why galaxies spin wrong

The Mechanism

Between 1968 and 1970, working at the Department of Terrestrial Magnetism of the *Carnegie Institution of Washington* in northwest Washington, D.C., the American astronomer *Vera Cooper Rubin* (born Philadelphia, *23 July 1928*; died Princeton, *25 December 2016*, aged 88) and the instrumentalist *W. Kent Ford Jr.* (born Indiana, *14 February 1931*) carried out a careful spectroscopic measurement of the *rotation curve* of the Andromeda Galaxy (*Messier 31* or *M31*) — the plot of how the orbital velocity of stars and gas around the centre of the galaxy varies with distance from that centre — and obtained a result that did not match any contemporary model of galaxy mass distribution. They published it as *"Rotation of the Andromeda Nebula from a Spectroscopic Survey of Emission Regions,"* *Astrophysical Journal* vol. 159, p. 379 (February 1970). The instrument that made the measurement possible was Ford's *image-tube spectrograph*, a device he had built at the DTM in the early 1960s and refined over the decade: an electron-multiplying tube that amplified the faint optical light from individual *H II regions* — clouds of ionised hydrogen gas in the spiral arms — by a factor of about *thirty*, allowing each cloud to yield a usable optical spectrum in a single exposure of *fifteen to forty-five minutes* on a *2.1-metre or smaller* telescope. The instrument was mounted on the *84-inch (2.1-metre) telescope* at *Kitt Peak National Observatory* in Arizona and on the *60-inch telescope* at *Mount Wilson Observatory* in California. Rubin and Ford had been collaborating on galaxy-rotation measurements since 1965, when Ford joined the DTM. M31 was chosen as the first major target: it is the nearest large spiral galaxy to the Milky Way (about 2.5 million light-years away), large enough on the sky that its spiral arms can be resolved and individual H II regions measured one at a time, and inclined at about 77° to the line of sight so that line-of-sight Doppler shifts give a clean readout of orbital velocity. They identified *67 H II regions* across the disk of M31 — running from a radius of *3 kiloparsecs* (close to the galactic centre) out to *24 kiloparsecs* (the visible edge of the galaxy) — and measured the radial velocity of each one from the redshift or blueshift of its emission lines. The expected result, from the standard pre-1970 understanding of galaxy mass distribution, was a *Keplerian falloff*: at small radii, the orbital velocity should rise approximately linearly with radius (a *rigid-body* signature, characteristic of a region inside the bulk of the visible mass); at large radii, where almost all of the visible mass is interior to the orbit, the orbital velocity should fall off as the square root of the inverse radius — exactly the same falloff that Neptune obeys relative to Mercury in the solar system. The observed result was different. From about 3 to 7 kiloparsecs the rotation curve rose, as expected; from 7 to 24 kiloparsecs it *stayed flat*, with the orbital velocity holding nearly constant at approximately *225-275 km/s* all the way to the visible edge of the disk. There was no falloff. The curve simply did not bend down. The implication was forced: at every radius from 7 kiloparsecs outward, the enclosed mass had to be growing in *direct proportion to the radius*. That is the signature of an *extended distribution of mass* that continues outward beyond the visible stars — much further out than the visible disk, in a roughly *isothermal halo* surrounding the galaxy. The mass interior to the outermost measured radius was *several times* the total mass of the visible stars and gas. Rubin and Ford were not the first to suggest that mass might be hiding in or around galaxies. The Swiss-American astronomer *Fritz Zwicky*, at Caltech, had in *1933* studied the *Coma cluster of galaxies* and concluded from the relative velocities of the cluster's member galaxies that there must be a large *"dunkle Materie"* (dark matter) holding the cluster together gravitationally; *Horace Babcock*, working under Edwin Hubble at Mount Wilson in *1939*, had measured a rotation curve of M31 out to about 6 kpc that was hard to reconcile with the visible-mass distribution; the radio astronomers *Morton Roberts* and *Arnold Rots* had, in the mid-1960s, used the 21-cm hydrogen line at *Green Bank* and *Westerbork* to extend the M31 rotation curve out to large radii and also seen it stay flat. What Rubin and Ford's 1970 paper added was a *precise, calibrated, optical measurement* extending to the visible edge of the galaxy, made with a single instrument across a uniformly-treated sample of H II regions, of a flatness that could not be dismissed as a measurement artefact or a peculiarity of one wavelength regime. The 1970 paper was followed by an eleven-year programme by Rubin, Ford, and *Norbert Thonnard* (joined later by *David Burstein*) to measure the optical rotation curves of *60 spiral galaxies* of different luminosities and Hubble types. The 1980 summary paper — *"Rotational Properties of 21 Sc Galaxies with a Large Range of Luminosities and Radii,"* *Astrophysical Journal* 238: 471-487 — reported flat rotation curves out to the visible edge in *every galaxy measured*. The result was no longer interpretable as a peculiarity of M31: the *flat-rotation-curve phenomenon* was a *universal property of disk galaxies*. The implication is the modern *dark-matter halo paradigm*: every disk galaxy is embedded in a roughly spherical halo of non-luminous mass, several times more massive than the visible stars and gas, that extends well beyond the visible edge of the disk and dominates the gravitational dynamics in the galaxy's outer regions. The composition of this dark mass has been the central open question of physical cosmology for fifty years and is the principal target of the current generation of underground direct-detection experiments (*XENONnT*, *LUX-ZEPLIN*, *PandaX-4T*) and the upcoming *Vera C. Rubin Observatory* — a flagship 8.4-metre survey telescope under construction in Cerro Pachón, Chile, named for Rubin in 2019 and entering science operations in 2026, that will measure the rotation curves and weak-lensing signatures of an estimated *17 billion* galaxies over its ten-year survey. Vera Rubin received the *National Medal of Science* in 1993 and the *Royal Astronomical Society Gold Medal* in 1996 (the first woman so honoured since *Caroline Herschel* in 1828). She was widely expected by her colleagues to receive the Nobel Prize in Physics for the rotation-curve work; the Nobel committee never awarded one. She died in 2016 at the age of 88. The 2025 U.S. quarter-dollar coin in the American Women Quarters series carries her portrait.