Resolution Limits?

[Charles Lesburg]

Hello,

I was wondering today about very high resolution crystal structures.
There are, in the PDB, many high resolution structures.  Of these
structures, there seem to be a disproportionate number solved at 1.54 A
resolution.  This value (due to its relatively high prevalence)
seems related to the wavelength of the radiation used to solve the
structure (Cu Ka=1.5418 A).  When the number of structures solved is
plotted against resolution, there seems to be a 'bell-curve' effect 
between 1.57 A and 1.51 A with the peak centered at 1.54 A.

However, isn't the maximum resolution for Cu Ka radiation actually
very close to *half* the wavelength (Ref: Stout & Jensen, 2nd ed., p37)?
So, could people be arbitrarily cutting off their data at 1.54 A when
it might actally diffract to 1.50 A or better?

I would appreciate any sort of comments that would put my thinking on
the right track.  Thanks.

Charles Lesburg
lesburg at chem.upenn.edu