Huh?
If the effect is real, how did these bands near the equator escape it? What accounts for THEIR rogue behavior?
Local gravitational effects randomizing the orientations perhaps.
The data does not provide any information about the orientations of the galactic axes per se. All that is given is a spherical distribution of observed ellipticities.
If the galaxies were physically identical, then the variation in observed ellipticity would be due solely to the galactic orientations. However, there is presumably a distribution in the physical ellipticity as well, and any sort of clustering of this variation will be interpreted by the methodology as clustering in orientation.