[math-fun] ln(x^2+1)

this morning i found that for f(x)=ln(x^2+1), that the (4n+2)th derivatives all vanish at x=1. it took me an hour to figure out why, using geometric series, but surely there must be an easier way to see this. does anyone have an elegant proof of this? erich