You need to get the second mirror in place before the wavefront/photons have a chance to get from where the second mirror is cutting off the light to the first mirror and back again.
Not to mention absolutely parallel to the first mirror.
And if the first mirror wasn't absolutely perpendicular to the light beam from the flashlight (which it can't be, unless the reflector and bulb are rather better made than any flashlight I've ever had) then even if it's placed well the light will escape very quickly.
All this reminds me of the Galilean experiment to measure the speed of light: Two people with covered lanterns stand some distance apart from each other. The first uncovers his lantern and starts counting. The second uncovers
his lantern the instant he sees the first lantern's light. The first person stops counting when he see's the second lantern's light.
A friend of mine in high school propsed that the experiment should work, given modern technology. Rather than covered lanterns we would use electric lamps. Rather than counting we would use a digital stopwatch. (At the time, a digital wristwatch with a stopwatch function was a new toy, and rather expensive at around 60 to 80 bucks for cheap ones. We were looking for a justifiable way to play with our new watches.)
Results were inconclusive indoors, just not enough distance, so we dragged extension cords outside and set it up 400 feet apart. He clicked his lamp on and started his timer, I clicked my lamp when I saw his, and he stopped the timer when he saw mine.
Speed of light, verified through several repeated trials: ~62.5 MPH +/-15%
We later found our science to be flawed. It never occurred to us to account for reflex times, we judged our reflexes to be insignificant compared to light speed.
Sorry, no funny pic to go with the story.
