People seem to have a built in need to sort things into bins or categories. In science, this sorting step is often a first step in looking for the underlying physics that is causing morphological differences between systems. After he established the extragalactic nature of some types of the nebulae, Edwin Hubble tried to define ways to classify these new galaxies according to their appearance, or galactic morphology. Using a large set of images and the standard scientific approach in a new area of research, he tried to classify different objects by type and then to arrange them in a system that showed smooth transitions from one type to another. The act of classification does not lead directly to a physical theory, but the hope was that the relationship between galaxy types would have a physical underpinning. For example, the galaxy types might be related by age, by gas content, by stellar populations, by mass differences, or by rotational differences. Hubble found three major types of galaxies in his study of the regions of space beyond the Milky Way: spiral, elliptical, and irregular galaxies. He also noted that many spiral galaxies also possess a central bar.
Our Milky Way is a barred-spiral galaxy with a disk of arms that connect to a bar, a bulge, and a halo. Like other similar systems, the disk has spiral arms with bright emission nebulae, and a rich interstellar medium of gust and gas. The dust forms an obscuring band when the disk is viewed edge-on. If we could view the disk face-on from a great height, the spiral arms would appear clearly outlined by luminous, young, blue stars. Roughly one-third of spiral galaxies have bright bar-like features in their central regions like our Milky Way's. In such cases, the spiral arms originate at the ends of the bar, rather than originating in the nucleus itself.
In both normal and barred spirals, Hubble noticed a gradual transition of morphological types. Among normal spirals, the sequence from Sa to Sb to Sc goes toward less tightly wound spiral arms and less prominent central bulges. The corresponding barred spirals are classified from SBa to SBb to SBc. Galaxies are flung at random orientations in space, so some disks are seen face-on and some edge-on. The size of the central bulge and the degree of winding of the spiral arms go hand in hand, so we can classify spirals according to Hubble's scheme even if they are viewed edge-on. It is also observed that spirals with more loosely wound arms have much more prominent luminous stars, star clusters, and emission nebulae that outline the arms. All spirals rotate in the sense that the arms trail, as observed for the Milky Way. The rotation of spirals can be mapped by Doppler shifts measured three different ways: using stellar absorption lines, using emission lines from HII regions, or using the 21-cm line of neutral hydrogen. This rotation cannot be observed when a disk is face-on, because the motion is transverse to the line of sight and there is no Doppler shift, but even a slight inclination make determination of rotation direction possible.
Hubble also observed galaxies with prominent bulges and disks that lack spiral arms. These are classified as S0 galaxies and are sometimes called lenticular galaxies due to their lens-like shape. Lenticular galaxies have lost most of their interstellar matter so have little ongoing star formation.
Thus far our discussion has ignored the third major galactic component: the halo. It turns out that galactic halos are too diffuse to be visible, even on deep images of galaxies. The difficulty in seeing halos is important since the halo contains most of the mass of any spiral galaxy! Hubble did not recognize S0 galaxies as a distinct class; modern astronomers consider them intermediate between spirals and elliptical. One way to think of an elliptical galaxy is that it's galaxy with a pure halo stellar population.
Spiral galaxies can vary in size by as much as a factor of ten. By contrast, however, elliptical galaxies have an enormous range in size. Elliptical galaxies range from tiny dwarfs not much larger than a globular cluster to giants that are three to four times larger than the Milky Way. In general, elliptical galaxies have smooth shapes like spheres or squashed spheres and no spiral arms. They are classified according to how round or flat they look and according to their size. The numerical scheme from E0 to E7 runs from circular to highly elongated galaxies. Since galaxies are three-dimensional objects distributed in space, our perspective from the Earth may not give us a true indication of the shape. An E7 elliptical is highly elongated, and so it must be a relatively flat galaxy seen edge-on. However, an E0 elliptical appears round but it need not be a spherical galaxy. Both a flattened distribution of stars viewed face-on and a spherical collection of stars would be classified as an E0. The vast majority of ellipticals show and contain only reddish stars, with little indication of gas, dust, or young, luminous stars, although a very small number are blue and still possess star formation. It was once thought that elliptical galaxies are flattened because they rotate, but ellipticals rotate too slowly to cause the observed flattening.
The third category of galaxies is the irregular galaxy. Irregular galaxies come in many shapes, and they are usually small. Some irregulars show a degree of spiral structure but without the high symmetry of true spirals. Others have chaotic morphologies without any obvious symmetry, perhaps resembling the splatter of a dead bug on a windshield. Many of these chaotic irregulars appear to have undergone collisions or to be in the process of merging with other systems. Irregulars have regions of intense star formation with conspicuous young stars. The Magellanic clouds are the only two examples visible to the naked eye (if you happen to live in the Southern Hemisphere). Finally, a number of galaxies have unusual morphologies and resist being shoehorned into Hubble's classification scheme. Peculiar galaxies may have loops or tails or other extended structures not seen in irregulars that arise from interactions with other galaxies or intergalactic gas.
Classification is an important first step in organizing the richness of the extra galactic universe, but it does not automatically lead to physical understanding. Hubble organized galaxies into a "tuning fork" diagram, in which the normal and barred spirals form parallel sequences and the S0 galaxies are a transition type between the spirals and elliptical. This system correlated well with certain galaxy properties. For example, compared to spirals, elliptical galaxies have older stars and smaller amounts of gas and dust. For some time it was believed that the Hubble classification implied an evolutionary sequence, in which spirals gradually used up their gas, the stars aged and faded, and the final result was an elliptical. This cannot be true, since spirals contain old halo populations, and many must be as old as elliptical. Morphology alone does not explain all the differences in galaxy type.