Hi David,
thanks for your accurate comment about the rough and ready data that Hubble used to make his graph of Velocity vs Distance for galaxies. I reproduce the graph from his 1929 paper, below.
I do not know if this was his first published velocity-distance graph.
People can obviously disagree about whether or not some scattered points on a graph constitute a specific trend......
Given that today's
best available distance estimates for various individual galaxies have a nominal accuracy of about 10 percent (or possibly somewhat worse, if we include baseline errors), the errors on the distance axis of Hubble's graph must have been very very very large!!
Edwin Hubble's result that Galaxy Distance is proportional to Galaxy Recession Velocity (after each galaxy velocity has been corrected for the motion of our Sun) was published in (1929), Proc. Natl. Acad. Sci.,
15, 168
In other words, he found that Galaxy Velocity =
A Constant multiplied by the Galaxy Distance
(which is usually called
Hubble's Law)
(we now call the constant in this equation the
Hubble Constant)
Here is the velocity-distance graph for the "extra-galactic nebulae"(note the interesting terminology!) from E.P. Hubble (1929) : : :
Rough as guts!! Indeed, it is quite possible to fit these data points with something other than a straight-line trend.
But Hubble was right, as we now know;
for instance, the points on the graph of distance vs velocity for Type 1a Supernovae out to a redshift of z = 1 , do closely follow a straight line......
(
There were people who strongly suspected that this linear relationship existed, well before Hubble published...... indeed these people already talked about what we now call the value of the Hubble Constant.
Hubble, whether fairly or unfairly lauded for this, clinched the case in the eyes of his peers.....his attitude to his competitors and his hogging of the credit, is a matter for the historians.)