Masol 1 paleonto-archeological locality

Masol 1 is located on the dome and on its western slopes (Fig. 3; Chapon Sao et al., 2016a and Gargani et al., 2016). The eponymous locality is the center of the anticline and is of significant interest as a result of cut marks found here on a fossil shaft tibia (Dambricourt Malassé et al., 2016a and Dambricourt Malassé et al., 2016b). Six sediment samples were collected at the site for the sedimentology and the paleomagnetic measurements (Chapon Sao et al., 2016b), 3 samples were collected in the silt and 3 others taken from the sandstones (Fig. 5). •

Member F is a light gray sandy silt ∼2.40 meters thick. It is very compact, contains centimetric rock debris and white micas, and laminations are visible. One sample was taken here (S1).

Masol 13 paleonto-archeological locality

Masol 13 is located on the eastern bank of the Patiali Rao River (Fig. 3) and like Masol 1, is of a significant interest due to the presence of cut marks on a bony splinter (Dambricourt Malassé et al., 2016b). The stratigraphic section is oriented east-west. Five samples were collected from four layers (Fig. 5). This locality contains, respectively from the bottom to the top: •

Member A: Fine, light brown colored sandstone, 3 to 4.50 meters thick and containing silt. One sample was collected at the top of the layer (S1, Fig. 5).

Masol 6 paleonto-archeological locality

Masol 6 is a large area located on the western flank of the Picchli Choe torrent, ∼380 meters from Masol 1 (Fig. 3). This sector is the second richest locality in fossils and stone tools after Masol 1, with numerous accumulations of cobbles, sometimes in dismantled conglomerates on the slopes (Gaillard et al., 2016 and Moigne et al., 2016) (Fig. 6). Until now the locality has not provided cut marks, but the area is above the bed of the Pichhli Choe and at 190 meters from its terrace T2 (Gargani et al., 2016) where a third bone with cut marks was collected (Dambricourt Malassé et al., 2016b).

The sampling was performed on a cliff divided into four members from the bottom to the top, T, U, V and W (Fig. 5) as follows: •

Member T: Yellow fine sandstone, friable, 0,46 meter thick (no sample).

The samples were studied in France at the National Museum of Natural History, Paris, in the Laboratory of Sedimentology, at the Department of Prehistory, Institute of Human Paleontology (IPH) and in the Department of Earth History. Materials with a grain size < 2 mm were analyzed using two main approaches: (1) grain size measurement and, (2) sand and clay mineralogical analysis. A mass of 100 grams of material was required to carry out the various sedimentological analyses for each sample. The samples were dried at a temperature of 40 °C, and then the sediment was separated into grains sized below 2 mm (gravels) and above 2 mm (sand).

The grain size (coarse sand (2–0.2 mm), fine sand (0.2–0.040 mm), silt (0.040 mm–0.002 mm) and clay (< 0.002 mm)) was measured using a Malvern Mastersizer 2000 laser granulometer which performed diffraction and scattering measurements. The study was carried out using distilled water for the silty samples and without water for the sandy samples. Obscuration varied between 10 and 20 units. The interpretations were made according to the classical works of Anderson (2007), Folk and Ward (1957), Miskovsky and Debard (2002), Pye and Blott (2004) and Rivière (1977).

For heavy minerals, the works of Duplaix (1958), Parfenoff et al. (1970) and Tourenq (2002) were consulted. The sample was first sieved from 2 to 0.040 mm in order to retain only the sand fraction. This was attacked by hydrogen peroxide to remove organic matter and the decalcified with hydrochloric acid. The samples were then washed, several times, with distilled water and dried in an oven at 40 °C. The separation of heavy minerals was performed using a bromoform density equal to 2.89. For each sample, 250 to 300 grains were studied under a petrological microscope to determine the species.

For the clay mineralogy, we applied the method developed by Holtzapffel (1985). This consists of removing from the sample, by chemical treatments, organic matter and carbonates, in order to deflocculate the clay. The fraction below 2 microns, which corresponds to clay, is extracted from the solution after a rest period of 1 hour 35 minutes. This mode of treatment allows oriented aggregates to be obtained. For each sample, three slides are prepared for three types of analysis: (1) aggregate N: without treatment, (2) aggregate G: sample exposed, during 24 hours, to the ethylene glycol vapor, and (3) aggregate CH: heated sample at 490 °C during 2 hours.

Analyses to determine the minerals from their crystallographic properties were conducted at the Department of Earth History. X-ray diffraction was performed on orientated aggregates using a diffractometer Bruker D2 Phaser Lynxeye (30 kV, 10 mA) with a Cu anti-cathode. A continuous scan from 3 to 65° was performed each 0.02° during 0.2 seconds for each step and the rotation speed was of 20 revolutions/min (0 to 80). The identification of crystalline phases was made using the software DIFFRAC.SUITE. For the determination of the clay minerals, the following works were consulted: Caillère et al. (1982), Larqué (2002), Millot (1964), Pédro (1965), Larqué (2002) and Robert (1975).

Masol 1 paleonto-archeological locality

The silty-sand member “F” at the base of this sequence (Fig. 7) contains 52% silt and 48% sand. The two samples (S2 and S3) collected from silty layers “member E”, contain 4% clay, 50% silt and 46% sand for sample S2, and 6% clay, 54% silt, 38% sand and 2% coarse sand for sample S3 (Fig. 7). The last three samples were collected in the micaceous sandstone (S4, member C; Fig. 5) and in the medium to coarse sandstones (S5 and S6, member B; Fig. 5). The three samples recorded a high rate of coarse sand (between 21 and 53%). Fine sand was also well represented (between 34 and 50%) and the silt decreased relative to the lower levels (13–30%) (Fig. 7).

Masol 13 paleonto-archeological locality

The S1 sample collected at the base of the section in a fine sandstone “Member A” is essentially constituted of fine sand (60%) and silt (27%), but also of coarse sand (9%) and clay (3%) (Fig. 7). Samples collected in the brown silt “member B” (S2) and brown clay “member C” (S3 and S4) as described during the field work, have the same low contents of clay (9%) and of coarse sand (1%) (Fig. 7). The percentages of silt and fine sand in S2 and S3 are identical (respectively 65% and 25%) (Fig. 7). In sample S4 (Fig. 7), silt abundance reaches 84%. The last sample (S5) collected in the sandstone at the top of the section (Member D) (Fig. 7) presents 49% fine sand and 39% silt, and there is an increase recorded in coarse sand (8%) and a decrease of clay (5%) compared to the S2, S3 and S4 samples.

Masol 6 paleonto-archeological locality

The first two samples (S1 and S2) collected in the pink silt “Member U” show completely different contents (Fig. 7). The S1 sample is very rich in silt (86%) and contains a large amount of clay (10%). Sample S2 contains a significant amount of sand (50% fine sand and 24% coarse sand) and a non-negligible amount of silt (26%). The two other samples (S3 and S4) collected in sandstone “Member V” contain identical proportions of fine sand (56%), silt (33%), coarse sand (7%) and clay (4%) (Fig. 7).

The X-ray diffraction (XRD) measurements were taken on clay fraction of the sediment using three protocols: (1) classical for diffraction, (2) with samples heated at 490 °C, and (3) with samples treated with ethylene glycol. This allows the identification of illite, chlorite, kaolinite and smectite through the observation of four main peaks in the XRD spectrum.

A peak can be seen at 10 Å (Fig. 8). If the peak shows constant intensity whatever the treatment used, it is due to the presence of illite. A second peak is seen at 14 Å and is due to the chlorite, when the thickness of the peak is constant for all three protocols used (Holtzapffel, 1985). The two peaks at 7.2 Å (for the main peak) and 3.58 Å (for the secondary peak), which disappear after heating at 490 °C, indicate the presence of kaolinite (Fig. 8). The peak at 14 Å (protocol without treatment), at 10 Å after heating at 490 °C and moving to 17 Å after its exposure to ethylene glycol, is characteristic of smectite (Lucas et al., 1959).

We first demonstrated the results in stratigraphic sections before synthesizing the characteristics of each locality.

Masol 1 paleonto-archeological locality

The illite content ranges between 48% and 53% in all the samples except for the micaceous fine sandstone, Sample S4, where it presents 20%. Its peak is long and narrow in almost all the samples (Fig. 8). This is a characteristic of a good crystalline network. In Sample S4, there is a significant quantity of smectite (41%) and of chlorite (33%). The kaolinite abundance ranges between 18 and 27% except in two samples (S4 and S1 with 6 and 8%, respectively). The smectite content ranges between 10 and 41%, and the chlorite content ranges between 10 and 33%.

Masol 13

The Masol 13 deposits are mainly characterized by the predominance of illite (41–76%). The well-defined peak of illite suggests a well-developed crystalline network (Caillère et al., 1982). The chlorite and the smectite are poorly represented in Samples S5, S4 and S1 (contents between 1 and 6%) but are significantly represented in Samples S2 and S3 (contents between 18 and 21%). Their peaks often take the form of domes in the spectrum, indicating the beginning of degradation of the crystal lattice (Dunoyer de Segonzac et al., 1968). The kaolinite abundance is constant and it is present in a significant proportion (between 16% and 22%) (Fig. 8).

Masol 6

The illite content ranges between 43 and 73% in all samples, the kaolinite between 14 and 22% (Fig. 8). The chlorite abundance ranges between 4 and 10% and the smectite abundance between 2 and 20%.

Clay mineral synthesis

The illite is very abundant in all the silty layers of the three localities, and is scarce in the micaceous fine sandstone (member C) of Masol 1. Smectite and chlorite are present mainly at Masol 1. •

An intense peak (D1) is seen at 10.1 Å due to the illite (Fig. 8). Such a peak at 5.00 Å suggests that the mineral (illite) is in its hydrated form.

Masol 1 paleonto-archeological locality

Tourmaline represents 40 to 71% of the heavy minerals. It is the most abundant mineral in all the samples. There is only one sample (S4, in the micaceous fine sandstone member) where garnet (25%) reaches values comparable to tourmaline (25%). The average abundance of sphene is 12%; it varies from 4% (in the medium to coarse sand member “B”, S5) to 23% (in the micaceous sandstone member C, S4). Apatite proportions range between 1 and 13% (average 8%) and zircon between 3 and 15% (average 7%). The other detected minerals are the staurolite (4%), the kyanite (3%) and the rutile (2%) (Fig. 9).

Silt members contain the highest amounts of tourmaline and apatite. Sandstone and sandy Members B and C contain much garnet, sphene, staurolite, kyanite and zircon (Fig. 9).

Masol 13 paleonto-archeological locality

Tourmaline abundance ranges between 31 and 47%. There is a decrease from the bottom to the top of the section. The proportion of garnet ranges between 42% in the medium sandstone (member D) (S5) and 6% in the brown silt (member B) (S2). Zircon abundance in this silt (S2) is 25%, 14% in clay (S3), 6% in fine sandstone (S1) and 8% in medium sandstone (S5). The zircon abundance decreases from the bottom to the top of the section. Sphene has a higher content in the sandstone (S1) (19%) than in the sandstone (S5) (9%) and remains constant at 12% in silt (S2) and clay (S3). Kyanite reaches 6% in S1 and is rare in S5. Staurolite abundance ranges between 4% in the lower part of the sequence (S1), and 2% in its upper part (S5; Fig. 9). Rutile shows higher rates in silt “S2” and clay “S3” (respectively 7% and 6%) than in sandstones (S5 with 1%). Epidote abundance is constant in S2 and S5 (1%) and reaches a slightly higher percentage (3%) in S5 (Fig. 9).

Tourmaline, sphene, zircon, rutile and staurolite decrease from the bottom (S1, Fig. 9) to the top (S5) of the locality. Conversely, garnet abundance increases from the bottom to the top of the locality.

Masol 6 paleonto-archeological locality

This locality is characterized by a significant abundance of garnet (56%) and a relative decrease in tourmaline (22%) in comparison to Masol 1 and Masol 13 (Fig. 9). The silt layer with a polyhedral structure “U” contains more staurolite, while the sandstone member “V” contains more sphene, zircon and apatite (Fig. 9).

Heavy mineral synthesis

Tourmaline and garnet constitute the most significant proportions in the mineralogical assemblage of Masol (> 66%). These two minerals indicate opposite evolutions. In sandstone members, the proportion of tourmaline decreases significantly and the garnet increases. In the silty and clay members, tourmaline represents more than half the mineralogical assemblage. Sphene and zircon average rates are similar. Sphene has a constant occurrence in all the layers. Zircon abundance is higher in the silt and clay members. Apatite proportions are higher in layers where grain sizes are smaller. Staurolite and kyanite frequencies are slightly higher in the sand and the sandstone layers than in the silty members. Epidote is absent from Masol 1. The highest proportions of rutile are observable in the silt and clay members.

Mica (muscovite and biotite) rates range from 4 to 90%. A rate of 90% is reached in medium to coarse sand member “B” in Masol 1. When considered, this abundance hides the relative proportion between the other minerals. We have therefore excluded the micas from the counting. Tourmaline is usually brown and subangular (only one grain is blue). Garnet is often altered. Many minerals demonstrate no marks of erosion. The mineralogical assemblage composition suggests a metamorphic source.

Singh et al. (2004) studied the mineralogical assemblage of the Cenozoic sequences in the Jammu region (Northwest Himalaya), the sediments of the Murree Group (Early Miocene of Northwest Indian subcontinent) (Fig. 10A), the sand of the Baspa River which originate in the Higher Himalaya before joining the Sutlej River at Karcham (Fig. 10B), and the sand of the Bhuzas stream flowing in the Lesser Himalaya before joining the Chenab river at Atholi (Fig. 10 C). The Bhuzas stream that flows from north-east to south-west crosses the main metamorphic formations before reaching the Chenab River. These metamorphic formations are gneiss and schist with veins of quartz, granites, amphibolites and pegmatites. The mineralogical assemblage contains sillimanite, garnet, kyanite, staurolite and biotite-mica.

The Baspa River, a tributary of the Sutlej, is generally oriented east-west before diverging to the north at Sangla. The Baspa successively crosses phyllitic formations, quartzite and granite, gneiss with psammites, porphyroblastic gneiss and schists, and in its downstream portion, granitic formations. The minerals contained in theses formations are garnet, kyanite, sillimanite and biotite-mica (Singh et al., 2004).

Table 1 allows a comparison of the mineralogical compositions of a) the rivers Baspa and Bhuzas, b) the Jammu region, c) the Murree Group, d) the Siwalik Group, and e) the localities of Masol. The studies of Singh et al. (2004) show the constant presence of garnet, tourmaline, epidote, staurolite, zircon and kyanite for a), b) c) and d). Hornblende is seen in the sand formation in the Jammu region, and sillimanite is absent from the Murree Group. Another interesting point is the absence of sphene, apatite and rutile in the rivers, but also in the Jammu region and the Murree Group. Sphene is also absent in Lower and Middle Siwalik, and apatite and rutile are absent in Siwalik group

The Siwalik deposits contain ten species of heavy minerals (Delcaillau, 1986 and Raju and Dehadrai, 1962). Garnet, tourmaline, epidote, staurolite, zircon and zoïsite are common to the three subgroups of the Lower, Middle and Upper Siwalik. Kyanite is absent in the Lower Siwalik (where it should have an exceptional presence according to Delcaillau) while hornblende, sillimanite, andalusite and sphene are present only in the Upper Siwalik. The Lower and the Middle Siwalik differ from the Upper Siwalik in the absence of several minerals such as the hornblende, the sillimanite, the andalusite and the sphene.

The comparison of the mineralogical assemblages of the various sites with these of the paleonto-archeological localities of Masol shows that this sequence from Masol contains the richest assemblage of minerals. They differ in the absence of sillimanite and the presence of apatite and rutile. The diversity of the mineralogical content is due to the various geological formations drained by the different rivers, and to the heterogeneous resistance of minerals to the transport processes. These observations are consistent with the recent conclusions of Dubille and Lavé (2015), despite the dissimilarity in the mineralogical content, however, the entire Upper Siwalik subgroup (Masol localities included) contains generally the same heavy minerals.

The minerals identified in the Masol Formation contain illite, kaolinite, chlorite and smectite as previously seen in the Late Neogene sediments of the Subathu subbasin and in the Siwalik by Suresh et al. (2004) and Delcaillau (1986), respectively. Sediment deposits and mineral content could be influenced by (i) rock mineral initial content, (ii) climate, (iii) topography, (iv) diagenesis and (v) transport dynamic. (i)

Rock mineral initial content: both illite and chlorite are produced by the physical alteration or transformation of crystalline rocks (Ghosh et al., 2003 and Kumar et al., 1999). Kaolinite results from magmatic rocks and smectite originates from the magmatic or volcanic rocks of the Lesser Himalayan hinterland (Raiverman and Suresh, 1997). In conclusion, the minerals seen in the three localities of the Masol Formation have been transported (1) from the crystalline rocks of the Higher Himalaya, composed of orthogneiss (magmatic) and paragneiss (metamorphic), and (2) from the volcanic and sedimentary rocks of the Lesser Himalaya.