The Ramapo Fault System in New York and Adjacent Northern New Jersey: a Case of Tectonic Heredity

Home , Newark Basin, Ramapo Fault

NICHOLAS M. RATCLIFFE Department of Geology, City College of City University of New York,, New York,, New Yor{ 10031 The Ramapo Fault System in New York and Adjacent Northern New Jersey: A Case of Tectonic Heredity

ABSTRACT by Page and others (1968) suggests that the fault may still be active. These observations The Ramapo fault system forms the north- suggest a possible tectonic longevity of some western boundary of the Newark Triassic 700 m.y. for the "Triassic" border fault system basin in New York and adjacent northern New here. Jersey, and is commonly attributed to Mesozoic This fracture system had deep crustal con- crustal fracturing. However, the detailed nections along which basic magma rose toward geology in the vicinity of this fault indicates an the surface in Late Precambrian, Late Ordo- earlier, complex tectonic ancestry, perhaps vician (Cortlandt), and Late Triassic time dating from late Precambrian time. Intense (Palisades sill and Triassic flows). The Ramapo cataclastic effects found in Precambrian to fault also marks the boundary between tectonic Middle Ordovician rocks on both sides of the blocks having different Paleozoic deformational Ramapo fault are absent from the bordering histories. These observations, along with the Triassic rocks of the footwall block. Significant remarkable longevity, suggest that this fault post-Middle Ordovician right-lateral-trans- zone may be part of a fundamental crustal current faulting is recorded by actual offset of fracture system that was operative during distinctive units and by the megascopic fabric formation of the Appalachian orogen. in mylonite zones. The total cumulative offset The ideas presented here require a degree since the late Precambrian on these faults is of tectonic permanence not commonly reported unknown but could be large. Right-lateral dis- in orogenic belts. Activity along this fracture placement of 4 km is indicated for the Canopus system in pre-Middle Ordovician time as Valley area. Intrusive relationships of probable documented here suggests strongly that the late Precambrian to Late Ordovician diorite northern end of the Reading Prong is not plutons and dikes, as well as pegmatites of allochthonous in the sense proposed by Isachsen probable Devonian age, help date the strike- (1964). slip movements as Paleozoic and older. Block faulting along the Ramapo fault in the INTRODUCTION Lower Ordovician affected Middle Ordovician The Triassic basin of New Jersey and New sedimentation and resulted in unconformable York (Newark basin) is a half-graben bordered relationships between a Middle Ordovician on the northwest by a northeast-trending gneiss-boulder conglomerate (Bucher, 1957) system of irregularly developed high-angle and Precambrian gneisses on the upthrown faults (Sanders, 1963; Van Houten, 1969). block west of the fault north of Peekskill, New Although the Precambrian rocks of the foot- York. wall commonly have contacts with the down- Repeated movements of the Highlands dropped Triassic, locally Paleozoic (Cambro- block during Late Triassic time produced the Ordovician) rocks lie adjacent to the border coarse clastic fanglomerates of the Hammer faults on the east (Bayley and others, 1914; Creek Formation adjacent to the Ramapo fault. Drake and others, 1961). Depositional, non- This syndepositional faulting was followed by faulted contacts of Triassic rocks lapping onto post-Brunswick downdropping along the Ram- Paleozoic or Precambrian rocks along the apo fault south of the Hudson River. Recent western and northern margin of the Newark seismic activity along this fault system reported basin have been cited (Drake and others, 1961;

Geological Society of America Bulletin, v. 82, p. 125-142, 5 figs., January 1971 125

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 126 N. RATCLIFFE—RAMAPO FAULT SYSTEM

Van Houten, 1969; Ratcliffe, 1968a, p. 200). here referred to collectively as the Ramapo Thus the northwestern boundary of the Newark fault system. basin Triassic, although commonly referred to The extension of this fault system from New as the Triassic border fault, actually is com- Jersey across the Hudson River to the north- prised of many closely spaced separate sub- east into the Highlands of New York was sug- parallel faults that commonly trend N. 30° to gested by Berkey and Rice (1919) and appears N. 50° E. This consistency in trend, except to be well substantiated (Bucher, 1957; Paige, where later northwest-trending cross faults 1956; Ohan, 1964). such as those near Flemington, New Jersey, The Ramapo fault system west of the Hud- offset the border faults (Sanders, 1963), gives son River clearly underwent repeated move- the impression of a rather continuous fault on ments in the Late Triassic, as recognized by regional compilations shown at a small scale other workers (Carlston, 1946; Sanders, 1963; (for example, Sanders, 1963, Fig. 7). The term Van Houten, 1969). However, estimates of the "border fault system" will be used to describe maximum dip-slip displacement on the border collectively these various faults that either fault in New Jersey in post-Brunswick time mark the northwest border of the Triassic range from 30,000 ft (Sanders, 1963, p. 510) to sediments or lie close to the depositional edge of 18,000 ft (Van Houten, 1969, p. 328). the basin. Sanders' recent synthesis of Triassic tec- In northern New Jersey and southeastern tonism (1963), which summarizes much of the New York State, the border fault system is earlier work by that author and others, sug- expressed by a fairly straight fault trace gests that Triassic deformation was much more marked by the topographic escarpment of the extensive and complex than previously realized. Ramapo Mountains for which the fault is According to Sanders (1963, Fig. 2), the initial named. The Ramapo fault proper extends from phase of Triassic subsidence extended over a Stony Point, New York, on the Hudson River, single belt 50 to 70 mi wide, covering broad southwest approximately 50 mi to Peapack, areas of southern Vermont, western Massachu- New Jersey (Fig. 1). Banked against the fault setts, western Connecticut, and southeastern are coarse deposits of Triassic conglomerate New York and New Jersey. The single deposi- containing abundant clasts of dolostone and tional basin (the broad-terrane hypothesis) was metaquartzite thought to have been locally modified by longitudinal crustal warping of derived by uplift and erosion of the Paleozoic 30,000 ft and subsequent erosion to produce the cover of the Hudson Highlands during Late now isolated Massachusetts-Connecticut and Triassic time (Carlston, 1946; Savage, 1968). Newark basins. The northern termination of the At the northern end of the Triassic basin near Newark basin at the Hudson River (Stony Stony Point, New York, the Triassic rocks Point) (Fig. 1) was ascribed to the effects of a probably rest unconformably on Paleozoic transverse upwarp known as the Danbury rocks (Ratcliffe, 1968a, p. 200) rather than anticline, said to have 30,000 ft of structural being downfaulted along an east-west fault as relief (Sanders, 1960, p. 129; 1963, p. 510). An shown by Fisher and others (1962) and Savage important corollary of Sanders' hypothesis is (1968). At this point, the Ramapo fault down- that the fault system that produced the initial drops both Triassic and Paleozoic rocks against Triassic graben subsidence and later post- the Precambrian on the west. This fault extends Brunswick faulting should extend northeast- across the Hudson River to the Peekskill ward beyond the present limits of the Newark Hollow area where Paleozoic rocks (Annsville basin along the Ramapo fault. According to Phyllite-Middle Ordovician) rest against Pre- Sanders, Triassic rocks are not preserved along cambrian gneiss (Berkey and Rice, 1919; Ohan, this fault owing to erosion produced by uplift 1964). From Annsville, a complex fracture zone on the Danbury anticline (1963, p. 518). along Canopus Creek extends northeasterly There is considerable evidence cited in the across the entire width of the Hudson High- present paper that the Ramapo fault system lands locally, bringing Paleozoic rocks ("Tren- does extend into the Hudson Highlands east of tonian" Limestone; see Bucher, 1957) or the the Hudson River. However, the differential Poughquag Quartzite of Early Cambrian age movement and igneous activity that took place in juxtaposition with older rocks (Berkey and along this fracture zone appears to be pre- Rice, 1919; Gordon, 1911) (Fig. 1). The Triassic in age. Many of the intense deforma- Ramapo fault proper and its northern extension tional effects along this ancient fracture system into the Hudson Highlands of New York are are here attributed to severe cataclasis along

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 INTRODUCTION 127

deep-seated, right-lateral, late Precambrian any direct evidence of Triassic or younger and Paleozoic transcurrent faults. To what movement east of the Hudson River. In the extent Triassic or younger faulting modified opinion of this author, Triassic movement was the geologic relations along the fracture zone is rather limited along the northern trace of the uncertain. However, there does not appear to be Ramapo fault system. These ideas conflict

MAFIC INTRU3VES - PEACH LAKE u " 2-CROTON FALLS

8 - LAKE HOPATCONG 3- CANOPUS CREEK 9- GREENWOOD LAKE 4-ROSETOWN 5-STONY POINT 6- TORMENT HILL

7- CORTLANDT

Figure 1. Regional geologic map showing the eugcosynclinal rocks in the Manhattan Prong; p€f, northern end of the Newark basin, the Ramapo fault Fordham Gneiss of probable Precambrian age; p€, and its possible extensions into the Hudson Highlands, gneisses of the Hudson Highlands Massif. Numbered as well as mafic intrusive centers of probable late Pre- localities are referred to in text. Heavy dashed lines cambrian and Late Ordovician-Early Silurian age. S-D, show faults considered pan of the Ramapo fault Silurian and Devonian rocks; <2-O, Cambrian and system; light dashed lines show other high-angle faults Ordovician mctasediments, including shelf and in the Hudson Highlands.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 128 N. RATCLIFFE—RAMAPO FAULT SYSTEM

in general with the broad-terrahe hypothesis that the crustal fractures along which Triassic espoused by Sanders (1963) regarding the rocks were downdropped were present as zones extent of the original Triassic basin and sug- of dislocation as long ago as the late Precam- gest that the Danbury anticline proposed by brian or Early Paleozoic. Reactivation of these Sanders (1960, p. 129) may not be required by faults at various times in the geologic past is the geologic data. thought to be responsible for the complex The purpose of this paper is not to deny the deformational effects seen in Paleozoic and importance of Triassic faulting but to suggest Precambrian rocks adjacent to the Ramapo IGNEOUS ROCKS

LAMPROPHYRE DI KES LARGELY TRIASSIC FANGLOMERATE CORTLANDT EQUIVALENT (HAMMER CREEK FM) ,AND BRUNSWICK FM. •f^- UNCONFORMITY^-— CORTLANDT PLUTONIC ROCKS INCLUDING NORTHERN PART OF THE ROSETOWN COMPLEX ANNSVILLE PHYLLITE OR MANHATTAN SCHIST PRE CORTLANDT INTRUSIVES OF SOUTH PLUTON- ROSETOWN //Y AND CANOPUS PLUTON TRENTON LIMESTONE SHOWN ONLY AT CANOPUS A' LOCALITY

WAPPINGER LIMESTONE -' ROSETOWN COMPLEX/ i» OR INWOOD MARBLE

POUGHQUAG QUARTZITE ~-' UNCONFORMITY <~

PRECAMBRIAN GNEISSES OF THE HUDSON HIGHLANDS OR FORDHAM GNEISS(p6F) IN MANHATTAN PRONG

INTRUSIVE 8RECCIA ROSETOWN COMPLEX

CREDIT. DODO,I965(POPOLOPEN FRIMPTCR.(THIELLS>

Figure 2. Generalized geologic map of the northern Hudson Highlands (fault patterns west of the Ramapo end of the Newark basin, showing the relationship of fault from Dodd, 1965, [Popolopen Lake quadrangle] the Ramapo fault system to the Rosetown and Canopus and Frimpter, 1967). See Figure 3 for location of plutons. Heavy lines signify faults of the Ramapo fault quadrangles. system; fine dashed lines other high-angle faults in the

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 CATACLASTIC DEFORMATION 129

fault and along the northeastern projection of and others, 1961). These authors thought this fault in the Hudson Highlands. Triassic faults locally followed the older zones The idea that pre-Triassic movement occur- of weakness. red on the Ramapo fracture system was sug- In the Tomkins Cove-Thiells area of New gested by Berkey and Rice (1919, p. 116). York, near the northern terminus of the Recent information, however, comes from Newark basin, both the Paleozoic rocks and several different areas of investigation, sum- the Precambrian gneisses have been extensively marized in the following: (1) study of the mylonitized (Fig. 2). Cataclastic deformation physical nature of the cataclasis exposed in is most pronounced, however, in the gneissic fault zones bordering the Triassic; (2) relation- rocks. Here a 200- to 500-ft-wide zone of ship of the fault system to zones of deep crushed rock, ranging from cataclasite to erosion during Middle Ordovician time and ultramylonite, parallels the fault along whose study of the facies of the limestone deposited trace Paleozoic and Triassic rocks are down- immediately above the Middle Ordovician dropped (Fig. 3). As the fault is approached, unconformity; (3) study of igneous rocks of the Middle Ordovician Annsville Phyllite probable late Precambrian to Ordovician or develops a cataclastic fabric that overprints Early Silurian age that have intruded fracture the regional axial plane foliation. Despite the zones in the footwall block parallel to the intense cataclasis in the Paleozoic and Pre- adjacent downdropped Triassic blocks; (4) cambrian rocks, exposures of the Triassic nature of the Triassic sediments and dep- fanglomerate within 50 ft of the fault are non- ositional environment adjacent to the border cataclastic and are undeformed except for small faults during Late Triassic time. west-dipping antithetic faults. These various lines of evidence indicate that The fault contact was exposed briefly during the geologic history of the rocks in the vicinity construction at a locality near Tomkins Cove of the Ramapo fault has been complex. (Fig. 3, Loc. 1). A punky-weathering zone of Evidently repeated movements were localized open-work fault gouge composed of irregularly along structural trends subparallel to the north- shaped and oriented fragments of Annsville west wall of the New Jersey-New York Triassic Phyllite marks the fault contact. Nonfolded half-graben. This correspondence in space is but rotated fragments of phyllite 1 to 2 inches probably not merely coincidental. The fact in diameter form a loosely compacted porous that large volumes of tholeiitic basaltic magma material. Adjacent outcrops of granite gneiss rose toward the surface in this area during the are mylonitized, containing seams up to several Late Ordovician (Cortlandt) and again in Late feet thick of jet black well-foliated ultra- Triassic time suggests some fundamental deep mylonite. Evidence of the extremely severe crustal control. shearing such as would be required to produce the mylonites and ultramylonites seen in the CATACLASTIC DEFORMATION: adjacent granite gneiss is not found at the fault EVIDENCE FOR REPEATED contact. MOVEMENT The nature of the older faulting that pro- Bayley and others (1914) pointed out the duced the mylonite is uncertain. However, significance of a zone of cataclasite and mylonite displacement of the Paleozoic rocks on the in the Hudson Highlands adjacent to the Ramapo fault of Tomkins Cove (Fig. 4) sug- Triassic border fault in New Jersey. Recent gests a component of dip-slip movement of investigators have found a zone of brecciated 3000 to 4000 ft, judging from the exposures rock approximately 1000 ft wide in gneisses of Poughquag Quartzite on the footwall. A west of the border fault in the vicinity of the significant component of right-lateral strike- Delaware River near Frenchtown, New slip movement as shown by the offset exposures Jersey (Drake and others, 1961; Drake, 1969). of Poughquag Quartzite (Fig. 2), however, At this locality, either Paleozoic or Triassic cannot be ruled out. Such transverse movement rocks are in fault contact with the older would also be consistent with intense mylonit- gneisses. It was suggested that the intense ization of the quartzite that is found on both cataclasis and mylonitization in rocks of the sides of this fault. footwall could not have been produced during Movement of the Triassic rocks, however, an episode of Triassic normal faulting, but was could have been considerably less than the caused by either wrench or reverse faulting possible 3000 to 4000 ft displacement on the during a Paleozoic structural event (Drake older fault, perhaps as little as 200 to 500 ft of

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 130 N. RATCLIFFE—RAMAPO FAULT SYSTEM

vertical net slip. It is significant that the Triassic its contact with the older rocks, thus suggesting fanglomerate is only in fault contact with the it is an erosional-depositional contact. footwall of the main border fault for a distance These observations suggest strongly that of 700 ft (Fig. 3). Paleozoic rocks form the the cataclastic effects seen in the rocks did not hanging wall along most of the fault. Moreover, all form at the same time. The final phase of the strike of the Triassic rocks (Fig. 3) parallels Triassic subsidence probably was responsible

CCIATED DIORITES OF SOUTH PLUTON (PRECAMBRIAN)

IGNEOUS FLOW xSTRUCTURE \ 45

FOLIATION ~—' '^—" '"X_^ /~\

ZONES OF CATACIASIS

TRIASSIC CONGL

Figure 3. Generalized geologic map of the Ramapo Precambrian granite gneiss; €pQ, Poughquag fault at the northern end of the Newark basin near Quartzite; €OW, Wappinger Limestone; OA, Stony Point, showing the location of the Rosctown Annsville Phyllite. Igneous units identified on map. pluton and line of section shown on Figure 4. Num- Geology mapped by Ratcliffe, 1967-1970; and Shuart, bered localities are referred to in text. Explanation: pG, 1968-69.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 EVIDENCE OF LOCAL UPLIFT 131

for the open-work breccia described above. The mylonite zones shown on Figure 5 are Such an interpretation is consistent with the all marked by strong development of minor observation that the Triassic rocks adjacent to folds showing right-lateral shear senses and the Ramapo fault are not cataclastically near vertical fold axes. Evidence for right- deformed. The Triassic fanglomerate uncon- lateral transcurrent faulting is best displayed formably overlies the Wappinger Limestone in along the western side of the Canopus Valley this area and, judging from surface dips, the marble belt in the vicinity of the Canopus bottom of the Triassic basin is only several pluton. Here, offset of a distinctive 1.5- to 3-ft- hundred feet beneath the surface at the border thick magnetite deposit, shown by a special fault (Fig. 4). Because no significant Triassic symbol on the map (Fig. 5, Loc. 2), suggests a downdropping can be proved at the northern right-lateral displacement of 4 km (2.48 mi). end of the Newark basin, it appears unlikely The age relationships of this fracturing will that the Triassic sediments ever extended much be discussed in the section dealing with the farther northeast than the present exposures of intrusive rocks. the Triassic unconformity, thus eliminating the necessity of the Danbury anticline of Sanders THE MIDDLE ORDOVICIAN (1963, p. 510). UNCONFORMITY: EVIDENCE OF Extensive mylonitization along a complex LOCAL UPLIFT fracture zone is found along the extension Bucher (1957) reported the remarkable of the Ramapo fault east of the Hudson River exposures at a road cut on Route 9 near the at Annsville and Canopus Hollow. It is signif- town of Annsville, New York, on the east side of icant that cataclastic deformation similar to the Hudson Valley just north of Peekskill (Fig. that ascribed to the older faulting at Tomkins 5, Loc. 1). At this locality, a limestone con- Cove marks the extension of the fault zone to taining subrounded as well as angular clasts of the northeast rather than the open work Precambrian gneiss, quartzite, and dolostone breccia of the youngest fault episode. Cata- rests unconformably on gneisses of the Hudson clastic deformation took place at various times Highlands. The limestone matrix was reported in the Canopus area, judging from the cross- to contain "cystid" fragments (Bucher, 1957, cutting relationships of mylonite zones. How- p. 669), and on the basis of correlation with a ever, the details are imperfectly known at similar fossiliferous rock that overlies the present. The field relationships as presently Wappinger Limestone at Tomkins Cove understood are presented in Figure 5. The area (Bucher, 1957) was assigned a Middle Ordovi- was mapped by the writer at a scale of 1 in.: cian age. Bucher's suggestion that this lime- 1000 ft during investigation which spanned a stone represents the base of the regionally three-week period. The geologic relationships recognized Middle Ordovician unconformity vary significantly from previous studies by is generally accepted (Ratcliffe and Knowles, Berkey and Rice (1919) and Ohan (1964). 1969).

Figure 4. Profile and section of the Ramapo fault fault zone into which diorites of the younger part of in the vicinity of the Rosetown pluton, showing the the Rosetown pluton intrude. See Figure 3 for line of offset of the Paleozoic rocks and the Late Ordovician section.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 IGNEOUS ROCKS CANOPUS PLUTON

ANNSVILLE PHYLLITE LATE.NONCATACLASTIC QTZMONZONITE HB- BIO PEGMATITE

MARBLE IN CANOPUS BIO.-HB.-AUGITE DIORITE FRACTURE ZONE CANOPUS PLUTON -

WAPPINGER LIMESTONE

6PQ

POUGHQUAG QUARTZITE

p6BG P6AG F

DENOTES AREAS OF EXTENSIVE MYLONITE

TRANSCURRENT FAULT

PEEKSKILL GRANITE (DEV?)

/CORTLANDT / COMPLEX '(LATE ORD.)

Figure 5. Geologic map showing the projection of p€GG, granodioritic gneiss; pCRG, rusty-weathering the Ramapo fault system northeast of the Hudson River fine-grained biotite granite gneiss; p€MT, magnetite along the Canopus Valley (Peekskill quadrangle, ore and associated calc silicates; p€U, undifferentiated New York), and the position of the Canopus pluton. gneisses. Index to locations: (1) unconformity at Subdivisions in the Precambrian: p€BG, biotite- Annsville, (2) offset magnetite deposits, (3) Annsville plagioclase-quartz paragneiss; p€)AG, amphibolitic contact at Peekskill Hollow. gneiss; p€HG, hornblcnde-biotitc granite gneiss;

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 IGNEOUS INTRUSIVES IN FAULT ZONES 133

The Route 9 exposure is located at the south- 4) (Ratcliffe and Knowles, 1969). The change western end of a long narrow inlier of carbonate in the basal limestone facies in a northwest rocks in the Canopus Valley that lies along the direction as the Ramapo fault is approached (at northeastern projection of the Ramapo fault the latitude of Tomkins Cove) suggests that from the Tomkins Cove area. an uplifted area existed in that direction during The carbonate rocks of Canopus Valley were Middle Ordovician time. This evidence and the assigned a "Trentonian" age by Bucher (1957) observable unconformity at Annsville suggest on the basis of the reported fossils at Annsville, that block faulting may have been active on and that interpretation is used in Figure 5. the Ramapo fracture system in pre-Middle However, study of the carbonate rocks by the Ordovician time. author suggests that this interpretation may Block faulting of this kind is believed to have be too simple, as some of the dolomitic rocks been important during the formation of the within the valley that appear to be more re- Middle Ordovician unconformity in the crystallized than the limestone at Annsville Middlebury synclinorium and its southern may eventually prove to be lower Paleozoic extension in western New England (Zen, 1968). Wappinger equivalents or even late Precam- A locality in the Pawlet quadrangle, Vermont, brian in age as suggested by Berkey and Rice described by Thompson (Thompson and (1919, p. 105-106). This problem is un- Shumaker, 1967, p. 87-89), is sufficiently resolved at present. similar to the relationships thought to exist at The Cambro-Ordovician sequence is well Peekskill to warrant discussion here. On the preserved in a syncline (Bucher, 1957, p. 669) north slopes of Dorset Mountain, Vermont, the in the adjacent Peekskill Hollow (Fig. 5). The Middle Ordovician Ira Formation uncon- fossiliferous base of the Annsville Phyllite formably overlies adjacent outcrops of Pre- (Ratcliffe and Knowles, 1969, p. 50) rests on cambrian Mount Holly Gneiss and Lower Wappinger Limestone approximately 1200 ft Ordovician Shelburne Marble. Because the above the Precambrian basement on the eastern Ira shows no signs of having been offset itself, side of the syncline (Fig. 5, Loc. 3). The con- and because the unconformity surface over- tact of the Annsville on the western limb of the laps rocks as young as the Bascom Formation syncline unconformably rests on Wappinger locally, a post-Lower Ordovician-pre-Ira nor- Limestone, Poughquag Quartzite, or Precam- mal fault was proposed to account for the field brian gneiss (Fig. 5). Thus the pre-Middle relationships. Locally in the vicinity of the Ordovician erosion can be shown to have cut fault, volcanic rocks, the Baker Brook vol- most deeply in the direction of the uncon- canics, form a unit immediately above the formity at Annsville. Local differential uplift unconformity. Evidently the pre-Ira Tinmouth in excess of 1200 ft is suggested. Bucher (1957, Disturbance, as it is locally known in Vermont p. 669) suggested that warping was responsible (Thompson and Shumaker, 1967, p. 88), in- for the truncation. However, the development volved block faulting as well as volcanic of coarse clastic debris (clasts as much as 1 ft activity that presumably reached the surface in diameter; Bucher, 1957, p. 668) calls for via the pre-Ira faults. considerable local topographic relief such as The Peekskill locality and the nearby Rose- could be accomplished in a block-faulted, town area also underwent block faulting, shallow-water marine environment. erosion, and basaltic magmatism (in the form Study of the matrix of the "Trenton" of dikes and small plutons), suggesting crustal limestone is particularly interesting. Both at instability similar to that active in New the Canopus locality (Fig. 2, Loc. 1) and at England at the same time. Tomkins Cove (Fig. 2, Loc. 2) the limestone is marked by distinctive beds 1/4 to 1/2 in. IGNEOUS INTRUSIVES IN FAULT thick, rich in sand-sized silicate detrital debris ZONES: RELATIONSHIP TO including quartz grains and some rock frag- REGIONAL FAULTS ments of gneiss, calc-silicate rock, and detrital Igneous rocks of the Cortlandt Complex feldspar (Bucher, 1957, p. 668; Ratcliffe, and other mafic rocks are extensively developed unpub. data). It is significant that this detrital to the east of the Ramapo fault (less than 1 mi debris is absent in the same basal limestone east) and along a fault that extends northeast exposed at Verplanck Point (Fig. 2, Loc. 3), from Peekskill, forming the southeastern Peekskill Hollow (Fig. 2, Loc. 5), and south of boundary of the Hudson Highlands against the the Cortlandt Complex at Crugers (Fig. 2, Loc. Manhattan Prong (Fig. 1). Only rarely, how-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 134 N. RATCLIFFE—RAMAPO FAULT SYSTEM

ever, are these mafic rocks intrusive into the dovician age for the diorite (Kemp, 1888; Highlands block. The Rosetown pluton (Kemp, Frimpter, 1967; Ratcliffe and Shuart, 1970). 1888; Shuart, 1969; Ratcliffe and Shuart, 1970) North and south of the pluton, unsheared several hundred feet west of the Ramapo fault lamprophyric dikes (largely spessartites) and at Tomkins Cove (Fig. 1, Loc. 4) and a small small pods of diorite intrude the shear zone. biotite diorite-monzonite complex (Fig. 1, These lamprophyric dikes intrude gneissic Loc. 3; Fig. 5) along the west wall of the rocks in the vicinity of the Ramapo fault near Canopus inlier are the exceptions. Tomkins Cove (Fig. 3) and, although foliated Both the Canopus and Rosetown localities near the fault, have not been as intensely de- are located along the trend of the Ramapo formed as the mylonitized country rocks. fault and have textures and structural relation- Igneous rocks from the younger part of the ships indicating that they were intruded during Rosetown pluton include biotite-hornblende a tectonic episode of brittle fracturing of the diorites, lamprophyres, and cortlandtite that basement rocks along the Ramapo fault match in detail lithic types found in the trend. Although these plutons were attributed Cortlandt intrusions at nearby Stony Point, to Cortlandt magmatism (Fisher and others, where the Wappinger-Annsville (Inwood- 1962; Isachsen, 1964, Fig. 5), the evidence Manhattan A) have been intruded (Ratcliffe, presented here and elsewhere (Ratcliffe and I968b). The proximity of these two intrusives Shuart, 1970; Ratcliffe and R. G. Senechal, in (Fig. 2), their maximum age, and the similarity prep.) suggests that the igneous rocks are in of igneous rock types may indicate that they part Precambrian and in part lower Paleozoic are approximately coeval. Based on a biotite in age. K-Ar age determination from norite of the main The Rosetown pluton (mapped in detail by body east of the Hudson River, the Cortlandt Shuart, 1969) has long been regarded as a west- Complex is thought to be of Late Ordovician- ward extension of the Cortlandt Complex Early Silurian age (Long and Kulp, 1962). (Kemp, 1888). The Rosetown (Fig. 3) is an Because the Cortlandt Complex is recognized elongate intrusive, the long axis of which trends to cut fossiliferous rocks of probable Middle N. 40° E., roughly parallel to the Ramapo Ordovician age (Ratcliffe and Knowles, 1969, border fault that lies immediately east of it. p. 53), the biotite age cited above is regarded as The pluton actually is composite, having an reliable. older funnel-shaped mass of hornblendite and Igneous flow-oriented hornblende from the hornblende diorite in the southeast (Shuart, south pluton at Rosetown yielded a K-Ar age 1969; Ratcliffe and Shuart, 1970). of 810 + 8 m.y., suggesting that the older part This older pluton and the surrounding Pre- of the pluton was Precambrian (Ratcliffe and cambrian gneisses have been fractured along a Shuart, 1970). Subsequent isotopic studies shear zone that trends N. 40° E., coplanar with (Ratcliffe and R. G. Senechal, in prep.), how- the nearby Ramapo border fault at Tomkins ever, suggest that the hornblende ages may be Cove. The fracture zone, cutting the older part excessively old, because hornblende from the of the pluton, coincides with the northerly diorite that includes a xenolith of Annsville projection of the Ramapo fault from a point 5 Phyllite (Middle Ordovician) yielded an mi north of the New Jersey-New York state anomalous 580 + 20 m.y. K-Ar age. Although line (Fig. 1). This fracture zone separates the data are not complete yet, these pre- blocks of Precambrian hornblende and biotite liminary results suggest that the older part of granite gneiss having different foliation trends. the Rosetown may not be as old as the 810- Foliation in the gneiss of the eastern block m.y. age reported by Ratcliffe and Shuart trends N. 40° E. parallel to Paleozoic structures (1970). It should be noted that because of the in the adjacent Manhattan Prong, a direction lack of fossils in the Annsville xenolith, the nearly at right angles to the strong northwest geologic relationships are not firm enough to trends in the western block (Fig. 3). Dioritic prove that the hornblende ages are erroneous. magma that formed the younger and major part Unlike the diorites within the fracture zone, of the Rosetown pluton (Figs. 3 and 4) rose the rocks of the southern intrusive center do to the surface along this fracture (Ratcliffe and not contain xenoliths of Paleozoic rocks and Shuart, 1970). Blocks of Precambrian gneiss, yield old isotopic ages, thus suggesting a rocks of the older pluton, Wappinger Lime- Precambrian age. The age of the fracturing stone, and Annsville Phyllite are found as related to the intrusion of the northern diorites xenoliths, thus requiring a post-Middle Or- that are typical Cortlandt rocks on geologic

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 IGNEOUS INTRUSIVES IN FAULT ZONES 135

grounds is expected to be post-Middle Or- town pluton crop out along the fracture zone dovician, a time consistent with the suspected on the west side of the Canopus Valley in the Late Ordovician-Early Silurian age of the main Peekskill 7-1/2 min. quadrangle (Fig. 4). The mass of the Cortlandt. Canopus pluton has intrusive relationships This argument and the observation that the locally with the bordering Precambrian gneisses. younger diorites have been intruded into an However, the eastern and western contacts of openwork fracture system suggest that normal the igneous rocks are extensively mylonitized faulting was active here in Late Ordovician- and faulted along N. 30° to N. 40° right-lateral Early Silurian time (Ratcliffe and Shuart, transcurrent faults. The igneous rocks of the 1970). The parallelism of this fracture zone with Canopus pluton were interpreted by Fisher the Ramapo fault indicates that a deep-seated and others (1962) as Cortlandt equivalents. fracture system was active here in the lower However, the Canopus pluton contains felsic Paleozoic, presumably at a time postdating rocks, monzonites, quartz monzonites, and the uplift that produced the unconformity at granodiorites that are rare or absent in the Canopus Hollow. Rosetown pluton. Moreover, biotite separated In the Popolopen Lake area northwest of from the Canopus diorite yielded a K-Ar the Rosetown Complex, Dodd (1965) mapped minimum age of 700 ± 28 m.y. (Ratcliffe and a series of north- to northeast-trending faults R. G. Senechal, in prep.). Steeply plunging (Fig. 2). On the basis of the comparative flow structures present in the igneous rocks stratigraphy in different blocks, he proposed a and the filling of subsidiary right-lateral shears normal-fault component for most of these and by late stage differentiates suggest that right- assigned them a Triassic age. However, abun- lateral strike-slip faulting was active at the time dant lamprophyric dikes, compositionally and of the intrusion in late Precambrian time. Thus, geographically closely tied to the diorites of some of the fracturing along the west side of the Rosetown pluton, fill northeast-trending Canopus Hollow may be late Precambrian in fractures in the gneiss. Characteristically, these age. dikes postdate the northeast-trending faults The marble of the Canopus Valley is thought (Dodd, 1965). If the dikes are Rosetown-Cort- to be "Trentonian" in age on the basis of landt equivalents, as the field and petrographic crinoid remains reported by Bucher (1957) data suggest (Ratcliffe, 1968b; Shuart, 1969), from a road cut on Route 9 (Fig. 5, Loc. 1). this faulting may also have been of Ordovician The recrystallization and mylonitization of this or older age. Thus, intrusive relationships of the marble in the Canopus Valley may have been diorite cited above and the lamprophyres the result of intense shearing within the fault indicate an age older than Triassic for some of zone during a second period of transcurrent the northeast-trending faults. Triassic reactiva- faulting that was accompained by intrusion of tion of older faults, however, cannot be ruled mafic and pegmatitic dikes noted by Berkey out for some of the faults shown on Figure 2, (1907, p. 369) and Ohan (1964). Locally the so that Dodd's age assignment may in part be mafic dikes are boudinaged and brecciated in correct. areas where the marble is cataclastkally de- Perhaps these faults, too, have undergone formed (Ohan, 1964). repeated movement during a long period of Faulting and cataclasis of the Annsville- time. The Timp Pass fault (Fig. 2), for example, Wappinger sequence on the eastern side of the is thought to be a Precambrian reverse fault at valley along the projection of the Ramapo fault its northern end (Lowe, 1950) but a Triassic from the Tomkins Cove area indicate post- normal fault along its southern trace (Dodd, Ordovician fracturing. Although this fault has 1965). Both authors could be correct if the been interpreted as a normal fault (Berkey and fault were reactivated as a hinged fault showing Rice, 1919), the intense cataclasis and the strong increasing net slip to the southwest, a mech- development of right-lateral minor folds having anism proposed for the Ramapo fault itself steeply plunging axes suggest transcurrent (see Van Houten, 1969, Fig. 11). Rejuvenation faulting. The offset of the Annsville Phyllite of old northeast-trending faults during the in a left-lateral sense shown on the map does Triassic has also been suggested by Offield not agree with the fabric within the mylonite (1967, p. 67-69) for the Goshen-Greenwood zone; thus this offset is probably apparent. The Lake area of the Hudson Highlands (Fig. 1, age of post-Precambrian faulting, accompanied Loc. 9). by extensive mylonitization and right-lateral Igneous rocks similar to those of the Rose- movement, can only be dated by the observa-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 136 N. RATCLIFFE—RAMAPO FAULT SYSTEM

tion that pegmatitle dikes in the mylonite zones Ramapo fault is hinged at a point north of and in the marble of the Canopus belt have not Tomkins Cove with increasingly greater net been deformed (Ohan, 1964). The youngest slip to the southwest (see Van Houten, 1969, "granitic" rock recognized in this area is the Fig. 11). Likewise, the lack of cataclasis in the Peekskill Granite, which suggests that even the Triassic rocks bordering the Ramapo fault late fracturing was pre-Peekskill. The age of (Fig. 3) and the severe mylonitization in the the Peekskill Granite is not known with footwall block suggest that post-Brunswick certainty, although concordant K-Ar and Rb-Sr movement was rather slight. muscovite ages of 355 + 15 m.y. (Long and The abundant igneous activity in the Newark Kulp, 1962) indicate a Devonian or older age. Basin indicates the presence of a deep crustal Xenoliths of Manhattan Schist and Cortlandt fracture zone beneath the point where the ultramafic rocks in the Peekskill reported by igneous rocks were intruded or where they Espeho and Mattson (1970) indicate that the reached the surface. The relationship of the granite is probably not older than Late Or- large intrusive Palisades phase to the east and dovician Early Silurian. the Watchung flows to the west of the basin This line of reasoning points to a Precam- suggests that the border fault at depth served brian to Early Paleozoic age for the important as an avenue of magma ascent, although fracturing in the Canopus Valley similar to that magma apparently never reached the surface adduced for the Rosetown area. The evidence along this fault. presented in these first three sections argues strongly for repeated pre-Triassic fault move- RECENT SEISMIC ACTIVITY ments in N. 30° to 40° E. trends at the north Page and others (1968) recently presented end of the Ramapo fault in the late Precam- evidence of seismic activity near the Ramapo brian, the Early Ordovician, and again in the fault. Four earthquake epicenters recorded Late Ordovician. since 1962 have been located on the southeastern side of the fault from Peekskill south to REPEATED FAULTING DURING LATE Pompton Lakes. Richter values reported fell TRIASSIC TIME: THE BORDER in the 1.0 to 2.0 range. Although accurate CONGLOMERATES hypocenter determinations were not available, Coarse Triassic border conglomerates locally it was reported that the four earthquakes flank the northwest edge of the Newark Basin could have originated at moderate depths (5 and belong to the Hammer Creek Formation to 10 km) on a southeast-dipping fault whose (Glaesser, 1966; Van Houten, 1969). Clasts of trace could be consistent with that of the Cambro-Ordovician dolostones and quartzites, Ramapo fault. Page and others (1968) suggest Siluro-Devonian rocks, and locally Precambrian that fault movements are taking place at the gneisses make up the bulk of the clasts in these present time along this old fracture, but they deposits (Savage, 1968; Van Houten, 1969). could not judge whether the crustal adjust- The distribution and composition of these ments were in response to tectonic stresses or fanglomerates indicate local derivation from were the result of unloading of glacial ice. In the uplifted Paleozoic cover of the Hudson the light of the long history of faulting in this Highlands (Carlston, 1946). The high relief area, from the late Precambrian to the Triassic, necessary to form such conglomerates suggests it may be reasonable to suggest that these repeated uplift of the Highlands, whereas the recent movements are also of tectonic origin. change in the character of the deposits from Minor seismic activity has recently been place to place may indicate differential uplift. recorded for the Lake Hopatcong area of north- In post-Brunswick time, faulting downdropped ern New Jersey (Fig. 1, Loc. 8) located in the the Triassic rocks along the present border fault. Dover and Stanhope quadrangles, 13 mi north- Near Stony Point, the Hammer Creek west of the Ramapo fault (Sbar and others, Formation evidently rests unconformably on 1970). The earthquake activity is interpreted the Paleozoic rocks (Inwood and Manhattan to have taken place along the inferred south- Schist) (Ratcliffe, 1968a, p. 200), thus exposing west extension of a N. 40° E.-trending normal the floor of the Triassic basin. The post-Bruns- fault that forms the northwest side of a graben wick faulting need not have produced large that downdrops Devonian rocks (Buddington displacements of the Triassic at the border and Baker, 1961). Omeld (1967, p. 67-68) fault near Tomkins Cove for the reasons suggested that this fault in New York State previously mentioned. This implies that the was an Ordovician reverse fault rej uvenated as a

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 RELATIONSHIP OF FAULT ACTIVITY TO PROPOSED ALLOCHTHON 137

Triassic normal fault. A first motion study of as the Middle Ordovician Martinsburg Forma- the earthquakes indicates normal faulting for tion have been overturned and moved west- the Lake Hopatcong tremors, and Sbar and ward as the cover of a large nappe cored with others (1970) suggest that this may be a case of Precambrian gneiss. Both Isachsen (1964) and reactivation of a Triassic fault. Drake (1969) favor the interpretation of an entirely rootless (allochthonous) Reading RELATIONSHIP OF FAULT ACTIVITY Prong. Workers such as Offield (1967, p. 68-70) TO THE PROPOSED READING PRONG and Smith (1969, p. 44-45) believe that the ALLOCHTHON overthrusting is quite limited along the The arguments for the tectonic longevity western edge of the Reading Prong in New of the Ramapo fracture system presented here York and adjacent New Jersey. Thus the have an important bearing on a major un- parautochthonous or allochthonous nature of resolved tectonic problem in the Hudson the Reading Prong is disputed at the present Highlands. Isachsen (1964) suggested that the time. entire Reading Prong was a large allochthon The structural arguments presented here emplaced in pre-Cortlandt time (Late Ordovi- and by Ratcliffe and Shuart (1970) suggest cian-Early Silurian) from a more southeasterly that the Ramapo fracture system has been source. The pre-Cortlandt age was required by operative in the rocks of the Hudson Highlands the fact that rocks of the Rosetown pluton since late Precambrian time. Movement in pre- attributed to the Cortlandt Complex intrude Middle Ordovician time is dated by the Middle the supposed allochthon (Isachsen, 1964, p. Ordovician unconformity at Annsville. Facies 825) and autochthon. The radiometric and in the limestone at the base of the Manhattan geologic investigations cited here and in Formation or Annsville Phyllite are not offset recent work (Shuart, 1969; Ratcliffe and by the supposed trailing edge of the proposed Shuart, 1970; Ratcliife and R. G. Senechal, in allochthon. Moreover, detrital debris in the prep.) equate some of the Rosetown rocks with rocks above the Middle Ordovician uncon- the Cortlandt and thus support in part formity coarsens and becomes more abundant Isachsen's interpretation regarding timing of as the Ramapo fault is approached, which the possible movement. However, the Pre- indicates uplift at the site of the present High- cambrian age of igneous activity at Canopus lands block in pre-Middle Ordovician time. and of the older part of the Rosetown pluton This evidence suggests strongly that the south- is not in agreement with Isachsen's (1964) age ern edge of the Hudson Highlands could not assignment. have been emplaced in Late Ordovician time as The idea of strong westward movement of proposed by Isachsen (1964). Precambrian massifs in western New England A stratigraphic argument based on a regional and the Reading Prong has gained support study of the Lower Cambrian quartzite from recent studies. In the Berkshire Massif of sequence seems pertinent here. Within the western Massachusetts (S. A. Norton, 1969; Reading Prong, the Lower Cambrian strata, Ratcliffe, 1969) strong remobilization of the having sedimentary (unconformable) contacts gneissic basement took place in the middle with the Precambrian rocks, are predominantly Paleozoic under metamorphic conditions rang- quartz-rich facies variously known as the ing from sillimanite to garnet grade. Large- Hardyston or Poughquag Quartzite. Locally scale nappes having Precambrian cores and silica- or hematite-cemented conglomerates involving rocks as young as Middle Ordovician form the base of the formation (Aaron, 1969; show surface displacement westward of at least Balk, 1936; Ohan, 1964; Ratcliffe, unpub. 4 mi and perhaps as much as 8 mi (Ratcliffe, data). This quartz-rich lithology is common in 1969; Ratcliffe and D. S. Harwood, unpub. the western outcrop belts of the Cheshire or data). The Berkshire Massif is regarded by the Poughquag quartzites in eastern New York author as parautochthonous in the sense that and western Massachusetts. More easterly it is only partially detached and traceable to a exposures of these formations or their equiv- root zone along its eastern edge (for a discussion alents, the Dalton and Hoosac Formations (in of the terms parautochthonous and alloch- western Massachusetts and Connecticut) (S. thonous nappes, see Hills, 1963, p. 250). A. Norton, 1969; Ratcliffe, 1969) and the The detailed work of Drake (1969) and his Lowerre Quartzite-in most of the Manhattan co-workers in the Pennsylvania part of the Prong-(M. F. Norton, 1959; Hall, 1968), differ Reading Prong indicates that rocks as young markedly from this western facies. In the cen-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 138 N. RATCLIFFE—RAMAPO FAULT SYSTEM

tral and eastern part of the Manhattan Prong, movement and igneous activity is indicated for feldspar-rich granulites, micaceous feldspathic the Ramapo fault system, spanning the time quartzites, and quartz-rich schistose rocks from the initial pre-Trenton disturbance to lithically similar to the Dalton Formation of post-Brunswick Late Triassic faulting. If the western Massachusetts represent the bulk of fracturing along the west side of the Canopus the lower Cambrian clastic sequence (Leo Hall, Valley is considered part of the same fracture 1970, personal commun.; Ratcliffe, unpub. system, as seems likely, activity in this zone data). The lack of the distinctive eastern sed- may have extended back into the Precambrian. imentary facies (Dalton-Hoosac) in the Read- Although no field data indicate that any post- ing Prong suggests strongly that its northern Mesozoic faulting took place, the seismic end is not allochthonous, because the eastern activity recently recorded by Page and others impure facies of the quartzite sequence, the (1968) suggests reactivation is currently under- Lowerre, crops out in the area of the proposed way and extends the possible tectonic life span root zone. to 600 or 700 m.y. Thus the structural and stratigraphic argu- In addition to its remarkable longevity, the ments presented here suggest strongly that the Ramapo fault system also marks the boundary Hudson Highlands is rooted along its south- between structurally different terranes, separat- eastern edge in New York State. However, ing the Reading Prong from both the Newark this interpretation does not necessarily require basin and the rocks of the Manhattan Prong. that the southern or western edge of the Read- This fracture system is known to have had deep ing Prong be rooted as well. A half klippe would crustal connections that allowed intrusion of equally well explain the geologic relationships basaltic magmas in late Precambrian, Ordovi- seen at the southern end of the Reading Prong cian, and again in Late Triassic time. All the (Smith, 1969, p. 45), whereas a parautoch- evidence suggests that the Ramapo fault thonous style of the Hudson Highlands, similar system and the Newark basin border fault in to that proposed here for the Berkshire general may be part of a fundamental crustal Massif, would satisfy some of the data in New fracture zone such as the North Pyrenean fault- York State. zone of southwestern France (de Sitter, 1959, p. 175-176). DISCUSSION A commonly raised and legitimate objection Many different lines of evidence indicate to the idea of old linear faults in orogenic belts that protracted faulting took place in the is the lack of subsequent deformation. That is, vicinity of the northwest border of the Newark why are the fault surfaces not folded? The basin from the late Precambrian to the Triassic answer may lie in the fact that the later regional and perhaps is continuing at present. Although structural grain is nearly colinear with the older it is not suggested that all movements were in faults. Therefore, the steeply dipping fault sur- response to the same tectonic forces, the de- faces are not likely to reflect the effects of later tailed analysis of the several localities discussed folding. Subsequent tectonic stresses may have points to repeated movement along a N. 30° to been taken up on these faults and thus would 40° E.-trending fracture system of regional reinforce the difference in structural style importance. Right-lateral transcurrent faulting between terranes adjacent to such a fault. not previously recognized seems to have been This interpretation seems likely in light of the of particular importance. Some fundamental evidence of significant right-lateral transcurrent crustal feature or structural grain predeter- faulting along the fracture system at Canopus mined the later movements. The geologic Hollow. This probably is the case at the record is not clear in this regard, but perhaps boundary between the Reading and Manhattan the original grain that developed in Precam- Prongs in the vicinity of the Ramapo fault. brian time is part of a major crustal fracture The effects of multiple Paleozoic meta- zone separating plates of lithosphere. Never- morphism and structural deformation so clear theless, this interesting geologic feature has in the Manhattan Prong adjacent to the border experienced a long tectonic life. Some mech- fault (Ratcliffe, 1968a, 1968b) and elsewhere anism allowing localization of tectonic activity in the Manhattan Prong (Hall, 1968; Langer has been operative here over a long period, a and Bowes, 1969; Bowes and Langer, 1969) are process sometimes referred to as tectonic difficult to detect in the Precambrian rocks of heredity. the Hudson Highlands west of the Hudson A 280-m.y. period of intermittent fault River (Lowe, 1950; Dodd, 1965; Dallmeyer,

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 REFERENCES CITED 139

1969). Evidently, Paleozoic structural over- scribed here are broadly consonant with the printing has been slight in rocks west of the plate-tectonic model for the origin of the north- Ramapo fault. ern Appalachians proposed by Bird and The acceptance of this explanation for the Dewey (1970). According to their scheme, the Ramapo fault system calls for a degree of northern Appalachians formed as a result of tectonic permanence not commonly dealt with late Precambrian to Ordovician expansion and in current global tectonic models. It seems that Ordovician to Devonian contraction of the the faults that produced the Newark basin in proto-Atlantic ocean basin. Items (2) and (3) Late Triassic time were not initiated by above could have formed during the extensional Mesozoic tectonics alone, as proposed by Van phase, and (4) during the contractional phase. Houten (1969, p. 330), but rather were merely Perhaps the intense right-lateral deformation on reactivated at that time. The initial formation the Ramapo fracture system (phase 4) is of this fracture system evidently extends at related to the disappearance of the old proto- least as far back into geologic past as the Middle Atlantic ocean rise beneath the eastern edge of Ordovician, and probably back to the Pre- the North American continent (Bird and cambrian. Dewey, 1970, Fig. 9) in much the same way The role that the Ramapo fracture system that the San Andreas fracture may be related played in Taconic and later orogenesis is un- to the present position of the East Pacific rise. known at present, but it seems certain that it The answers to these and other questions was a major zone of crustal weakness and move- must await the results of detailed field and ment during the formation of the Appalachian isotopic investigations. Although the inter- orogen. Such major tectonic features as the pretations in this paper are necessarily specula- bend in the trend of the Appalachian Mountain tive, the writer hopes that the ideas presented belt at the Hudson River, the extent of the will suggest fruitful areas of investigation to Taconic orthotectonic belt, and the westward other field geologists. limits of Acadian reheating may somehow be ACKNOWLEDGMENTS related to this fault system. The detailed movements on the Ramapo This investigation was partially funded by fracture system are imperfectly understood the City College Research Foundation. John at present. However, the geologic relationships M. Bird, Leo M. Hall, Kurt E. Lowe, Ely described here suggest the following tectonic Mencher, Henry W. Dodge, Jr., John Sanders, history: (1) late Precambrian, right-lateral and E-an Zen kindly read the manuscript. faulting concomitant with intrusion of the Their helpful suggestions are greatly appre- Canopus pluton, followed by post-tectonic ciated. I am indebted to Ronald G. Senechal for intrusion of the older part of the Rosetown allowing me to mention results of a radiometric pluton; (2) Early Ordovician block faulting study now in progress. with oceanward block moved down, erosion on the western block, producing the Middle REFERENCES CITED Ordovician unconformity at Annsville; (3) Aaron, J. M. Petrology and origin of the Hardyston renewed normal(?) faulting in Rosetown area Quartzite (Lower Cambrian) in eastern Penn- and intrusion of diorites of younger part of the sylvania and western New Jersey: in Geology Rosetown pluton in the Middle to Late Or- of selected areas in New Jersey and eastern Pennsylvania and guidebook of excursions dovician; (4) intense right-lateral transcurrent Seymour Subitzky, ed.), Rutgers Univ. faulting in post-Middle Ordovician to pre- Press, New Brunswick, p. 21-33, 1969. Middle Devonian time, producing cataclasis of Balk, Robert. Structural and petrologic studies in Precambrian to Middle Ordovician rocks along Dutchess County, New York; Part I. Geo- fracture zone. This movement may have been logic structure of sedimentary rocks: Geol. Soc. in part synchronous with Taconic deformation Amer., Bull., Vol. 47, p. 685-774, 1936. and metamorphism in the adjacent Manhattan Bayley, W. S.; Salisbury, R. D.; and Kummell, Prong; (5) Late Triassic rejuvenation of the H. B. Description of the Raritan quadrangle, old fracture system as a system of hinged New Jersey: U.S. Geol. Surv., Geol. Atlas, Folio 191, 32 p., 1914. normal faults, producing the Newark deposi- Berkey, C. P. Structural and stratigraphic features tional basin. Block faulting continued during of the basal gneisses of the Highlands: N.Y. the Late Triassic and possibly into the Jurassic. State Mus. and Sci. Service Bull., Vol. 607, Except for the suspected late Precambrian 377 p., 1907. right-lateral movement, the movements de- Berkey, C. P.; and Rice, Marion. Geology of the

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 140 N. RATCLIFFE—RAMAPO FAULT SYSTEM

West Point quadrangle, New York: N.Y. State tional environments of Triassic sediments in Mus. and Sci. Service Bull., No. 225-226, the Newark-Gettysburg basin: Pennsylvania 152 p., 1919. Geol. Surv., Bull., 4th Ser., Vo. G-43, 168 p., Bird, J. M.; and Dewey, J. F. Lithosphere plate- 1966. continental margin tectonics and the evolution Gordon, C. E. Geology of the Poughkeepsie of the Appalachian orogen: Geol. Soc. Amer., quadrangle: N.Y. State Mus. and Sci. Service Bull., Vol. 81, p. 1031-1060, 1970. Bull., No. 148, 121 p., 1911. Bowes, D. R.; and Langer, A. M. Polyphase Hall, L. M. Times of origin and deformation of deformation in the schists of the Bronx, New bedrock in the Manhattan Prong: p. 117-127 York City: in Symposium of the New York in Studies of Appalachian geology: Northern City Group of Formations (E. Alexandrov, and maritime (E-an Zen, W. S. White, J. B. ed.), Geol. Bull, of Queens College Press, p. Hadley, and J. B. Thompson, Jr., eds.), Inter- 17-32, 1969. science, New York, 475 p., 1968. Bucher, W. H. Taconic klippe—A stratigraphic- Hills, S. E. Elements of structural geology: John structural problem: Geol. Soc. Amer., Bull., Wiley & Sons, Inc., New York, 483 p., 1963. Vol. 68, p. 657-674, 1957. Isachsen, Y. W. Extent and configuration of the Buddington, A. F.; and Baker, D. R. Geological Precambrian in the northeastern United map of the geology of the Franklin and part States: N.Y. Acad. Sci. Trans., Ser. 2, Vol. 26, of the Hamburg quadrangles, New Jersey: p. 812-829, 1964. U.S. Geol. Surv., Misc. Geol. Inv. Map 1-346, Kemp, J. F. On the Rosetown extension of the 1961. Cortlandt Series: Amer. J. Sci., 3rd Ser., Vol. Carlston, C. W. Appalachian drainage on the High- 36, p. 247-253, 1888. land border sediments of the Newark series: Langer, A. M.; and Bowes, D. R. Polyphase Geol. Soc. Amer., Bull., v. 57, p. 997-1032, deformation in the Manhattan Formation of 1946. Manhattan Island, New York City: in Igneous Dallmeyer, R. D. Multiple deformation of Precam- and metamorphic geology (L. H. Larsen, ed.), brian gneisses at Bear Mountain, New York Geol. Soc. Amer. Mem. 115, p. 361-377, 1969. [abstr.l: Geol. Soc. Amer., Abstr. 1969, Part 1, Long, L. L.; and Kulp, J. L. Isotopic age study of (Northeast. Sect.) p. 10, 1969. the metamorphic history of the Manhattan and de Sitter, L. U. Structural geology: McGraw-Hill Reading Prongs: Geol. Soc. Amer., Bull., Vol. Book Co., Inc., New York, 552 p., 1959. 73, p. 969-996, 1962. Dodd, R. T., Jr. Precambrian geology of the Popo- Lowe, K. E. Storm King Granite at Bear Moun- lopen Lake quadrangle, southeastern New tain, New York: Geol. Soc. Amer., Bull., York: N.Y. State Mus. and Sci. Service Map Vol. 61, p. 137-190, 1950. and Chart Ser. No. 6, 39 p., 1965. Norton, M. F. Stratigraphic position of the Drake, A. A., Jr. Precambrian and lower Paleozoic Lowerre Quartzite: N.Y. Acad. Sci. Ann., Vol. geology of the Delaware Valley, New Jersey- 80, p. 1148-1158, 1959. Pennsylvania: in Geology of selected areas in Norton, S. A. Unconformities at the northern end New Jersey and eastern Pennsylvania and of the Berkshire Highlands: p. 21-1 to 21-20 guidebook of excursions (Seymour Subitzky, in Guidebook for field trips in New York, ed.), Rutgers Univ. Press, New Brunswick, Massachusetts, and Vermont (J. M. Bird, ed.), p. 51-131, 1969. New England Intercoll. Geol. Cong., 61st Drake, A. A., Jr.; McLaughlin, D. B.; and Davis, Annu. Meet., 1969. R. E. Geology of the Frenchtown quadrangle, Offield, T. W. Bedrock geology of the Goshen- New Jersey-Pennsylvania: U.S. Geol. Surv. Greenwood Lake area, New York: N.Y. Geol. Quad. Map G.Q.-133, 1961. State Mus. and Sci. Service Map and Chart Espeho, Anibal; and Mattson, P. M. Textural Ser. No. 9, 78 p., 1967. evidence for the synorogenic intrusion of the Ohan, A. A. Geology of the southwestern section Peekskill Granite [abstr.]: Geol. Soc. Amer., of the Lower Paleozoic inlier, Peekskill, New Abstr. 1970, Part 1, (Northeast. Sect.) p. 10 York: M.S. thesis, New York Univ., New (incorrectly printed as written by Espheo York, 133 p., 1964. Anibal and P. M. Mattson), 1970. Page, R. A.; Molnar, P. H.; and Oliver, Jack. Fisher, D. W.; Isachsen, Y. W.; Rickard, L. V.; Seismicity in the vicinity of the Ramapo Broughton, J. G.; and Offield, T. W., fault, New Jersey-New York: Seismol. Soc. (comp, and ed.) Geologic map of New York, Amer., Bull., Vol. 58, No. 2, p. 681-687, 1968. 1961: N. Y. State Mus. and Sci. Service Paige, Sidney. Cambro-Ordovician age of the "In- Map and Chart Ser. No. 5, 5 sheets (scale, wood" Limestone and "Manhattan" Schist 1:250,000), 1961. near Peekskill, New York: Geol. Soc. Amer., Frimpter, M. H. Geology of the Thiells quadrangle, Bull., Vol. 67, p. 391-394, 1956. New York: Ph.D. thesis, Boston Univ., Bos- Ratcliffe, N. M. Stratigraphic and structural rela- ton, Massachusetts, 142 p., 1967. tions along the western border of the Cortlandt Glaesser, J. D. Provenance, dispersal and deposi- Intrusives: p. 197-220 in Guidebook to field

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021 REFERENCES CITED 141

excursions (R. M. Finks, ed.), N.Y. State Geol. Sbar, M. L.; Rynn, J.M.W.; Gumper, F. J.; and Ass., 40th Annu. Meet., Flushing, New York, Lahr, J. C. An earthquake sequence and focal 253 p., 1968a. mechanism solution, Lake Hopatcong, north- Ratcliffe, N. M. Contact relations of the Cortlandt ern New Jersey: Seismol. Soc. Amer., Bull., Complex at Stony Point, New York, and their Vol. 60, No. 4, p. 1231-1234, 1970. regional implications: Geol. Soc. Amer., Shuart, S. R. Petrology of the Rosetown igneous Bull., Vol. 79, p. 777-786, 1968b. complex: M.S. thesis, City College of New Ratcliffe, N. M. Structural and stratigraphic rela- York, New York, 82 p., 1969. tions along the Precambrian front in south- Smith, B. L. The Precambrian geology of the western Massachusetts: p. 1-1 to 1-21 in central and northeastern parts of the New Guidebook for field trips in New York, Massa- Jersey Highlands: in Geology of selected areas chusetts, and Vermont (J. M. Bird, ed.), New in New Jersey and eastern Pennsylvania and England Intercoll. Geol. Cong., 61st Annu. guidebook of excursions (Seymour Subitzky, Meet., 1969. ed.), Rutgers Univ. Press, New Brunswick, Ratcliffe, N. M.; and Knowles, R. R. Stratigraphic p. 35-50, 1969. relations along the western edge of the Cort- Thompson, J. B,, Jr.; and Shumaker, R. C. landt Intrusives and their bearing on the Bedrock geology of the Pawlet quadrangle, Inwood-Manhattan problem: p. 49-64 in Vermont: Vermont Geol. Surv., Bull., No. Symposium of the New York City Group of 30, 98 p., 1967. Formations (E. Alexandrov, ed.), Geol. Bull, Van Houten, F. B. Late Triassic Newark Group, of Queens College Press, 83 p., 1969. north central New Jersey and adjacent Penn- Ratcliffe, N. M.; and Shuart, S. G, The Rosetown sylvania and New York: in Geology of selected complex and its bearing on the antiquity of areas in New Jersey and eastern Pennsylvania the Ramapo fault system [abstr.]: Geol. Soc. and guidebook of excursions (Seymour Su- Amer., Abstr. 1970, Part 1, (Northeast. Sect.) bitzky, ed.), Rutgers Univ. Press, New Bruns- p. 32-33, 1970. wick, p. 314-347, 1969. Sanders, J. E. Structural history of Triassic Zen, E-an. Nature of the Ordovician orogeny in the rocks of the Connecticut Valley belt and its Taconic area: p. 129-139 in Studies of Appala- regional implications: N.Y. Acad. Sci. Trans., chian geology: northern and maritime (E-an Ser. 2, Vol. 23, p. 119-132, 1960. Zen, W. S. White, J. B. Hadley, and J. B. Sanders, J. E. Late Triassic tectonic history of Thompson, Jr., ed.), Interscience, New York, northeastern United States: Amer. J. Sci., Vol. 475 p., 1968. 261, p. 501-524, 1963. Savage, E. L. The Triassic rocks of the northern Newark basin: p. 49-68 in Guidebook to field excursions (R. M. Finks, ed.), N.Y. State MANUSCRIPT RECEIVED BY THE SOCIETY FEBRUARY Geol. Ass., 40th Annu. Meet., Flushing, N.Y., 24, 1970 253 p., 1968. REVISED MANUSCRIPT RECEIVED JUNE 4, 1970

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/82/1/125/3428212/i0016-7606-82-1-125.pdf by guest on 27 September 2021