Type locality and history of name in Indiana: The original name Tar Springs Sandstone is generally attributed to D. D. Owen (1857, p. 85-87). The name is taken from a natural feature in Breckinridge County, Ky., near which the unit, as described by Butts (1917, p. 103-105), is 100 to 150 feet (30 to 45 m) thick and consists of massive cliff-forming sandstone, shaly sandstone, and shale. The Tar Springs Sandstone is a unit in the standard Chesterian section (Swann, 1963, p. 37-38; Willman and others, 1975, p. 158-159).
The name Tar Springs Sandstone was first applied in Indiana only to a crossbedded cliff-forming sandstone that is as much as 90 feet (27 m) thick in some places but is absent from others (Malott, 1925, p. 106-108). The same name was later applied to a rather indeterminate unit that includes other rocks as well as the prominent sandstone (Malott, Esarey, and Bieberman, 1948). The unit was designated the Tar Springs Formation by Gray and others (1957, p. 6 and pl. 2), and the lower boundary was established at the top of the main or massive part of the Glen Dean Limestone by Gray, Jenkins, and Weidman (1960, p. 30). As a result of this redefinition, a considerable (but variable) thickness of fossiliferous rocks was excluded from the Glen Dean and included in the Tar Springs see the article "Glen Dean Limestone" for a discussion of the biostratigraphic correlation of these rocks. Finally, the upper boundary was fixed at the base of the Vienna Limestone Member of the Branchville Formation and outcrop aspects of the formation were discussed by Gray (1978, p. 5-8).
Description: In Indiana the Tar Springs Formation is primarily shale, but it also contains scattered thin beds of limestone and massive local lenses of sandstone that on outcrop are differentiated as the Tick Ridge Sandstone Member. The formation is commonly about 65 feet (20 m) thick but reaches 100 feet (30 m) in places where the Tick Ridge member is prominent. It is recognized on the surface from southwestern Orange County to the Ohio River and is known in the subsurface from central Martin County southwestward. Although the base of the sandstone member is uneven, in most places the formation overlies the Glen Dean Limestone conformably (but see Swann, 1963, p. 37-38, for a discussion of locally disconformable relationships).
Where the Vienna Limestone Member is present, it overlies the Tar Springs Formation conformably where the limestone is depositionally absent, higher Chesterian rocks (Branchville Formation of outcrop usage) overlie the Tar Springs, also apparently conformably, but the distinction is difficult to make. Northward, however, these formations are truncated by the Mississippian-Pennsylvanian disconformity, and the Tar Springs is disconformably overlain by the Mansfield Formation (Morrowan).
History of boundary problems and concepts of naming formations in upper part of Chesterian Series: For many years the exact positions of most of the boundaries between formations in the upper part of the Chesterian Series were not well understood. In large part this uncertainty resulted from the fact that no detailed work was done on these rocks for many years following Malott's (1925) pioneering study. In that paper Malott did not recognize the need for including all the rocks in his scheme of formal nomenclature, for he stated (p. 106):
The shale masses count but little in the stratigraphic expression of the upper Chester, and may not be used in the study of the structural details which the region of their occurrence expresses. They are mere fillers or intervals between the outstanding limestone and sandstone units which alone are given names and definite status.
Later, Malott and Esarey (1940, p. 7), in describing the section at Sulphur, stated that "the sandstone above the Vienna . . . is not the Waltersburg," the shale interval between them being only 10 feet (3 m) rather than the more common (as then understood) 50 to 60 feet (15 to 18 m), and ".. . the Tar Springs is poorly represented as a sandstone." Still later, however, Malott, Esarey, and Bieberman (1948, stops 2-5, pl. 2) apparently recognized the necessity of extending formational names to the entire section some of the shale intervals were assigned to the limestone formations, however, and some to the sandstones, and it is difficult to determine what philosophy governed the choice.
Boundaries, nomenclature, and lithologic content of outcropping upper Chesterian rocks in southernmost Indiana were reconsidered by Gray (1978). Subsurface usage has never been fully studied, however, and there some difficulties and uncertainties remain.