In July 1991, Susan Vuke-Foster (Montana Bureau of Mines and Geology) and Ed Belt discovered an angular unconformity 10 to 25 m (32.8 to 82.0 ft) above the base of strata previously mapped (Bergantino, 1980) as the Tongue River Member of the Fort Union Formation. Seven students, directed by the writers in the summer of 1992, made careful field examinations of all the Paleocene strata in a 90-km2 (35-mi2) study area north and southeast of Ekalaka, in Carter and southern Fallon Counties, Montana. Paleocurrent vectors were summarized in azimuthal histograms and microfault poles were plotted on stereonets using a computer in Ekalaka. The results of these analyses were then field checked. These efforts produced five senior honors theses from Amherst, Carleton, and Whitman Colleges in May 1993.

The unconformity was found to extend throughout most of the study area and to be angular, where blocks (Clark, 1993) of underlying rippled sandstone had rotated from the horizontal, but to be disconformable where it had not rotated. The rippled sandstone is a mappable unit 20 to 30 m (65.6 to 98.4 ft) thick, covering an area of 340 km2 (135 mi2). It is lithologically distinguishable from the conformably underlying facies of the Ludlow Member of the Fort Union Formation and from the newly restricted overlying Tongue River Member, with which it is both conformable and unconformable. Most, but not all, of the strata of the Tongue River in the study area are represented by the massive, cliff-forming Medicine Rocks Sandstone.

Late Torrejonian (middle Paleocene) mammalian remains occur near the base of the Medicine Rocks Sandstone (e.g., Mehling Site; Archibald and others, 1987) within the study area. Hence the unconformity mentioned above is pre-late Torrejonian in age. Pollen from the P-l pollen zone (Nichols and Brown, 1992) was identified by Doug Nichols (U.S. Geological Survey, written comm., 1993) from the youngest coal bed in the Ludlow Member below the unnamed rippled sandstone. Nichols and Brown (1992, and earlier work cited therein) indicated that the P-l pollen zone spans the Puercan (early Paleocene) and early Torrejonian. Hence the unconformity is post-early Torrejonian.

Paleodrainage directions, based on channelbelt sandstones in the Ludlow Member of the Fort Union Formation, trend northeast through most of the eastern margin of the Powder River Basin (Brown, 1993). Similar trends are also found in the Ekalaka study area (Abell, 1993; Cole, 1993), in the Cave Hills district, South Dakota (Goodrum, 1983), and between these two areas (Susan Vuke-Foster, unpubl. map data, 1990). Hence the intervening Black Hills district was not a significant obstruction to the direction of fluvial drainage trends during early Paleocene. In contrast, paleodrainage directions trend to the east and southeast in the strata of the same age in the Miles City coal field (Belt and others, 1992; Brown, 1993), and to the south and southeast along the Little Missouri River of southwestern North Dakota (Belt and others, 1984).

The apparent conflict between the northeast and southeast trends in paleodrainage can be resolved by a westward embayment in the margin of the Cannonball Sea along the North Dakota-South Dakota border. This event may have become more pronounced during the transgression of the Oyster tongue of the Cannonball Sea in North Dakota (Belt and others, 1984). Perhaps the unnamed rippled sandstone, which also contains Skolithos-likc burrows, is a marginal marine deposit related to the transgression.

The pollen dates (P-l pollen zone, Doug Nichols, written comm., 1993), from coals in the upper Ludlow and in the so-called Tongue River sandstones (Roger Colton, U.S. Geological Survey, unpubl. geologic map) on the Mill Iron Road crossing of Box Elder Creek, suggest the need for a revision of the stratigraphy there and east of the Box Elder. The writers subscribe to the suggestion (Goodrum, 1983) that Sandstone D and Sandstone E in the North Cave Hills are at least partially marine (see also Cvancara and Hoganson, 1993) and ought to be mapped as the Cannonball Member of the Fort Union Formation, rather than as Tongue River Member (as per Pipperingos and others, 1965). Ophiomorpha and Skolithos burrows were found in Sandstone F by our group in the North Cave Hills area (guided by Jack Redden of the South Dakota School of Mines, 1992). None of the North Cave Hills deposits have been pollen dated (Nichols, oral comm., 1992) to aid in the chronostratigraphic correlation of these strata.

Paleodrainage directions from channelbelts of the Tongue River Member above the unconformity trend southeast throughout the Ekalaka study area. These trends and their relationship to various possible causes for the unconformity need to be resolved in terms of 1) possible movement along the Miles City Arch related to uplift in the Black Hills, 2) the welldated and widespread middle Paleocene (mid-Torrejonian) unconformity in the Bighorn Basin (Hicks, 1993), and 3) the relevance of eustasy within the Cannonball Sea.