The Tama Kosi/Rolwaling area of east-central Nepal is underlain by the exhumed mid-crustal core of the Himalaya. The geology of the area consists of Greater Himalayan sequence phyllitic schist, paragneiss, and orthogneiss that generally increase in metamorphic grade from biotite ± garnet assemblages to sillimanite-grade migmatite up structural section. All metamorphic rocks are pervasively deformed and commonly record top-to-the-south sense shear. The top of the Greater Himalayan sequence in the mapped area is marked by an undeformed, pegmatitic leucogranite stock. Relationships in adjacent areas constrain the age of the leucogranite and the deformation structures it crosscuts, including the top-to-the-south sense deformation, to be older than middle Miocene. The lower portion of the exhumed midcrustal package has been subject to late-stage folding during the formation of the Tama Kosi window, a structural culmination that may reflect out-of-sequence adjustment of the orogenic wedge. The geology of the mapped area appears similar to that observed in the adjacent, better-studied Everest region.


Much has changed in our understanding of orogenesis since the first reconnaissance mapping of the Himalaya. The ideas put forth to explain the varied evolution of the orogen, such as midcrustal flow (e.g., Bird, 1991; Grujic et al., 1996; Beaumont et al., 2001), critical taper (e.g., DeCelles et al., 1998a, 1998b), and plateau collapse (e.g., England and Houseman, 1988), have all been based on geologic map interpretation. There are still some areas along the mountain belt, however, that have not been mapped. One of those areas is the uppermost portion of the Tama Kosi valley in east-central Nepal (Fig. 1). In order to properly evaluate the evolution of the Himalaya and understand the processes responsible for its formation, it is critical that all areas along the length of the mountain chain be investigated, at least at a reconnaissance scale.

The Tama Kosi valley is situated between the Cho Oyu/Everest/Makalu massifs to the east and the Kathmandu klippe/nappe to the west (Fig. 1). Recent work in these areas serves to highlight stark differences between them. In the Kathmandu region, the extruded midcrustal core is folded and preserved far into the orogenic foreland in the form of a klippe or nappe (e.g., Johnson et al., 2000). Furthermore, there may be evidence for a merger of the two major, antithetic fault systems that bound the mid-crustal core (Webb et al., 2011). In contrast, the geology of the Everest region reflects a deeply eroded, formerly ductily extruded mid-crustal channel and associated leucogranitic bodies (e.g., Searle et al., 2006; Jessup et al., 2006; Cottle et al., 2009; Streule et al., 2010). This makes the Tama Kosi valley area important not only to help complete the geologic map of the Himalaya, but also for potential assessments of lateral variation during the evolution of the mountain belt. This preliminary study presents the basic lithology, structure, and metamorphism recorded in the Tama Kosi area and interprets those findings within the current conceptual framework of the orogen.



The geology of the lower and middle portions of the Tama (also written as Tamba) Kosi valley (Fig. 1) was first reported on in the late 1960s (Ishida, 1969). This early work outlined the basic lithologic framework of the area and how it is related to the adjacent Everest/Makalu region. Ishida subdivided the area into a series of tectonic units termed “formations” (Ishida, 1969) or “zones” (Ishida and Ohta, 1973), which were in turn assigned to either the Himalayan gneiss group or the Midlands metasediment group (Fig. 2). The units of Ishida (1969) and Ishida and Ohta (1973) are quite similar in description to the tectonostratigraphy reported by Schelling (1992) who revisited and expanded the scope of their early reconnaissance work.

Schelling (1992) separated the geology of the lower and middle portion of the Tama Kosi into the more traditional Greater Himalayan sequence (Higher Himalayan Crystallines) and Lesser Himalayan sequence lithotectonic assemblages. His “Higher Himalayan Crystallines” approximately correspond to Ishida and Ohta's (1973) Himalayan gneisses (Fig. 2) and include a series of sillimanite-bearing paragneiss and orthogneiss units that display varying degrees of partial melting, granitic intrusion, and migmatization. The rocks within Schelling's (1992) Lesser Himalayan sequence comprise much of Ishida and Ohta's (1973) Midland metasedimentary group (Fig. 2). The lithologies mapped by Schelling as the Lesser Himalaya include locally graphitic-rich, garnet ± staurolite ± kyanite schist, and orthogneiss, commonly K-feldspar augen-bearing.


Differences in the structural interpretation between previous studies of the lower and middle Tama Kosi valley and area are more significant than those for the tectonostratigraphy. Ishida (1969) mapped thrust faults between almost every “formation” or “zone” (Fig. 2), separating them into tectonically bound rock units. Each of these faults is considered to be an originally north-dipping structure that accommodated top-to-the-south sense displacement. Subsequent folding of some of the faults has since modified their dip direction and, paired with erosion, favored the development of tectonic windows (Ishida, 1969). In contrast, Schelling (1992) considered the contacts between most lithotectonic units to be gradational, at least in the Higher Himalayan Crystallines. He mapped a major thrust discontinuity, the Main Central thrust, at the base of the Higher Himalayan Crystallines that juxtaposed them on top of the Lesser Himalayan sequence (Fig. 2). While Schelling (1992) considered his Main Central thrust to be the major structural discontinuity in the region, he also recognized intense shearing of his mapped Lesser Himalayan sequence rocks below the structure. He considered these rocks, which include all units below the Main Central thrust outlined in Figure 2, to be part of the Lesser Himalayan Shear Zone (Schelling, 1992). He notes that it is akin to the Zone des Ecailles of Bordet (1961), the “MCT zone” of Arita (1983), and the “Nappes Inferieurs” of Brunel (1986) and Brunel and Kienast (1986). The rocks that comprise this zone are characterized by mylonitic deformation structures (Schelling, 1992) and record inverse metamorphism with low-grade rocks at low structural levels and higher-grade rocks at higher structural levels (Ishida and Ohta, 1973; Schelling, 1992).


The present study builds on the previous work by Schelling (1992) and Ishida and Ohta (1973) and extends into areas not yet reported on geologically. This study presents the results of detailed geologic mapping along the Tama Kosi river from Dolhaka in the south to the Nepal-Tibet border and the river's headwaters in the north (previous studies did not include areas north of Lamabagar; Fig. 3). This mapping was also extended into the tributary Rolwaling and Khare valleys (Fig. 3). These new observations allow for direct comparison of the Tama Kosi region to the present knowledge of the Everest region (e.g., Jessup et al., 2006; Goscombe et al., 2006) and the Kathmandu klippe/nappe (e.g., Johnson et al., 2000; Webb et al., 2011), which is significantly different from the interpretations of either Ishida (1969) or Schelling (1992).

All rocks in the study area have been metamorphosed; in general, metamorphic grade increases northward. The structurally lowest unit encountered in the mapped area is a locally augeniferous orthogneiss (equivalent to the Tama Kosi window formation, Fig. 2; for brevity only the equivalent unit of Ishida [1969] is noted; readers are referred to Figure 2 for the equivalent unit of Schelling [1992]). The quartz + feldspar + biotite + muscovite ± garnet orthogneiss crops out along the Tama Kosi River just south of the town of Shigati and extends northwards to Suri Dhoban (Fig. 3). It is at least 300 m thick; however, its total thickness is unconstrained. The orthogneiss forms the core of the Tama Kosi window (Fig. 4), a late-stage structural culmination that will be discussed in a later section of this study. The unit contains garnet-bearing anatexite locally (Fig. 5A); however, it does not typically make up more than 5% of the rock unit by volume.

The orthogneiss is overlain by a quartz + feldspar + muscovite + biotite ± garnet ± chlorite phyllitic schist (equivalent to the Dolakha formation, Fig. 2). The unit also contains local coarser-grained intercalations of metasandstone (Fig. 5B). The thickness of the phyllitic schist along the southern side of the Tama Kosi window is unconstrained (Fig. 3); however, along the northern side it is ∼800 m (Fig. 4). While the original relationship between the underlying orthogneiss and the phyllitic schist is not known, orthogneiss units in similar structural positions along the Himalayan front (e.g., Ulleri augen gneiss) have been interpreted as intercalated volcanoclastics (Le Fort, 1975), metasomatized granitic basement (Arita, 1983), or granitic laccoliths (Le Fort, 1989). The contact between the units now is tectonic.

A second orthogneiss unit occurs structurally above the phyllitic schist (equivalent to the Melung augen gneiss, Fig. 2). This upper orthogneiss is also augeniferous (Fig. 5C) and has a very similar mineralogy to the lower unit, though it is typically coarser grained. It is ∼800 m thick along the river valley and folds over the top of the Tama Kosi window (Fig. 4) to crop out in the town of Dolakha to the south (Fig. 3). It is not known if this rock unit is related to the orthogneiss found structurally lower. It may represent a thrust repeated section, multiple original intrusions or layers, or have a completely different progeny.

The upper orthogneiss is structurally overlain by a pelitic schist unit consisting of quartz + muscovite + biotite + feldspar + garnet ± staurolite ± kyanite ± chlorite (equivalent to the Jiri formation, Fig. 2). It also contains a thin, laterally continuous graphitic schist layer. This layer serves as a marker horizon and was observed at four different locations at approximately the same structural position (Fig. 3). The pelitic schist is at least 800 m thick along the river valley (Figs. 3 and 4); however, its interpreted map pattern indicates that it may be significantly thicker in adjacent areas. It contains a small volume (5%–10%) of leucogranitic anatexite, which typically occurs as thin discontinuous deformed layers (Fig. 5D). It is not known if this pelitic schist is related to the structurally lower phyllitic schist; however, there is a conspicuous absence of a graphitic layer within the lower rock unit.

A thin quartzite layer with subordinate muscovite structurally overlies the pelitic schist (equivalent to the lower Solo formation; Fig. 2). This quartzite unit is at most 300 m thick along the river valley and can be followed laterally across adjacent ridges and into other valleys (Fig. 3). The quartzite unit marks a distinct change in character of the rocks in the mapped area with rocks that contain a significant volume (>20%) proportion of anatextite above it and rocks that do not below it.

The next structurally higher rock unit in the mapped area is an ∼5200-m-thick quartz + biotite + muscovite + feldspar + garnet + sillimanite ± chlorite migmatitic paragneiss (equivalent to part of the Solo formation, Figs. 2 and 5E) with subordinate intercalated quartzite and calc-silicate gneiss (Fig. 5F). The migmatite neosome consists of quartz + feldspar ± muscovite ± tourmaline leucosome layers and associated biotite-rich melanosome halos that generally increase in volume up structural section (up to 50% leucosome toward the top of the unit). In addition to stratiform leucogranite, these rocks also contain evidence of later in situ development of pegmatitic anatexite pods, which will be discussed in more detail later in this study.

The migmatitic paragneiss transitions up structural section into a medium- to finely crystalline quartz + feldspar + biotite ± muscovite ± sillimanite ± garnet gneiss (equivalent to the upper portion of the Solo formation; Fig. 2). The transition between the two units is not exposed and as such the nature of it is not known. The unit is ∼1.5 km thick along the line of section (Fig. 4), has a well-developed gneissic foliation, and is variably migmatitic with leucosome formation localized within more fertile, mica-rich layers (Fig. 5G). In one location, ∼2–4 km to the west of the village of Bedding (Fig. 3), sillimanite and quartz occur together as 1–2-cm-diameter flattened nodules.

The next structurally higher unit is a distinctive feldspar + quartz + biotite ± muscovite granitic augen orthogneiss (equivalent to part of the Khumbu formation; Fig. 2) that contains K-feldspar augen that are up to 15 cm in diameter (Fig. 5H). It is 1.5–2 km thick along the line of section (Fig. 4).

The augen orthogneiss unit is structurally overlain by quartz + feldspar + biotite gneiss that is locally migmatitic (part of the Khumbu formation, Fig. 2). The thickness of this unit is not constrained as it is intruded by a large granitic complex higher up in the structural section (Fig. 4). This rock unit is characterized by a well-defined foliation that is commonly isoclinally folded, at least at the outcrop scale (Fig. 5I). The foliation may reflect transposed compositional bedding or gneissic layering. Unlike the rocks below it, this unit does not contain a typical high-grade mineral assemblage, which may be a reflection of its protolith.

The contact with the structurally higher intrusion is characterized by a network of dykes and sills that increase in density toward the main granite body (Fig. 4). The intrusion [equivalent to the Rolwaling-Khumbu granites of Schelling (1992); Ishida (1969) reported no equivalent units; Fig. 2] consists of at least two distinct phases. A subordinate older medium- to coarsely crystalline quartz + feldspar + biotite + muscovite granite is only observed locally and is most commonly associated with contact with the underlying metasediments. The much more voluminous phase is a pegmatitic feldspar + quartz + muscovite ± sillimanite leucogranite. The ages of these phases are interpreted to be Cenozoic due to the structural position of the granite body, its relationship with surrounding rocks, and the ages of granite bodies in similar positions in adjacent areas (e.g., Searle et al., 2003; Viskupic et al., 2005; Streule et al., 2010).



The rocks exposed in the upper Tama Kosi and Rolwaling valleys are pervasively, ductily deformed; all planar and linear features have been transposed into parallelism. The transposition foliation that resulted from that deformation is the dominant structural fabric preserved in the mapped area.

Rocks at lower structural levels are typically more phyllitic or schistose in nature (Fig. 3; see above descriptions) than those at higher structural levels and commonly record the development of secondary foliations, including both S and C′ planes (Fig. 6A). These secondary foliations consistently indicate a top-to-the-south sense of shear. Orthogneiss at lower structural levels also preserves secondary foliations that record pervasive top-to-the-south sense deformation (Figs. 6B and 6C). In addition, local anatexite pods within these orthogneiss units commonly act as sigma-type porphyroclasts, which record a similar shear sense (Fig. 6D). Lineations are rarely observed at lower structural levels (Fig. 3), but become more common at the structural level of the village of Suri Dhoban and above (Fig. 3). There, they are defined by the alignment of micaceous minerals and mineral aggregates and typically plunge moderately toward the north.

Rock units in the structurally higher portion of the exhumed metamorphic core (i.e., north of Jagat, Fig. 3) display well-developed gneissic foliations defined by mineral segregation, plastically deformed quartz, and aligned mica grains. These rocks do not typically exhibit secondary foliations, though they are present locally in more mica-rich lenses. Lineations are well developed within the paragneiss and orthogneiss at this structural level. They are commonly defined by elongate, plastically deformed quartz and aligned micaceous minerals and typically plunge moderately toward the north (Fig. 3). Leucosome in these rocks generally occurs in layers parallel to the dominant foliation (Fig. 6E) without distinct asymmetry, however, localized asymmetric lenses record top-to-the-south sense deformation (Fig. 6F). Local top-to-the-west sense of shear may be recorded by large K-feldspar augen in deformed orthogneiss (Fig. 6G) just to the east of the village of Na (Fig. 3), however, this sense of shear was not observed elsewhere in the map area (Fig. 3). The South Tibetan detachment system, a major top-to-the-north sense structure that marks the top of the exhumed metamorphic core along the length of the orogen (Burchfiel et al., 1992; Yin, 2006; Godin et al., 2006a), occurs to the north of the present study area (Jessup and Cottle, 2010); no top-to-the-north sense shear related to that structure was observed.

Local isoclinal folding is preserved at the outcrop and cliff-side scale at upper structural levels in the mapped area (Fig. 5I). In most cases the limbs of the isocline and the fold axial plane are parallel to the dominant foliation (Fig. 6H), which indicates layer perpendicular shortening and layer parallel elongation.

All rocks south of approximately Jagat (Fig. 3) in the mapped area are affected by late, km-scale open folding (Fig. 4). This folding postdates all ductile shear and the development of the transposition foliation. The folding is most apparent just south of Suri Dhoban (Fig. 3) where the tectonic foliation to the north of the village dips to the north and the foliation immediately to the south of the village dips away from the river valley on either side. Farther south, near Shigati (Fig. 3), the foliation changes again and dips southward. The change in foliation outlines the Tama Kosi window of Ishida (1969) and Ishida and Ohta (1973), which is similar to structural culminations observed elsewhere in Nepal (e.g., Godin et al., 2006b). It has been interpreted to reflect interference folding of one local fold event with a vertical axial plane parallel to the NNE trend of the river and another regional event with a vertical axial plane perpendicular to the first (Ishida and Ohta, 1973).

Metamorphism and Crustal Melting

All rocks exposed in the study area have been metamorphosed to at least greenschist facies. In general, metamorphic grade increases structurally up section from south to north and from lower to higher structural levels until just north of the town of Jagat (Fig. 3), where sillimanite grade is reached. This inverted metamorphic sequence is typical of the Himalayan metamorphic front and has been variably attributed to shear heating (Arita, 1983; Harrison et al., 1998), tectonic assembly/ductile shear (Jamieson et al., 1996; Hubbard, 1996; Stephenson et al., 2001), or post–over thrusting conductive heating (Le Fort, 1975). Anatexite within the mapped area follows a similar pattern with volume increasing up structural section, perhaps indicating a change in peak temperatures.

In the southernmost part of the map area, near the town of Dolakha, the metamorphic assemblage typically includes garnet + biotite + muscovite. Evidence of crustal melting in this portion of the map area occurs in the form of local leucogranite lenses that are commonly boudinaged (Figs. 5A and 6B). The leucogranite that forms the lenses typically contains quartz + feldspar + muscovite ± tourmaline; however, at one location, ∼2.5 km north of Shigati (Fig. 3), garnet is also observed as part of the leucogranite (Fig. 7A). Melt products at this structural level do not comprise a significant portion of the total rock volume (∼5%).

The same metamorphic mineral assemblage is maintained northward until approximately halfway between the villages of Suri Dhoban and Jagat (Fig. 3), where staurolite is observed (Fig. 7B). Anatexite at this structural level is more voluminous (10%–15%) than at lower levels. It typically forms discontinuous stratiform lenses intercalated with the country rock (Fig. 8A).

Kyanite is first observed approximately at the structural level of the village of Jagat in kyanite + garnet + biotite + muscovite migmatitic gneiss (Fig. 7C). The volume of leucogranite in the rocks at this level is markedly higher than at lower levels, comprising up to 30%. The leucogranite here forms semicontinuous layers within the host paragneiss and is often rimmed by a thin biotite-rich layer in the host rock (Fig. 8B).

Sillimanite is recognized ∼1 km south of the confluence of the Tama Kosi and Rolwaling rivers (Fig. 3). The first occurrence is within a migmatitic quartz + feldspar + garnet + biotite + sillimanite + muscovite paragneiss (Fig. 7D). Sillimanite is found throughout the rest of the upper portion of the exhumed metamorphic core when the protolith allows, including locally within the leucogranite at the structurally highest levels observed (Fig. 7E). The presence of sillimanite (Figs. 7E and 7F) within the leucogranite indicates that it was either intruded under sillimanite-grade conditions or it was metamorphosed at sillimanite grade after its intrusion. Two significant anatexite phases are recognized within the sillimanite-bearing metamorphic core of the study area. Continuous, foliation-parallel leucosome here comprises at least 40% of the rock by volume (Fig. 8C). The stratiform leucogranite and surrounding paragneiss are intruded locally by pegmatitic pods (Figs. 8C and 8D). The crosscutting quartz + feldspar + muscovite ± tourmaline pegmatites are entirely undeformed and appear to be the result of in situ melt formation (Fig. 8D). The nature of those pegmatitic pods is similar to that exhibited by the granitic stock observed at the highest structural levels in the mapped area where it intrudes migmatitic paragneiss (Fig. 8E).

The stock is a multiphase intrusion with an earlier medium- to coarsely crystalline granite phase and a later, more voluminous, very coarse to pegmatitic phase (Fig. 8F). The stock appears to be largely undeformed. A well-developed sill and feeder dyke network can be followed from below the stock into the main body (Fig. 8G). This feeder network cuts across foliation, often at a high angle, and does not record evidence of significant deformation (Fig. 8H).


The main goals of this preliminary study were to examine the geology of the Tama Kosi valley and adjacent areas and interpret those findings within the framework of our contemporary understanding of the region. This understanding now includes the recognition of channel flow as a potential major process that has accommodated convergence at mid-crustal levels (Beaumont et al., 2001, 2004), which has been interpreted to explain the geology in the Everest region (e.g., Searle et al., 2006) as well as the potential recognition of the “tip” of the extruded mid-crustal core of the orogen observed within the Kathmandu nappe structure (e.g., Webb et al., 2011). Neither of these significant contributions had been made at the time of the last study of the areas examined in the present paper.

This investigation confirms that the rocks within the upper Tama Kosi and Rolwaling valley are part of the exhumed mid-crustal core of the orogen. Like other portions of the Himalayan metamorphic front (e.g., Larson and Godin, 2009; Searle et al., 2008; Larson et al., 2010a), the lower part is characterized by pervasive top-to-the-south sense shearing defined by secondary foliations and inverted metamorphism, ranging from garnet + biotite assemblages at lowest levels to kyanite + garnet + biotite ± muscovite at higher levels. There is a possibility of structural duplication of two orthogneiss units and vertical thickening of the lower part of the metamorphic core; however, confirming or refuting that will be the subject of future work. The upper portion of the exhumed metamorphic core in the map area is dominated by sillimanite-grade migmatitic paragneiss and orthogneiss, which is intruded at the highest structural levels by a two-phase leucogranite stock.

The previous studies that examined portions of the present mapped area generated significantly different interpretations of its structural history (Fig. 2; Ishida and Ohta, 1973; Schelling, 1992). The findings of this study do not support the initial structural interpretations of Ishida (1969) and Ishida and Ohta (1973), who mapped the metamorphic core as a series of discrete thrust-separated lithotectonic units. The metamorphic core exposed in the study area records evidence of pervasive ductile strain distributed throughout the entire exhumed mid-crust. There is no evidence of significant localized high-strain zones. Furthermore, in contrast to the work of Schelling (1992), which maps the Main Central thrust, the base of the Greater Himalayan sequence (the exhumed metamorphic core), approximately at the village of Jagat (Fig. 3), this study does not recognize the structure in the mapped area. All rocks in the mapped area are interpreted to be in the hanging wall of the Main Central thrust as defined by Searle et al. (2008), who state that the structure separates rocks that record metamorphism and cooling related to Himalayan orogenesis in its hanging wall (the Greater Himalayan sequence) from those that do not in its footwall (the Lesser Himalayan sequence). Therefore, all rocks within the mapped area are part of the Greater Himalayan sequence. The change in geology that occurs near Jagat does not correspond to an abrupt structural or metamorphic break, but instead reflects a change in lithology perhaps across a structure akin to the Himalayan Unconformity of Goscombe et al. (2006) mapped in the adjacent Everest region. The change in geology near Jagat also may reflect a change in displacement and distortion, or structural style, similar to that observed in the Manaslu-Himal Chuli Himalaya, where the upper part of the Greater Himalayan sequence records extending flow (Price, 1972)—vertical thinning and horizontal extension—consistent with foliation parallel isoclinal folds and the lower part records compressing flow—vertical thickening and horizontal shortening (Price, 1972)—consistent with the possibility of structural repetition of units (Larson et al., 2010a). At a first order, these deformational characteristics are also consistent with the spatial and temporal evolution of channel flow models (e.g., Jamieson et al., 2004; Larson et al., 2011). While more investigative work is necessary to fully explore and evaluate this possibility, this is consistent with the original observations of Ishida (1969) and Schelling (1992) that the geology of the Tama Kosi area is similar to that of the Everest region.

The evolution of the Tama Kosi window postdates the pervasive deformation associated with the development of the dominant foliations in the mapped area. Similar culminations in Nepal and Tibet have been interpreted to reflect out-of-sequence rebuilding of the orogenic wedge after the early to middle Miocene southward extrusion of the mid-crust (Godin et al., 2006b; Larson et al., 2010b). In this scenario, the Tama Kosi window would have been developed to facilitate the foreland migration of deformation from the Main Central thrust to the Main Boundary thrust after the metamorphic core was extruded (e.g., Larson et al., 2010b). Alternatively, the Tama Kosi window may be due to tectonic inversion of local, structurally controlled sedimentary thickness variations (Long et al., 2011), lateral ramp structures (Johnson, 1994), or localized increased erosion along river valley bottoms (Montgomery and Stolar, 2006). Detailed geochronologic constraints will help discriminate between these models.

Inferences about timing constraints on the tectonometamorphic evolution of the Greater Himalaya series in the mapped area can be made based on work in adjacent areas. The South Tibetan detachment system exposed on the northern flank of the Lapche range just north of the study area in Tibet (the study area includes the southern flank of the same range) is characterized by pervasively deformed marble, paragneiss, and leucogranite that records ductile top-to-the-north sense shear; the leucogranite commonly occurs as ultramylonite layers with feldspar porphyroclasts (Jessup and Cottle, 2010). Movement across the South Tibetan detachment system in that part of the Himalaya ceased between 16 and 13 Ma (Jessup and Cottle, 2010). If the leucogranite deformed in the detachment system on the northern flank of the Lapche range is related to similar intrusive rocks observed in the present study on the southern flank of the range, then the intrusive stock mapped in this study must be older than ca. 16–13 Ma. This provides an important time constraint for the history of the Greater Himalayan sequence in the mapped area. The top-to-the-south sense shear deformation migmatite development in the rocks crosscut by the stock and related anatexite pods must be older than middle Miocene. This is consistent with deformational and metamorphic ages from the Everest region that outline a protracted history from ca. 39 Ma to ca. 16 Ma (Cottle et al., 2009). Moreover, regional scale folding and development of the Tama Kosi window, which is interpreted to be younger than the last movement across the South Tibetan detachment system, likely occurred during the middle to late Miocene.


The upper Tama Kosi/Rolwaling valley region of Nepal is characterized by pervasively deformed, exhumed, mid-crustal rocks that commonly preserve a top-to-the-south shear sense. The lower part of the metamorphic core records an inverted metamorphic sequence from biotite-garnet to kyanite grade, while the upper portion is typically at sillimanite grade and contains a significant proportion of anatexite. In contrast to previous studies, no major discrete structures are mapped in the area; all rocks mapped are interpreted to be in the hanging wall of the Main Central thrust. The geometry of the Greater Himalayan sequence was altered by later large-scale folding as reflected in the development of the Tama Kosi window. The relationship between a leucogranite stock at high structural levels and deformation related to the South Tibetan detachment in adjacent areas indicates the age of that intrusion is likely older than ca. 13–16 Ma. That constraint also limits the age of south-directed ductile deformation crosscut by the leucogranite to pre–mid-Miocene and later folding to be younger than ca. 13 Ma.

This study was funded by a University of Saskatchewan Faculty Start-up Grant. Alicia, Josh, and Dale Larson are thanked for their assistance and company in the field. This paper benefited from a discussion with K. Ansdell, reviews by C.J. Warren and an anonymous reviewer, and editorial direction from M. Williams. Logistical support was provided by Teke, Pradap, Kajiman, Lakpa, Some, Sete, Dawa, Karma, and Teke Bahadur Tamang; Phum, Beg, and Gajindra Shresta; Dorha Dahal; and Forba Sherpa.