An analysis is presented of the low-frequency (ω≲0.6 cycles per day) current fluctuations in Lake Ontario, using measurements made at 15 and 75 m during the 1972--1973 winter. The data, collected as part of the International Field Year for the Great Lakes (IFYGL), were obtained at nine stations: one near midlake and the other eight spaced around the lake, about 50 km apart and 12 km from shore. Let U denote the alongshore velocity (positive leaving the shore to the right) and V the cross-shore velocity (positive offshore). It is found that (1) the standard deviation of V is approximately constant: (2) the standard deviation of U, which is about 2.4 times greater than that of V may be depth independent and is less at the southers shore stations than at the shallower northern stations where the flow appears to be directly wind driven; (3) the velocity fluctuations are roughly aligned with the local orientation of the coastline: (4) the U fluctuations show much better mutual correlation than the V fluctuations, having an alongshore coherence length of about 250 km (nearly half the basin perimete): (5) the U signal shows high coherence for periods of about 10--25 and 4--5 days: (6) the computation of lagged correlations shows the U fluctuations to propagate in a counterclockwise sense from the northern shore at a rate of about 42 km d 1(0.5m s-1), which implies a period of about 13 days for a complete circuit around the lake. These results provide additional evidence for the existence of topographic Rossby waves (shelf waves) in Lake Ontario, the calculated period being consistent with results of previous studies which used data from the summer and fall of 1972.