In 1912 Vesto Slipher began taking spectra of nebulae. Apart from a few now known to be close to the Milky Way they showed redshifts indicating recession velocities up to 1800 km/s.

[Warning: Do not confuse redshift with reddening. Redshift changes the wavelengths of spectral lines and the peak of the energy distribution to longer wavelengths while reddening selectively absorbs blue light more strongly than red. Both mechanisms shift the peak of the observed energy distribution to the red but reddening does not affect the wavelengths of spectral lines.]

During the 1920's Edwin Hubble and Milton Humason (who drove a mule train up the trail from Pasadena to Mount Wilson then became night assistant and eventually became a full astronomer on the Mount Wilson Observatory staff) obtained spectra with the 100 inch telescope and estimated the distance as well (resolving in 1923 the problem of whether the nebulae were extragalactic). By 1927 Hubble and Humason found that the distance r was proportional to the recession velocity v or redshift Dl (v=cDl/l) and is now known as Hubble's law:

v=Hr

where H, often written Ho or H(now) since H=H(t) is a function of time, is the constant of proportionality known as the Hubble constant. Hubble's value was 550 km/s/Mpc, nearly an order of magnitude greater than today's 40 to 100 km/s/Mpc range.

The Hubble law is a result of an expanding universe and in fact led Einstein to remark that the "cosmological constant L" he introduced to keep his early models of the universe based on his new theory of relativity from expanding to be "my greatest mistake". I find a lot of people do not like the idea of an expanding universe but even from a Newtonian aspect it is the most logical world and an expanding universe leads straight to Hubble's law.

A typical model used to demonstrate the expansion is to stretch a rubber ruler (the 1 and the 9 separate eight times faster than the 1 and the 2… the separation speed is proportional to the separation. An expanding plum pudding is another example. I like an expanding balloon with coins glued on to represent clusters of galaxies.

The unit of the Hubble constant is reciprocal time. In fact, the reciprocal of Ho (assume the units are always km/s/Mpc) is just the maximum age of the universe (no deceleration),

To=978Gigayear/Ho

so if Ho=100 then To=9.78Gyr which is uncomfortably close to globular cluster ages. If the universe has "critical mass", just enough matter and energy to just halt the expansion after an infinite time, then Tc=2/3To which is even worse. If Ho=50 then Tc=13Gyr, a bit young but acceptable.

And don't get the idea that we can see the entire universe because v/c can never reach unity. It just means we cannot "see" or communicate with what is beyond infinite redshift. That is our horizon.

Now we need to look at models of the universe, cosmological models.