In Puharich's classic work (fig 12) he described the appearance of radial pearl trains of bubbles forming
between the cathode and anode in his electrolysis cells. Another researcher (Andy Davies) has posted microscopic imagery of HHO bubbles and release.
A bubble appears to be trapped in an water eddy caused by the
presence of four larger bubbles imho. Also electrostatic forces on the outside of the bubble
may be slowing or preventing or slowing passage beyond the larger bubbles' "log jam"
The use of use of software to measure bubble diameter in a dynamic situation is
interesting. A calibration slide or background could be used to determine volume and
changes of the volume of the free floating bubbles.