2. The notion of dark matter

The notion of "dark matter" emerged from observations of large astronomical objects such as galaxies and clusters of galaxies, which displayed gravitational effects that could not be accounted for by the visible matter: stars, gas, and so on, assuming the validity of Newton's universal law of gravitation.

It was concluded that such observations could only be described satisfactorily if there existed stronger gravitational fields than those provided by the visible matter and Newtonian gravity. Such gravitational fields required either more mass or an appropriate modification of Newton's universal law of gravitation.

Early preliminary evidence for such a "mass discrepancy" was observed in 1933 by Zwicky [1] for the Coma cluster of galaxies. He estimated that the cluster contained considerably more "dark matter" than the visible galactic matter in order to account for the fast motions of the galaxies within the cluster and also to hold the cluster together.

Additional preliminary evidence for the mass discrepancy was found by Babcock [2] in 1939 and Rubin and Ford [3] in 1970 by measuring the rotation curve of the Andromeda galaxy, the nearest spiral galaxy to the Milky Way. The rotation curve of a galaxy is the dependence of the orbital velocity of the visible matter in the galaxy on its radial distance from the center of the galaxy. However, neither Babcock nor Rubin and Ford attributed their observations of an increase in mass toward the edge of the galaxy to any missing mass.

In 1970, Freeman [4] found rotation curves for several galaxies that disagreed with expectation based upon the assumption that the galaxies consisted of stars, gas, and nothing else. Freeman suggested that these galaxies, like the Coma cluster observed much earlier by Zwicky, contained considerably more invisible "dark matter" than the luminous matter. In 1973, Roberts and Rots [5], using 21-cm line data, obtained neutral hydrogen rotation curves of three nearby spiral galaxies. These rotation curves extended to considerably larger distances from the centers of the galaxies than the corresponding rotation curves for the stars. In each case, the complete rotation curve was essentially "flat" out to the edge of the 21-cm data.

In 1974, Ostriker et al. [6] stated that the current observed rotation curves strongly indicated that the mass of a spiral galaxy increases approximately linearly with radius to about 1 Mpc so that the ratio of the total mass to the observed visible mass was large. They concluded that the rotation curves could most plausibly be understood if the spiral galaxy was embedded in a giant spherical halo of invisible "dark matter."

Further evidence for the dark matter hypothesis in many spiral galaxies was obtained in the 1970s by Rubin et al. [7], who measured high-quality optical rotation curves for the luminous matter and Bosma [8], who compiled 21-cm rotation curves for the neutral hydrogen gas that extended far beyond the luminous matter of each galaxy. In all these cases, the complete rotation curve was essentially "flat" out to the edge of the 21-cm data.

By 1980, the conclusive observation from the rotation curves of spiral galaxies was that there existed a major "mass discrepancy" that was greater if larger distance scales were involved. This implied that if Newton's universal law of gravitation was approximately valid, as in the Solar System, considerably more mass was required to be present in each galaxy. This invisible matter was termed "dark matter" with the introduction of the dark matter hypothesis: each spiral galaxy was embedded in a huge spherical halo of dark matter.

Thus, the notion of "dark matter" essentially emerged from the observed rotation curves of spiral galaxies that provided convincing evidence for a mass discrepancy within the galaxy. The only alternative to "dark matter" seemed to be a significant modification of Newton's universal law of gravitation to provide the required stronger gravitational field at larger distance scales. However, at that time, such a modification of Newtonian gravity was not considered a viable alternative.
