Cold Dark Matter (1984): How Galaxies Grew From the Bottom Up

The cold dark matter model was set out in a 1984 Nature paper by George Blumenthal, S. M. Faber, Joel Primack and Martin Rees. They proposed that dark matter consists of slow-moving “cold” particles, so that structure in the universe formed from the bottom up: small clumps first, merging over time into galaxies, groups and clusters.

The question before the paper

By the early 1980s, flat rotation curves had convinced most astronomers that galaxies sit inside massive halos of dark matter. But what kind of matter? One popular idea was “hot” dark matter, such as light neutrinos, which would have zipped around at nearly the speed of light in the early universe and smoothed out small clumps. The question was which kind of dark matter could build the galaxies and clusters we actually see.

What the paper found

• Cold beats hot. The authors argued that dark matter is made of particles that moved slowly in the early universe, as opposed to fast “hot” particles like light neutrinos.
• Bottom-up growth. With cold dark matter, small halos form first and merge hierarchically into larger and larger ones.
• Natural galaxy masses. This picture explains the characteristic masses of galaxies and the existence of galaxy groups and clusters without special tuning.

How they did it

The paper is a theoretical synthesis. The authors followed how small density ripples in the early universe would grow under gravity if the dark matter were cold and slow, rather than hot and fast, and compared the resulting picture with the galaxies, groups and clusters astronomers observe.

Why it mattered

Cold dark matter, combined with a cosmological constant after the 1998 discovery of cosmic acceleration, became the standard Lambda-CDM model of cosmology. Almost every cosmological simulation still uses the framework this paper set out. Those simulations produced the universal NFW density profile of dark matter halos, and the Bullet Cluster gave a direct picture of dark matter behaving as a collisionless substance, as the model assumes.

What we know now

The model’s large-scale predictions have held up well, from the cosmic microwave background measured by Planck to galaxy surveys. Its small-scale predictions, for dwarf galaxies and the cores of halos, are where cold dark matter is being tested most intensely today. ESA’s Euclid mission, surveying up to two billion galaxies, is one of the new tests. The particle itself has not been found: direct searches such as LZ keep ruling out popular candidates.

Who wrote it

George Blumenthal is first author, with S. M. Faber, Joel Primack and Martin Rees. The paper appeared in Nature in October 1984 and has about 1,100 citations in Semantic Scholar as of October 2026. An earlier strand of the same idea, galaxies forming inside dark halos, goes back to White and Rees in 1978.

Read the original

Blumenthal, G. R., Faber, S. M., Primack, J. R., and Rees, M. J. (1984). “Formation of galaxies and large-scale structure with cold dark matter.” Nature 311, 517. doi:10.1038/311517a0. The journal version requires a subscription.