Eighteen interplanetary sector boundary crossings observed at 1 AU during the period January 1971 to January 1974 by the magnetometer on the Imp 6 spacecraft was discussed. The events were examined on many different time scales ranging from days on either side of the boundary to high-resolution measurements of 12.5 vectors per second. Two categories of boundaries were found, one group being relatively thin (averaging ?104 km) and the other being thick (averaging ?106 km). In many cases the field vector rotated in a plane from polarity to the other. Only two of the transitions were null sheets. Using the minimum variance analysis to determine the normals to the plane of rotationa and assuming that this is the same as the normal to the sector boundary surface, it was found that the normals were close to (?45 ¿) the ecliptic plane. The high inclination of the sector boundary surfaces during 1971--1973 verifies a published prediction and may be related to the presence of large equatorial coronal holes at this time. An analysis of tangential discontinuities contained in 4-day periods about our events showed that their orientations were generally not related to the orientations of the sector boundary surface, but rather their characteristics were about the same as those for discontinuities outside the sector boundaries. Magnetic holes were found in thick sector boundaries, at a rate about 3 times that elsewhere. The holes were especially prevalent near stream interfaces, suggesting that they might be related to the convergence and/or slip of adjacent solar wind streams.