John
Can't recall ever having read the definitive history. However, it will be found in some dusty ICAO tome of the 50s which led to its embodiment in the Design Standards.
That was a challenge, so I looked out my treasured copy of the 1953 ICAO Final Report of the Standing Committee on Performance. It's there, but you aren't going to believe this!
They were trying to establish statistical margins for a range of aircraft to define how much height should be allowed so that at any point on the Net Take-off Flight Path the probability of being below the NFP would be equal to the design incident probability. To do this they calculated the gross/net margins at a range of heights for a "few typical aeroplanes" and found that the resulting NFP was curved. To simplify things without introducing too much conservatism they decided to approximate the curved path by two straight lines and guess what? the convenient break point was 400 ft.
This matched some unspecified administrative feasibility and got enshrined into the standards.
Subsequently, since some of the aircraft they studied couldn't reach 1000 ft (OEI) before they ran out of time at TOP, the rules were amended to allow a change of configuration after 400 ft.
Incidentally, that same report says that the datum height at the end of the Take-off Distance Required is clearly a point on the NFP, and that since the height of the ground beyond this point is uncertain no credit for ground effect on performance can be given.
Amazing how myths develop isn't it
