By the early 1900s, telescopes had revealed hundreds of faint, fuzzy "nebulae," many of them shaped like spirals. Were they small clouds inside the Milky Way — or entire galaxies far beyond it, "island universes" of their own? In 1920 the question got its own public showdown, the Great Debate, between Harlow Shapley (who argued the Milky Way was everything) and Heber Curtis (who argued for separate galaxies). Neither side could prove its case, because nobody could measure the distance to a spiral.
In 1923–24, Edwin Hubble, using the largest telescope in the world at Mount Wilson, found a Cepheid variable in the Andromeda spiral — a kind of pulsating star whose pulsing reveals its true brightness, and so its distance. Andromeda turned out to be far outside the Milky Way. Today we know it's about 2.5 million light-years away and even larger than our own galaxy.
Hubble sorted galaxies by what they look like. His scheme, drawn as a tuning fork, puts the smooth ellipticals on the handle, from round to flattened. At the joint are the lenticulars (S0), which have a disk but no spiral arms. The fork then splits into two prongs: ordinary spirals (S) and barred spirals (SB), whose arms start at the ends of a straight bar of stars through the center. Galaxies with no regular shape at all — irregulars — sit off to the side.
Click any galaxy on the fork — or the irregular off to the side — to see what defines its class.
Bars are common: roughly two-thirds of nearby spirals have one, including the Milky Way.
Each class has a number or letter that pins it down more precisely. For an elliptical, measure its long axis a and short axis b on the image and compute:
A perfectly round galaxy (b = a) is E0. One whose short axis is half its long axis is E5. The flattest ellipticals ever seen are about E7. For spirals, the letter after S or SB comes from two clues that usually go together: a means a big bulge and tightly wound arms, c means a small bulge and loose, open arms, and b is in between. Galaxies that fall between classes get two letters, like the Milky Way's SBbc.
Shape a galaxy and watch its class update. The real skill is running this in reverse: look at a galaxy and work out the class.
A galaxy's shape turns out to be a clue to its contents. Think back to 5.1.1: the Milky Way's disk is full of gas, dust, and young blue stars, while its halo holds only old stars. The galaxy types split the same way:
| Type | Stars | Gas & dust | New stars forming? | Size range |
|---|---|---|---|---|
| Elliptical | Old, reddish-yellow | Very little | Almost none | From tiny dwarfs to the largest galaxies known |
| Spiral / barred | Old in the bulge and halo; young and blue in the arms | Plenty, in the disk | Yes, in the arms | Mid-sized to large |
| Irregular | Many young, blue stars | Lots | Often very active | Usually small |
The reason is fuel. New stars form from cold gas. Ellipticals have used up or lost almost all of theirs, so all that's left are old stars quietly aging. Spirals and irregulars still have gas, so they keep making new ones. The biggest galaxies of all are giant ellipticals like M87, sitting at the centers of crowded clusters.
Galaxies don't travel along the tuning fork, but they can change type — usually by running into each other. Galaxies live in groups and clusters, and over billions of years they collide. Stars are so far apart that they almost never hit each other, but gravity scrambles their orbits and squeezes the gas into bursts of new stars. When two big spirals fully merge, their neat, flat disks are wrecked, and the result often settles into an elliptical. That's one reason ellipticals are most common in crowded galaxy clusters.
Our own neighborhood is the Local Group: three large spirals (the Milky Way, Andromeda, and Triangulum) plus dozens of small dwarf galaxies, including the irregular Large and Small Magellanic Clouds. Andromeda is heading toward us at about 110 km/s. In roughly 4 to 5 billion years it will pass close to the Milky Way, and the two will very likely merge into one large elliptical-like galaxy — though recent measurements suggest the merger isn't guaranteed.