Select Historical Papers on Cosmology

These papers are relevant to the observational discovery of the expansion law, homogeneity, and isotropy of the universe.

Early Review Papers

Comments on expansion law

Everyone relied on the same set of velocities, a complete set of which are listed in Stromberg (1925) (~46 in total).

Shapley & Shapley (1919) noted a correlation in brightness and redshift: "The speed of spiral nebulae is dependant to some extent upon apparent brightness, indicating a relation of speed to distance or, possibly, to mass."

Wirtz (1924) assumed that galaxy diameters are "standard rulers" and found a loose correlation between velocity and diameter. He did not list his set of galaxies nor did he make any figures. He did not have any absolute distance calibration.

Stromberg (1925) assumed that galaxy absolute magnitudes are "standard candles". He did not find any significant correlation between velocity and apparent magnitude.

Lundmark (1924, 1925) combined both diameter and magnitude information to estimate galaxy distances, thus combining the "standard ruler" and "standard candle" assumptions. He used a variety of distance indicators (novae; Opik's method) to set the distance scale. He found a quadratic polynomial relation between distance and velocity with a large constant offset. (Beware: Lundmark's distances from apparent magnitudes are a complete mess.) He also tried a linear relationship but rejected it (line 20).

Lemaitre (1927) assumed that galaxy absolute magnitudes are "standard candles". He used Hubble's 1926 calibration but ignore Hubble's caveats about the standard candle assumption. He assumed a linear relationship between velocity and distance and made the first estimate of the slope (625 km/s/Mpc) (actually the second estimate, but, unlike Lundmark, he did not reject it.)

Robertson (1928) largely repeated Lemaitre's analysis and made the second estimate of the slope (463 km/s/Mpc).

Hubble (1929) assumed that the brightest stars in late-type spirals are "standard candles". (Additionally, for the Virgo cluster, he assumed that the cluster galaxy luminosity function matches the luminosity function of nearby field galaxies.) He used his distances to 7 nearby galaxies to calibrate the brightest star intrinsic magnitude. He derived a linear relation between velocity and distance and made the third estimate of the slope (500 km/s/Mpc).

Comments on Homogeneity

Hubble (1926) made the first test of homogeneity by counting galaxies as a function of apparent magnitude and comparing with a Euclidean model.

Comments on Isotropy

Fath (1914, AJ, 28, 75) made one of the first uniform photographic surveys over a substantial portion of the sky. Seares (1925, Ap. J., 62, 168) made a detailed analysis of Fath's counts, determining corrections for limiting mag., aberration, etc. The biggest confounding effect was dust extinction, but the effect of large-scale structure was also understood. Because both bright and faint galaxies (presumably at different distances) showed the same pattern of number count variations, the interpretation of it being due to dust came to be favored. Seares claimed to find that counts in the South are 3/4 of those in the North. Hubble (1926) claimed that this is due to the Virgo cluster in the North:

"When the influence of the cluster in Virgo is eliminated the density appears to be roughly uniform for all latitudes greater than about 25 degrees."