Don,
I found the Derr article - it was listed in the Bibliography - Derr (BSSA, Dec 1973).
It contains the photographs you discussed in your note. The three color photos were
taken on

Feb. 7, 1966 (0421 JST or Feb 6 19:21 UT)
Feb 12, 1966 (0417 JST or Feb 11 19:21 UT)
and
Sept 26, 1966 (03:25 JST or Sept 25 18:25 UT)

Checking these times against the geomagnetic storms we find that geomagnetic
storms (and aurora borealis) occurred were occurring at the following times:
Sept 25 - major geomagnetic storm began about 18:30 UT (1966)
Feb 11 - strong geomagnetic storm peaked about 11:30 UT (1966)
Feb 6 - strong geomagnetic storm was finishing up about 19:00 UT (1966)

Black and white pictures included in the Derr article also purporting to show
EQL were taken on :
Jan 21, 1966 (18:20 UT) - a strong geomagnetic storm peaked at about 18:30 UT
on Jan 21, 1966

Sept 19, 1966 (09:21 UT) - a strong geomagnetic storm was at maximum at 09:30 UT
on Sept. 19, 1966

Jan 12, 1967 (09:30 UT) - a very strong solar flare occurred at 09:30 on Jan 12, 1967
leading to a major geomagnetic storm on Jan 14.

Dec 4, 1965 (14:48 UT) - a strong geomagnetic storm began about 14:30 UT on
Dec 4, 1965.

OK, so we can see that each of these photographs was taken at the time of
strong or major geomagnetic storms (and presumably associated auroral activity
which could well have been seen in Japan). Does that mean that these are pictures
of Aurora, not of EQL? Not necessarily. Since earthquakes did apparently occur
at or near the time of the pictures - the Matsushiro swarm went on for months
and included thousands of earthquakes. A statistical study of the swarm showed
that a strong correlation between geomagnetic activity and earthquakes existed
during the swarm - the first such specific correlation reported in literature that I am
aware of. Perhaps the geomagnetic storm somehow interacted (maybe by
inducing electrical currents in the earth) with the faults responsible for the swarm
triggering EQL at the time of the geomagnetic storms.

Even so, I would need a good deal more evidence that these are really rare pictures
of earthquake lights and not of more mundane aurora before I could accept
that hypothesis.