Two practical problems in the use of time-lapse photography for the measurement of speed were encountered during the recent surge of West Fork Glacier in the central Alaska Range, Alaska, U.S.A. The first is severe rotational camera instability; we show how natural, unsurveyed features on the valley wall can be used to make the necessary corrections. The second problem is the computation of absolute speed when many different, unsurveyed glacier-surface features are used as targets. We give a method for connecting the data obtained from different targets, and for determining the scale using limited information obtained by surveying. Severe systematic errors can occur unless the angle between the axis of the lens and the direction of horizontal motion is determined.