From: STEVE HARDING <Steve.Harding@nottingham.ac.uk>
  To  : rasmb@bbri.harvard.edu
  Date: Tue, 4 May 1999 08:48:43 GMT0BST

Re: On the determination of D

Dear RASMBers
I strongly agree with Hiroshi Fujita's comments concerning the 
preference of DLS over the UC for diffusion coefficient measurements.
For the polydisperse sort of biological systems we have been working 
mainly with in the past - polysaccharides and mucin glycoproteins, UC 
methods have not been applicable: more recently even with the modern 
software and moderately polydisperse protein mixtures we have 
been looking at results have been unreliable. 

With DLS you still have to be very careful though - 
sample clarity is the main problem.  The single fixed angle 
instruments are also only valid for near-spherical systems.  Any 
departure from sphericity you need multi-angle measurements: in this 
regard the "Dynamic Zimm plot" procedure developed by W. Burchard and 
coworkers at Freiburg (which extrapolates out complications from 
rotational diffusion and concentration dependence) is highly useful.

Steve Harding.