Start with an existing black hole and an event horizon radius R at time T. Say the black hole is being "fed" an infinite series of golf balls, one after the other, which are all stamped numerically such that the current golf ball external to the event horizon is 1.0 * 10^32. See linked img: http://i1373.photobucket.com/albums/ag380/rjbeery/golfball_black_holes_zps339d1899.png Now, starting at time T, run the clock backwards to T_past until R_past = R/2. What does the scene look like? Do golf balls with numbers less than 1.0 * 10^32 appear? If they do then there is a time T_crossover such that T_past < T_crossover < T where we could have witnessed the event horizon expand due to matter crossing it. In my understanding of GR, this cannot happen because golf balls external to the event horizon remain theoretically observable (with perfect instrumentation) forever. But in this thought experiment the black hole at time T is made of nothing but golf balls numbered 1 through (1.0*10^32)-1. I find it difficult not to view this as a contradiction, so what is the resolution?