Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Mantle plume wikipedia , lookup
Great Lakes tectonic zone wikipedia , lookup
Late Heavy Bombardment wikipedia , lookup
Geology of Great Britain wikipedia , lookup
Clastic rock wikipedia , lookup
Baltic Shield wikipedia , lookup
Algoman orogeny wikipedia , lookup
Large igneous province wikipedia , lookup
Geochemistry and petrogenesis of extrusive rocks, dykes and high-level plutonic rocks on the island of Oldra, Solund-Stavfjord Ophiolite Complex, western Norway HANNE LONE RYTIV AD, HARALD FURNES, KJELL PETIER SKJERLIE & RINN ROLFSEN Ryttvad, H. L., Furnes, H., Skjeriie, K. P. & Rolfsen, R.: Geochemistry and petrogenesis of extrusive rocks, dykes and high-leve! plutonic rocks on the island of Oldra, Solund-Stavfjord Ophiolite Complex, western Norway. Norsk Geologisk Tidsskrift, Vol. 80, pp. 97-110, Oslo 2000. ISSN 0029-196X. The island of Oldra, part of the Late Ordovician Solund-Stavfjord Ophiolite Complex of the western Norwegian Caledonides, comprises extrusive rocks, a sheeted dyke complex and high-leve! gabbros. The metabasalts are of N-MORB affinity, and their Nd isotopic composition (eNd=+7.8 to +8.4) indicates generation from a relatively homogeneous, bul strongly depleted mantle.Eighty percent of the metabasalts classify as FeTi-basalts (FeO'/MgO >l. 75, Ti02 >2 wt% ), but pseudostratigraphically their distribution varies significantly (51% of the extrusive rocks and 92% of the sheeted dykes). The contents of FeO', MgO, AI203, Cr, Ni and Zr indicate that differences in depth of melting (30 to 12 kbar), crystal fractionation of olivine, plagioclase and clinopyroxene at different levels in the mantle and crust ( <8 kbar) and magma mixing affected the composition of the rocks. The MgO-Zr relations of the established chemostratigraphy of the extrusive rocks indicate that magma mixing caused significant scatter among magma compositions. Hanne Lone Ryttvad, Norsk Hydro ASA, Sandslivn. 90, N0-5049 Bergen, Norway; Harald Furnes, Geologisk Institutt, A/legt. 4/, N0-5007 Bergen, Norway (e-mail. [email protected]); Kjell Petter Skjerlie, Institutt for Geologi, N0-9037 Tromsø, Norway; Rinn Rolfsen, Esso Norge AS, Grensevn. 6, N0-4313 Sandnes, Norway Introduction Field relationships The Solund-Stavfjord Ophiolite Complex (SSOC) occurs in the Upper Allochthon of the Norwegian Caledonides in westernNorway (Furnes et al. 1990), where it crops out on a number of islands and skerries in the Solund-Stavfjord area (Fig. 1). A quartz diorite from within a high-level gabbro provided a U-Ph zircon age of 443 ± 3 Ma (Dunning & Pedersen 1988). The part of the SSOC presented here, from the island of Oldra, represents crust that may have been generated in a propagating rift setting (Skjerlie et al. 1989; Dilek et al. 1997), and its geochemical evolution is dominated by FeTi-basalts, a typical feature displayed at short distances (0-100 km) behind active propagating rift tips (Furnes et al. 1998). The ophiolite pseudostratigraphic units on Oldra are, from bottom to top: vari-textured and fine-to-medium grained gabbro, a sheeted dyke complex and extrusive rocks. The rocks have suffered lower greenschist-facies metarnorphism of ocean-floor and subsequent late Cale donian origin (Fonneland 1997). The aims of this study were to establish the stratigraphy of the extrusive rocks in order to study the magmatic evolution with time, and to propose a petrogenetic model based on field observations and the geochemical evolution of the metabasalts. The distribution of extrusive rocks, sheeted dykes and gabbros, each separated by steeply dipping shear zones striking at approximately 040°, is shown in Fig. l. Extrusive rocks The extrusive rocks comprise pillow lavas, massive lava flows and hyaloclastite breccias, described in detail by Furnes (1972, 1974). The shape of the pillow lavas, convex roofs and cusps at the bases, in addition to well-defined drain-out structures, shows the way up in the volcanic sequence. Dykes appear in close association with extrusive rocks. A mixed zone of dykes and screens of extrusive rocks represents a transition zone between the sheeted dyke complex and the volcanic sequence. The pillow lavas are aphyric to moderately phyric (<l% and 2-10%, respectively). Sheeted dykes The sheeted dyke complex is dominated by parallel to subparallel, regular dykes. Most of the dykes (65%) have not been split and range in width from 1.5 to 190 cm, with an average thickness of 71 cm (Ryttvad 1997). This 98 H. L. Ryttvad et al. NORSK GEOLOGISK TIDSSKRIFT 80 (2000) A 10km � Western Gneiss Region [=:J } Undiff. Precambrian � Caledonian nappes bd Oslo Graben Devonian basinal strata Domain Domain 11 Domain Il l Undifferent1ated Ill D Q J+ +l � � � Sognefjorden Solund-Stavfjord ophiolite Complex and 1ts sed1mentary cover ( Slavenes Group ) Sunnfjord Melange Herland Group ( HG ) Høyvik Group Dalsfjord suite Undifferentiated mylonties and gneisses of the western Gneiss Region. Eclogite-bearing rocks ( e ) are indicated. Nordfjord-Sogn Detachment Fault Other normal faults Extrusive rocks (pillow lavas, massive flows, hyaloclastite breccia) Transition zone Sheeted dyke complex ,4. A � A'-,8 \ �� Vari-textured and fine- to medium-grained gabbro Strike and dip of dyke Strike and dip of foliation Younging direction of pillow lava Profile line Shearzone Sense of shear 500m Fig. l. A. Simplified geological map of part of western Norway, showing the occurrence of the Solund-Stavfjord Ophiolite Complex (SSOC) within the Norwegian Caledonides (modified from Dilek et al. 1997). B. Simplified geological map of Oldra. suggests that pre-existing dykes may have offered struc tural guidance to succeeding dykes, favouring adjoining intrusions (Cann 1974). Where younger dykes split older dykes, they provide a way to study the evolution of magma through time (e.g. Robson & Cann 1982). At least six generations of magma intrusions, revealed by a series of dyke-splittings, occur within the sheeted dyke complex. The dykes are invariably aphyric. High-leve/ plutonic rocks On the basis of textural variations, two types of plutonic Fine to medium FMG16 FMG28 FMG29 FMG18 FMG17 14.10 13.36 13.68 13.38 13.84 5.16 14.14 4.38 4.12 4.86 14.70 4.67 4.61 4.66 4.92 4.34 4.58 14.50 4.19 4.44 3.96 Dykes from the sheeted dyke complex 12.33 47.94 3.05 0171 13.21 2.51 49.07 DS 2.78 12.66 48.53 0169 2.86 12.58 48.13 0173 2.71 12.73 48.45 0172 13.82 2.57 47.71 013 12.79 48.29 2.69 0181 12.87 2.62 48.37 0162 12.87 2.75 48.40 0157 2.71 13.06 48.04 D44 13.01 48.97 2.55 065 13.14 2.57 48.18 055 2.76 13.53 48.34 09 13.91 2.55 47.05 070 13.87 2.44 47.99 054 2.32 13.68 47.94 050 grained gabbro 2.31 49.20 2.26 48.87 2.10 48.77 2.18 47.31 2.31 47.38 3.59 4.21 3.55 3.56 3.15 12.37 13.26 12.93 13.28 14.21 Dykes from the transition zone 2.87 47.53 TZ!02 TZ97 48.16 2.36 47.38 2.56 TZ108 2.22 TZI07 47.93 1.86 47.52 TZ95 3.12 4.80 4.41 4.23 4.09 4.04 14.33 4.20 13.52 3.83 13.13 14.67 13.40 3.36 3.68 12.51 13.48 13.06 14.34 13.08 14.06 15.30 14.04 13.49 13.22 Fe203 Pillow Javas and massive flows 47.98 EK140 2.73 48.63 EK16 2.58 2.51 EK146 48.02 2.41 45.26 EK1 2.37 EK132 47.46 2.30 EKS 47.31 2.83 43.91 EK3 2.24 EK12 47.90 2.48 49.05 EK145 48.00 2.30 EKI25 Sample Ai203 Si02 8.85 8.35 8.17 9.32 9.72 10.04 n.a. 9.85 10.62 9.21 n.a. 9.34 8.73 9.02 8.83 8.51 8.92 n.a. 9.96 8.48 8.49 10.72 8.56 10.22 9.00 8.10 10.31 n.a. 8.92 n.a. 9.30 n.a. n.a. n.a. 8.38 8.06 FeO 0.14 0.18 0.17 0.19 0.19 0.26 0.25 0.32 0.32 0.28 0.26 0.27 0.23 0.23 0.21 0.25 0.25 0.27 0.25 0.21 0.24 0.23 0.22 0.24 0.22 0.20 0.25 0.20 0.22 0.20 0.21 0.20 0.21 0.22 0.20 0.20 MnO 6.84 6.87 6.89 7.01 7.02 5.54 6.13 6.20 6.30 6.33 6.41 6.44 6.48 6.58 6.61 6.67 6.69 6.78 6.98 7.09 7.11 6.42 6.56 7.09 7.15 7.80 6.12 6.21 6.37 6.61 6.73 6.76 6.77 6.77 7.14 7.17 MgO 9.07 10.47 10.46 9.66 8.96 9.38 8.93 9.28 8.49 9.65 9.21 9.01 9.87 9.48 9.88 9.23 10.21 8.93 8.30 9.75 10.00 10.03 9.37 9.70 10.69 11.89 10.13 10.57 10.78 10.67 11.39 10.90 9.83 11.51 7.96 12.02 CaO 3.57 2.85 3.15 3.08 3.17 3.17 3.03 3.18 3.40 3.14 2.88 3.09 3.33 3.47 3.18 3.56 3.03 3.26 3.18 3.13 3.07 2.48 2.79 2.75 2.55 2.14 2.48 3.32 2.82 2.53 2.37 2.66 2.29 2.20 3.84 2.48 Na20 Representative major and trace element analyses of volcanic rocks, dykes and gabbros from Oldra. Ti02 Table l. 0.12 0.12 0.13 0.11 0.12 0.05 0.08 0.05 0.04 0.05 0.02 0.03 0.03 0.04 0.05 0.04 0.05 0.06 0.03 0.04 0.05 0.03 0.03 0.05 0.04 0.03 0.17 0.18 0.09 0.27 0.15 0.12 0.18 0.09 0.07 0.06 K20 0.20 0.22 0.19 0.20 0.22 0.30 0.23 0.24 0.26 0.24 0.24 0.25 0.25 0.25 0.24 0.23 0.25 0.25 0.22 0.21 0.21 0.24 0.17 0.21 0.15 0.13 0.24 0.22 0.22 0.22 0.20 0.20 0.27 0.19 0.21 0.19 P205 1.38 1.01 1.00 1.56 2.04 1.77 2.31 1.54 1.79 1.65 1.99 2.02 1.80 1.52 1.44 1.62 1.55 1.95 2.39 1.93 1.88 2.50 2.81 2.24 2.28 2.04 1.74 1.91 1.88 2.95 1.72 1.73 2.88 1.88 2.41 1.71 LOi 98.90 99.36 99.12 98.23 99.06 98.99 99.89 99.01 98.91 99.30 99.81 98.89 99.19 99.27 99.17 98.98 99.42 100.63 99.01 99.58 98.95 99.01 98.50 98.92 99.07 99.07 98.70 101.63 99.09 98.98 98.81 99.37 99.14 100.44 98.59 99.09 Sum 396 372 336 375 388 457 389 433 445 406 392 427 407 431 429 403 414 431 418 396 397 456 391 420 387 348 412 403 407 400 404 366 448 364 438 388 V 241 145 238 205 174 81 !59 134 151 150 213 161 256 249 191 301 248 186 321 299 299 142 271 221 215 319 95 223 211 328 258 265 238 259 256 297 Cr 67 59 60 73 59 44 51 58 55 53 65 58 66 68 58 65 70 61 89 88 77 57 73 62 50 95 51 67 64 78 74 75 91 83 71 79 Ni 121 157 150 140 121 128 133 121 95 108 147 104 157 119 114 114 126 100 110 125 127 127 147 142 147 119 140 159 146 173 168 130 174 135 130 166 Sr 59 57 55 59 57 84 71 74 79 70 74 72 70 76 72 68 70 77 71 66 62 76 63 69 55 48 77 67 66 69 64 62 79 61 70 59 y 166 155 142 166 156 228 185 201 213 191 186 196 189 199 193 184 180 201 187 173 163 203 167 178 152 114 196 181 176 170 165 153 202 145 173 159 Zr 9 9 7 9 9 Il 8 li Il 9 8 Il 10 12 lO !O 10 8 lO 10 8 10 9 9 9 7 lO 8 9 7 lO 7 9 8 10 lO Nb 8 \0 \0 Q .§ � � -· (5• ;:::- � � � � � ?V:! ;::s � � § -.::> 00 o � � ;<l Cll Cll ..., o Cl i'ii � � t"" z o ;<l Cll � Cl 100 H. L. Ryttvad et al. NORSK GEOLOGISK TJDSSKRIFf 80 (2000) Table 2. Representative rare earth elements and Hf analyses of volcanic rocks (EK), dykes (TZ and D) and gabbro (FMG) from Oldra. The numbers in brackets are extrapolated. Sample La Ce Nd Sm Eu Gd Th Ho Yb Lu Hf EK 3 EK 140 TZ 9S TZ 108 D 70 FMG 18 21 12.6 8.7 16.6 13.1 9 [31] 21.8 [16] [26] 18.S 23.7 l S.l 14.8 8.S 10.7 10.2 [13] S.S S.4 2.9 4.9 3.7 4.4 1.09 0.83 0.74 0.6 0.79 1.08 8.4 8.7 [4.5) [7) [4.2] [S.S] [1.4] [l.S] 0.9 [1.3] 0.8 [l] 2 2.2 [1.4] 1.8 [1.3] l.S 4.8 S.2 3.8 4.3 [3.4] 3.6 0.9 l 0.7 0.9 0.6 0.7 2.9 4.3 2.7 3.7 4.S 3.4 rocks can be distinguished: fine- to medium-grained gabbro (FMG) and vari-textured gabbro (VTG). The aphyric, FMG occurs as irregular bodies intruded by regular, 10 to 40 cm-wide, and irregular, 0.5 to 15 cm wide, mafic dykes. Some of the dykes are rooted in irregular gabbro bodies. This close, rooted association of dykes and irregular magma bodies has been interpreted as the root zone of a sheeted dyke complex (Pedersen 1986; Nicolas & Boudier 1991). The VTG varies in grain size from medium grained to pegmatitic. Primary plagioclase, clinopyroxene and mag netite are partly preserved, and secondary minerals are actinolite, epidote, chlorite and leucoxene. Plagioclase rich layers within the VTG indicate that it partly represents a cumulate rock. The geochernistry of the VTG will therefore not be considered here. Samples were dissolved in a rnixture of HF and HN03• REE were separated by specific extraction chromatogra phy using the method described by Pin et al. (1994). Sm and Nd were subsequently separated using a low-pressure ion-exchange chromatographic set-up, with HDEHP coated Tefion powder as the ion-exchange resin (Richard et al. 1976). Sm and Nd were loaded on a double filament and analysed in static mode and their concentrations were 5 deterrnined using a mixed 1 CNd/149Sm spike. Nd isotopic ratios were corrected for mass fractionation using a 146Nd/144Nd ratio of 0.7219. Repeated measurements of 3 the JM Nd-standard yielded an average 14 Nd/144Nd ratio of 0.511113 ± 15 (2a) (n = 62). The typical Nd blank leve! in the laboratory is 5 pg. Geochemistry Analytical techniques Major- and trace-element analyses were performed on an X-ray fiuorescence spectrometer (XRF) at the University of Bergen. The glass-bead technique of Padfield & Gray (1971) was used for the major elements and pressed powder pellets for the trace elements, using international basalt standards with recommended or certified values from Govindaraju (1994) for calibration. The rare earth elements (REE) and Hf were analysed by instrumental neutron activation analysis (INAA) using a coaxial HP Ge detector and the analytical methods described by Brunfelt & Steinnes (1969). Sm and Nd isotopes were analysed on a Finnegan 262 mass-spectrometer at the University of Bergen. All chemical processing was carried out in a clean-room environment with HEPA filtered air supply and positive pressure. The reagents were either purified in two-bottle Tefion stills or passed through ion-exchange columns. Table 3. The geochemical study is based on major- and trace element analyses of the extrusive rocks, the transition zone dykes, the sheeted dykes and the FMG (collectively referred to as metabasalts). A total of 193 samples (54, 18, 102 and 19 from the volcanic zone, the transition zone, the dyke complex and the FMG, respectively) were analysed. Of these, six representative samples were analysed for REE, Hf, and Sm and Nd isotopes. A selection of representative major- and trace-element analyses are shown in Table l. The REE and Hf analyses are presented in Table 2, and Sm-Nd isotopes in Table 3. Alteration effects For co-magmatic metabasaltic rocks, Coish (1977) pro posed that elements which systematically co-vary with Zr have not been mobilized to a great extent. For the Oldra metabasalts, Ti02 displays nearly linear, and Fe2031 and MgO relatively good co-variation with Zr. Si02, A}z03, Representative Nd isotope analyses of volcanic rocks (EK), dykes (TZ and D) and gabbro (FMG) from Oldra. Sample Sm ppm Nd ppm 147Sm/144Nd EK 3 EK 140 TZ 9S TZ 108 D 70 FMG 18 6.92 6.Sl 4.02 S.82 6.0S S.l8 20.40 19.11 11.13 17.14 17.78 1S.62 0.21 0.21 0.22 0.21 0.21 0.20 143Nd/144Nd O.S l 3086 +10.51307S +10.513132 +10.513103 +10.513088 +1O.S13063 +1- s 6 8 9 s 6 eNd (1=440) 8.09 7.84 8.2S 8.42 8.11 7.92 NORSK GEOLOGISK TIDSSKRIFT 80 (2000) 100 r-------� 3 .. o-r----;::====::::=:--� o Extrusive rocks Transition zone dykes • Gl .:= .. 'O 5 2,5 c3 10 .... .li: u a_ --o-EK3 --Å-070 1 Ce Nd o Sheeted dykes o • Fine- to mediurn-grained gabbro (FMG) Ferrobasalt ��==�==��==+=�--��- La o -o-EK140 �TZ108 -FMG18 -+-TZ95 Sm Eu Gd Tb Ho Yb Lu o -------- • �o � ---- 10 r-------� B o m a: N-MORB o :::E .... .li: g --o-EK3 a: --Å-070 0,1 Sr Nb Ce P.Os Zr Hf o • o o 1,0+-----+----+--l 7,5 8 6,5 7 5,5 6 MgOwt% -D-EK140 �TZ108 -FMG18 -+-TZ95 101 Solund-Stavjjord Ophiolite Complex Sm Ti02 Y Yb Cr Fig. 2. Chondrite-normalized (A) and MORB-normalized (B) multi-element dia grams of representative samples from the extrusive rocks, dykes and gabbros. Chondrite-normalization values after Haskin et al. (1968), and MORB-normali zation values after Pearce (1980) (in ppm, unless otherwise stated: Sr=120, Nb=4, Ce=10, P205=0.12 (wt%), Zr=90, Hf=2.4, Sm=3.3, Ti02=1.5 (wt% ), Y=30, Yb=3.4, Cr=250). Normalized values indicated for Ce, Nd, Gd, Tb, Ho and Yb, for which no data are given in Table 2, are extrapolated. CaO, Cr andNi co-vary poorly with Zr, a phenomenon that may be due to alteration and metamorphism, as well as to petrogenetic processes (as will be discussed below). Experimental studies of water-basalt interactions over a range of different pressure and temperature conditions show that Fe, Ca and Si may be removed from basalts, and Na and Mg may be added, whereas Al, Ti, Zr, Cr and Ni are relatively immobile (e.g. Thompson 1973; Coish 1977; Staudigel & Hart 1983). Thus, we assume that the composition of the rocks reftects that of the original magma. Classification Chondrite-normalized REE diagrams and MORB-normal ized multi-element diagrams for the metabasalts of Oldra are presented in Fig. 2. Except for a slight to moderate enrichment of La, the REE define a flat pattem, mostly 15 to 35 times chondritic values. The samples also show small to moderate, negative Eu anomalies, which suggest plagioclase fractionation. The MORB-normalized multi element diagram defines relatively horizontal pattems with ratios from l to 3. This suggests that the rocks represent melts with compositions similar to moderately fractionated N-MORBs. Eighty per cent of the metabasalts classify as FeTi basalts (Fig. 3) by using the criteria Fe01/Mg0 > 1.75, which separate ferrobasalts (Fe01/Mg0 > 1.75, Ti02 > 2 Fig. 3. MgO-FeO'/MgO for the metabasalts of Oldra. The horizontal, dotted line FeO'!MgO=1.75 separates ferrobasalts (above) from N-MORBs (below) (from Sinton et al. 1983). wt%) from N-MORBs (FeOt/MgO < 1.75) (Melson et al. 1976; Sinton et al. 1983). The remaining 20% are of N-MORB composition. Of the extrusive rocks, 51% are FeTi-basalts, whereas 83% of the transition zone dykes, 92% of the sheeted dykes and 79% of the FMG are of FeTi basalt composition. The rocks of FeTi-basalt composition have FeOt/MgO ratios from 1.75 to 2.65 and Ti02 contents ranging from 2.1 to 3.1 wt%. The metabasalts ofN-MORB composition have Fe01/Mg0 ratios from 1.39 to 1.74, and Ti02 contents of 1.9 to 2.6 wt%. Bulk rock composition The MgO content of the metabasalts ranges from 5.6 to 7.9 wt%, with the majority (93.8%) in the 6.1 to 7.3 wt% interval (Fig. 4). The concentrations of other elements vary significantly at a given MgO content. The regression lines (least squares linear) of the major and trace elements, all definin � poor fits, are shown in Fig. 4. The best regression lines (r between 0.225 and 0.383) are obtained for Ti02, Fe203\ P205 , Y and Zr, whose concentrations increase with decreasing MgO content (Fig. 4). The concentrations of A1z03, Cr and Ni decrease with decreasing MgO content, and the elements show less good � results (� between 0.194 and 0.283). Plots of Si02 and CaO versus decreasing MgO content show a relative scatter, but the regression lines indicate an overall slight increase in the Si02 concentration and decrease in the CaO concentration with decreasing MgO content. Chemical variations in bulk rock composition with time The chemical data for the extrusive rocks versus strati- 102 H. L. Ryttvad et al. NORSK GEOLOGISK TIDSSKR!Fr 80 52 T-------, 4 -r-------. no. wt"J. 32._:>�� 0 • o 2 r"= 0.03 5 16 0 r' = 0.225 9�� r" = 0.361 100 90 80 7 -rP-�� Crppm a> i -- 0 -- ::, otP ro ::.e� o o � 80 50 .... 40 5,5 � r" = 0.026 % o o o • o • o Nippm 120 . 6,5 Yppm o 250 200 o o�-JI• =0.354 6 o .... =0. 222 • :2-. Å o o �o��._ 11 0,2 o o•o CaO wt"k 13 200 Å • 11 0,3 300 o ; 15 f?._"_t,� � -- r" =0.194 400 • -�- Fe,o.• wt"l. 13 0, 1 • r' = 0.383 17 o (2000) 7 MgO wt«'.tO 7,5 150 Zrppm o_�l&� o o () fl%5 100 8 o o • .:> • -�- .... =0.351 5,5 6 6,5 7 MgO wt«'.tO 7,5 8 Fig. 4. Bowen variation diagrams showing MgO versus selected major and trace elements for extrusive rocks (white diamonds), transition zone dykes (lilled trian gles), sheeted dykes (white circles), and FMG (lilled squares). The dotted lines are regression lines and il indicates the lit of the regression lines to the data points. Best lit is il l, while il O indicates no correlation. = = graphic height (decreasing age) are shown in Fig. 5. Several elements show co-variation throughout the profile. One group that co-varies is that of Cr and Ni, and to some extent Ah03. A second group is Ti02o Fe203\ P205, Y and Zr. The two groups display concentration variations that plot in zigzag pattems which are relatively good mirror images of each other. Diagrams demonstrating the correlation between Ah03 and Ni, Cr and Ni, Zr and Ti02, and Zr and Y for the extrusive rocks and the dykes are shown in Fig. 6. The Si02 concentration is nearly constant throughout the profile, whereas the MgO and CaO contents display distinct trends toward increase and decrease, without good correlation, as also demonstrated in Fig. 4. Sm and Nd isotope systematics Six samples that span the Cr and Zr range of the metabasalts were analysed for Sm and Nd isotopes. The NORSK GEOLOGISK TIDSSKRIFT 80 Profile (2000) Solund-Stavjjord Ophiolite Complex 103 0�-+-r�--���T-��--r-�-+����_L-+--���J_��L---�-r�� 12 14 45 47 49 2,0 2,5 3,0 13 14 15 9 11 0,20 0,25 150 250 6,5 7,0 7,5 65 85 60 70 150170190 SiO. Al203 Fe•O.' MgO CaO p,o, Cr Ni y Zr Fig. 5. Selected major and trace elements versus stratigraphic height (decreasing age) for the extrusive rocks. The profile shows the abundance and distribution of pillow Javas (open rectangles), massive tlows (filled rectangles) and hyaloclastite breccias (dotted rectangles). Sm/Nd ratios show a narrow range, from 0.3316 to 0.3612, and eNd va1ues from +7.8 to +8.4 (Table 3). Discussion Source region The similarities observed in the multi-element diagrams for the Oldra metabasalts (Fig. 2) indicate that these rocks represent melts that derived from a mantle source region which, compositionally, was relatively homogeneous. The relative flat pattems without sign of negativeNb anomalies argue for a typical MORB source, and the slightly negative Eu anomalies and generally lower Cr values than average N-MORB show that the melts are moderately fractionated. The Nd isotopic data (Table 3) show only minor variations in the eNd (t=440 Ma) values (+7.8 to +8.4) and a common source is thus strongly suggested. In Fig. 7 the eNd data of the SSOC metabasalts are shown in relation to the depleted mantle growth curve of Nelson & De Paolo (1985). They all plot above the growth curve, showing that the rocks represent melts that were generated from a depleted mantle. Petrogenetic processes Langmuir et al. (1992) showed that differences in the extent and depth of fractional melting cause variations in the geochemical compositions of the primitive melt. Magmas generated by fractional melting at progressively higher pressures become increasingly enriched in FeO and MgO and Partial metting and crystal fractionation. - depleted in Na20, and crystal fractionation at different levels in the mantle causes further variations in the chemistry of the melts. The melts that gave rise to the metabasalts of Oldra were modified by crystal fractionation (Fig. 4). The co-variation of MgO and Ni reflects fractional crystallization of olivine and the decrease in A}z03 concentration with decreasing MgO content may indicate crystallization and fractiona tion of plagioclase. This is supported by the Ni versus A}z03 relationships shown in Fig. 6a. The decrease in Cr concentration with decreasing MgO content may indicate fractional crystallization of clinopyroxene. However, with decreasing MgO content, the Cr content decreases sig nificantly, while CaO shows a minor decrease (Fig. 4), which is not compatible with clinopyroxene fractionation only. The MgO-CaO relations show poor correlation; hence, only parts of the data are likely to be attributed to clinopyroxene fractionation. It is thus 1ike1y that the decrease in the Cr content with fractionation (e.g. Fig. 6b) could partly be related to fractional crystallization of Cr-spinel. We thus conclude that the primary liquids were dominantly modified by fractional crystallization of olivine, plagioclase and Cr-spinel, i.e. typical mineral phases on the N-MORB liquidus at low pressures, and olivine, plagioclase and clinopyroxene at higher pressures. To obtain information on the pressures at which the Oldra metabasalts were generated, and at which pressures fractional crystallization took place, the compositions of pooled melts generated at 12, 20, 30 and 40 kbar were considered (Langmuir et al. 1992). The Alz03 and Fe01 wt% concentrations for the Oldra metabasalts in the 6-8 wt% MgO range are shown in Fig. 8. The chemical data in Fig. 8a, b are displayed in diagrams showing the liquid 104 H. L. Ryttvad et al. 100 NORSK GEOLOGISK TIDSSKRIFT 80 90 100 <>o<> A 80 80 z 70 60 z o 40 30 60 14 13 12 11 AI203 15 30 16 • 150 50 90 c 3 .� 40 • 3,5 N 70 50 50 o i= B 90 o (2000) 250 Cr 350 o 80 70 2,5 > 60 2 50 o 1,5 <) 50 o 40 150 100 Zr Fig. 6. 200 30 250 eDykes � 50 o Extrusives 100 150 Zr 200 250 Selected element-element diagrams for Al203-Ni, Cr-Ni, Zr-Ti02 and Zr-Y. lines of descent (LLD) for a melt generated at rv20 kbar. The Fe01 versus MgO data from the Oldra plot within the LLDs are defined by fractional crystallization at l atm and 8 kbar (Fig. 8b). It should be made clear that the mineral proportions are markedly different at different pressures. The proportions of plagioclase:olivine:clinopyroxene used to construct the LLDs at MgO contents comparable with the metabasalts on Oldra, are hence 1:0.27:0.56, 1:0.23:0.7, and 1:0.25:5.25 at l atm, 4 kbar and 8 kbar, respectively (Langmuir et al. 1992). This indicates that melts generated at rv20 kbar and modified by crystal fractionation at pressures between 8 kbar and l atm reproduce the Fe01 versus MgO data from Oldra. How ever, crystal fractionation of magma generated at rv20 kbar cannot explain the scatter in the Al203 data at a given MgO content, because crystal fractionation at different pressures of magma generated at rv20 kbar results in coinciding LLDs for Ah03 at MgO < 8.5 wt% (Fig. 8a). The Ah03 versus MgO data indicate that magmas were generated by fractional metting over a range of pressures. Melts generated at 12 kbar reproduce the Fe01 versus MgO data when modified by crystal fractionation in the pressure range rv4 to rv8 kbar (Fig. 8c). A melt generated at 30 kbar must fractionate in the pressure range rv4 kbar to l atm to reproduce parts of the data (Fig. 8d), and a melt generated at 40 kbar only touches the Fe01 versus MgO data when fractionating at rv l atm (Fig. 8e). This shows that the metabasalts of Oldra probably represent melts generated at a range of pressures in the mantle, that were subsequently modified by crystal fractionation at various levels in the mantle and crust. Mixing of magmas. - To illustrate that the rocks, ranging from N-MORB to FeTi-basalt compositions, may repre8,5 • 8 ...... o .., .., -- ... � 7,5 7 Depleted l I mantle • • l �------�,�C�H�U�Rr-L--� 6,5 6 400 420 440 460 480 500 Age (Ma) Fig. 7. eNd and model depleted mantle growth curve of Nelson & DePaolo (1985), onto which the metabasalts from Oldra are plotted. NORSK GEOLOGISK TIDSSKRIFT 80 17 16 � 15 � o"'14 ;i13 12 (2000) Solund-Stavfjord Ophiolite Complex a ,..,. melts from 20 kbar 17 15 � ,..,.� 13 � 11 r-e b .._ Fe0'1Mg0=1.75 melts from 20 kbar 9 7 17 15 � i13 �11 r-e c � 9 melts from 12 kbar 17 15 � i13 �11 d � � 9 melts from 30 kbar 17 15 � i13 �11 e � � 9 7 melts from 40 kbar 6 7 8 9 MgOwt% Fig. 10 11 12 105 sent mixed magmas, the concentrations of the compatible elements Cr and Ni are plotted versus the concentration of the incompatible element Zr (Fig. 9). The two fractionation curves are calculated based on data from Tviberg (Skjerlie 1988), which represent the most primitive high-MgO basalts and the most fractionated FeTi-basalts (Fig. 9). The fractionation curves have been calculated from an assumed primary magma, and the distribution coefficients used for the modelling are shown in Table 4 (for further details, see text Fig. 9). Compared to the Tviberg metabasalts, whose Cr and Zr contents range from approximately 30--500 ppm to 50-300 ppm, respectively, those on Oldra show relatively limited Cr, Ni and Zr concentrations. They plot above the calculated fractionation curves, and display scatter that is not consistent with fractional crystallization alone. The element concentrations of the rocks can be explained by mixing of different magmas, creating hybrid compositions (Rhodes et al. 1979). Samples lying above the mixing line can be explained by mixing of magmas whose composi tions extend beyond the chosen maximum values of Cr, Ni and Zr (Fig. 9). It should be noted, however, that mainly samples in the Ni-Zr diagram plot above the mixing line. This feature can be explained by the presence of small amounts of olivine phenocrysts in the rocks, observed as pseudomorphs in thin sections, that have increased the Ni content considerably (see Table 4). As discussed above, the scatter in the Al203 and Fe01 data can, to a significant extent, be attributed to melt generation at different pressures. If the total degree of partial melting differs between various batches of compo site melts, this will have an effect on the content of incompatible elements. Thus, the smaller the degree of partial metting, the higher the content of an incompatible element, such as for example Zr (e.g. Cox et al. 1979). If we, on the other hand, make the assumption that the total degree of partial melting represented by each batch of composite melt is comparable, then the scatter displayed by the Cr-Zr and Ni-Zr diagrams can largely be explained in terms of mixing of liquids. Sparks et al. (1980) showed that magmas with approxi mately 9-10 wt% MgO occupy a density minimum compared with FeO-rich, evolved magmas like FeTi basalts. Thus, primitive magmas are more buoyant and can physically mix with more fractionated magma. However, the ability of two magma batches to mix is controlled by the density gaps between the melts; a large gap makes mixing difficult. Thus, if a magma batch is relatively primitive, substantial mixing with a more evolved liquid can be minimal (Huppert et al. 1986). Consequently, a relatively primitive magma batch may ascend rapidly and may be the first to erupt, followed by denser, more evolved magma. 8. The MgO versus Ah03 (A) and FeO' (B toE) data for the Oldra metabasalts (193 samples) displayed in diagrams showing the calculated liquid lines of des cent (LLD) for melts generated by fractional melting at 20 kbar (a and b), 12 kbar (c), 30 kbar (d) and 40 kbar (e), and subsequently modified by fractional crystal lization of olivine, plagioclase and clinopyroxene at l atm, 4 and 8 kbar (moditied from Langmuir et al. 1992). The dotted lines in (a) represent calculated LLDs for melts generated by fractional melting at pressures higher and lower than 20 kbar, which are subsequently modified by crystal fractionation of clinopyroxene at l atm, 4 and 8 kbar. The diagonal line FeO'/MgO 1.75 separates ferrobasalts (above, FeO'!MgO > 1.75) from N-MORBs (below, FeO'!MgO < 1.75). = H. L. Ryttvad et al. 106 b 2,335 t',, .... l .... '' ........ \ 3 300 \ .... ...... .... . � �·. •-=- . Mixing line . . '.... 2 • _ Fractionation curve .... .... ....... .... .... - - -- .('"' e "' ........ .... .... � .�· .. .... - ----- 1 ":.� a Crppm b �.... l 1 3\ \ e Cl. Cl. .. N 200 100 . ......... \ '' .... .... ' '-- B Mixing line . ........ � • - � ........ � --• Fractionatio 2 ', . l . • � ........... �urve _ . . .... ....� .... .......... --.,------'-a Cr versus Z:t (A) and Ni versus Zr (B) for the Oldra metabasalts. The fractionation curves are calculated from a primitive magma containing 900 ppm Cr, 350 ppm Ni and 70 ppm Z:t. Between the assumed primary magma and 1: 16% fractionation of olivine and cbromite; between l and 2: 40% fractionation of olivine (0.16) + plagioclase (0.84) + cbromite (trace); between 2 and 3: 45% fractionation of olivine (0.16) + plagioclase (0.38) + clinopyroxene (0.48). a and b represent primitive high-MgO basalt and FeTi-basalt, respectively, from Tviberg. Modified from Skjerlie (1988). Partition coefficients used in calcula tions are from Cox et al. (1979) and Pearce & Norry (1979). 9. o o o 2,315 oo o o � 2,305 " A A 2,300 1,30 , , •• 00 2,20 1,90 1,60 line FeO'!MgO = 1.75 separates ferrobasalts (right) from N-MORBs (left). El 250 � -7 Modell 250 • c. c. .. N inpatmagma ----. � � input magma 250 • • 150 Model3 • l intrudedl erupted magma • • - El c. c. .. N Model4 250 " 150 intrudtd/ eruptedmagma El c. c. .. N 250 150 " o <> . e cbamber magma l ;- Olivine Clinopyroxene Plagioclase Z:t o.o1• 0.1• om• • From Pearce & Norry (1979). From Cox et al. (1979). b Cr Ni 5,5 6.5 MgOwt% 7,5 l MgOwt% l � \ increase Zrppm l t-----1 t-----1 MgOwt% Zrppm \ MgOwt% i ,.-_. rncrease MgOwt% Input magma 4,5 Distribution coefficients used in fractional crystallization. Zrppm t-----1 -�. ' MgOwt% . rncrease Mode15 't(-.1 ,, t-----1 ,.-_. rncrease 0e l t-----1 . 1ncrease ? intruded/erupted magma El o • .... ..... 150 \ 1ncrease chamber m•gm• c. c. � • • 150 magma El Modell " c. c. 50 Mineral 2,80 2,50 FeO'/MgO chamber magma Iron-rich, differentiated basaltic liquids are rare arnong erupted rocks but are found in small quantities at divergent plate margins (Brooks et al. 1991). Brooks et al. (1991) suggested that erupted N MORB may represent a mixture of primitive liquid and differentiated, iron-rich liquid which exists at depth, but normally does not reach the surface because of its high density. Pseudostratigraphically, there is a significant difference in the distribution of the FeTi-basalts on Oldra. Fifty-one percent of the extrusive rocks are FeTi-basalts, whereas Table 4. A o 2,320 Nippm Emplacement of FeTi-basalts. 6> 83% of the transition zone dykes and 92% of the sheeted dykes are of FeTi-basalt composition. Liquid densities at 1300°C, using partial molar volumes (Bottinga et al. 1983; Lange & Carmichael 1987) have been calculated (anhy- • 0 +-----+-----�--�--� 80 40 60 100 20 120 o Fig. o Fig. JO. The FeO'/MgO ratios versus density for the extrusive rocks (0), transi tion zone dykes (A), sheeted dykes (0) and FMG ( . ) . The vertical, dotted / .... 2,325 � 2,310 ---+---�--� .... 0 +--300 200 100 400 o 500 300 o = .. .... (2000) o 2,330 A \:��� � .... ..... 100 NORSK GEOLOGISK TIDSSKRIFr 80 ,.-_. 1ncrease \ t-----1 increase Zrppm i . rncrease Zrppm 8.5 Fig. Il. The MgO-Z:t relations (l to 5) that occur in ascending order from one sample to another in the extrusive rock profile are shown by solid arrows on the right side of the diagram. To the left, the respective processes that cause varia tions (replenishment of magma, crystal fractionation and/or mixing) are illu strated. Curved, dotted arrows crystal fractionation; dotted lines chords connecting two mixing magmas. The hybrid magma produced by magma mix ing is situated on the chord. The location of the hybrid on the chord depends on the quantity of the two magmas that mix. = = NORSK GEOLOGISK TIDSSKR!Ff 80 (2000) Solund-Stavjjord Ophiolite Complex A ,�-+-----.... - - - - - � -1 atm B o ExtJUsivO � 107 rockS --- -1 -1 -2 30 10 ������ -3 -1 .5 1 00 -30 -o Distance from spreading centre (km) -o - 15 . Distance from spreading centre (km) 5 - Fig. 12. Cartoon depicting a petrogenetic model for the Oldra metabasalts. A. Solid circles are melts generated by fractional metting at pressures from 30 to 12 kbar. Solid arrows indicate upward transport and crystal fractionation of olivine, plagioclase, clinopyroxene and minor Cr-spinel from melts at pressures <8 kbar. During ascent, individual batches of magma mix, as indicated by the ovals. B. Enlargement of part of A. Melts of FeTi-basalt (dotted arrows) and N-MORB (solid arrows) composition mix. Ovals indicate areas of magma mixing and crystal fractionation. Low-density N-MORBs (solid diapir in oval) may ascend rapidly, without sub stantial mixing with high-density FeTi-basalts. drous) for the Oldra metabasalts, and show good correla tion with the Fe01/Mg0 ratios (Fig. 10). The higher density of the FeTi-basalts compared with the N-MORBs affects the ability of the magma to erupt, and may explain the lower abundance of FeTi-basalts in the extrusive rock sequence compared to the sheeted dyke complex. Petrogenetic processes and time The stratigraphically sampled extrusive rocks (Fig. 5) provide a good opportunity to study the geochemical evolution of the magmas with time. In this account, the concentrations of MgO and Zr are considered, and the presence of magma chambers is assumed. Based on the behaviour of these two elements, five models will be shown below. The petrogenetic processes considered are crystal fractionation of olivine, plagioclase and clino pyroxene, and magma mixing. In Fig. 11 we show the five models that may account for the different MgO-Zr relationships. In model l, MgO decreases and Zr increases owing to crystal fractionation of the magma. In model 2, MgO increases and Zr decreases owing to mixing of the chamber magma with a more primitive magma. In model 3, both MgO and Zr increase because of magma mixing. In model 4, both elements decrease owing to magma mixing, and in model 5, the MgO and Zr contents are relatively constant. The latter is caused by crystal fractionation of magma l to magma 2. Prior to intrusion/eruption, chamber magma 2 mixes with a more primitive magma, and the hybrid magma that intrudes/erupts has a composition relatively sim ilar to that of chamber magma l. Thus, the MgO-Zr relations shown in Fig. 11 indicate that crystal fractionation was an important process modifying the chamber magmas. How ever, magma mixing, in which the compositions of the chamber magmas and the input magmas may have been considerably different, seems to have been the most important modifying process affecting the chemistry of the intruded and erupted magmas. Petrogenetic model Grove et al. ( 1992) discussed three petrogenetic models for MORB genesis. Model l - Partial melting and later fractional crystallizationlmixing events are separated by an aggregationlmixing step; Model 2 - Fractional melts are moved to a shallower, cooler mantle, where they are modified by differentiation befare aggregationlmixing beneath the spreading centre; and Model 3 - Fractional melting, aggregationlmixing and differentiation occur throughout the upper mantle, and the pressure ranges of aggregationlmixing and differentiation overlap. Aggrega tion of melts may, thus, occur at many levels during the melt production and modification process, i.e. in the region of melt production and in the region of melt modification. The separation of melt production and melt modification processes, as suggested in Model l, may represent an oversimplification of the more complex Model 3. Further more, Model 2 may be viewed as an end member case. Grove et al. ( 1 992) concluded that Model 3 is the most likely to occur because it closely approximates the way in 108 H. L. Ryttvad et al. NORSK GEOLOGISK TIDSSKRIFT 80 which melt modification by fractional crystallization fits into the overall evolution of MORB magmas. We suggest a petrogenetic model for the rocks of Oldra (Fig. 12) that is fairly similar to Model 3 of Grove et al. (1992). Accumulated fractional melting occurred at pressures from ""'30 to ""'12 kbar from a strongly depleted mantle (Fig. 12A). Furthermore, on their way to the surface the melts were modified by fractional crystallization of olivine, plagioclase, clinopyroxene and minor Cr-spinel and by mixing, the extent of mixing depending on the densities of the magmas. Low-density N-MORBs erupted more commonly than high-density FeTi-basalts, and therefore constitute "'50% of the extrusive rocks, whereas the FeTi-basalts constitute the majority of the transition zone dykes and the sheeted dykes (Fig. 12B). In the crust, fractional crystallization of plagioclase and clinopyroxene generated cumulates, as represented by the VTG (Fig. 12A). The gabbro-dyke transitions observed in the FMG are interpreted to be the root zone of the sheeted dyke complex and the roof zone of the inferred chamber(s). S ummary The island of Oldra, which exposes part of the Late Ordovician Solund-Stavfjord Ophiolite Complex (SSOC), contains the following units: vari-textured and fine- to medium-grained gabbro, metabasalts defining a sheeted dyke complex, and extrusive rocks. On the basis of detailed field and geochemical studies, the following conclusive remarks can be made: l. The rocks are of N-MORB affinity. The majority classify as FeTi-basalts (FeOt/MgO > 1.75, Ti02 > 2 0 wt% ), resulting partly from crystal fractionation of olivine, plagioclase, clinopyroxene, and minor Cr spinel from N-MORB type melts. 2. Pseudostratigraphically, there is a significant difference in the distribution of the FeTi-basalts. Approximately 50% of the extrusive rocks are FeTi-basalts, whereas ca. 90% of the sheeted dykes are of FeTi-basalt composi tion. This difference is explained by the higher density of the FeTi-basalts compared with the less fractionated N-MORBs. 3. The Nd isotopic data show minor variations in the eNd (t = 440 Ma) values (+7.8 to +8.4), and a homogeneous, strongly depleted source is suggested. 4. The FeO\ MgO and Ah03 data suggest that the metabasalts were generated by partial melting at pressures from ""'30 to ""'12 kbar. The melts were subsequently modified by fractional crystallization at a range of pressures (probably <8 kbar) in the mantle and crust. 5. The Cr-Zr and Ni-Zr relationships of the metabasalts suggest mixing of magma batches that ranged from N MORB to FeTi-basalt compositions. 6. In addition to crystal fractionation, magma mixing was . (2000) apparently an important modifying process affecting the chemistry of the magmas. - Financial support to carry out this study was provided by the University of Bergen and the Research Council of Norway. We !hank Dr Y. Dilek for constructive discussions during fieldwork. We also !hank the reviewers C. Bames, E.-R. Neumann and D. Roberts for constructive criticism. Acknowledgements. Manuscript received January 1999 References Bottinga, Y., Richet, P. & Weill, D. F. 1983: Calculation of density and thermal expansion coefficient of silicate liquids. Bulletin de Mineralogie 106, 129-138. Brooks, C. K., Larsen, L. M. & Nielsen, T. F. D. 1991: Importance of iron-rich tholeiitic magmas at divergent plate margins: a reappraisal. Geology 19, 269272. Brunfelt, A. O. & Steinnes, E. 1969: Instrumental activation analysis of silicate rocks with epithermal neutrons. Annals of Chemical Acta 48, 13-24. Cann, J. R. 1974: A model for oceanic crustal structure developed. Geophysical Journal of the Royal Astronomical Society 39, 169-187. Coish, R. A. 1977: Ocean floor metamorphism in the Betts Cove Ophiolite, Newfoundland. Contributions to Mineralogy and Petrology 60, 277-302. Cox, K. G., Bell, J. D. & Pankhurst, R. J. 1979: The Interpretation of Igneous Rocks. 398 pp. Chapman & Hall. Dilek, Y., Furnes, H. & Skjerlie, K. P. 1997: Propagating rift tectonics of a Caledonian marginal basin: multi-stage seafloor spreading history of the Solund-Stavfjord ophiolite in western Norway. Tectonophysics 280, 213-238. Dunning, G. R. & Pedersen, R. B. 1988: U/Pb ages of ophiolites and arc-related plutons of the Norwegian Caledonides: implications for the development of Iapetus. Contributions to Mineralogy and Petrology 98, 13-23. Fonneland, H. C. 1997: The relationship between hydrothermal history, tectonism, and volcanic products in construction of a part of the Solund-Stavfjord Ophiolite Complex, Western Norway. Cand. Scient. Thesis, University of Bergen, 84 pp. Furnes, H. 1972: Meta-hyaloclastite breccias associated with Ordovician pillow lavas in the Solund area, West Norway. Norsk Geologisk Tidsskrift 52, 385407. Furnes, H. 1974: Structural and metamorphic history of the Lower Paleozoic metavolcanics and associated sediments in the Solund area, Sogn. Norges geologiske undersøkelse 302, 33-74. Furnes, H., Dilek, Y., Skjerlie, K. P., Ryttvad, H. L. & Fonneland, H. C. 1998: Variations in basaltic geochemistry along a propagating rift of the Late Ordovician marginal basin of the West Norwegian Caledonides. Terra Nova, in press. Furnes, H., Skjerlie, K. P., Pedersen, R. B., Andersen, T. B., Stillman, C. J., Suthren, R. J., Tysseland, M. & Garmann, L. B. 1 990: The Solund-Stavfjord Ophiolite Complex and associated rocks, West Norwegian Caledonides: geology, geochemistry and tectonic environment. Geological Magazine 127, 209-224. Govindaraju, K. 1994: 1994 compilation of working values and sample description for 383 geostandards. Geostandards Newsletter 18, Special Issue. Grove, T. L., Kinzler, R. J. & Bryan, W. B. 1992: Fractionation of Mid-Ocean Ridge Basalt (MORB). In Morgan, J. P., Blackman, D. K. & Sinton, J. M. (eds.): Mantle Flow and Me/t Generation at Mid-Ocean Ridges. Geophysical Monograph 7 1 , 281-3 1 0. Haskin, L. A., Haskin, M. A., Frey, F. A. & Wildeman, T. R. 1968: Relative and abso1ute terrestria1 abundances of the rare earths. In Ahrens, L. H. (ed.): Origin and Distribution of the Elements, 889-912. Pergamon Press, New York. Huppert, H. E., Sparks, R. S. J., Whitehead, J. A. & Hal1worth, J. 1986: Rep1enishment of magma chambers with light inputs. Journal of Geophysical Research 91, 6113--{;122. Lange, R. A. & Carmichae1, I. S.E. 1987: Densities of Na20 - K20 - CaO - MgO - FeO - FezOr A120r Ti02 - Si02 1iquidus: new measurements and derived partial molar properties. Geochimica et Cosmochimica Acta 51, 2931-2946. Langmuir, C. H., Klein,E. M. & Plank, T. 1992: Petrological systematics of mid ocean ridge basalts: constraints on melt generation beneath ocean ridges. In Morgan, J. P., Blackman, D. K. & Sinton, J. M. (eds.): Mantle Flow and Melt Generation at Mid-Ocean Ridges. Geophysical Monograph 1 1 , 183-280. Melson, W. G., Vallier, T. L., Wright, T. L., Byerly, G. & Nelson, J. 1976: Chemical diversity of abyssal volcanic glass erupted along the Pacific, Atlantic, and Indian Ocean seafloor spreading centers. In Sutton, G., Manghari, M.H. & NORSK GEOLOGISK TIDSSKRIFf 80 (2000) Moberly, R. (eds.): The Geophysics of the Pacific Ocean Basin and its Margins. Geophysical Monograph 12, 351-367. Nelson, B. K. &DePaolo,D. J. 1985: Rapid production of continental crust 1.7 to 1.9 b.a. ago: Nd isotopic evidence from the basement of the North American mid-continent. Geological Society of America Bulletin 96, 746-754. Nicolas, A. & Boudier, F. 1991: Rooting of the sheeted dike complex in the Oman ophiolite. In Peters, T., Nicolas, A. & Coleman, R. G. (eds.): Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Series in Petrology and Structural Geology 5, 39-54. Padfield, T. & Gray, A. 1971: Major element rock analyses by X-ray ftuorescence - a simple fusion method. NV Philips, Eindhoven, Analytical Equipment FS 35. Pearce, J. A. 1980: Geochemical evidence for the genesis and eruptive setting of Javas from Tethyen ophiolites. Proceedings of the International Ophiolite Symposium, Cyprus 1979, 261-272. Pearce, J. A. & Norry, M. J. 1979: Petrogenetic implications of Ti, 'h:, Y and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology 69, 33-47. Pedersen, R. B. 1986: The nature and significance of magma chamber margins in ophiolites: examples from the Norwegian Caledonides. Earth and Planetary Science Letters 77, 100-112. Pin, C., Briot, D., Bassin, C. & Poitrasson, F. 1994: Concomitant separation of strontium and samarium-neodymium for isotope analyses in silicate samples, based on specific extraction chromatography. Analitica Chimica Acta 298, 209-217. Rhodes, J. M., Dungan, M. A., Blanchard, D. P. & Long, P. E. 1979: Magma mixing at mid-ocean ridges: evidence from basalts drilled near 22°N on the Mid Atlantic Ridge. Tectonophysics 55, 35-61. 143 1 Richard, P., Shimizu, N. & AIIegre, C. J. 1976: Nd/ �d, a natura! tracer: an application to oceanic basalts. Earth and Planetary Science Letters 3 l, 269278. Solund-Stavjjord Ophiolite Complex 109 Robson,D. & Cann, J. R. 1982: A geochemical model of mid-ocean ridge magma chambers. Earth and Planetary Science Letters 60, 93-104. Rytrvad, H. L. 1997: Geochemistry and petrogenesis of extrusive rocks, dikes and high-leve! plutonic rocks in the Solund-Stavfjord Ophiolite Complex, Western Norway. Cand. Scient. Thesis, University of Bergen, 75 pp. Shaw, D. M. 1970: Trace element fractionation during anatexis. Geochimica et Cosmochimica Acta 34, 234-243. Sinton, J. M., Wilson, D. S., Christie, D. M., Hey, R. N. & Delaney, J. R. 1983: Petrologic consequences of rift propagation on oceanic spreading ridges. Earth and Planetary Science Letters 62, 193-207. Skjerlie, K. P. 1988: Geologiske relasjoner mellom gabbro og gangkompleks, petrogenese og tektonomagmatisk modellutvikling i området Tviberg - NV Atløy, Solund-Stavfjorden Ofiolittkompleks. Cand. Scient. Thesis, University of Bergen, 176. Skjerlie, K. P., Furnes, H. & Johansen, R. J. 1989: Magmatic development and tectonomagmatic models for the Solund-Stavfjord Ophiolite Complex, West Norwegian Caledonides. Lithos 23, 137-151. Sparks, R. S. J., Meyer, P. & Sigurdsson, H. 1980:Density variation amongst mid ocean ridge basalts: implications for magma mix.ing and the scarcity of primitive Javas. Earth and Planetary Science Letters 46, 419-430. Staudigel, H. & Hart, S. R. 1983: Alteration of basaltic glass: mechanism and significance for the oceanic crust-seawater budget. Geochimica et Cosmochi mica Acta 47, 37-50. Tamey, J., Saunders, A.D., Mattey, D. P., Wood, D. A. & Marsh, N. G. 1981: Geochemical aspects of back-are spreading in the Scotia Sea and Western Pacific. Phi/osophical Transactions of the Royal Society, London, A300, 263285. Thompson, G. 1973: A geochemical study of the low temperature interaction of sea-water and oceanic igneous rocks. Transactions of the American Geophy sical Union 54, 1015-1019.