>
>   I agree, but what it is happening there is not improving the
>   "geometry" of the system but simply ignoring the weaker fluorescence
>   at the top and at the botton that makes the bead appear longer.
>   I believe that if the x-y-z resolutions are matched, then the amount
>   of elongation indicates how "confocal" the system is.
>
>
 
I agree, so please allow me to make the original question of Shanti more clear.
When looking at an x-z section through a 10um flourescent bead, we see an
ellipse that measures 13um(z) and 10um(x) on its major and minor axes
respectively.  Doing a similar experiment in a straight reflectance
(non-fluorescent) mode, this elongation is not apparent.  Obviously, as
mentioned above, the saystem is less "confocal" in this fluorescent mode. This
is to be expected from theory (Several sources Wilson et al.) but not to
this degree, I don't think. Has anyone else noticed such a difference in fluor.
and non-fluor. confocal modes?  The 3um elongation is simply not caused by
the fact that the xy res. is 150nm and the z res. being 200nm. It looks more
like the z res. is on the order of 1um!!
 
  -Keith Bartels.