From: Jeffrey C. Hansen <hansen@bioc09.uthscsa.edu>
  To  : rasmb@bbri.org
  Date: Fri, 30 Apr 1999 08:19:08 -0500

Re: Molecular weight (Mw)

Hi Yarabe,

Walter's comments were on the mark and I will re-emphasize the caveat
regarding sample homogeneity.  To ensure that your sample truly is
homogeneous and ideal, you should analyze your velocity data by both the van
Holde/Weischet method and one of the Lamm equation-based direct boundary
fitting methods (using a single ideal component model.)  If homogeneous, the
latter method will directly and accurately yield both s and D, and therefore
M.  Since these methods unfortunately don't come with the Beckman software,
you can contact me or Bo Demeler if you need further assistance.

(I see that just as I was about to send this off Bo has responded similarly
but in more detail.  I'll send this anyway to emphasize the points). 

Good luck,

Jeff Hansen