How would you tell the difference between a unit cell shift and a wavelength shift when collecting diffraction data at a synchrotron beamline? Well, all the cell length would scale by the wavelength, so that would be one hint that the wavelength changed. If a got longer and c got shorter, then it would be less likely to be a wavelength shift. The crystal-to-detector distance can easily change if the crystal rotation device (i.e. goniometer) rotates the crystal without keeping it at the exact same crystal-to-detector distance. This could easily happen if the crystal is not centered at the rotation point or if more crystal volume rotated into the beam during your experiment. However in a typical experiment the crystal is not going to move by 5 mm and remain in the X-ray beam, so you would not expect your distance to change by 5 mm. Jim _____ From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Jacob Keller Sent: Tuesday, September 09, 2008 3:08 PM To: [log in to unmask] Subject: [ccp4bb] Are cell parameter shifts real? . The bottom line questions are: 1. given that there really are cases of cell shifts, and that there are also probably experimental artifactual changes, how is one to decide what to do? 2. Can there be (or what is?) a plausible mechanism for these shifts? Jacob Keller