A short-lived display of relatively high quality occurred in Berkshire, UK, in the morning of 4th March, 2017. A routine check at 9:35 local time suggested the presence of faint 18° and 23° parhelia, in addition to the usual 22° stuff, so I collected my camera, tripod and blocker and set up the system in a nearby park area. After just 25 minutes the display was all over, but there had been enough time to capture three 30-frame series for stacking.
In addition to the 18° and 23° parhelia, both visible to the eye during the display, the processed stacks suggest parts of 20° and 35° (and possibly 24°) circular halos. Unfortunately, the region of 9° halo is over-exposed in my photos. However, as far as my visually-based observations are to be trusted, there simply were no halos 9° from the Sun.
Given that we are dealing with a cirrus display here, I find the presence of both supralateral and infralateral arcs indicative of fairly ideal column crystals. This is confirmed by the side-view stack, as faint Wegener and subhelic arcs can be identified. The latter crosses the parhelic circle near the tall tree at the bottom. Just slightly further away from the Sun, there is a 120° parhelion too.