A new experiment for testing bendy light.
According to the flatter's theory, the amount of upward bend of light increases with distance travelled. Therefore, a beam of light travelling in one direction cannot possibly coincide with a beam of light travelling in the opposite direction. So here's how to test this.
Let's get two tall buildings, however far apart you want them to be, as long as you can see one from the other. On top of one you mount a green laser, on top of the other you mount a red laser.
You then point the lasers directly at each other, so the red laser hits the origin of the green laser, and vice versa.
According to round earth/straight light theory, both beams are following the same path to reach their destinations.
According to flat earth theory, the beams will have to be following different paths to reach their destinations, as the further from the origin the beam gets the greater the upward bend in the path.
So how can we tell what path the beams are following?
At any point along the beam, one can hold up a disc of translucent white plastic (either by crane, lowering from a helicopter, man with long stick, another building between the laser towers etc - doesn't matter how exactly).
If the light is travelling straight, wherever along the beam one holds up the disc, you will see both red and green light superimposed.
If the light is following a path of increasing upward curvature the further it goes, you will find that part way along the beam, the red and green lasers would not be superimposed on the disc.
If anyone has trouble following this, I'm sure Robosteve will draw you a diagram as he tells me he's an expert on experimental methods with light beams.