Unless I'm mistaken most of the early flights were conducted using dead reckoning in the Northern 'hemisphere', where FET and RET don't deviate too greatly. Later flights also used dead reckoning and frequent stopovers to refuel, and again most were either over land (again where FET and RET don't disagree that greatly) or in the North.
Southern 'hemisphere' flights that crossed the oceans came later (although please correct me if I'm wrong on this point) and there aren't enough of them to conclusively prove that there are no viable FE solutions to the ocean sizes before GPS became widely used.
First and foremost, kudos on the style... You're the Man!
There's definite validity to your points. The SH (being ambiguous about what the H stands for -- hemidisc? hemisphere?) is less populated has less land mass and greater distances between its three populated continents. (Asia all but abuts North America in contrast.) Furthermore Great Circle routes don't take commercial traffic any where near the far south as the SH continents are rather distant from that area. Use Google Earth to look 'down' from the RE SP, and you'll be amazed how different the SP region is from the NP.
However, with commercial flights now regularly scheduled and measured between various cities and doing a little math, assuming that no one is breaking the sound barrier on a commercial flight, FE is falsified easily. There have been several convincing posts. I suggest SEARCH on Chile, South Africa, Perth, etc. If you want to do the 'armchair' experiment: use a travel site to see the on-time arrival statistics for various pairings of SH cities. Be sure to eliminate the Jet Stream effect by looking at the 'return' flights. Be sure to ignore Tom Bishop when he says that the flights aren't always on time. If the average flight time is less than FE can handle, FE is falsified.
Thanks again!