Recently, I published a catalog of 2,474 central stellar velocity dispersions for 1,563 galaxies. The reference is McElroy, 1995, ApJS, 100, 105. Offprints are available. And if you send me email ( email@example.com ), I will send you the neat electronic version, which is much easier to sort and search than the paper one.
Of what use are these measurements, I hear you mentally asking. For one thing, it is thought that the absolute luminosity of a galaxy is proportional to the fourth power of the velocity dispersion (the "so-called" Faber-Jackson relation). Ok, I have the dispersions, let's get the luminosities (from de Vaucouleurs Third Reference Catalog), plot them up and see what we get.
If we just do a straight fit to this, by golly we get an exponent of about 3.9 +- 0.1. Close enough to four. There is a lot of scatter. This indicates that a third parameter is probably needed.
Another interesting thing to do is look at velocity dispersion as a function of some other galaxy parameter. How about morphological type? Let's use "T". This is a parameterization I think by de Vaucouleurs (at least, it's in his Third Reference Catalog). T=-5 corresponds to E7-E5, T=0 is E0/S0, T=5 is Sc, and interpolate from there (actually, see de Vaucouleurs Third Reference Catalog for a coherent explanation). So when you plot up dispersion versus T you get this:
Note that sigma declines with T. Not a big surprise. There is, however, an unexpected "hump" at about T=1, around Sa galaxies. I think it indicates an anomolous lack of low dispersion early-type spiral galaxies (which you don't see at T=0 because you are mixing in E0's there).
There's no place like home page.