Spatial, temporal, mineralogical, and compositional variations in Mesozoic kimberlitic magmatism in New York State

Lithos 212–215 (2015) 298–310

Contents lists available at ScienceDirect

Lithos journal homepage: www.elsevier.com/locate/lithos

Spatial, temporal, mineralogical, and compositional variations in Mesozoic kimberlitic magmatism in New York State David G. Bailey a,⁎, Marian V. Lupulescu b a b

Geosciences Department, Hamilton College, Clinton, NY, USA New York State Museum, Albany, NY, USA

a r t i c l e

i n f o

Article history: Received 25 July 2014 Accepted 20 November 2014 Available online 27 November 2014 Keywords: New York Mesozoic magmatism Kimberlite Mineralogy Geochemistry Continental rifting

a b s t r a c t Mesozoic kimberlitic magmatism was geographically widespread across central New York State, and nearly 90 distinct intrusions have been discovered since the first “serpentinite body” was described over 175 years ago. Most of the intrusions are narrow (b30 cm wide), near vertical, north–south oriented dikes, although three larger, irregular diatremes are also known. Previous studies assumed that all of the intrusions were genetically and temporally related, and often examined only a small sub-set of the intrusions. By combining modern samples with historic samples in the collections of the New York State Museum and Hamilton College, we were able to obtain detailed mineralogical and geochemical data on samples from 27 distinct intrusions. The intrusions can be divided into four distinct groups on the basis of both mineralogy and geochemistry, and previously published radiometric age dates suggest that these four groups may also have distinct emplacement ages. Group A intrusions are exposed on the western margin of Cayuga Lake near Ithaca, and are characterized by olivine and phlogopite macrocrysts in a serpentine and phlogopite-rich matrix. These intrusions are relatively Ti-rich and contain abundant perovskite grains in the groundmass that yielded U–Pb crystallization ages of ~ 146 Ma (Heaman and Kjarsgaard, 2000). Group B intrusions are exposed over a relatively large area surrounding Ithaca, and are characterized by having a diverse macrocryst assemblage that includes pyrope, diopside, and spinel in addition to olivine and phlogopite. These intrusions are the most incompatible and REE enriched, and are chemically similar to the Kirkland Lake kimberlites in eastern Ontario. Intrusion ages for this group cluster between 125 and 110 Ma. Group C intrusions are all found within the city of Syracuse, and are similar to the Group B intrusions in both mineralogy and chemistry. They appear to be somewhat older, with intrusion ages of 135–125 Ma. Finally, Group D intrusions are geographically distant from the other three groups, being exposed in East Canada Creek nearly 100 km east of the Syracuse dikes. They are characterized mineralogically by abundant olivine and sparse, but large, phlogopite macrocrysts, and chemically by having the lowest incompatible element and rare earth element (REE) concentrations, and the highest 87Sr/86Sr ratios. Intrusion ages for these dikes are poorly constrained, but appear to be contemporaneous with, or slightly older than, the ~146 Ma Group A intrusions. All of the kimberlitic intrusions in central New York State have initial Sr and Nd isotope ratios close to bulk earth. This fact, combined with the observed macrocryst assemblages and incompatible trace element ratios, indicates that these magmas were derived primarily from an asthenospheric, garnet lherzolite source. Episodic intrusion of small volume, volatile-rich kimberlitic magmas into the Paleozoic sedimentary platform rocks of central New York appears to have occurred along ancient crustal structures that were reactivated by the far field stresses related to the opening of the North Atlantic Ocean. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Magmatism was widespread in northeastern North America during the Mesozoic, and dominated by three distinct episodes and styles of activity (Fig. 1): 1) extensive tholeiitic basalt flows and sills of the

⁎ Corresponding author. Tel.: +1 315 859 4142. E-mail addresses: [email protected] (D.G. Bailey), [email protected] (M.V. Lupulescu).

http://dx.doi.org/10.1016/j.lithos.2014.11.022 0024-4937/© 2014 Elsevier B.V. All rights reserved.

Circum-Atlantic Magmatic Province (CAMP) (~ 200 Ma) (Kontak, 2008; Marzoli et al., 2011; Murphy et al., 2011); 2) shallow, relatively small, alkalic intrusions of the Monteregian Hills and larger New England–Quebec Province (~ 130–110 Ma) (Eby, 1987; McHone and Butler, 1984); and 3) larger, dominantly granitic to syenitic intrusions of the White Mountain Magma Series (200–155 and 130–110 Ma) (Eby et al., 1992; McHone, 1996). Small volume, ultramafic and ultrapotassic intrusions are also widespread in the region, but they have not been as extensively studied, and their petrogenesis is poorly understood. These intrusions provide the only direct evidence of the

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

nature of the mantle underlying eastern North America, and are an important part of the tectonic and magmatic history of the region. The rocks of this study lie within a broad belt of mica-bearing ultramafic intrusions that runs along the western flank of the Appalachian Mountains from Alabama north into Ontario and Quebec. While these rocks exhibit a wide range of mineral assemblages and textures, all have been referred to as kimberlites by previous researchers (e.g. Basu and Rubury, 1980; Bikerman, 1997; Bolivar, 1976; Meyer, 1976). Currently, there are two theories that have been put forth to explain the origin of the kimberlitic rocks in New York State: 1) they are part of a chain of small, alkaline intrusions in eastern North America related to passage of the North American plate over the Great Meteor hot spot (Heaman and Kjarsgaard, 2000); or 2) they are part of a belt of kimberlitic intrusions along the western flanks of the Appalachian Mountains that were intruded along old structures that were reactivated by crustal extension related to rifting and opening of the Atlantic Basin (Parrish and Lavin, 1982). This paper presents the results of the first comprehensive study of the kimberlitic rocks of New York State. The goals of the study are: 1) to identify and document mineralogical and chemical variations within and between intrusions, 2) to identify any spatial or temporal patterns in the distribution of these rocks, and 3) to constrain the conditions and tectonic setting under which these rocks formed. 1.1. Kimberlitic rocks of New York 1.1.1. Geographic distribution Approximately 90 kimberlitic dikes and small diatremes have been found in New York State. Detailed information on the number of dikes in New York, their location, size, and orientation are in Bailey and Lupulescu (2007), Foster (1970), Kay and Foster (1986), and Martens

299

(1924). The exact number of intrusions is difficult to ascertain for a number of reasons. First, the field descriptions provided by previous researchers vary in quality and specificity, and many of the earliest localities have been “lost” or buried. Second, many of the dikes are extremely small (widths of 1–3 cm) and they occur in irregular, anastomizing swarms; the criteria for identifying and counting individual dikes were not clearly described by any of the previous researchers. Finally, many of the dikes that have been given distinct names and described in the literature are undoubtedly along-strike exposures of the same dike (e.g. the Green Street, James Street, Foot Street, and Butternut Street dikes in downtown Syracuse). Most of the kimberlitic intrusions occur within an elongate NNE–SSW area between the cities of Ithaca and Syracuse. The Montgomery County and Ogdensburg dikes are the only occurrences outside of this area. The vast majority (nearly 80) of the known intrusions are found in the vicinity of Ithaca, primarily in the ravines that feed into Cayuga Lake. The irregular spatial distribution of known kimberlitic intrusions in New York is probably due, in part, to the variable effects of glacial erosion and deposition, with the greatest number of exposures in the deep, glacially cut gorges of west-central New York. It is highly probable that many other dikes are present throughout the state but are still undiscovered, being buried by thick blankets of glacial till and/or alluvial sediment. 1.1.2. Previous work The unusual serpentine-rich rocks of upstate New York State were first reported in 1837 in one of the earliest publications of the newly formed New York State Geological Survey (Vanuxem, 1837). This report, along with two subsequent reports (Vanuxem, 1839, 1842), contains the earliest known scientific description of kimberlitic rocks. Hunt (1858) published the first extended description and partial chemical analysis of these rocks, documenting their ultramafic character.

Fig. 1. Locations of Mesozoic igneous activity in northeastern North America (excluding CAMP basalts). Stars denote locations of known kimberlitic intrusions; hatched field denotes location of Monteregian Hills igneous activity; stippled field denotes location of White Mountains Magmatic Province. Line represents approximate age/location of proposed Great Meteor Hotspot track (Heaman and Kjarsgaard, 2000).

300

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

Fig. 2. a) Photograph of 1 m wide, coherent, and relatively resistant Group A dike (Glenwood Creek); b) Photograph of typical ~10 cm wide Group B (Taughannock Creek) dike with weathered margins.

Following Lewis's (1887) seminal report documenting the association of diamonds with “mica peridotites”, there was an explosion in the number of papers published on all types of peridotitic rocks, including those in New York. Between 1887 and 1909 over 25 articles on the “serpentine dikes” and “peridotites” of central New York were published. By the 1930s, dozens of additional dikes had been discovered, and even one small diatreme that was prospected for diamonds (Filmer, 1939). Zartman et al. (1967) published the first K/Ar and Rb/Sr ages on phlogopite grains extracted from two dikes. The K/Ar ages ranged from 145 to 493 Ma, and the Rb/Sr ages ranged from 118 to 146 Ma. Zartman et al. recognized that the early Paleozoic ages were clearly incompatible with the known stratigraphic relationships, and attributed the old K/Ar ages to excess radiogenic argon and/or the retention of argon by old xenocrystic phlogopite. They concluded that the New York intrusions were of Late Jurassic to Early Cretaceous age, and subsequent K/Ar studies by Basu et al. (1984) and Watson (1979) confirmed this, with most of the reported ages between 120 and 150 Ma. The most recent radiometric dates on these rocks are three high precision U–Pb ages obtained on groundmass perovskite grains that yielded Early Cretaceous ages (144.8 +/− 3.2, 146.7 +/− 2.4, and 147.5 +/− 3.0) (Heaman and Kjarsgaard, 2000). Two paleomagnetic studies revealed a complex history of emplacement times and temperatures, with normal and reversed pole positions, and a previously unrecognized late Jurassic–early Cretaceous virtual geomagnetic pole position at 58°N, 203°E (DeJournett and Schmidt, 1975; Van Fossen and Kent, 1993). Most of the recent work on these rocks has been focused on the mineralogy of the dikes in the Ithaca region, and on their disaggregated xenoliths (Kay, 1990; Kay and Foster, 1986; Kay et al., 1983; Snedden, 1983; Snedden and Kay, 1981a, b). While there have been numerous studies on individual dikes or small clusters of dikes, there have been no modern petrologic studies on the New York kimberlites as a group. The goal of this paper is to provide a comprehensive review and analysis of the mineralogy, geochemistry, and spatial and temporal distribution of these unusual rocks. 1.1.3. Nomenclature and classification For most igneous rocks, classification is straightforward, based primarily on modal mineralogy, rock texture, and/or rock chemistry

(Le Maitre et al., 2002). Unfortunately, because kimberlites are part of a broad clan of complexly related igneous rocks, simple classification criteria do not exist. Kimberlites are currently divided into two groups (Skinner, 1989; Smith et al., 1985). Group I kimberlites are the analogue of the rocks originally found and described at Kimberley, South Africa (the “basaltic kimberlites” of Wagner (1914)). Group II kimberlites (also commonly referred to as orangeites) are the equivalent of the micaceous kimberlites of the Orange Free State, South Africa (the “lamprophyric kimberlites” of Wagner (1914)). The two groups of kimberlites display subtle differences in their mineralogical and chemical compositions, although there are currently no simple criteria for distinguishing rocks from the two groups; classification currently relies on subtle differences in overall mineral assemblages and/or mineral compositional trends (Mitchell, 1995; Skinner, 1989; Smith et al., 1985; Tainton and Browning, 1991; Tappe et al., 2005; Woolley et al., 1996). In general, the kimberlitic rocks of New York State exhibit mineralogical and chemical features common to both groups, but none exhibit the diagnostic features of either group (e.g. monticellite is common in Group I kimberlites, but absent in Group II, and Zr silicates are common in Group II kimberlites, but very rare in Group I; unfortunately, neither phase was observed in any of our samples). Because of the lack of distinctive mineral assemblages, and the lack of extensive mineral compositional data, we refer to these rocks using the adjective “kimberlitic” to emphasize that they have not been formally classified.

1.1.4. Field characteristics The characteristic feature of most of the kimberlitic dikes in New York is their relatively consistent N–S strike and near vertical orientation. In all instances, except one, they cut the Paleozoic sedimentary rocks of central NY, which range in age from the Middle Ordovician Trenton Group in Montgomery County to the Upper Devonian Genesee Group in Tompkins County. The one instance of a dike cutting rocks of the Precambrian basement is at Big Nose along the Mohawk River; unfortunately, this dike is no longer exposed. Estimates of uplift and cooling rates for central and western NY suggest emplacement depths of 2–4 km for the current exposures (Miller and Duddy, 1989).

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

Most of the dikes are quite narrow, varying in width from 1 cm to slightly over 2 m (Fig. 2). The only exceptions to this are three large masses, possibly diatremes, exposed along Poyer Creek north of Ithaca, at Dewitt Reservoir east of Syracuse, and the large composite intrusion referred to as the Green Street dike in downtown Syracuse. Exposures also vary considerably in color, texture, and coherency. The large (~ 1.5 m wide) Williams Brook/Glenwood Creek dike near Ithaca is a very dark gray to black on fresh surfaces, and very hard and coherent (Fig. 3a). In contrast, some dikes are yellowish-green and composed largely of soft, unconsolidated clay and hydroxide minerals. Most intrusions are brown to green on fresh surfaces, and are inequigranular with large (typically 2–6 mm) round, dark green pseudomorphs of serpentine after olivine. In some, the dike margins are more resistant to weathering, whereas in others, they are more easily eroded (Fig. 3b). Crustal xenoliths are found in many of the intrusions, in particular the Dewitt Reservoir and Green Street intrusions in Syracuse. Angular xenoliths of the local Paleozoic country rock are the most common, however, rounded xenoliths of deeper crustal metamorphic lithologies, primarily gneisses, are also common. Distinct mantle xenoliths are rare. While many of the large macrocrysts are undoubtedly from disaggregated mantle xenoliths, only a few relatively small (b1 cm) multimineralic ultramafic xenoliths have been observed; almost all are of broadly wehrlitic composition. Maps showing detailed locations, along with a table documenting the dimensions and orientations of all known intrusions, are available in a previously published field trip guidebook (Bailey and Lupulescu, 2007).

301

2. Samples and analytical methods Samples from twenty-seven distinct intrusions were collected, and over 150 thin sections were prepared for petrographic examination. Twenty representative samples were then prepared for SEM/EDS analysis for further identification and characterization of mineral phases. Five samples were selected for quantitative mineral analysis by wavelength-dispersive electron microprobe using a JEOL Superprobe 733 in the Dept. of Earth and Environmental Sciences at Rensselaer Polytechnic Institute, Troy, New York. The operating conditions were 15 kV accelerating voltage, 15 nA and a 20 μm beam. Standards used were kyanite (Si, Al), synthetic forsterite (Mg), synthetic fayalite (Fe), synthetic diopside (Ca), jadeite (Na), rutile (Ti), synthetic tephroite (Mn), orthoclase (K), chromite (Cr), topaz (F) and sodalite (Cl). Fe was reported as FeO from the electron-microprobe analysis. The data were reduced using a ZAF correction routine. In addition, garnet and pyroxene macrocrysts were extracted from one dike and were analyzed for rare earth element (REE) concentrations by LA-ICP-MS at McGill University, Montreal, Quebec. Forty-three hand samples from the collections of the New York State Museum and Hamilton College were selected for whole-rock analysis. Some were recently collected, while others are old samples from dikes no longer exposed or whose localities were lost. Friable and porous samples were avoided; the densest most coherent samples were selected for analysis whenever possible. The hand samples were crushed in a hardened steel jaw crusher, and then pulverized to a fine powder using a tungsten carbide ring mill. Obvious (N3 mm) crustal xenoliths were avoided and/or removed from the chipped rocks prior to pulverizing.

Fig. 3. Photomicrographs of macrocrysts. a) Rounded olivine partly replaced by serpentine (plane polarized light, sample GS-1); b) rounded pyrope with large opaque reaction coronae (PPL, sample TC-1); c) rounded and deformed phlogopite (cross polarized light, sample EC-1a); and d) corroded red-brown and brown spinel macrocrysts (plane polarized light, sample CC-1).

302

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

Rock powders were sent to Washington State University (WSU) for major and trace element analysis by X-ray fluorescence spectrometry. The samples were prepared following WSU's in-house process for ultramafic rocks, and analyzed on an automated Thermo-ARL XRF spectrometer using 40 international standards for calibration. Splits of the powder were also sent to Washington State University for rare earth and additional trace element analyses by inductively coupled plasma mass spectrometry (ICP-MS). Detailed sample preparation protocols, analytical methods, and estimates of analytical precision and accuracy are available at http://soe.wsu.edu/facilities/geolab/ technotes/. Five whole-rock samples were selected for Sr and Nd isotope analyses at the Isotope Geology and Geochronology Research Facility at Carleton University, Ottawa, Ontario. Samples were crushed to b 3 mm granules from which 100 g of matrix material free from macrocrysts and xenoliths was handpicked and then pulverized using a tungsten carbide ring mill. Detailed analytical methods are described in Cousens et al. (2008).

Group C intrusions are petrographically similar to those in Group B in that they exhibit a diverse macrocryst assemblage, but they differ by having a more serpentine-rich matrix, and by the presence of clinopyroxene and ilmenite in the matrix. These intrusions are all found in the vicinity of Syracuse, NY. Group D consists of two dikes exposed in East Canada Creek, Montgomery County, south of Dolgeville, NY. These dikes are characterized by abundant olivine macrocrysts and sparse, but large (up to 1.5 cm long) phlogopite macrocrysts, the absence of garnet or pyroxene macrocrysts, and by the presence of clinopyroxene, ilmenite, and unidentified Ca–Fe silicate and titanate phases in the matrix. In addition, there are two small (b 10 cm wide), geographically isolated dikes that are not assigned to any of the above groups. One is the easternmost known intrusion (the “Big Nose” dike along the Mohawk River), and the other is the northernmost (the Eel Weir dike at Ogdensburg). These two dikes are poorly characterized because neither is currently exposed, and available samples are limited and highly weathered. 3.1.1. Macrocrysts

3. Results 3.1. Petrography & mineral chemistry The kimberlitic rocks of New York State are mineralogically and texturally variable, yet all exhibit the characteristic kimberlitic texture of macrocrystic olivine and phlogopite. While nearly every intrusion is petrographically distinct, the intrusions can be divided into four major groups on the basis of macrocryst and matrix phase assemblages (Table 1). Group A intrusions are characterized by the presence of abundant, large olivine macrocrysts (usually serpentinized), and by the absence of garnet, clinopyroxene, and spinel macrocrysts. All the intrusions belonging to this group are exposed along the western margin of Cayuga Lake north of Ithaca; these exposures are almost certainly portions of a single dike exposed over a distance of ~5 km. Group B intrusions are characterized by having a diverse macrocryst assemblage of olivine, phlogopite, garnet, clinopyroxene, and multiple spinels. These intrusions are also all found in the Ithaca region, but are more widely distributed geographically, and exhibit a more varied petrography than those in Group A. Table 1 Mineral phases observed in New York State kimberlitic rocks.

Macrocrysts Olivine Phlogopite Garnet Clinopyroxene Spinel Chromite Magnetite Matrix phases Serpentine Calcite Phlogopite Clinopyroxene Unidentified Ca–Fe silicate Perovskite Unidentified titanate Ilmenite Magnetite/chromite Apatite Fe sulfides and/or Fe–Ni sulfides Barite/celestine

Group A

Group B

Group C

Group D

A U

A A C C C C

A C C U C C

A C

C A C

C A A U

C

A C C U U C

C U U U

U C C U U

C C

A C A U U A U C C U

C U C C U U

A = Abundant (N10%); C = Common (1–10%); U = Uncommon (b1%); Blank = not observed.

3.1.1.1. Olivine. Olivine is, for the most part, completely serpentinized, although partly unaltered macrocrysts have been found in dikes belonging to all four petrographic groups (Fig. 3a). Most of the olivine macrocrysts range in size from a few millimeters to about one centimeter, although one 3.6 mm long macrocryst was found in one of the Group C dikes (Euclid Ave.). Generally, large and small olivine crystals coexist in the same rock. Most of the macrocrysts are rounded, and a few exhibit deformation lamellae and are clearly of xenocrystic origin; euhedral olivine phenocrysts also exist, but are uncommon. The serpentine replacing the olivine macrocrysts varies in color from pale yellow to a very dark olive brown, and is fibrous to scaly in habit. Other products of the alteration of olivine are calcite, magnetite, chlorite and millerite. Chemically, the olivine macrocrysts are relatively homogeneous (Fo85–92), contain moderate NiO contents (0.20 to 0.57 wt.%), and relatively low concentrations of CaO (b 0.12 wt.%). The Fo85–92 compositions are in the range of olivines from both mantle peridotites (Dawson, 1980) and of phenocrysts in ultrabasic magmas. Olivine macrocrysts from the four petrographic groups appear to vary slightly in composition, with mean compositions ranging from ~ Fo92 in Groups B and C, to ~ Fo90 in Group A, and ~ Fo88 in Group D. More data are needed to confirm this preliminary observation. 3.1.1.2. Garnet. Garnet macrocrysts have been found only in intrusions belonging to Groups B and C. In all occurrences, the garnet macrocrysts are rounded and have variably developed kelyphitic coronae (Fig. 3b). Most macrocrysts are only 1–3 mm in diameter, although garnets over 5 mm have been reported (Hopkins, 1914). In contrast to the relatively uniform compositions of the olivine macrocrysts, the garnets are variable in both color (pink to orange to yellow) and chemical composition. Electron microprobe data reveal three major compositional groups: i) High Cr (N 4 wt.% Cr2O3) “G9” pyropes, ii) Low Cr (1–4 wt.% Cr2O3) “G9” pyropes, and iii) Cr-poor “G4” and “G3” magnesian almandines (Fig. 4a). A few macrocrysts of high-titanium, ferro-magnesian grossular were also found in one of the Group B dikes. The Cr-bearing pyropes are compositionally similar to garnets in garnet lherzolites, one of the most common mantle xenoliths found in kimberlites (Dawson and Stephens, 1975). Unfortunately, there is considerable compositional overlap with garnets in garnet peridotites and garnet pyroxenites, so these sources cannot be excluded on the basis of bulk composition. Trace element compositions of the high Cr garnet macrocrysts obtained through laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) reveal high concentrations of the HREE (Lucn = 9.65 to 11.14) and extremely low concentrations of the LREE (La/Nbcn = 0.03) (Fig. 4b). The Sc/Yb and Ti/Sc ratios of these garnet macrocrysts

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

303

to 12.16), and depleted in the HREE (LuN = 0.15 to 0.45) yielding subchondritic ratios of Sm/NdN (0.83 to 0.86) and Lu/HfN (0.02 to 0.14) (Fig. 4b). The trace element data are insufficient to characterize or uniquely identify the source of the low-Cr diopside macrocrysts. 3.1.1.4. Phlogopite. Phlogopite macrocrysts are found in all of the New York kimberlites, although the modal abundance is highly variable (trace to 15%). Phlogopite macrocrysts are most pronounced in the Group D dikes, where flakes can reach 1.5 cm in length. Rounding and strong deformation characterize all of the phlogopite macrocrysts, clearly indicating their xenocrystic origin (Fig. 3c). Chlorite, calcite, and Fe–Ti oxides replace some crystals along the margins and along (001) cleavage planes. Some dikes display sub-parallel alignment of phlogopite macrocrysts near dike margins. Microprobe analyses reveal subtle compositional differences between phlogopite macrocrysts from the four petrographic groups. Macrocrysts from Groups A and C have TiO2 concentrations N 2.5 wt.%, whereas those from Groups B and D typically have TiO2 concentrations between 1.5 and 2.5 wt.%. In addition, phlogopite macrocrysts from Group B intrusions are the only crystals with Cr2O3 concentrations consistently over 0.25 wt.%. These compositions are all within the range of mica compositions commonly observed in other kimberlitic rocks (Mitchell, 1986). 3.1.1.5. Spinels. Spinel macrocrysts are found in all of the New York kimberlites, with the exception of the Group D dikes. Most macrocrysts are relatively small (b1 mm), embayed, and/or rimmed by magnetite. Typical of spinels in kimberlites and related igneous rocks, they are extraordinarily variable in size, color, abundance, and composition (Barnes and Roeder, 2001; Mitchell, 1986), even in samples from the same intrusion (Figs. 3d and 5). Group A dikes only contain Cr and Fe-rich spinel macrocrysts, whereas Groups B and C have the most Mg-rich, and most diverse, spinel macrocryst assemblages. The most common spinels, both as macrocrysts and in the matrix, are

Fig. 4. a) Cr2O3 vs. CaO concentrations in garnet macrocrysts from NY kimberlites. Compositional field boundaries from Grutter et al. (2004). b) Chondrite normalized REE concentrations for Cr-pyrope and diopside macrocrysts from a Group B dike. Normalizing values from McDonough and Sun (1995).

are close to those of C1-chondrites (McDonough and Sun, 1995). All of these features strongly suggest that the Cr-rich pyrope macrocrysts are xenocrysts derived from a garnet-lherzolite source. The relatively high CaO content of the G9 pyrope macrocrysts indicates that these intrusions are unlikely to have been derived from a diamond-bearing mantle source (Grutter et al., 2004), a conclusion supported by the lack of any verified discoveries of diamonds in over 150 years of study. The Cr-poor, magnesian almandines and grossular macrocrysts are most likely derived from disaggregated xenoliths that come from shallower mantle or crustal sources. 3.1.1.3. Diopside. Diopside macrocrysts are common in the Group B intrusions near Ithaca, are scarce in the Syracuse Group C intrusions, and absent in all other. They are all b4 mm in diameter, and range in color from dark emerald green to a pale yellow-green. They are typically rounded and highly fractured, and occasionally exhibit reaction coronae. Microprobe data reveal two groups of clinopyroxenes: low Cr-diopsides (with b0.1 wt.% Cr2O3) and high Cr-diopsides (with 0.5 to 1.60 wt.% Cr2O3). The high Cr-diopsides also are relatively sodiumand aluminum-rich (~1.8 wt.% Na2O, and ~2.8 wt.% Al2O3, respectively), similar to clinopyroxenes in garnet lherzolites (Shimizu, 1975). Most of the low Cr-diopsides have low Na2O (b 0.5 wt.%) and low Al2O3 (b 1 wt.%), are relatively enriched in the LREE (La/YbN = 10.19

Fig. 5. Cr/(Cr + Al) vs. Fe2+/(Fe2+ + Mg) for spinels in kimberlitic rocks of NY State. Spinel compositional fields are from Roeder and Schulze (2008): Chr = chromite, Xen = peridotite xenocrysts, Mum = magnesio-ulvospinel-magnetite.

304

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

deep red-brown Mg and Cr-rich spinels typical of primitive mantle derived magmas. The wide range of compositions observed is typical of kimberlites and reflects the xenocrystic nature of the vast majority of spinel macrocrysts (Barnes and Roeder, 2001; Roeder and Schulze, 2008).

intrusions. An unusual Ca–Fe silicate, possibly ferro-akermanite, has been identified (by SEM/EDS) in the matrix of some of the Group A and Group C intrusions. 3.1.2.9. Apatite. Apatite is common as elongate (up to 1 mm) prismatic crystals in most of the intrusions.

3.1.2. Matrix phases 3.1.2.1. Calcite. Calcite is a major component of the groundmass of many of the intrusions; in places, it also replaces the cores of olivine macrocrysts. Calcite could come from two sources; primary calcite derived from the CO2-rich kimberlitic fluid, or secondary calcite derived largely from the surrounding calcareous shales and limestones. No detailed chemical or isotopic studies of the carbonate phases have been conducted. 3.1.2.2. Serpentine. Serpentine is the most abundant matrix phase in the Group A and Group C intrusions, and the second most abundant phase (after calcite) in Groups B and D. It varies in color from pale yellow to brown to deep olive green, and in habit from massive to fibrous. A few microprobe and SEM/EDS analyses reveal moderate chemical variability, with Al2O3 contents ranging from 0.5 to 5.0 wt.%, and FeO* contents ranging from 2.5 to 12 wt.%. 3.1.2.3. Phlogopite. All of the intrusions contain groundmass phlogopite, but it is most abundant in the Group A dikes. Many of the matrix grains are euhedral and display strong optical and chemical zoning having a Ba-rich core (up to 3.2 wt.% BaO) with Ba-depleted rims (b 0.2 wt.% BaO). The relatively Ba-rich micas (those with N 0.1 a.p.f.u. Ba) are found only in the Group C intrusions near Syracuse. 3.1.2.4. Perovskite. Perovskite is found in all the dikes except those belonging to Group D; it is abundant only in the Group A dikes. It has a characteristic square cross-section, is yellow-brown in color, and occurs as small (b 50 μm), pseudo-octahedral crystals in the phlogopite, serpentine and calcite dominated matrix. Clusters of small crystals also can be found as rims on magnetite grains. These two occurrences represent two different generations of perovskite: primary perovskite that crystallized from the kimberlitic fluid and secondary perovskite that formed as a post-magmatic reaction rim on magnetite. The chemical composition of both is in the range of perovskites from other kimberlitic rocks (Mitchell, 1986). In the Group D intrusions, it appears that the primary perovskite grains have been replaced by unidentified titanates, Ti-bearing silicates, and oxides. 3.1.2.5. Ilmenite. Ilmenite has been observed in the groundmass of all of the dikes, except those belonging to Group B. It typically occurs as small (b 0.1 mm) irregular, often rounded and embayed grains. Nice, tabular, euhedral crystals are only found in the matrix of Group A intrusions. Unlike the large MgO-rich ilmenite macrocrysts commonly found in Group I kimberlites (Mitchell, 1986), these matrix ilmenites are small and have relatively low MgO contents (b7.0 wt.%).

3.1.2.10. Sulfide and sulfate. An assortment of sulfide and sulfate minerals are found in the dikes, including pyrite, pyrhottite, pentlandite, chlorbartonite, millerite, barite and celestine. Most of these are clearly of secondary origin. 3.2. Whole-rock geochemistry The analysis and interpretation of the chemical composition of kimberlitic rocks is complicated by the fact that they are, by nature, hybrid rocks containing complex mixtures of mantle and crustal derived material (Mitchell, 2008), and by the fact that almost all kimberlites have experienced extensive post-emplacement hydrothermal and/or groundwater alteration. Because of these complications whole-rock compositions almost certainly do not represent, or even approximate, magmatic liquid compositions, thereby limiting our ability to understand the mineralogical and chemical nature of the mantle source of kimberlitic magmas and their subsequent evolution. Nevertheless, whole-rock chemistry does provide important information that allows us to categorize and classify these unusual rocks, and to constrain the geological processes involved in their formation. Not surprisingly, whole-rock major element compositions, while broadly similar to those of other kimberlitic rocks (Fig. 6), are highly variable, with SiO2 concentrations ranging from 15 to 40 wt.%, and MgO concentrations from 5 to 30 wt.% (Supplementary Table S1; Fig. 7). The wide range of compositions primarily reflects the relative abundance of the three major matrix phases — serpentine, calcite, and phlogopite. The strong correlation of increasing CaO with decreasing MgO (and to a lesser extent with SiO2) largely reflects the covariation of serpentine with calcite in the matrix. The samples with the lowest MgO concentrations are all Group D dikes that intrude a dolostone and have an unusually high percentage of secondary calcite in the matrix and as olivine macrocryst pseudomorphs. Because of the complex, hybrid nature of kimberlitic rocks, a number of chemical filters have been proposed to screen whole-rock analyses for the chemical effects of crustal contamination and/or weathering. Two that have been found to be useful are the contamination index [C.I. = (SiO2 + Al2O3 + Na2O) / (MgO + 2 ∗ K2O)] (Ilupin and Lutz, 1971), and the molar Si/Mg ratio (Clement, 1982). When applied to the kimberlitic rocks of New York, these indices suggest that all have

3.1.2.6. Chromite & magnetite. Chromite & magnetite are common as small (b0.05 mm) grains disseminated throughout the groundmass of all of the intrusions. Much of the very fine magnetite is associated with secondary serpentine. 3.1.2.7. Clinopyroxene. Clinopyroxene is uncommon, but is found in the groundmass of all the intrusions, with the exception of Group B. The crystals tend to be blocky to tabular, and intimately intergrown with phlogopite. Compositionally, they are all diopsides; compared to the large pyroxene macrocrysts, they have low Na, Al and Cr contents. 3.1.2.8. Melilite. Melilite, despite being reported in previous studies (Smyth, 1893), has not been positively identified in any of the

Fig. 6. Whole-rock concentrations of TiO2 vs K2O in kimberlitic rocks of NY State. Fields of Group I and Group II kimberlite compositions from Smith et al. (1985) and Mitchell (1995).

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

305

45

10

35

Al2O3 5

SiO2 25

15 4

0 15

10

FeO*

2

TiO2

5

0 40

0 3

30 2

Na2O

20 CaO 1 10

0

0 4

2

3

P2O5

K2O 2 1 1

0

0

10

20

30

MgO

40 0

10

20

30

0 40

MgO

Fig. 7. Fenner diagram of whole-rock, major element oxides (in weight percent) in kimberlitic rocks of NY State plotted against MgO. Because of the extremely variable, and generally high, volatile contents of these rocks (LOI ranging from 2 to 30 wt.%), the major element oxides were normalized to an average volatile content of 15% prior to preparing the scatter plots. Symbols: solid circle = Group A; solid square = Group B; solid diamond = Group C; X = Group D; + = Big Nose dike; solid triangle = Eel Weir dike.

been moderately to highly contaminated, with the Group D dikes being the most contaminated (see Bailey and Lupulescu (2007) for a more detailed discussion). Whereas major element scatter plots reveal few meaningful patterns or clusters, plots of high field strength (HFS) minor and trace elements show that the four petrographic groups described above are compositionally distinct (Fig. 8). All have steep and smooth REE profiles (Fig. 9), typical of nearly all kimberlites and ultramafic lamprophyres (Mitchell, 1986). Group B intrusions have significantly higher REE concentrations than all other groups, and Group A intrusions have the highest La/Yb ratios. All are extremely incompatible element enriched relative to primitive mantle, with similar, but slightly different profiles (Fig. 10). Group B kimberlites are the most incompatible element enriched, and are compositionally similar to diamond bearing kimberlites from the Kirkland Lake region of Ontario (MacBride, 2005), with relatively low K/Nb and V/Nb ratios, and high Ba/Rb ratios. Separates of matrix material from samples of all four groups all yielded Nd and Sr isotope ratios close to those of bulk earth (Fig. 11),

Fig. 8. Ti/Nb vs. Zr/Y ratios in kimberlitic rocks of NY State illustrating that the four petrographic groups recognized are compositionally distinct.

306

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

Fig. 9. Average chondrite-normalized REE concentrations in the four recognized groups. Number of analyses for each average provided in legend. Normalizing values from McDonough and Sun (1995).

with the exception of one sample from Group D that has an anomalously high 87Sr/86Sr ratio. These dikes are the most altered of all the intrusions, with abundant secondary calcite in their matrices; the Sr isotope ratios were very likely modified by this process, despite the high initial Sr concentrations of the magmas. Early isotopic studies of archetypal kimberlites and orangeites revealed distinct isotopic compositions of the two groups, with Group I kimberlites (both on and off-craton) having isotopic compositions consistent with derivation from a slightly to nondepleted source, often interpreted to be a convecting asthenospheric mantle source, and orangeites from an enriched source, often interpreted to be a lithospheric mantle source (Mitchell, 1995; Smith, 1983). On the basis of near zero εNd values, Basu et al. (1984) concluded that the New York kimberlitic rocks were derived from a primitive, chondritic, mantle source, consistent with derivation from an asthenospheric mantle source, similar to the proposed source of Group I kimberlites.

Fig. 11. 87Sr/86Sr vs. 143Nd/144Nd ratios of matrix material from NY State kimberlitic intrusions. Field of Group I and II kimberlite compositions is from Mitchell (1986).

4. Ages of intrusions Many attempts have been made to date the intrusions in New York using techniques ranging from fission track dating of apatite (Miller and Duddy, 1989), whole-rock K–Ar (Basu et al., 1984), K–Ar and Rb–Sr dating of phlogopite macrocrysts (Watson, 1979; Zartman et al., 1967), and most recently, U–Pb dating of matrix perovskite (Heaman and Kjarsgaard, 2000). Ages obtained ranged from 493 Ma (Zartman et al., 1967) to 104 Ma (Miller and Duddy, 1989), with the vast majority indicating an early Cretaceous intrusion age. The few Paleozoic ages were on phlogopite macrocrysts that are undoubtedly of xenocrystic origin and, therefore, do not record intrusion ages; these ages often exceeded the age of the bedrock that the dikes intruded. Fig. 12a shows the published age data for the kimberlitic rocks of New York, excluding the anomalous Paleozoic ages and the one apatite

Fig. 10. Primitive mantle normalized trace element averages of the four compositional groups. Number of analyses for each average is provided in the legend. Data for Kirkland Lake kimberlite are from MacBride (2005).

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

307

Fig. 12. a) Radiometric age measurements by method. b) Radiometric age measurements by Group. Data sources: Basu et al. (1984); Heaman and Kjarsgaard (2000); Watson (1979); Zartman (1988); Zartman et al. (1967).

fission track date that was excluded because of the uncertainties in estimates of intrusion depth and cooling rate. Despite large uncertainties in much of the data, it is clear that all of the intrusions are Cretaceous in age. When the data are sorted by mineralogical and chemical group (Fig. 12b), it appears that the four groups may have distinct intrusion ages. The westernmost (Group A) and easternmost (Group D) intrusions appear to be the oldest, with intrusion ages between 145 and 150 Ma. Most of the intrusions belonging to Group B, which is the largest and most diverse group, appear to be younger, with intrusion ages ~ 120 Ma (Fig. 12b). Two dikes from this group yielded older ages of ~ 150 Ma. This may indicate a separate event, or anomalously old ages due to xenocrystic phlogopite (both are K–Ar

ages, one obtained on a phlogopite macrocryst, the other on a wholerock sample). Two Group C dikes from the Syracuse area yielded ages of ~130 Ma. Unfortunately, of the four groups, the only one whose intrusion age is tightly constrained is Group A. The three high precision U–Pb ages on matrix perovskite grains yielded an average intrusion age of 146 +/− 3 Ma (Heaman and Kjarsgaard, 2000). Unfortunately, no similar high precision ages have yet been obtained on samples from any of the other intrusions. Attempts to extract perovskite grains from dikes from the other groups have been unsuccessful due either to the extremely small size of the grains (typically b 20 μm in the Group B and C intrusions), or the lack of unaltered perovskite (Group D

308

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

intrusions). 40Ar/39Ar analysis of groundmass phlogopite grains may be the best method for obtaining precise emplacement ages on the remaining intrusions. On the basis of existing data, it appears that there were two distinct episodes of kimberlitic magmatism in central New York. The first occurred between 145 and 150 Ma, and was extremely limited in both volume and geographic extent. This event is represented by the Group A intrusions along the western shore of Cayuga Lake, which are very likely all part of a single 0.5 to 1.5 m wide dike and a small central diatreme, and two 20–30 cm wide dikes exposed on the eastern bank of East Canada Creek. While rocks from these two groups are petrographically and chemically distinct, they do share a number of features: they both contain only olivine and phlogopite macrocrysts (lacking garnet, pyroxene, or spinel macrocrysts), and they both have relatively low concentrations of most incompatible elements (Fig. 10), and relatively high Ti/Nb ratios (N125) (Fig. 8). After a hiatus of ~25 Ma, kimberlitic magmatism resumed in central New York, generating the vast majority of the known intrusions (Groups B and C) between 130 and 110 Ma, contemporaneous with the alkaline magmatism of the Monteregian Hills in southern Quebec, and the syenitic magmatism of the White Mountains in central New Hampshire. Group B contains the largest number of intrusions, and the intrusions are found across a fairly large area of south-central New York. This group is also the most petrographically and chemically diverse, and probably could be subdivided into distinct sub-groups with additional data. Group C intrusions are confined to the greater Syracuse area and are compositionally relatively homogeneous. Group B and C intrusions are similar in that they both exhibit diverse macrocryst assemblages, and have relatively low Ti/Nb ratios (50–100). 5. Discussion While a detailed petrogenetic model for this diverse group of kimberlitic rocks is beyond the scope of this paper, a few preliminary observations can be made: 1) the Sr and Nd isotope ratios for all of the intrusions are nearly chondritic, suggesting a mantle source that had not experienced either long-term depletion or enrichment, 2) the steep and smooth REE profiles are consistent with derivation by small degrees of melting of a garnet lherzolite source, and 3) many of the incompatible element ratios (e.g. Nb/U shown in Fig. 13) are strongly indicative of involvement of an ocean island basalt (OIB)-like asthenospheric mantle source. On the basis of similar isotopic and trace element evidence, an asthenospheric mantle plume source has been proposed for South African Group I kimberlites (Hoffman et al., 1986; Le Roex, 1986; Smith, 1983). In contrast, the unusually high Mg-numbers and Ni contents of Group I kimberlites, combined with their extremely low

HREE concentrations indicate a relatively depleted mantle source (Le Roex et al., 2003). Three different petrogenetic models have been proposed to explain these unusual features: 1) low degrees of melting of a cryptically metasomatized asthenospheric mantle (Becker and Le Roex, 2006); 2) high degrees of melting of a metasomatically veined asthenospheric mantle (Mitchell, 2004); and 3) low degrees of melting of a depleted lithospheric mantle source metasomatized by asthenospheric melts or fluids (Le Roex et al., 2003). Regardless of which model ultimately explains the origin of kimberlitic rocks in general, it appears that an OIB-like asthenospheric mantle played a significant role in the formation of the kimberlitic rocks of New York State. The tectonic setting and processes responsible for generating these kimberlitic magmas are also not fully understood. Mesozoic magmatism in northeastern North America was initiated and dominated by T–J boundary tholeiitic flood basalts associated with the rifting and break up of Pangea (Marzoli et al., 2011; Puffer, 1992). Subsequent magmatic activity continued for over 100 Ma, was compositionally diverse (ranging from kimberlitic to syenitic), and geographically widespread (from central Quebec to western New York and Arkansas). Based upon a crudely linear distribution of Mesozoic igneous rocks in eastern Canada and New England with generally decreasing ages from northwest to southeast (Fig. 1), it has been argued that most of these rocks, including the New York kimberlites, were generated by passage of North America over the Great Meteor hotspot (Crough, 1981; Heaman and Kjarsgaard, 2000). While this model explains many of the observed occurrences of Mesozoic magmatic activity in eastern North America, it fails to account for the Jurassic (~ 200– 155 Ma) White Mountain Magma Series, the geographically widespread Early Cretaceous (~ 130–110 Ma) alkaline magmatism of the Monteregian Hills and White Mountains, as well as the belt of kimberlitic intrusions distributed along the western flank of the Appalachian Mountains from Quebec to Tennessee, remote from the path of the proposed hot spot track. Parrish and Lavin (1982) noted that many of the kimberlitic intrusions in the eastern United States were spatially associated with large crustal growth faults. They proposed a model for kimberlite emplacement that called for the reactivation of old crustal structures along the western margin of the Appalachian Mountains due to the far field effects of rifting and opening of the Atlantic Ocean basin. Many of the kimberlites also lie along, or near, the long recognized Alabama–New York magnetic lineament (King and Zietz, 1978). This lineament is interpreted to be a fundamental and ancient crustal boundary in eastern North America, and one that records right lateral displacement during the Neoproterozoic to Cambrian (Steltenpohl et al., 2010). Steltenpohl et al. (2010) noted that modern seismic activity in eastern Tennessee is concentrated along the NY–AL lineament, indicating that this structure is still accommodating intraplate stresses. This ancient and major crustal boundary very likely played the same role in the Mesozoic, accommodating intraplate stresses associated with both rifting and the isostatic adjustments of the evolving Appalachian Mountain belt. In an early study on the correlation of kimberlitic magmatism in South Africa with plate tectonic activity, Sykes (1978) concluded that “most kimberlites in South Africa seem to have been emplaced along pre-existing zones of weakness that were reactivated during the early opening of the South Atlantic.” More recent studies have confirmed the overall correlation of kimberlitic magmatism with continental rifting (Heaman, 2003). Kimberlitic magmatism in eastern North America may similarly be due to the distal lithospheric stresses generated during the opening of the North Atlantic during the Mesozoic. 6. Summary

Fig. 13. Nb/U vs. Nb concentrations in New York State kimberlitic intrusions. Fields of mid-ocean ridge basalt (MORB), OIB, and Group I and Group II kimberlite compositions from Le Roex et al. (2003).

While kimberlitic magmatism in eastern North America was geographically widespread, it was volumetrically insignificant. Intrusions were small dikes or diatremes, each with its own distinct mineralogical and chemical signature. A large number (~ 90) of these

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

intrusions are spread across central New York. Past studies assumed that these intrusions were related and recorded a single magmatic event. While these intrusions do share many features, they are, in detail, mineralogically and chemically distinct. Four major groups are recognized on the basis of petrographic and chemical differences. The four groups are each geographically confined, and possibly temporally distinct. Overall, it appears there were two major episodes of kimberlitic magmatism in central New York, the first between 150 and 145 Ma, and the second between 130 and 115 Ma. The first episode generated magmas carrying only olivine and phlogopite macrocrysts, and with relatively high Ti/Nb ratios. Beyond these similarities, the Group A and Group D intrusions are markedly different, with distinct petrographic features and major and trace element compositions. Most of the kimberlitic magmatism in New York appears to have occurred ~ 125 Ma in the vicinities of both Ithaca (Group B) and Syracuse (Group C). These intrusions share many petrographic and chemical similarities, including a complex macrocryst assemblage and similar Ti/Nb ratios. They differ substantially, however, in the relatively elevated REE and incompatible element concentrations in the Group B intrusions. The dominant Group B intrusions are petrographically and chemically similar to the diamondiferous Kirkland Lake kimberlites in eastern Ontario. All of the New York intrusions have macrocryst assemblages as well as geochemical and isotopic characteristics consistent with derivation from an asthenospheric, garnet lherzolite source. Far field stresses related to the opening of the Atlantic Ocean reactivated major crustal structures and provided pathways for small volume, volatile-rich kimberlitic magmas to ascend and intrude the Paleozoic sedimentary platform rocks of central New York. More precise age constraints and comprehensive isotopic studies are needed on all of the eastern North American kimberlites to better understand their petrogenesis and the complex tectonic history of the region. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.lithos.2014.11.022. Acknowledgments We thank the editor, J. Chiarenzelli, and R. Darling for constructive reviews. We also thank K. Bart and D. Tewksbury for help with SEM analysis and final figure preparation, respectively. References Bailey, D.G., Lupulescu, M.V., 2007. Kimberlitic Rocks of Central New York. Field Trip Guidebook 79. New York State Geological Association, pp. 53–81. Barnes, S.J., Roeder, P.L., 2001. The range of spinel compositions in terrestrial mafic and ultramafic rocks. Journal of Petrology 42, 2279–2302. Basu, A.R., Rubury, E., 1980. Tectonic significance of kimberlite dikes in central New York. Geological Society of America, Abstracts with Programs 12, 23. Basu, A.R., Rubury, E., Mehnert, H.H., Tatsumoto, M., 1984. Sm–Nd, K–Ar and petrologic study of some kimberlites from eastern United States and their implication for mantle evolution. Contributions to Mineralogy and Petrology 86, 35–44. Becker, M., Le Roex, A.P., 2006. Geochemistry of South African on- and off-craton group I and group II kimberlites: petrogenesis and source region evolution. Journal of Petrology 47, 673–703. Bikerman, M., 1997. New phlogopite K–Ar dates and the age of southwestern Pennsylvania kimberlite dikes. Northeastern Geology and Environmental Sciences 19, 302–308. Bolivar, S.L., 1976. Geochemistry of the Elliott County, Kentucky kimberlites. Eos, Transactions American Geophysical Union 57, 761. Clement, C.R., 1982. A Comparative Geological Study of Some Major Kimberlite Pipes in the Northern Cape and Orange Free State. University of Cape Town, Cape Town. Cousens, B.L., Prytulak, J., Henry, C.D., Alcazar, A., Brownrigg, T., 2008. The geology, geochronology, and geochemistry of the Miocene–Pliocene ancestral Cascades arc, northern Sierra Nevada, California and Nevada: the roles of the upper mantle, subducting slab, and the Sierra Nevada lithosphere. Geosphere 4, 829–853. Crough, S.T., 1981. Mesozoic hot spot epeirogeny in eastern North America. Geology 9, 2–6. Dawson, J.B., 1980. Kimberlites and their xenoliths. Springer-Verlag, Berlin. Dawson, J.B., Stephens, W.E., 1975. Statistical classification of garnets from kimberlites and associated xenoliths. Journal of Geology 83, 589–607.

309

DeJournett, J.D., Schmidt, V.A., 1975. Paleomagnetism of some peridotite dikes near Ithaca, New York. Eos, Transactions American Geophysical Union 56, 354. Eby, G.N., 1987. The Monteregian Hills and White Mountain alkaline igneous provinces, eastern North America. Geological Society Special Publications 30, 433–447. Eby, G.N., Krueger, H.W., Creasy, J.W., 1992. Geology, geochronology, and geochemistry of the White Mountain Batholith, New Hampshire. Special Paper — Geological Society of America 268, 379–397. Filmer, E.A., 1939. New peridotite dikes of Ithaca. Pan-American Geologist 72, 207–214. Foster, B.P., 1970. Study of the Kimberlite–Alnoite Dikes in Central New York (Finger Lakes Region). SUNY at Buffalo, Buffalo, NY (Master's). Grutter, H.S., Gurney, J.J., Menzies, A.H., Winter, F., 2004. An updated classification scheme for mantle-derived garnet, for use by diamond explorers. Lithos 77, 841–857. Heaman, L.M., 2003. The timing of kimberlite magmatism in North America; implications for global kimberlite genesis and diamond exploration. Lithos 71, 153–184. Heaman, L.M., Kjarsgaard, B.A., 2000. Timing of eastern North American kimberlite magmatism; continental extension of the Great Meteor Hotspot track? Earth and Planetary Science Letters 178, 253–268. Hoffman, A.W., Jochum, K.P., Seufert, M., White, W.M., 1986. Nb and Pb in oceanic basalts: new constraints on mantle evolution. Earth and Planetary Science Letters 79, 33–45. Hopkins, T.C., 1914. The Geology of the Syracuse Quadrangle [New York], Bulletin. New York State Museum, Albany, NY, p. 80. Hunt, T.S., 1858. Contributions to the history of ophiolites. Part II. American Journal of Science 26, 234–240. Ilupin, I.P., Lutz, B.G., 1971. The chemical composition of kimberlite and questions on the origin of kimberlite magmas. Sovietskaya Geologiya 6, 61–73. Kay, S.M., 1990. Central New York kimberlites; evidence for an Early Cretaceous thermal disturbance in the Appalachian Basin. Geological Society of America, Abstracts with Programs 22, 27. Kay, S.M., Foster, B.P., 1986. Kimberlites of the Finger Lakes Region. Field Trip Guidebook 58. New York State Geological Association, pp. 219–238. Kay, S.M., Snedden, W.T., Foster, B.P., Kay, R.W., 1983. Upper mantle and crustal fragments in the Ithaca kimberlites. Journal of Geology 91, 277–290. King, E.R., Zietz, I., 1978. The New York–Alabama lineament; geophysical evidence for a major crustal boundary in the basement beneath the Appalachian basin. Geology 6, 312–318. Kontak, D.J., 2008. On the edge of CAMP; geology and volcanology of the Jurassic North Mountain Basalt, Nova Scotia. Lithos 101, 74–101. Le Maitre, R.W., Streckeisen, A., Zanettin, B., Le Bas, M.J., Bonin, B., Bateman, P., et al., 2002. Igneous Rocks; a Classification and Glossary of Terms; Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks Edition: 2. 2nd ed. Cambridge University Press, Cambridge. Le Roex, A.P., 1986. Geochemical correlation between southern African kimberlites and South Atlantic hotspots. Nature 324, 243–245. Le Roex, A.P., Bell, D.R., Davis, P., 2003. Petrogenesis of Group I kimberlites from Kimberley, South Africa; evidence from bulk-rock geochemistry. Journal of Petrology 44, 2261–2286. Lewis, H.C., 1887. On diamantiferous peridotite and the genesis of diamond. Geological Magazine 4, 22–24. MacBride, L.M., 2005. A Comparative Study of the Petrology, Mineralogy, and Geochemistry of Kimberlite-like and Carbonatitic Rocks from Kontozero (Kola Peninsula, Russia) and Bona Fide Kimberlites, Geology. University of Manitoba, Manitoba, p. 364. Martens, J.H.C., 1924. Igneous rocks of Ithaca, New York, and vicinity. Geological Society of America Bulletin 35, 305–320. Marzoli, A., Jourdan, F., Puffer, J.H., Cuppone, T., Tanner, L.H., Weems, R.E., et al., 2011. Timing and duration of the Central Atlantic magmatic province in the Newark and Culpeper Basins, Eastern USA. Lithos [Oslo] 122, 175–188. McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chemical Geology 120, 223–253. McHone, J.G., 1996. Constraints on the mantle plume model for Mesozoic alkaline intrusions in northeastern North America. Canadian Mineralogist 34 (Part 2), 325–334. McHone, J.G., Butler, J.R., 1984. Mesozoic igneous provinces of New England and the opening of the North Atlantic Ocean. Geological Society of America Bulletin 95, 757–765. Meyer, H.O.A., 1976. Kimberlites of the continental United States: a review. Journal of Geology 84, 377–402. Miller, D.S., Duddy, I.R., 1989. Early Cretaceous uplift and erosion of the northern Appalachian Basin, New York, based on apatite fission track analysis. Earth and Planetary Science Letters 93, 35–49. Mitchell, R.H., 1986. Kimberlites: Mineralogy, Geochemistry, and Petrology. Plenum Press, New York. Mitchell, R.H., 1995. Kimberlites, Orangeites, and Related Tocks. Plenum Press, New York. Mitchell, R.H., 2004. Experimental studies at 5–12 GPa of the Ondermatjie hypabyssal kimberlite. Lithos 76, 551–564. Mitchell, R.H., 2008. Petrology of hypabyssal kimberlites: relevance to primary magma compositions. Journal of Volcanology & Geothermal Research 174, 1–8. Murphy, J.B., Dostal, J., Gutierrez-Alonso, G., Keppie, J.D., 2011. Early Jurassic magmatism on the northern margin of CAMP; derivation from a Proterozoic sub-continental lithospheric mantle. Lithos [Oslo] 123, 158–164. Parrish, J.B., Lavin, P.M., 1982. Tectonic model for kimberlite emplacement in the Appalachian Plateau of Pennsylvania. Geology [Boulder] 10, 344–347. Puffer, J.H., 1992. Eastern North America flood basalts. In: Puffer, J.H., Ragland, P.C. (Eds.), Eastern North American Mesozoic Magmatism. Geological Society of America, pp. 95–118.

310

D.G. Bailey, M.V. Lupulescu / Lithos 212–215 (2015) 298–310

Roeder, P.L., Schulze, D.J., 2008. Crystallization of groundmass spinel in kimberlite. Journal of Petrology 49, 1473–1495. Shimizu, N., 1975. Rare earth elements in garnets and clinopyroxene from garnet lherzolite nodules in kimberlites. Earth and Planetary Science Letters 25, 26–32. Skinner, E.M.W., 1989. Contrasting group I and group II kimberlite petrology; towards a genetic model for kimberlites. In: Ross, J., Jaques, A.L., Ferguson, J., Green, D.H., O'Reilly, S.Y., Danchin, R.V., Janse, A.J.A. (Eds.), Fourth International Kimberlite Conference. Geological Society of Australia, Sydney, N.S.W., Australia, Perth, Australia, pp. 528–544. Smith, C.B., 1983. Pb, Sr, and Nd isotopic evidence for sources of African Cretaceous kimberlites. Nature 304, 51–54. Smith, C.B., Gurney, J.J., Skinner, E.M.W., Clement, C.R., Ebrahim, N., 1985. Geochemical character of southern African kimberlites: a new approach based upon isotopic constraints. Transactions Geological Society of South Africa 88, 267–280. Smyth, C.H., 1893. Alnoite containing an uncommon variety of melilite. American Journal of Science 46, 104–107. Snedden, W.T., 1983. Mineralogy and Setting of the Ithaca Kimberlites, Geology. Cornell University, Ithaca, NY, p. 89. Snedden, W.T., Kay, S.M., 1981a. Initial stages of kimberlite eruption; evidence from mantle minerals in Ithaca kimberlites. Geological Society of America, Abstracts with Programs 13, 557. Snedden, W.T., Kay, S.M., 1981b. Mineral chemistry of kimberlite and included xenocrysts, Ithaca, New York. Geological Society of America, Abstracts with Programs 13, 178. Steltenpohl, M.G., Zietz, I., Horton, J.W.J., Daniels, D.L., 2010. New York–Alabama lineament: a buried right-slip fault bordering the Appalachians and mid-continent North America. Geology 38, 571–574. Sykes, L.R., 1978. Intraplate seismicity, reactivation of pre-existing zones of weakness, alkaline magmatism and other tectonism postdating continental fragmentation. Reviews of Geophysics and Space Physics 16, 621–688.

Tainton, K., Browning, P., 1991. The relationship between Group-2 kimberlites and lamproites; an example from the northern Cape Province, South Africa. European Union of Geosciences Terra Abstracts 3, 17–18. Tappe, S., Foley, S.F., Jenner, G.A., Kjarsgaard, B.A., 2005. Integrating ultramafic lamprophyres into the IUGS classification of igneous rocks: rationale and implications. Journal of Petrology 46, 1893–1900. Van Fossen, M.C., Kent, D.V., 1993. A palaeomagnetic study of 143 Ma kimberlite dikes in central New York State. Geophysical Journal International 113, 175–185. Vanuxem, L., 1837. First Annual Report of the Geological Survey of the Third District, New York, Albany. Vanuxem, L., 1839. Third Annual Report of the Geological Survey of the Third District, New York. Vanuxem, L., 1842. Geology of New York; Part III, Comprising the Survey of the Third Geological District. W. and A. White and J. Visscher, Albany, NY. Wagner, P.A., 1914. The Diamond Fields of South Africa, Johannesburg. Watson, K.D., 1979. Kimberlites of eastern North America. In: Wyllie, P.J. (Ed.), Ultramafic and Related Rocks, 2nd ed. Robert E. Krieger Publ. Co., Huntington, NY, pp. 312–323. Woolley, A.R., Bergman, S.C., Edgar, A.D., LeBas, M.J., Mitchell, R.H., Rock, N.M.S., et al., 1996. Classification of lamprophyres, lamproites, kimberlites, and the kalsilitic, melilitic, and leucitic rocks. Canadian Mineralogist 34, 175–186. Zartman, R.E., 1988. Three decades of geochronologic studies in the New England Appalachians. Geological Society of America Bulletin 100, 1168–1180. Zartman, R.E., Brock, M.R., Heyl, A.V., Thomas, H.H., 1967. K–Ar and Rb–Sr ages of some alkalic intrusive rocks from central and eastern United States. American Journal of Science 265, 848–870.