Doing my rounds, collecting what little is left of Fluorescents here in Colorado Springs,

I was told that every single 4ft T8 (I took the remaining 4ft T12's)tube was collected by... the local shopping mall (Citadel Mall) because apparently that old mall's fluorescent systems will melt, yes dripping melt, LED replacement/retrofit tubes. I actually get it because the main mall core was last touched over 20 years ago, the ballasts (including compact fluorescent CARTRIDGE, or those dual or quad pin CFL bulbs) weren't touched unless they threw a fuse or breaker

which leads me to the final question: why did 4ft fluorescent fixtures become the standard?

Reasonable size lumen package (small enough to not be that of a problem to design with, large enough that the production costs were not that high) combined with reasonable size (fit for standard ceiling patterns) and good system efficiency made them favorites of mainly commercial applications, where the total cost of the light was the top priority. The high demand from commercial sector made the production volume to ramp up. And large production volumes led to lower costs because of economy of scale and brought more RD money to get even better performance from their systems. And lower costs made them even more favored.

Until the F40T12 became the virtually only workhorse of industrial lighting.
The F32T8 is then "just" a more efficient replacement, even when it meant a need for different ballasts (the OCV of the then most common RS ballasts was not enough to reliably start the F32, so a new ballast was needed anyway, so the spec evolved to maintain the lumen equivalency), the equivalent lumen output meant they perfectly fit into older systems (with just the ballast replacement).

In Europe then common glowbottle preheat start systems were able to ignite the T8 perfectly fine, so the evolution went to make the F36T8 electrically compatible with the F40T12 ballasts, with the assumption part of the lamp positions will be skipped to get the desired power saving from the system.

According to the book of Zwikker (1950) i posted earlier this week here, it was because the lamp was developed in Europe where 220v was the common voltage. For the most reliable and stable operation, roughly half the voltage needed to be dropped over the lamp, and the other half over the choke ballast. A length of 120cm and the matching diameter resulted in the best intersection between light output, efficiency and reliable striking.
There is a HUGE amount of 'Why did they do it in specifically that way, when they developed the fluorescent lamp' type of information in that book.

War time production standardisation also has a lot to do with it. Because a lot of light was needed in factories, the fluorescent lamp factories started cranking out as many standardized parts as possible. So the 40w 120cm tube with the matching driver/ballast became one of those parts. The UK used bayonet caps instead of the standard 2 pins, because bayonet caps were more standardized there and therefore easier to mass produce. That's why they kept on using bayonet cap fluorescent lamps for a significant time after the war was over.

It also mentions that because of the low mains voltage in the USA, 15 and 20w lamps were developed with lamp voltages around 55v, so they could operate on basic cheap choke ballasts. Those took a while before becoming popular in Europe because the 40w one was already the big standard at that time. The book does mention that the 20w ones are possibly the ones that will take over living room lighting.

This is all speaking from the 1950 point in time. Halophosphate phosphors were the big new thing. Eventually, despite the higher cost and slight drop in efficiency, in the USA people also massively started installing 40w fluorescents with leakage transformer type drivers.
These were also used in areas in The Netherlands where mains was still 110v. It wasn't until the mid 1960s that the last old city centres where mains was 110v, were converted to 220v.

Of course in the USA where 3 phase power was already available in big factories, you could run choke ballasts and 40 or 36w tubes between two 110/127v phases on 208v/277v or whatever it is. I am not sure whether they actually did that or not, nor about the details of 3 phase usage in the USA. Perhaps someone from over there can chime in about that.

Wow, MELT!
Someone finally designed an LED-proof fixture!

Interestingly, Japan’s 4 foot fluorescent lighting has been a heavy mix of preheat, rapid start, and electronically ballasted fixtures.

The preheat fixtures were intended for use with FL40 (F40T12), FL40S (F40T10), and FL40SS/37 (F37T9) fluorescent tubes.

The rapid start fixtures were intended for use with FLR40/M (F40T12), FLR40S/M (F40T10), and FLR40S/M/36 (F36T10) tubes.

Lastly, the electronically ballasted fixtures were intended for FHF32 (F32T8) tubes, but some ballasts also listed the older magnetically ballasted tubes as well.