<<Reply from a confocal company employee>> Coming from England, it took me the longest time to figure out what was so bad about a football-shaped sampling volume! My guess is that there are two things going on here: 1/ The point spread function will indeed elongate very small structures in the Z direction to make them football shaped, but this will not change the size of larger structures, except perhaps to make their tops and bottoms more fuzzy. 2/ If the aspect ratio of your pixels is a little out of whack, the whole image gets stretched in the Z-dimension. Using the refractive index correction should help, but you should really find out what is happening before you hide it. Given the sample, I doubt if a real refractive mismatch effect could give so large an error. The first confocal system I ever used had an XY aspect ratio of about 1:1.4, which had to be "corrected" with the vertical hold of the monitor. I suggest you try an X-Z image through the beads and see if it still gives the same effect. XZ images are built directly from scanning in X and Z rather than being assembled in software from a large number of slices. If the error disappears, the problem is in the rendering. If not, suspect the Z-drive or the accuracy of reported X-Y pixel size. Have you tried larger beads? Molecular Probes has some very nice FocalCheck beads which I helped them develop for just this kind of problem. They are 15 microns in diameter with two or more fluorophores embedded under their outer surface to give a sharp ring structure when seen in cross section. If your voxels are cubes, they will appear as circles in any direction. If your voxels are elongate, they will become footballs. I just couldn't resist this online plug, but if you would like to talk to me off-line, I will see if I can rustle up a slide for you to try. It's hip to be square! David Carter Applications Scientist Meridian Instruments 2310 Science Parkway Okemos Michigan 48864 Tel. (517) 349 7200 Fax. (517) 349 5967 Personal e-mail [log in to unmask]