serie NOVA TERRA nº 49

92 sometimes almost intact and even preserving the original sedimenta- ry features ( Díez Fernández, 2011 ). In the low deformed sections, the lower metasedimentary sequence consists of a thick pile of meta- greywackes alternating with minor layers of metapelites, graphitic schists, calc-silicate lenses and quartzites ( Fig. 2 a). The meta- greywackes preserve Bouma sequences, crossed bedding, erosive contacts and normal graded bedding. Their protoliths were clast- supported sedimentary rocks containing angular feldspar fragments (the most abundant clasts), quartz and detrital micas in a clay rich matrix with carbonaceous material. No conglomeratic levels have been found. Metapelitic horizons are common, and carbonaceous matter within them can be so abundant as to form graphitic horizons. Quartzite layers occur in the upper part of the sequence ( Fig. 2 a). The original thickness of the lower sequence cannot be calculated due to the intense ductile deformation accompanying the Variscan subduc- tion and subsequent exhumation, but a minimum present thickness of 4 km can be estimated, although this value probably represents less than half of the original thickness. The maximum depositional age, as revealed by the detrital zircon input, is latest Neoproterozoic (c. 560 Ma; Díez Fernández et al., 2010 ). The upper sequence appears strongly deformed. No signi fi cant de- formation partitioning associated with the subduction – exhumation process can be seen in this series during the subduction – exhumation process. This sequence consists of mica schists alternating with minor lenses of amphibolites, graphite schists, metacherts, calc-silicate lenses, metagreywackes and quartzites ( Fig. 2 b). The maximum de- positional age of this sequence is Late Cambrian (c. 500 Ma; Díez Fernández et al., 2010 ). 4. Whole rock geochemistry The chemical composition of sedimentary rocks depends on nu- merous factors, including the nature of the source areas and the sub- sequent processes affecting them, such as weathering, diagenesis or metamorphism. Likewise, the abundance of some elements, such as rare earth elements (REE), Hf, Ti, Cr, Co, Zr, Nb, Ta, Y, Th and Sc, is pre- served in sedimentary rocks in spite of the weathering processes. These elements have very low residence times in oceanic waters, being transferred almost quantitatively to sedimentary rocks. Thus, they provide excellent discriminating factors for determining the provenance and tectonic setting of sedimentary rocks ( Bathia, 1983; Bathia and Crook, 1986; Hegde and Chavadi, 2009; Roser and Korsch, 1986, 1988; Taylor and McLennan, 1985 ). The effects of homogenization in sedimentary processes result in a relatively uni- form distribution of REE in detrital rocks, whose pattern re fl ects the REE abundance in the upper crust. Moreover, the Th/Sc, La/Sc, Ti/Zr, Zr/Sc and La/Th ratios has been frequently used to investigate prove- nance according to the different compatibility of these elements dur- ing magmatic crystallization ( McLennan et al., 1993 ), and also in relation to the PAAS abundances (Post Archean Australian Shale; Taylor and McLennan, 1985 ). Twenty three samples from the sedimentary series that consti- tutes the basal units of the allochthonous complexes of NW Iberia were collected in order to study their geochemistry, provenance and tectonic setting. Samples were collected mainly along the northern sector of the Malpica – Tui Complex although a few samples were taken in the NW part of the Órdenes Complex (see location of the samples in Figs. 1 b and 2 ). Eighteen samples belong to the lower se- quence and they include weakly sheared metagreywackes and some albite schists developed after intense deformation and metamor- phism of the greywackic types. Five samples represent the upper se- quence including the Cean schists (Malpica – Tui Complex) and the Santiago schists (NW Órdenes Complex). Sample preparation was carried out at Universidad Complutense de Madrid, and whole rock major and trace elements analyses were performed at Activation Laboratories Ltd (Actlabs) in Canada. Fusion with lithium metaborate/tetraborate was used for sample digestion, and the analytical techniques for major and trace element determina- tion were ICP-OES and ICP-MS, respectively. The chemical analyses results are shown in Tables 1 and 2 . 4.1. Composition and classi fi cation 4.1.1. Lower sequence The greywackes and albitic schists show a homogeneous major el- ement composition ( Table 1 ). The SiO 2 content shows the largest var- iation (60.3 – 74.9 wt.%), with average value of 70.65 wt.%. Only 4 samples have SiO 2 contents lower than 70 wt.% (B-3, B-5, B-9, B-23), mainly compensated by larger values of Al 2 O 3 and MgO. The Na 2 O con- tent is relatively high and homogeneous (2.5 – 4.1 wt.%; average 2.9 wt.%), and also can be considered homogeneous the content in K 2 O (2.0 – 3.4 wt.%; average 2.6 wt.%) and MgO (1.1 – 3.4 wt.%; average 1.7 wt.%). The compositional range of the other major elements is: CaO (0.5 – 2.9 wt.%; average 1.0 wt.%), Al 2 O 3 (11.0 – 17.8 wt.%; average 13.3 wt.%), Fe 2 O 3 (3.3 – 6.8 wt.%; average 4.32 wt.%), MnO (0.03 – 0.14 wt.%), TiO 2 (0.50 – 0.83 wt.% ) and P 2 O 5 (0.13 – 0.20 wt.%). SiO 2 shows marked negative correlation with Al 2 O 3 , Fe 2 O 3 , MnO and MgO, and moderate negative correlation with CaO, Na 2 O, K 2 O, TiO 2 and P 2 O 5 . The relatively low SiO 2 /Al 2 O 3 ratio of these rocks indicates an im- mature character, as con fi rmed by the Al 2 O 3 /Na 2 O (3.6 – 5.8) and Al 2 O 3 / TiO 2 (16.8 – 23.9) ratios, which lay within the typical range of the upper continental crust (3.9 and 30.4, respectively; Taylor and McLennam, 1985 ). The Fe 2 O 3 /K 2 O ratio can be applied to distinguish between mature and immature compositions of unconsolidated fi ne- to coarse- grained sediments (particularly in arkoses). According to the chemi- cal classi fi cation diagram published by Herron (1988) , most of the protoliths of the metasedimentary rocks of the lower sequence can be classi fi ed as greywackes, but some few samples (Xareira meta- greywackes; Díez Fernández, 2011 ) plot in the boundary with the litharenite fi eld ( Fig. 3 a). According to the K 2 O/Na 2 O ratio, most of the analysed rocks have quartz-intermediate compositions ( Crook, 1974 ). Table 2 shows the results of the REE analyses. The samples have similar REE contents, with Σ REE values ranging between 112 ppm (sample B-22, with marked depletion in LREE) and 213 ppm (the highest values in some samples with the highest contents in La and Ce). The samples also show similar chondrite-normalized ( Nakamura, 1974 ) fractionation patterns ( Fig. 3 b), with a (La/Yb) N ratio ranging be- tween 7.1 and 14.4. The LREE (La – Sm) show a moderate enrichment in relation to the HREE, which display almost fl at patterns with (Gd/Yb) N ratios ranging between 1.0 and 1.7. All the samples show slight but signi fi cant negative Eu anomaly, variable between 0.57 and 0.77 (calcu- lated according to Taylor and McLennan, 1985 ). Eu anomalies in sedi- mentary rocks are usually interpreted as inheritance from the igneous rock source. Despite some differences in abundance between samples, the REE patterns of the greywackes are similar to those of PAAS, which are considered representative of upper continental crust. The analysed samples show an average Th/Sc value of 1.2, larger than the PAAS value (Th/Sc=0.9) and probably suggesting a source area with predominance of felsic rocks. This interpretation is also based in the high La contents in relation to Sc, with average La/ Sc=4.27 above the PAAS ratio (La/Sc=2.4). The same interpretation can be obtained from the Ti/Zr and Zr/Sc ratios, respectively lower and higher than PAAS, and also from the low contents in Cr (average 73 ppm) and Ni (average 26 ppm). Moreover, the La/Th ratio ranging from 2.4 to 4.7 with an average value of 3.5, higher than PAAS (2.6), also suggests the abundance of felsic rocks in the source areas. The relatively high Hf contents (average 6.2; PAAS=5) can be also con- sidered an indication of source areas located in the surrounding area of a passive margin ( Hegde and Chavadi, 2009 ). 199 J.M. Fuenlabrada et al. / Lithos 148 (2012) 196 – 208