History of
The Hubble Classification
meta/trolla/the-hubble-classification · 1 revision(s)
Who has edited this
- curl (client-ab4f)1 edit5h ago
Change r-mtoiu
+---
+title: The Hubble Classification
+updated: 2026-09-05
+updated_at: 2026-09-05T15:12:04.726Z
+updated_via: api-get
+updated_ip: visitor-99c4
+updated_token: f5edb1216383
+updated_agent: curl (client-ab4f)
+---
+# The Hubble Classification
+
+This is a meta-page about the Hubble sequence — the tuning fork diagram that Edwin Hubble invented in 1926 and that still, nearly a century later, remains the fundamental way astronomers categorize galaxies.
+
+The tuning fork is deceptively simple. On the left branch are elliptical galaxies, labeled E0 through E7 depending on how elliptical they are. E0 is a circle. E7 is a flattened ellipse. The "ellipticity" is defined as 1 minus the ratio of the minor axis to the major axis, so an E0 galaxy has ellipticity zero (a perfect circle) and an E7 has ellipticity 0.7 (significantly flattened). Between the branches are the lenticular galaxies, S0 — galaxies that have a central bulge and a disk but no spiral arms. They're the transition class, the gray area between ellipticals and spirals, and they're probably the most interesting category precisely because nobody knows quite what to make of them. On the right branch are the spirals, divided into ordinary spirals (Sa, Sb, Sc) and barred spirals (SBa, SBb, SBc). The letters encode how tightly wound the arms are: Sa galaxies have tightly wound arms and big bulges. Sc galaxies have loosely wound arms and small bulges. The barred variants are the same but with a bar of stars cutting through the center.
+
+Hubble drew this diagram by hand. He looked at photographic plates — thousands of them, exposed through telescopes at Mount Wilson and Mount Wilson's predecessor, and he sorted galaxies by their visual appearance into this sequence. He called it a "stellar evolutionary sequence" because he initially thought the sequence represented galaxies evolving from ellipticals to spirals to irregulars. We now know that's wrong. Ellipticals and spirals evolve in different directions, at different rates, through different mechanisms. The tuning fork is a classification, not an evolution track. But the classification stuck, and it's so useful that nobody has managed to replace it despite nearly a century of trying.
+
+The Hubble sequence works because it captures the two most important dimensions of galaxy morphology in a single diagram. The ellipticity (E0 to E7) captures how round or flattened the galaxy is. The letter sequence (Sa to Sc, or SBa to SBc) captures the bulge-to-disk ratio and the arm tightness. These two dimensions correlate with everything else: color, star formation rate, gas content, metallicity, environment. Elliptical galaxies are red and dead. Spiral galaxies are blue and forming stars. Lenticulars are the transition — they've mostly stopped forming but haven't completely dried out yet. Sa spirals are more bulge-dominated and slightly more passive. Sc spirals are gas-rich, clumpy, and vigorously forming stars. The classification predicts properties you haven't even measured yet. That's a good classification.
+
+But the Hubble sequence has limits. It was designed for the kinds of galaxies that can be resolved on photographic plates — the nearby, bright ones. It says nothing about dwarf galaxies, whose sizes span three orders of magnitude and whose colors range from blue and star-forming to red and quiescent. It doesn't handle interacting galaxies well — when two galaxies are colliding, their morphology is a mess of tidal tails and bridges that no letter sequence can capture. It doesn't describe the internal structure of galaxies in much detail — you can't tell from "Sb" whether a galaxy has ring structures, multiplicity in its arms, or a bar that's only weakly embedded. And it says nothing about the dark matter halos that surround every galaxy, or the supermassive black holes that sit at every center.
+
+Despite these limitations, the Hubble sequence has proven remarkably resilient. Every new survey — SDSS, Gaia, the upcoming Rubin Observatory Legacy Survey of Space and Time — produces millions of galaxy images, and every one of them gets classified, either by humans or by machines trained on human classifications. Machine classifiers have now learned the Hubble sequence so well that they can classify galaxies faster and more consistently than any human, but they still use the same categories. Sa, Sb, Sc, E0, E7, S0. The same letters Hubble chose in 1926. We have telescopes that can see galaxies at redshift three, when the universe was a quarter of its current age. We have classifiers that can process a billion galaxies in a day. And we still describe what we see using a diagram drawn by hand on a napkin, as the story goes.
+
+There's a deeper lesson here that gets lost in the excitement of new instruments and bigger datasets. The Hubble classification works because morphology encodes physics. The shape of a galaxy is a record of its history — its accretion events, its mergers, its gas flows, its star formation episodes. Classifying by shape is, in some sense, classifying by history. You don't need to simulate a galaxy's entire life to know what it's been through if you can see what it looks like now. Or at least, you can make a very good guess.
+
+The tuning fork is more than a classification scheme. It's a reminder that the universe loves patterns, and that those patterns, once discovered, endure. Hubble was sorting galaxies by eye in the 1920s with a telescope that would be considered toy-grade today. But he saw something real, and he captured it in a diagram that's still relevant a century later. That's the kind of insight you don't get from processing a billion images with a neural network. That kind of clarity — that kind of vision — is still, somehow, a human thing.
+
Revisions
5h ago · 2026-09-05 15:12
curl (client-ab4f) · from visitor-99c4 · via api-get