1 Since the mid-1800s (Jackson 1845; Clarke 1886), the small village of Litchfield in south-central Maine (Fig. 1) has been known as a collecting locality for specimens of cancrinite, sodalite, nepheline, and lepidomelane (Fe-rich biotite). The host rock to these minerals, a coarse-grained nepheline syenite (Fig. 2) that was assigned the name “litchfieldite” by Bayley (1892), is part of a composite alkalic plutonic complex (Barker 1965) that occupies an area of approximately 7 km² (Fig. 3). While these unusual rocks have attracted the interest of mineral collectors for over a century and a half, extremely poor exposure has discouraged detailed study; thus the age and tectonic significance of these unusual rocks have remained elusive.

2 Silica-undersaturated, nepheline-bearing plutonic rocks are uncommon in the geological record and are generally attributed to small degrees of partial melting of previously metasomatized mantle in anorogenic tectonic settings (e.g., Woolley 2001; Martin et al. 2012). Burke et al. (2003) have also proposed that the presence of these alkaline plutonic rocks may be indicators of ancient suture zones. Detailed studies of these unusual rock types, when they are present, can thus be used to provide important constraints on the tectonic setting of a region at the time of their formation. Additionally, these rocks have received renewed attention due to their tendency to contain relatively high abundances of rare earth and high field strength elements (Kogarko 1990; Sørensen 1992; Möller and Williams-Jones 2016).

3 In the last detailed study completed on alkaline rocks of the Litchfield pluton, Barker (1965) suggested that they are associated with the Mesozoic White Mountain magma series. However, limited major element geochemical data are available from these rocks, and no isotopic igneous crystallization ages have been published. The purpose of this paper is to report new U-Pb zircon ages and complete whole-rock chemical analyses from a limited number of samples of the Litchfield pluton in south-central Maine. These data are then used to determine the igneous crystallization age of the pluton and speculate on the processes responsible for the generation of this unusual magma composition. Additionally, the new age constraints provided by our U-Pb work allow us to make connections with other alkaline igneous rocks of broadly similar age in the northern Appalachian orogen and place constraints on the tectonic setting of the region at the time of intrusion.

4 The Litchfield pluton occupies a topographic basin in an area of very poor bedrock exposure. Although these unusual rocks had been the subject of directed mineralogical studies since the late 19^th century (Clarke 1886; Bayley 1892), field relationships were first mapped in detail by Barker (1965), and more recently at a scale of 1:24 000 by West and Ellenberger (2010). Bedrock outcrops are most abundant in the northern part of the pluton, but the map pattern of the pluton is based primarily on the spatial distribution of alkalic granitoid boulders. Although contacts between the plutonic rocks of the Litchfield pluton and the surrounding country rocks are not exposed, map patterns in the surrounding country rocks appear to be truncated by the inferred outline of the pluton (Barker 1965; West and Ellenberger 2010).

5 Hence the Litchfield pluton is interpreted to intrude Late Ordovician (?) to Silurian metamorphosed turbidites (mostly thin-bedded schist, granofels, quartzite, and more rarely impure carbonate rocks) of the Central Maine Sequence (Osberg 1988; Marvinney et al. 2010). These country rocks were multiply deformed and metamorphosed to amphibolite-facies conditions during the late Silurian-Early Devonian Acadian orogeny (Gerbi and West 2007). While many of the alkalic rocks of the Litchfield pluton show weak to moderate preferred orientation of dark colored minerals, thin sections reveal little to no evidence for significant penetrative deformation or metamorphic recrystallization. Additionally, Barker (1965) reported that foliation in the plutonic rocks is locally at right angles to the foliation in adjacent metasedimentary rocks. Thus, while contacts cannot be observed directly, field relationships indicate the intrusion post-dated regional deformation and metamorphism associated with the Acadian orogeny.

6 In the general area of the Litchfield pluton, constraints on the timing of amphibolite-facies metamorphism are provided by a U-Pb age of 381 ± 4 Ma on metamorphic zircon overgrowths in the Hornbeam Hill pluton located approximately 15 km south of the present study area (West and Cubley 2006; Gerbi and West 2007). A ⁴⁰Ar/³⁹Ar hornblende cooling age of 349 ± 5 Ma from country rocks exposed approximately 10 km east of the Litchfield pluton indicates the time of regional cooling below approximately 500°C following this metamorphism (West et al. 1988). Finally, a large number of ⁴⁰Ar/³⁹Ar biotite ages from both plutons and country rocks in the immediate area constrain the time of regional cooling below approximately 300°C in the vicinity of the Litchfield pluton to have occurred approximately 240 to 250 million years ago (Dallmeyer 1979; 1989). Dallmeyer (1989, p. 427) reported that “Dallmeyer (1979) presented a 227 ± 5 Ma ⁴⁰Ar/³⁹Ar plateau age for biotite from nepheline syenite” near Litchfield, Maine. However, a review of Dallmeyer (1979) did not reveal these data.

7 Although the unusual rocks near Litchfield, Maine, had been known to mineral collectors for decades prior, Clarke (1886) provided the first detailed mineralogical descriptions of nepheline syenite from the Litchfield pluton. This work included major element mineral chemistry for cancrinite, nepheline, and biotite from these rocks. Of note, Clark (1886) showed that biotite in this rock, long referred to as “lepidomelane,” is high in both Fe₂O₃ (19.49 wt.%) and FeO (14.10 wt.%), low in MgO (1.01 wt.%), and completely lacking in TiO₂. While no detailed mineral chemistry is presented here, analysis of biotite using energy dispersive spectrometry on a scanning electron microscope confirmed that biotite from these rocks is high in iron and contains virtually no titanium or magnesium.

8 Bayley (1892) provided the first whole-rock chemical analysis (major elements only) from nepheline syenite of the Litchfield pluton and these old results are consistent with the modern results presented later in this paper. Upon review of the unusual aspects of the cancrinite-bearing nepheline syenite, Bayley (1892, p. 243) reported: “Its peculiarities are so strongly marked that the rock seems worthy of a distinctive varietal name, for which no more appropriate one can be found than litchfieldite, derived from the familiar locality – Litchfield – whence nearly all the specimens were obtained.” The rock name “litchfieldite” is included in the most recent IUGS Subcommission on the Systematics of Igneous rocks (Le Maitre et al. 2002, p. 105) and is therein defined as “a coarse-grained, somewhat foliated variety of nepheline syenite consisting of K-feldspar, albite, nepheline, cancrinite, sodalite and lepiodomelane.”

9 Daly (1918) published the first systematic field descriptions of rocks associated with the Litchfield pluton, and was the first to describe other alkalic plutonic rocks in the area (e.g., nepheline-deficient, amphibole-bearing syenite). No detailed petrologic descriptions, geochemical data, or geologic map accompanied this work. Daly (1918, p. 469), based upon this limited field-based investigation, concluded that “ . . . it is tolerably certain that there is no large mass of litchfieldite in the region. The celebrated boulders seem to have been derived from short, sill-like pods injected into the prevailing crystalline schists.” Additionally, Daly (1918) concluded that the unusual alkaline rocks were formed through variable assimilation of limestone in granite magma.

10 In the nearly 100 years since the publication of Daly’s work, only the detailed field and petrologic study of Barker (1965) and bedrock mapping of West and Ellenberger (2010) have provided additional data on the alkalic plutonic rocks near Litchfield, Maine. Barker’s study revealed three main rock types within what he interpreted to be a composite pluton: (1) fine-grained, leucocratic syenite that contains limited amounts of Na-rich pyroxene (aegirineaugite) and amphibole (riebeckite-arfvedsonite) with no nepheline or quartz; (2) medium- to coarse- grained, magnetite-bearing biotite syenite that similarly contains no nepheline or quartz; and (3) coarse-grained, foliated, biotite nepheline syenite that includes the “type variety” of litchfieldite described by Bayley (1892) and the rock type that is found in most petrologic collections (shown in Figure 2). Barker (1965) also recognized rare cross-cutting dykes of mafic syenite and albite-rich pegmatite, but these rocks types could not be located in the field during the present study.

11 Barker (1965) interpreted the three main alkalic rock types described above to be included within a single oval-shaped composite pluton occupying an area of approximately 7 km². Although West and Ellenberger’s (2010) later detailed mapping recognized Barker’s (1965) three main alkalic rock types in the field and confirmed an overall map pattern consistent with that portrayed in Barker (1965), they did not subdivide the pluton into different phases due to the limited availability of in situ outcrops. Barker (1965) interpreted the different alkalic rock types in the Litchfield pluton to have been produced by differentiation of an initial nepheline syenite magma; furthermore, he suggested that, in places, country rock assimilation and metasomatism resulted in the late-stage textural and mineralogical variety observed within individual phases. Finally, Barker (1965) correlated rocks of the Litchfield pluton with three other nepheline syenite occurrences in New England that are now known to be Mesozoic in age (i.e., Cuttingsville, Vermont: 100 Ma, Armstrong and Stump 1971; Red Hill, New Hampshire: 194 Ma, Foland and Faul 1977; and Pleasant Mountain, Maine: 112 Ma Foland and Faul, 1977).

12 As discussed in all previous studies, the paucity of fresh in situ exposures greatly hinders detailed petrologic study of the Litchfield alkalic rocks. Despite this, representative samples of each of the three major rock types identified by Barker (1965) and confirmed in this study were collected. Bayley (1892) provided detailed petrographic descriptions of the litchfieldite, and Barker (1965) provided multiple modal analyses of each of the rock types within the pluton and the reader is referred to these works for details. The type “litchfieldite” (Fig. 2) is coarse-grained and rich in K-feldspar and albite (combined about 70%); it contains conspicuous yellow granular-textured cancrinite, greasy-grey coarse-grained nepheline, and less common fine-grained blue sodalite. Accessory minerals include coarse-grained biotite that is typically segregated into streaks within the lighter colored minerals, and euhedral zircon prisms up to 5 mm in length. Representative photomicrographs of litchfieldite are shown in Figures 4a-c. Nepheline comprises up to 20% of the rock and forms subhedral to anhedral grains up to 3 cm across. Based upon old chemical analyses (Clark 1886; Barker 1965), the nepheline has a compositional range of Ne 70-80 . Cancrinite is anhedral, occurs in abundances of up to 10%, and shows no apparent textural affinity for other minerals. Sodalite is fine grained (< 3 mm) and occurs in abundances of less than 1%.

13 Fine-grained, nepheline- and cancrinite-absent syenite is most commonly found in the northern part of the pluton and is characterized by accessory Na-rich amphibole and pyroxene (Fig. 4d). These rocks are notably leucrocratic (color index <5) with the sizes of individual minerals rarely exceeding 5 mm across. Finally, coarse-grained nephelineand cancrinite-absent biotite syenite (± magnetite) that texturally appears similar to the litchfieldite occurs in the central and southern portions of the pluton.

14 Elongate, prismatic, and clear zircons displaying igneous zoning were separated from a sample of cancrinite-bearing nepheline syenite (P-603) from the Litchfield pluton. U-Pb isotopic analyses were carried out on these zircons at the United States Geological Survey-Stanford University sensitive high-resolution ion microprobe (SHRIMP) during a session in March of 2011. Bradley et al. (2014) described analytical techniques used at this lab. A total of 13 spot analyses were completed and all individual spot results are reported in Table 1 and shown on Figure 5 as ²⁰⁶Pb/²³⁸U ages with 1 sigma uncertainty.

15 Biotite was also separated from sample P-603 and it was analyzed by the ⁴⁰Ar/³⁹Ar technique as a single total fusion increment. The sample was encapsulated and irradiated in the United States Geological Survey TIRGA reactor in Denver, Colorado (Dalrymple et al. 1981) with the MMhbhornblende age standard (Alexander et al. 1978; 519.4 ± 2.5 Ma) used as a flux monitor. The sample and flux monitors were analyzed at the United States Geological Survey argon dating laboratory in Reston, Virginia, and the details of analytical methods employed are discussed in Kunk et al. (2005). Decay constants were those recommended by Steiger and Jäger (1977). The analytical results of this experiment are shown in Table 2 and the total fusion age is reported with two-sigma uncertainty.

16 Six samples representing a range of bulk compositions identified through field and petrographic observations were selected for whole-rock chemical analysis (locations and brief hand sample descriptions are provided in Appendix A). These samples were crushed in a porcelain jaw crusher and powdered in a tungsten carbide shatterbox. We recognize that the use of tungsten carbide has the potential to introduce small amounts of tantalum contamination; however, we do not consider this to be significant in this study given the relatively high concentrations of tantalum in the samples analyzed. A full suite of major- and trace-element analyses (Table 3) was completed at Acme Analytical Laboratories Ltd. in Vancouver, British Columbia. Major elements were measured using ICP-emission spectrometry and trace elements were measured using ICP-mass spectrometry. Both methods used lithium metaborate fusion techniques to prepare the samples for ICP analysis.

19 The whole rock major and trace element compositions of representative samples of the Litchfield pluton are presented in Table 3. The rocks are intermediate in terms of their SiO₂ contents (58 to 65 wt.%) and all contain high concentrations of both K₂O (4 to 6 wt.%) and Na₂O (6 to 9.5 wt.%). The rocks are also notably depleted in MgO (5 of 6 samples ≤0.1 wt. %), and also have low concentrations of TiO₂ (5 of 6 below 0.2 wt.%) and CaO (0.26 to 1.27 wt.%). On a total alkali versus silica diagram (Fig. 6a) the rocks are classified as nepheline syenite and syenite, and on a SiO₂ versus K₂O diagram, compositions plot in the alkaline field (Fig. 6b). All of the samples analyzed are nepheline normative (range = 1.0 to 17.5% normative nepheline). Although there is variability in the major element abundances between the different rock types, Harker diagrams (Fig. 7) reveal no discernable fractionation trends.

20 In terms of trace elements, the rocks are generally characterized by relatively high concentrations of Rb (67 to 158 ppm), Nb (32 to 68 ppm), and Ga (24 to 38 ppm), and highly variable contents of Zr (28 to1809 ppm), Sr (2 to 810 ppm), and Ba (15 to 1482 ppm). All samples are essentially devoid of trace elements compatible with mafic silicates (e.g., Cr and Ni ≤ 2 ppm). As with the major elements, Harker diagrams of trace elements (Fig. 7) reveal no discernable fractionation trends.

21 A plot of chondrite-normalized rare earth elements (REE) (Fig. 8a) reveals two groupings: one group (n = 3) with significant light REE (LREE) enrichment (~ 90 to 200 times chondrite abundance), and another group (n = 3) with a much less pronounced LREE enrichment (~ 7 to 10 times chondrite abundance). Five of the six samples display a positive Eu anomaly with only the aegerine-riebeckitebearing syenite sample displaying a negative anomaly. The largest positive Eu anomalies are in samples with the lowest total rare earth element contents. The samples have similar primitive mantle-normalized trace element patterns (Fig. 8b) with most samples showing notable enrichment in Ta, Nb, Zr, and Hf, and all have a negative Ti anomaly.

1. Recent detailed mapping and the results of this study confirm the earlier work of Barker (1965) and show that the 7 km² Litchfield pluton in south-central Maine is a composite strongly alkalic intrusion that cuts previously deformed and metamorphosed Silurian turbidites of the Central Maine basin. Rock types within the pluton include Na-amphibole and pyroxene bearing syenite, biotite syenite, and nepheline syenite that includes the type variety of “litchfieldite” that is found in many petrologic collections (coarse-grained, nepheline syenite containing cancrinite, sodalite, and lepiodomelane).
2. A new U-Pb zircon age of 321 ± 2 Ma from the nepheline syenite is interpreted to date igneous crystallization of the pluton. A ⁴⁰Ar/³⁹Ar biotite age of 239 ± 1 Ma is similar to previously published mica ages from surrounding country rocks and indicates the region cooled below approximately 300°C in the Early Triassic.
3. Whole rock geochemistry of a select number of samples from the pluton reveal alkalic, silica-undersaturated, metaluminous compositions that are consistent with fractionation of an initial alkalic basalt parental magma. Geochemical signatures are consistent with magma generation in an anorogenic tectonic setting.
4. The age and geochemical characteristics of the alkalic rocks near Litchfield, Maine are consistent with the generation of a small volume of alkalic magma beneath south-central Maine during a period of Carboniferous transcurrent tectonism in the northern Appalachian orogen.