Bald Friar metabasalt and Kennett Square amphibolite: two Iapetan Ocean floor basalts

Robert C. Smith, II (retired) , Pennsylvania Geological Survey, DCNR, 3240 Schoolhouse Road, Middletown, PA 17057-3534


Geologic mapping of the Piedmont of Pennsylvania has largely utilized units defined prior to the development of the plate tectonic paradigm. Widespread acceptance of the pioneering work of Harold Williams in Newfoundland suggests that recognition of terrane boundaries is critical for understanding the geology throughout the Appalachians. One technique to locate such boundaries is to compare the immobile and incompatible trace element contents of mafic volcanic units to modern, better understood analogs. Such potential affinities can then be combined with traditional field data on the mafic volcanics and their host units. Finally, the areal distribution of a given population of mafic rocks can be used to propose new geologic units. This has been attempted in the Piedmont and has resulted in two first order groupings: One consists of mafic volcanics generated from Neoproterozoic rifting in Laurentia through early drifting. These mafic volcanics are generally referred to as Catoctin Metabasalt, but include compositions that progress from continental initial rifting, represented by Catoctin Metabasalt itself, through drifting, represented by the Holtwood Metabasalt and others. This relatively early Catoctin Metabasalt grouping is discussed in an article centered on the White Clay Creek Amphibolite (Smith and Barnes, 2004), one of the subpopulations that required a more restricted name. The younger major grouping consists of pieces of the oceanic floor of Iapetus that appear to have formed either after the mantle magma source had lost all memory of Catoctin rifting or from an entirely different mantle source. These oceanic basalts were likely docked during the Ordovician and thrust onto Laurentia during the Silurian. This chapter covers two such oceanic formations (Bald Friar Metabasalt and Kennett Square Amphibolite) and mentions others that are not yet well defined. The boundary between the two first order groupings is analogous to the Baie Verte-Brompton plate boundary of the northern Appalachians. Unfortunately, the plate boundary is less impressive in Pennsylvania because the original thrust surfaces were later obscured and “repeated” by folding during the Alleghanian (Smith, 1993) and, in places, overprinted by late Alleghanian lateral shearing (Smith and Faill, 1994).

The Bald Friar Metabasalt (BFM) and Kennett Square Amphibolite (KSA) are two mafic units occurring in the Piedmont of southeastern Pennsylvania (Figure 1). In addition, the WCCA occurs in the Piedmont of northwestern Delaware and the BFM in northern Maryland. Additional data for the KSA and BFM have become available since they were proposed by Smith and Barnes (1994). The KSA is not believed to be directly related to the BFM.

Host rocks to the BFM at the type locality, Cecil County, Maryland, appear to have been included in the Sykesville Formation (Higgins and Conant, 1986) whereas the host to the southwest at Liberty Reservoir, Carroll County, Maryland, is better exposed and has been mapped as the Morgan Run Formation (Muller et al., 1989). According to Muller et al., the Morgan Run Formation together with the overlying Sykesville Formation constitute the Liberty Complex. The Morgan Run Formation itself consists of pelitic schist and, especially near its base, mafic- and ultramafic-bearing metagraywacke, all located in an oceanic and trench fill setting. It is the metagraywacke of the Morgan Run Formation that hosts the BFM at Liberty Reservoir. Muller et al. note that the Liberty Complex is underlain by a low angle thrust, now folded, similar to the situation noted by Smith (1993) at the Peach Bottom structure. Far to the northeast in Québec, the host to what are likely BFM equivalents is the Caldwell Group (Bédard and Stevenson, 1999).

The Bald Friar Metabasalt was proposed by Smith and Barnes as a formal, field-recognizable and mappable unit. It typically occurs as 2.5 ± 1m-thick fragments in mélange associated with ultramafic fragments of the Baltimore Mafic Complex, locally carbonated to listwaenite or steatized. Except for pillows (Figure 2), which have partial internal concentric banding, the BFM is typically a laminated, dark green chlorite-actinolite-epidote metabasalt that is noticeably denser than most other mafic volcanic rocks in the region. Chlorite seems to be more abundant than actinolite at the more northwesterly locations such as MORT, MORTNE, TLPPOP, TLPPOP2, and PB1 and actinolite is more abundant at those more to the southeast. Titanite is the most abundant Ti-bearing accessory mineral at most localities, but polymorphs of TiO 2 are abundant at the three BRANDY-series samples (Stop 14, Smith, 1994).

The type locality for the Bald Friar Metabasalt is herein designated as the four known fragments of BFM exposed in the railroad cut and adjacent areas on the northwest side of Bald Friar, Cecil County, Maryland, in the Conowingo Dam 7½' quadrangle. A stratigraphic section has little meaning in a mélange, but fragments of BFM occur from 550 m WNW of the crest of Bald Friar to 700 m northwest to 450 m slightly north of west. The two most accessible fragments are located in the railroad cut in what was mapped by Higgins and Conant as Sykesville Formation. Sample SYKFM (Appendix I, Table 1) is a greenish gray laminated metabasalt collected from the railroad cut 70 to 80 cm above the base of a 1.6 to 2.05 m-thick slab that is 14 ± 1 m-long (39º45'24”N, 76º13'05”W). Foliation in the metabasalt trends N77ºE, 47ºS. This sample was collected 33 m NW of sample SYKFM2, 220 m southeast of Mile Post 14 and 550 m northwest of the crest of Bald Friar. Sample SYKFM2 is a greenish, laminated metabasalt containing possible sheared vesicles, from the lower limb of a small chevron fold, the hinge of which trends N10 o E. This fragment is 2.5 ± 1 m-thick and exposed for 14 m along the horizontal at track level (39º42'28”N, 76º13'04”W). The sample was collected from a point 7.5 m southeast of where the base is exposed, 530 m WNW of the crest of Bald Friar. Sample BFMZ2 is from a concentrically banded 20x8 cm section through a chlorite-epidote-actinolite pillow. The base of this pillow, which froze in an anomalous orientation of N12ºE, 50ºSE, was left in outcrop for future generations. S o in this set of outcrops trends N63ºE, 50ºSE. All pillows suggest a right-side-up orientation. This pillow was collected from the middle of a 3.2 ± 0.2 m-thick flow 8 m above water level in the pond and about 9.5 m below the bed containing isolated epidote-bearing pillows. This latter is 95 ± 5 m along the railroad tracks NW of sample SYKFM. Sampled pillow BFMZ2 (39º42'35”N, 76º13'08”W) is located S54ºE of milepost 14 and 0.68 km NW of the crest of Bald Friar. Sample BFMZ1 is a dark green chlorite-epidote-actinolite metabasalt having pahoehoe-like surfaces that trend N61ºE, 63ºSE and that exhibit ropy linear (extrusion?) structures that plunge NE at 84º. (As submarine chill surfaces, this sample was subjected to “pillow enrichment disease.”) Collected from 0.5 m above present water level, 0.5 m above the base of a 2.8 m-thick metabasalt fragment. A well- slickensided fault surface 1.1 m to the east of the sample site trends N78 o E, 70 o N and contains slickenlines that plunge to the west at 20º. Apparent S 1 in this outcrop trends N75ºE, 70ºS. This outcrop is located on the SW side of a creek bed and/or eroded talc exploration trench approximately 40 m S50ºE of the main hollow at an elevation of about 85 ± 5 m, 0.45 km NNW of the crest of Bald Friar.

Bald Friar Metabasalt was examined at Stops 7 and 14 during the 59th Field Conference of Pennsylvania Geologists. At Stop 14, ultramafic fragments associated with the BFM on the north side of the Peach Bottom structure ranged from an attenuated, 7 cm-wide fragment to a ~2 m-wide talc-magnesite schist zone to a 9.5 m-wide listwaenite and a 1.9 m-wide talc-magnesite schist zone on the south side (Smith, 1994, Figure 73 and Table 13-A and B). At Stop 7, in addition to BFM in both a road cut and railroad cut below, an apparent Mesozoic fault cutting the section in the railroad cut was noted (Figure 61 by R. C. Smith with assistance from R. T. Faill, 1994).

Two new sets of BFM outcrops have been recognized. One chloritic outcrop was reported by Gil Wiswall and Hal Bosbyshell, West Chester University (personal communications, 1994), as occurring north of Mortonville, Chester County, just prior to the 1994 Field Conference. It was since sampled as MORTN has been confirmed to be BFM. Gale Blackmer, Pennsylvania Geological Survey, later found similar outcrops 0.8 km to the northeast. One was sampled as MORTNE and it, too, is BFM based on the analyses included in Table 1. On a weekend visit with his wife to the Mineral Hill area, Carroll County, Maryland, the BFM was recognized by the author in a section (Table 1) on the north shore of the Morgan Run Arm of Liberty Reservoir, Maryland. Samples from there (LRMRA4 and LRMRA6) were found to be BFM. As shown on Figure 1 of Smith and Barnes (2004), the known lateral extent, albeit highly discontinuous, of the BFM is now 143 km. In addition to BFM and serpentinites, the Liberty Reservoir section and area includes presumably related amphibolites. They have not been noted in Pennsylvania and are only informally proposed herein as a working lithodeme until such time as other localities are recognized. Nevertheless, data for selected trace elements are included in Table 2 and more complete analyses are presented in Appendix I. The amphibolites at Liberty Reservoir might represent fractionated, more slowly cooled intrusive magmas than the BFM.


The Bald Friar Metabasalt is a back arc basin basalt (BABB), chemically similar to N-OFB, derived from the floor of the Iapetus Ocean. It occurs in a tectonic mélange containing ultramafic clasts likely derived from the Baltimore Mafic Complex. Where it is not tectonically attenuated, it and associated ultramafic fragments, and steatized and carbonated fragments derived there from, make an excellent marker for the Morgan Run Formation of the Liberty Complex. The work of Candela et al. (1989) indirectly suggests that detrital chromite might make a suitable marker in sheared formations. Because of the excellent exposure at Liberty Reservoir during water draw downs, it is suggested that the name of the well-described Morgan Run Formation (Muller et al., 1989) be considered for extension into York, Lancaster, and Chester counties, Pennsylvania. With some caution, such a tectonostratigraphic unit could likely be used to map the low-angle Baie Verte-Brompton equivalent obduction thrust fault over the Brandywine massifs, especially along their margins where it was later sheared. It might also prove useful in many other areas of the Piedmont of southeastern Pennsylvania where its appearance is repeated by folding.

The association of BFM with an island arc complex does not make it an island arc basalt. The lack of negative Nb and Ta anomalies on mantle-normalized diagrams in particular clearly shows this. The fact that fragments of BFM fell into sediments derived from a continent does not make it a continental basalt. All that is required for this association is that the closure of Iapetus transport oceanic material toward a continent. As reported by Candela et al. (1989) the BFM is associated with ultramafics at early stages. This is based, in part, on the fact the both ultramafic and mafic source rocks were needed for the metals in the hydrothermal solutions introduced while still on hot seafloor. Thus, it would be quite misleading to consider BFM part of Catoctin rifting as opposed to generation of the oceanic floor of Iapetus.

Thus, the Baltimore Mafic Complex was a sub-arc portion of the floor of Iapetus, the Bald Friar Metabasalt an initially somewhat distal portion of the back arc now found with the BMC in mélange, and the informal Conowingo Creek Metabasalt a forearc boninitic basalt. Such, arc-related ophiolites and boninites are well known farther north in the Appalachians (Bédard et al. 1998 and Kim and Jacobi, 2002).

Kennett Square Amphibolite was part of the floor of Iapetus derived from a spreading center having its lower TiO 2 , La, and Nb end to the east, suggesting that the Wilmington Complex is more likely to have been the source than the Baltimore Mafic Complex. As noted by Smith and Barnes (1994, p.69), amphibolite samples BRANDYB and DILW, which have an apparent affinity to the Wilmington Complex, are found very close to the east end of the “snail race.” They were likely thrust onto the Brandywine massifs during the Silurian. This obduction likely left an attenuated, garnet-free veneer of arc-related granites, amphibolites, and ultramafic over much of the otherwise garnet-bearing Brandywine massifs, not to mention in the higher angle shear zone along the margins. (These shears are likely the result of obduction of the same mélange folded during the Alleghanian and laterally sheared during the late Alleghanian. That they became shear zones is likely due to steatitization of ultramafic clasts in the original mélange.) This obduction likely occurred after the widespread, rift-related thermal pulse associated with the 432.6-Ma Sword Mountain Olivine Melilitite (Smith, Foland, and Nickelsen, 2004) of the extreme northern part of the Ridge and Valley, Maryland, the related 424 ± 20 Ma Beemerville Igneous Complex of northwestern New Jersey, and the 442 ± 7 Ma baddeleyite-bearing refractory oxides in the Baltimore Mafic Complex, Lancaster County, Pennsylvania (Smith and Barnes, in review). The full range of this Silurian extensional event is still being evaluated, but likely includes the Beckett Quarry volcanic arc granite on the margin of the Connecticut Valley trough which may extend northward as the Gaspe synclinorium. Karabinos (2004) related the Beckett Quarry granite, Connecticut, to an extensional environment and dated it at 432 ± 3 Ma using 206 Pb/ 238 U dating of zircons. To the north, Castonguay and Tremblay (2003) To the south, Trujillo and Sinha (2004) document similar such dates in mantle-derived gabbros to diorites in an extensional setting at several localities in Virginia.

This paper was included with the 2004 Field Conference of Pennsylvania Geologists guidebook .