On this page
Edwin Hubble is the name most students link with the expanding universe. But the speeds in Hubble's famous 1929 graph were not his own measurements. They came from spectra taken by Milton Humason and by Vesto Slipher, working at an observatory in Arizona. This post goes through Slipher's own reports, the Doppler physics he used, and a simple calculation that explains how he got usable spectra from such faint objects.
Who was Vesto Slipher?
The American National Biography gives his life as 1875 to 1969. He worked at Lowell Observatory in Flagstaff, Arizona, where his reports were signed, and later directed it, as his Physics Today obituary notes in its title.
What did he measure in 1912?
His report is Lowell Observatory Bulletin No. 58, The Radial Velocity of the Andromeda Nebula. The problem was faintness. The light of a spiral nebula, spread into a spectrum, is extremely weak. His first plate, on 17 September 1912, needed an exposure of 6 hours and 50 minutes.
Four plates, taken between September and December 1912, gave velocities of −284, −296, −308 and −301 km/s. He rounded the mean to 300 km/s. The negative sign means Andromeda is coming toward us. He called it "the greatest hitherto observed" and concluded that "the Andromeda Nebula is approaching the solar system with a velocity of about 300 kilometers per second."
How does a spectrum give a speed?
Light from a source moving toward you is squeezed to shorter wavelengths. Moving away, it is stretched to longer ones. For speeds much less than light,
Δλ / λ = v / c
Worked example for Andromeda:
Pick a line. Hydrogen's Hγ line is at about 434.0 nm.
Speed. v = −300 km/s, c = 299,792 km/s.
Shift. Δλ = 434.0 × (−300 / 299,792) ≈ −0.43 nm.
Meaning. The line appears about 0.43 nm toward the blue. That is a tiny shift. Slipher measured his plates with a Hartmann spectrocomparator magnifying fifteen times.
Common mistakes: getting the sign the wrong way round (negative velocity means approaching, a blueshift), and using Δλ/λ = v/c for speeds close to light, where it breaks down. Our post on Christian Doppler covers the effect itself, and our post on the Doppler effect in sound has exam-style practice.
How did he make a faint spectrum usable?
In a report to the American Astronomical Society, Spectrographic Observations of Nebulae, published in 1915, Slipher explains a key change. For an extended object like a nebula, no choice of telescope makes the spectrum brighter, but the spectrograph's camera does. He replaced an 18.5-inch camera lens with a 3¼-inch one, which made the instrument "fully 30-fold" faster.
You can check his figure. If the camera's aperture stays about the same, the brightness of the image of an extended source goes as 1/f², where f is the focal length:
(18.5 / 3.25)² = (5.69)² ≈ 32
That is "fully 30-fold".
What did the 15 spirals show?
The same report lists 15 spiral nebulae. Two, including Andromeda, were approaching at about 300 km/s. Most of the others were moving away: NGC 1068 and NGC 4594 at +1,100 km/s and NGC 4565 at +1,000 km/s. The average speed was about 400 km/s, and the average was positive by about 325 km/s. He noted the spirals moved about 25 times faster than the average star.
He also saw that the lines in NGC 4594 were tilted, which meant one side was approaching and the other receding: the nebula was rotating.
How did this lead to the expanding universe?
In 1914 it was not yet settled whether the spirals were separate galaxies far outside our own. Once distances were measured, the speeds could be compared with them. As our post on Edwin Hubble explains, Hubble's 1929 relation combined his own distances with velocities measured by Slipher and by Milton Humason.
What can a student take from how Slipher worked?
Fix the real bottleneck. The limit was the camera, not the telescope.
Repeat the measurement. Four plates gave a mean he could trust.
Report what you see, even if it is odd. Speeds far above any measured before went straight into print.
If Doppler shift sums still feel unsure, Fermi's AI tutor can work out what you already know and build a plan for the session.
Next step: find the shift of the 656.3 nm Hα line for NGC 4594 moving away at 1,100 km/s, and say whether it moves to the red or the blue.
