As extended objects rather than point sources, galaxies show a wide variety of forms, some due to intrinsic structures, others due to the way the galaxy is oriented to the line of sight. The random orientations, and the wide spread of distances, are the principal factors that can complicate interpretations of galaxy morphology. If we could view every galaxy along its principal axis of rotation, and from the same distance, then fairer comparisons would be possible. Nevertheless, morphologies seen in face-on galaxies can also often be recognized in more inclined galaxies (Figure 1). It is only for the highest inclinations that morphology switches from face-on radial structure to vertical structure. In general we either know the planar structure in a galaxy, or we know its vertical structure, but we usually cannot know both well from analysis of images alone.
Figure 1. Four galaxies of likely similar face-on morphology viewed at different inclinations (number below each image). The galaxies are (left to right): NGC 1433, NGC 3351, NGC 4274, and NGC 5792. Images are from the dVA (filters B and g).
Galaxy morphology began to get interesting when the "Leviathan of Parsonstown", the 72-inch meridian-based telescope built in the 1840s by William Parsons, Third Earl of Rosse, on the grounds of Birr Castle in Ireland, revealed spiral patterns in many of the brighter Herschel and Messier "nebulae." The nature of these nebulae as galaxies wasn't fully known at the time, but the general suspicion was that they were star systems ("island universes") like the Milky Way, only too distant to be easily resolved into their individual stars. In fact, one of Parsons' motivations for building the "Leviathan" was to try and resolve the nebulae to prove this idea. The telescope did not convincingly do this, but the discovery of spiral structure itself was very important because such structure added to the mystique of the nebulae. The spiral form was not a random pattern and had to be significant in what it meant. The telescope was not capable of photography, and observers were only able to render what they saw with it in the form of sketches. The most famous sketch, that of M51 and its companion NGC 5195, has been widely reproduced in introductory astronomy textbooks.
While visual observations could reveal some important aspects of galaxy morphology, early galaxy classification was based on photographic plates taken in the blue region of the spectrum. Silver bromide dry emulsion plates were the staple of astronomy beginning in the 1870s and were relatively more sensitive to blue light than to red light. Later, photographs taken with Kodak 103a-O and IIa-O plates became the standard for galaxy classification. In this part of the spectrum, massive star clusters, dominated by spectral class O and B stars, are prominent and often seen to line the spiral arms of galaxies. These clusters, together with extinction due to interstellar dust, can give blue light images a great deal of detailed structure for classification. It is these types of photographs which led to the galaxy classification systems in use today.
In such photographs, we see many galaxies as a mix of structures. Inclined galaxies reveal the ubiquitous disk shape, the most highly flattened subcomponent of any galaxy. Studies of Doppler wavelength shifts in the spectra of disk objects (like HII regions and integrated star light) reveal that disks rotate differentially. If a galaxy is spiral, the disk is usually where the arms are found, and also where the bulk of interstellar matter is found. The radial luminosity profile of a disk is usually exponential, with departures from an exponential being due to the presence of other structures.
In the central area of a disk-shaped galaxy, there is also often a bright and sometimes less flattened mass concentration in the form of a bulge. The nature of bulges and how they form has been a topic of much recent research, and is discussed further in section 9. Disk galaxies range from virtually bulge-less to bulge-dominated. In the center there may also be a conspicuous nucleus, a bright central concentration that was usually lost to overexposure in photographs. Nuclei may be dominated by ordinary star light, or may be active, meaning their spectra show evidence of violent gas motions.
Bars are the most important internal perturbations seen in disk-shaped galaxies. A bar is an elongated mass often made of old stars crossing the center. If spiral structure is present, the arms usually begin near the ends of the bar. Although most easily recognized in the face-on view, bars have generated great interest recently in the unique ways they can also be detected in the edge-on view. Not all bars are made exclusively of old stars. In some bulge-less galaxies, the bar has considerable gas and recent star formation.
Related to bars are elongated disk features known as ovals. Ovals usually differ from bars in lacking higher order Fourier components (i.e., have azimuthal intensity distributions that vary mainly as 2), but nevertheless can be major perturbations in a galactic disk. The entire disk of a galaxy may be oval, or a part of it may be oval. Oval disks are most easily detected if there is considerable light or structure at larger radii.
Rings are prominent features in some galaxies. Often defined by recent star formation, rings may be completely closed features or may be partial or open, the latter called "pseudorings." Rings can be narrow and sharp or broad and diffuse. It is particularly interesting that several kinds of rings are seen, and that some galaxies can have as many as four recognizeable ring features. Nuclear rings are the smallest rings and are typically seen in the centers of barred galaxies. Inner rings are intermediate-scale features that often envelop the bar in a barred galaxy. Outer rings are large, low surface brightness features that typically lie at about twice the radius of a bar. Other kinds of rings, called accretion rings, polar rings, and collisional rings, are also known but are much rarer than the inner, outer, and nuclear rings of barred galaxies. The latter kinds of rings are also not exclusive to barred galaxies, but may be found also in nonbarred galaxies.
Lenses are features, made usually of old stars, that have a shallow brightness gradient interior to a sharp edge. They are commonly seen in Hubble's disk-shaped S0 class (section 5.2). If a bar is present, the bar may fill a lens in one dimension. Lenses may be round or slightly elliptical in shape. If elliptical in shape they would also be considered ovals.
Nuclear bars are the small bars occasionally seen in the centers of barred galaxies, often lying within a nuclear ring. When present in a barred galaxy, the main bar is called the "primary bar" and the nuclear bar is called the "secondary bar." It is possible for a nuclear bar to exist in the absence of a primary bar.
Dust lanes are often seen in optical images of spiral galaxies, and may appear extremely regular and organized. They are most readily evident in edge-on or highly inclined disk galaxies, but are still detectable in the face-on view, often on the leading edges of bars or the concave sides of strong inner spiral arms.
Spiral arms may also show considerable morphological variation. Spirals may be regular 1, 2, 3, or 4-armed patterns, and may also be higher order multi-armed patterns. Spirals may be tightly wrapped (low pitch angle) or very open (high pitch angle.) A grand-design spiral is a well-defined global pattern, often detectable as smooth variations in the stellar density of old disk stars. A flocculent spiral is made of small pieces of spiral structure that appear sheared by differential rotation. Their appearance can be strongly affected by dust, such that at longer wavelengths a flocculent spiral may appear more grand-design. Pseudorings can be thought of as variable pitch angle spirals which close on themselves, as opposed to continuously opening, constant pitch angle, logarithmic spirals.
There are also numerous structures outside the scope of traditional galaxy classification, often connected with strong interactions between galaxies. Plus, the above described features are not necessarily applicable or relevant to what we see in very distant galaxies. Accounting for all of the observed features of nearby galaxies, and attempting to connect what we see nearby to what is seen at high redshift, is a major goal of morphological studies.