Hi David Hunt

As it is a hand-cast gel, the most probable cause is the gel itself. So it is advised to retry with new gel before going with other causes. Let the gel to polymerase overnight in +4 then run your experiment and wash the wells thoroughly.

"Smiling" on gels (upward curvature of the bands near the edge of the gel) is probably caused by uneven heating. Reeducing the voltage may help, as will avoiding using the edge lanes.

Narrowing and widening of lanes is caused by differences in ionic strength (and maybe pH) between adjacent wells. This can be avoided by making sure all the samples have similar ionic strengths, or by precipitating the protein in the samples to prepare the gel samples.

I would suggest pHing your buffers (just in case). If possible, you can load your ladder in the center and avoid the edge wells. I agree that a longer gel polymerization time and a lower voltage should help too. Those are some pretty tall bands, maybe a thicker stacking gel could help.

Hi David Hunt , this is quite normal. As Adam wrote, this is caused by uneven conditions in neighbouring wells. Thus, if you have samples with variable composition (like elution fractions from chromatography), try to use smaller volume and dilute them a little with something uniform (like your buffer, but without additives).