can be found such that the errors of calculations, respecting magnitudes exhibited by

′ which assume that

′ is motionless in

, tend to the limit zero, provided that the time-lapse of the durations in

within which