (Atlas, p. 29), and NGC 1232 (Atlas, p. 32).

This feature is undoubtedly important and, although not recognized in Hubble's notation, does form the basis of a division of galaxies into strains or groups from Sa to Sc, where the characteristics can be traced through the entire sequence. Such groupings are discussed in the Hubble Atlas (cf. the section on Sc galaxies where families with similarly shaped arms are isolated).

De Vaucouleurs has proposed to recognize these characteristics explicitly in the notation by adding symbols in (for massive) and f (for filamentary) after the type designation, with an additional symbol to indicate how many arms are present. For example, NGC 1232 (Atlas, p. 32), which has filamentary and highly branched segmented structures starting from two main arms of the rs variety, would have a complete notation SAB(rs)2+Cf , where c denotes the arm character, and 2+ denotes branching from two main arms.

By now the notation has become quite detailed, and is as complete as is likely to be useful. No classification system can describe the infinite variations among galaxies: this was Reynolds's objection. Most classifiers would agree with part of Baade's statement (1963, p. 19) that beyond this stage ``if you want to study the variations on the theme Sc [for example], you simply have to take plates and examine them - only then do you get the full story. No code system can replace this. The code finally becomes so complicated that only direct inspection of plates helps.''

But, as de Vaucouleurs points out, the virtue of the extended notation is that symbols can be progressively dropped until the notation becomes as simple as Hubble's original system. The classifier can exclude as much detail as he wishes and still remain within the standard classification at any desired level of complexity.

Next Contents Previous