Previous paleomagnetic studies of Eocene rocks deposited in the Talkeetna Mountains on the northern edge of the combined Peninsular, Wrangellia, and Alexander terranes (the southern Alaska superterrane (SAS)) yield a plaeolatitude of 76 ¿N, while Late Cretaceous rocks on the southern edge of the SAS record a 32 ¿N paleolatitude. At face value, these drastically different paleolatitudes imply that Wrangellia moved 44¿ to the north in only 20 m.y., requiring a northward component of velocity of about 24 cm/yr. Alternatively, the discrepancy might be explained by postulating an unrecognized tectonic boundary separating the two localities, allowing different emplacement ages. We have tested the unrecognized tectonic boundary hypothesis (allowing post-Eocene emplacement of the SAS) by sampling Eocene volcanic rocks deposited near the southern edge of the SAS in the Talkeetna Mountains. Our data set comprises measurements from 97 oriented cores collected from 26 lava flows and two tuffs distributed over a distance of 50 km. A primary remanence is indicated by positive fold tests and penecontemporaneous intrusion tests. Age control is provided by nine new K-Ar whole rock age determinations ranging from 38.8 to 53.6 Ma. The mean paleomagnetic pole calculated for this study is not significantly different from the pole derived from the Eocene flows to the north and gives a paleolatitude of 78¿, thus ruling out the hypothesis of an unrecognized suture accommodating major post-Eocene displacement. This result has prompted a reevaluation of the data which suggest rapid movement, including the precision of the age control, the adequacy of averaging of secular variation, and the possibility of sedimentary inclination error. The bulk of the evidence suggests northward displacement from lower latitudes; however, the error limits are such that the dramatic velocities previously postulated are not required. ¿ American Geophysical Union 1990