Re my previous post of this title:
Harry Greenblatt just pointed out that I should have said that it is the
reciprocal cell edges a* and b* that are non-equivalent in P4, not the
real-space cell edges a and b. Hope that didn't confuse too many people.