In a plane the measure of an angle in radians is given by the ratio of arc length to radius. Can you devise an analogous measure of angle in three dimensions? Does your measure behave sensibly in limiting cases?