Geol. Mag. 138 (3), 2001, pp. 309–323. Printed in the United Kingdom © 2001 Cambridge University Press
309
The youngest Neoproterozoic mafic dyke suite in the Arabian Shield: mildly alkaline dolerites from South Jordan – their geochemistry and petrogenesis G. JARRAR* Department of Geology, University of Jordan, 11942 Amman, Jordan
(Received 6 June 2000; accepted 1 February 2001)
Abstract – The Arabian–Nubian Shield evolved through a sequence of tectonomagmatic cycles, which took place during Neoproterozoic time (1000–540 Ma). Dyke emplacement constitutes one of the conspicuous features of the Arabian–Nubian Shield, with mafic dykes being the most abundant. The investigated dykes represent the youngest Neoproterozoic mafic dykes and have been dated in Jordan at 545 ± 13 Ma. Geochemically the studied dykes are mildly alkaline, are enriched in large ion lithophile elements (LILE) and high field strength cations (HFSC), show moderate enrichment of REE, and lack Nb anomaly. These features are consistent with a predominantly extensional continental tectonic setting. Crystallization temperatures of the suite fall between 1050 and 800 °C to as low as 650 °C as deduced from pyroxene thermometry. The investigated dykes were derived from a metasomatized lithospheric mantle by 5 % modal batch partial melting of phlogopite-bearing spinel lherzolite, according to geochemical modelling. The intra-suite geochemical features are explicable by 64 % fractional crystallization of olivine, pyroxene, plagioclase and titanomagnetite and possibly other accessories like apatite at a later stage. The cumulate produced from this fractionation of the investigated dyke suite contributed to the formation of the mafic lower crust of the Arabian–Nubian Shield. Elemental ratios and petrographic evidence indicate possible minor crustal contamination of the suite. The youngest mafic dykes show striking geochemical similarities to the same generation of dolerite dykes in the adjacent countries, to transitional young basalt suites of the Main East African Rift, and to Quaternary Jordanian basalts. The youngest mafic dyke suite, the rhyolites of the Aheimir suite, and St Katherina rhyolites of Sinai represent the last igneous activity in the Arabian–Nubian Shield before the onset of the Cambrian at about 545 Ma ago.
1. Introduction The Arabian–Nubian Shield formed during Neoproterozoic time (1000–540 Ma); it is considered one of the most voluminous events of juvenile crust formation and extends over an area of about 2 million square kilometres on both sides of the Red Sea (Bentor, 1985; Stern, 1994). The major crust-forming events took place between 900 and 600 Ma and gave rise to voluminous calc-alkaline granitoids (e.g. Bentor, 1985; Jarrar, 1985; Stoeser & Camp, 1985; Stern, 1994). The final stage (600–540 Ma) of cratonization was dominated by NW-trending strike-slip faults (the Najd fault system), strong extension which was accompanied by bimodal magmatic activity and the deposition of molasse type sediments (e.g. Stern, 1985; Jarrar, Wachendorf & Zellmer, 1991; Jarrar, Wachendorf & Saffarini, 1992). A remarkable feature of this stage is the intrusion of dykes ranging from mafic to felsic including composite dykes with a distinct bimodality (e.g. Jarrar et al. unpub. data; Kessel, Stein & Navon, 1998). Six episodes of mafic dykes have been described in Saudi Arabia (Greenwood, 1981) while a maximum of three phases are recognized * Email:
[email protected]
in Sinai, Eastern Desert and South Jordan (Eyal & Eyal, 1987; Friz-Töpfer, 1991; Jarrar et al. unpub. data), other than the three phases of syntectonic dykes (schistose dykes) mentioned by Eyal & Eyal (1987). The latest of these events is the mafic dyke swarm which cross-cuts all basement rocks and is the only dyke generation which cuts the alkaline red granites (Jarrar et al. unpub. data, 568–560 Ma). Saffarini & Jarrar (1991, 1992) carried out geochemical and petrographical investigation of the investigated dyke suite in the area of Wadi Um Salab, Quweira area. They characterized these dykes as alkali within-plate basalts, which formed by 5–15 % partial melting of carbon-bearing peridotite, from a source with accessory amphibole and Fe–Ti oxides. The equivalents of the Jordanian youngest mafic dyke suite were classified as transitional between alkali basalts and tholeiites in the area of Timna (Beyth et al.1995), and transitional to mildly alkaline in Southern Sinai (Friz-Töpfer, 1991). Other authors classified these rocks as having tholeiitic character (Pudlo & Franz, 1994; Kessel, Stein & Navon, 1998). The widespread distribution of this suite of dykes implies that the mantle-derived magmas, which led to their formation, must have had regional proportions. Therefore a detailed study of petrogenesis of this suite
310 may increase our understanding of the closing stage in the evolution of the Arabian–Nubian Shield. Major and trace element compositions as well as mineral chemistry of these dykes are discussed, and a regional correlation is offered. 2. Geological setting The investigated dykes cut the Humrat and Mubarak granites and other alkali feldspar- and syenogranites, and the composite dykes in southern Jordan. The Humrat and Mubarak granites have been dated at 568 and 560 Ma, respectively (Brook, Ibrahim & McCourt, 1990), and the composite dykes in Wadi Rahma have been dated at 575 ± 6 Ma (Jarrar et al. unpub. data). One of the investigated dykes in Wadi Rahma was dated using the K–Ar technique at 545 ± 13 Ma (Jarrar, Wachendorf & Saffarini, 1992). Beyth et al. (1995) obtained two K–Ar ages for two equivalent dykes from Mt Timna with an average of 543.9 ± 10.7 Ma. The same dykes gave a 40Ar/39Ar plateau age of 531.7 ± 4.6 (Beyth & Heimann, 1999) and are unconformably overlain by Cambrian sediments .The rhyolites of the Aheimir suite which also pre-date the Cambrian were also dated using Rb–Sr at 553 ± 11 Ma (Jarrar, 1992). The Aheimir suite is the upper part of the volcano-sedimentary succession that conformably overlies the Saramuj Conglomerate. The ages of 600 and 595 Ma have been assigned to the latter (Jarrar, Wachendorf & Zachmann, 1993). The rhyolites and the studied youngest mafic dykes are both truncated by the Cambrian/Precambrian Unconformity. Therefore, the 545 Ma age seems to be a reasonable estimate for this suite, which is in accordance with the suggested Precambrian–Cambrian boundary at 545 Ma (Tucker & McKerrow, 1995; Landing et al. 1998). The youngest mafic dykes are black to green coloured, range in thickness from 1–25 metres, trend NE to ENE, are medium- to fine-grained, and locally have phenocrysts of plagioclase, pyroxene and pseudomorphs after olivine. The dykes in the studied suite from the Jordanian–Saudi border in the south to Wadi Fidan in the north, share abundant pyrite dissemination that can be used to distinguish them from other mafic dykes. Moreover, the investigated dykes were found to be almost the only dyke generation intruding the younger red granites of the Humrat, Mubarak and the Feinan. Forty-two dykes from the dyke suite under consideration were sampled for the purpose of this investigation in the following areas (Fig. 1): Wadi Shuraiyh (south of Aqaba), Wadi Yutum (east of Aqaba), the area south of Quweira (Wadi Um Salab, Jabal Marsad), Wadi Filk, Wadi Rahma and Jabal Mubarak. 3. Analytical techniques Whole-rock major- and trace-element analyses were carried out using Inductively Coupled Plasma Optical
G . JA R R A R
Emission Spectroscopy (ICP-OES) at the Institut für Geowissenschaften of the Technical University of Braunschweig, Germany. Rare earth elements for three samples were determined after separation using the resin Ag50WX12 and Ag5WX8 sulphonated polystrene cation exchanger in hydrogen form following the procedure of Zachmann (1988). The accuracy is between 0.5 and 1 wt % for the major elements and between 1 and 10 % for trace elements. All mineral analyses of plagioclase, pyroxene, oxides and sulphides were performed on a Cameca Camebax electron microprobe at the Institute of Physical Chemistry and Electrochemistry, University of Hanover, Germany. Typical working conditions were 15 kV and 10 nA, with a beam size from 1–3 µm. 4. Petrography and mineral chemistry The studied dolerites contain plagioclase, pyroxene, titanomagnetite and pseudomorphs after olivine (in the least evolved samples) as major phases. Minor phases include pyrite, alkali feldspar, apatite and quartz. The Quweira dolerites (Fig. 1) contain three types of ocellar structures that have been interpreted as fractionated melt segregation into gas filled vesicles in addition to deuteric stage filling with secondary minerals (Saffarini & Jarrar, 1992). The following is a discussion of the major mineral phases of the investigated mafic suite. 4.a. Pyroxene
Clinopyroxene was found in these dykes. It displays different habits, ranging from phenocrysts to small grains interstitially distributed between plagioclase laths. It ranges from beige to slightly brown pleochroic crystals, which are rich in titanium. Occasionally the pyroxene occurs also as poikilitic plates (3–5 mm) enclosing plagioclase laths. Some of the clinopyroxenes display branching forms. Furthermore, some of the pyroxene microcrystals were found growing on quartz xenocrysts. Analyses of representative pyroxenes are given in Table 1. On the conventional Di–Hd–En–Fs pyroxene quadrilateral (Fig. 2a) they plot in the fields of augite and salite (Morimoto, 1988). The contents of Al2O3 and TiO2 are similar to those in clinopyroxenes from the alkaline rocks (Leterrier et al. 1982). On the plot Al2O3 vs. SiO2 (Le Bas, 1962) the ten pyroxenes fall in the field of alkaline basalts (diagram not shown). Furthermore, the clinopyroxenes plot in the field of alkali basalts on the Ti vs. (Ca+Na) discrimination diagram (Fig. 3d) developed by Leterrier et al. (1982). A single pyroxene assemblage can be used to obtain minimum crystallization temperatures (Lindsley, 1983). The analysed pyroxenes are plotted on Lindsley’s pyroxene projection and yield crystallization temperatures between 1050 and 800 °C; two
311
Neoproterozoic mafic dyke suite, Jordan
Figure 1. (a) Index map of the northernmost Arabian–Nubian Shield. Localities in the adjacent areas (WF: Wadi Feiran, Timna, Amram) with the equivalent dykes are also indicated. (b) Simplified geological map of the Pan-African exposures in southern Jordan. The incomplete ellipses delineate areas where investigated dykes are sampled. Symbols between brackets after each locality are used in the figures except where otherwise indicated.
pyroxene grains document temperatures as low as 700–600 °C (Fig. 2b). 4.b. Feldspars
Plagioclase is the main feldspar encountered in the studied dykes. Typically, plagioclase occurs as laths in the groundmass, 0.1–0.5 mm in size. In addition, phenocrysts up to 4 mm in length are common. The chemistry of the feldspars varies from andesine to labradorite (An 45 to 61 %). Feldspar compositions are shown in Figure 2c and representative analyses with their chemical formulas are given in Table 1. 4.c. Opaques 4.c.1. Titanomagnetites
Titanomagnetite occurs as euhedral cubes, disseminated grains, skeletal and/or network grains and rarely as phenocrysts. Representative analyses with their
chemical formulae are listed in Table 1. The analyses are plotted on the FeO–TiO2–Fe2O3 ternary system (Buddington & Lindsley, 1964). All of the analyses fall on or close to the ulvospinel–magnetite tie-line (Fig. 2d). The compositional range is Usp37Mt67 to Usp74Mt26. Since ilmenite–hematite solid-solutions do not coexist with the ulvospinel–magnetite, the Fe–Ti geothermobarometric evaluation of the oxide phases cannot be applied. 4.c.2. Sulphides
Pyrite is the sole sulphide mineral in the investigated dykes. It is present as euhedral cubes and anhedral disseminated granular aggregates and as droplet-like grains in all investigated dykes. Some samples contain millimetre-thick veins of pyrite. Modal abundance of the sulphides ranges from a few grains to a large number of visible granules. The sulphur content of these dykes was determined for one of the localities to range between 100 and 1000 ppm (Saffarini & Jarrar, 1992).
46.55 3.37 4.29 12.07 0.29 11.63 20.91 0.57 99.68
SiO2 TiO2 Al2O3 Fe2O3 FeO MnO MgO ∑
0.09 19.07 2.62 25.58 46.86 0.87 0.07 95.16
33/SJ8 0.80 19.21 0.52 26.97 47.24 0.88 0.41 96.02
SJ14b 0.50 16.81 0.74 31.16 45.26 0.21 0.21 94.89
7/SJ18
93.54
0.12 23.62 3.45 14.64 51.36 0.34
3/SJ14
1.79 0.21 2.00 0.09 0.36 0.01 0.65 0.84 0.04 2.00
1.81 0.19 2.00 0.07 0.34 0.01 0.75 0.80 0.04 2.01
*Number of analyses/sample number.
1.78 0.22 2.00 0.06 0.33 0.01 0.75 0.81 0.04 2.01
1.79 0.19 1.98 0.10 0.38 0.01 0.66 0.85 0.04 2.03
0.00 0.56 0.12 0.75 1.53 0.03 0.00
0.03 0.56 0.02 0.79 1.54 0.03 0.02
0.02 0.50 0.03 0.93 1.50 0.01 0.01
0.00 0.70 0.16 0.43 1.69 0.01 0.00
Si Ti Al Fe3+ Fe2+ Mn Mg
47.51 2.52 4.30 10.93 0.20 13.30 19.88 0.52 99.16
8/SJ18
Si IV Al IV T site Ti VI Fe Mn Mg Ca Na O site
46.72 3.28 4.79 11.43 0.28 11.66 20.83 0.60 99.59
7/SJ14
Formulae calculated on the basis of 3 cations
46.95 2.57 4.97 11.26 0.19 12.77 20.68 0.55 99.94
9/SJ18
Titanomagnetites
Formulae calculated on the basis of 6 oxygens
SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O ∑
34/SJ8*
Clinopyroxenes
Table 1. Selected chemical analyses of constituent minerals and their formulae for the investigated dykes
65.84 20.22 1.68 8.47 3.42 99.63
10/SJ4 53.54 28.97 12.45 4.55 0.22 99.73
1/SJ14 66.33 20.23 1.34 11.41 0.13 99.44
P6
Al IV Si IV Ca Na K
1.54 2.44 0.59 0.43 0.02
1.06 2.93 0.08 0.73 0.19
1.55 2.43 0.61 0.40 0.01
1.05 2.93 0.06 0.98 0.01
Formulae calculated on the basis of 8 oxygens
SiO2 53.91 Al2O3 28.77 CaO 12.20 Na2O 4.87 K2O 0.29 ∑ 100.04
30/SJ8
Feldspars
Fe S Co Ni Cu Zn
46.65 54.33 0.34 0.07 0.13 0.18
1/SJ14
46.47 53.91 0.05 0.03 0.11 0.00
P8/SJ8
46.12 53.55 0.12 0.08 0.09 0.00
4/SJ14
Sulphides
45.64 53.93 0.01 0.00 0.22 0.00
SJ14b
312 G . JA R R A R
313
Neoproterozoic mafic dyke suite, Jordan
Figure 2. (a) The di–hd–en–fs pyroxene quadrilateral showing the compositions of the pyroxenes of the studied dykes. (b) Pyroxene compositions projected on Lindsley (1983) graphical geothermometer. Temperature contours are in °C. (c) Feldspar compositions plotted on the An–Ab–Or ternary diagram. (d) Titanomagnetite compositions plotted on the TiO2–FeO–Fe2O3 ternary after Buddington & Lindsley (1964).
The analysed grains were found to consist of pyrite FeS2, along with a maximum of 0.22 % Cu, 0.18 % Zn, 0.21 % Ni and 0.31 % Co. Representative pyrite analyses are given in Table 1. 5. Geochemistry The whole-rock major and trace element geochemical data of 42 dykes are presented as averages and one standard deviation in Table 2 for the six sampling localities shown in Figure 1. In addition, the data set comprises the full chemical data for representative samples and elemental ratios of petrogenetic significance discussed below. A complete data set of bulk and mineral chemistry is obtainable from the author upon request. 5.a. Classification
The great majority of the studied dykes fall in the field of basalts on the Total Alkali vs. Silica classification diagram (Fig. 3a) of Le Maitre (1989). Using the Total Alkali vs. Silica classification system alone, these dykes would be placed either in the alkaline or tholeiitic suites, depending on which dividing line is used.
The Macdonald & Katsura (1964) divider places almost all samples in the alkaline field while the Irvine & Baragar (1971) dividing line places the rocks into both alkaline and subalkaline fields. The presence of quartz xenocrysts jacketed by pyroxene microcrystals explains the appearance of quartz in the CIPW norms, which lends these rocks the tholeiitic character. On the normative colour index vs. anorthite content diagram the studied rocks plot in the field of alkali basalt (diagram not shown). For this reason, the high field strength cations (HFSC) classification schemes in addition to the chemistry of the pyroxenes are preferred. Diagrams involving the elements Zr, TiO2, Y, Nb and P2O5 (Fig. 3b,c) exclusively place the investigated dykes in the alkaline basaltic fields. Furthermore, the clinopyroxenes from these dykes are typical for alkali basalts. Based on the above classification schemes the studied dyke suite has an alkaline rather than a tholeiitic character in spite of the fact that many dykes are quartz normative. 5.b. Major, trace and rare earth elements
Plots of major and trace element abundances vs. Mg no. of the investigated dykes are shown in Figure 4.
50.95 1.77
51.43 2.04
50.80 3.25
Filk SD, n = 2
Rahma SD, n = 4
Mubarak SD, n = 2
1.99 1.42
2.13 0.41
2.92 0.16
2.60 0.51
16.83 3.87
14.95 0.72
13.61 0.50
14.79 0.84
14.36 0.91
2.57 1.73
2.79 0.54
3.37 0.09
3.29 0.34
3.27 0.52
3.02 0.44
6.92 4.67
7.61 1.62
9.09 0.25
8.89 0.91
8.81 1.41
8.16 1.19
FeO
0.15 0.06
0.16 0.03
0.19 0.01
0.18 0.02
0.23 0.06
0.17 0.03
MnO
3.31 2.52
4.44 0.31
4.73 0.57
5.41 0.84
4.82 1.32
5.75 1.24
MgO
6.54 3.04
6.68 0.27
7.16 1.38
7.68 0.73
7.33 1.64
7.92 0.62
CaO
Sm 6.13 11.53 6.60
Eu 1.87 3.47 2.20
Gd 3.93 8.73 4.80
Dy 4.87 7.80 4.20
Tm 0.27 0.47 0.30
3.58 2.71 2.65 2.09 3.34 3.04
5.50 3.86
3.07 0.21
3.09 0.73
3.00 0.37
2.67 0.51
3.03 0.32
Na2O
Yb 2.53 3.20 1.70
1.50 0.67 0.81 0.41 0.97 1.71
3.18 4.00
1.85 0.23
1.12 0.74
1.16 0.39
1.17 1.02
1.02 0.42
K2O
ΣREE 86.99 169.30 101.80
37 57 38 49 41 49
38 3
35 23
42 3
45 5
42 6
48 8
Mg no.
La/Nb 0.65(0.50) 1 1.8 0.9 1.85
Lu 0.40 0.47 0.40
1.25 0.68 1.20 0.60 1.60 1.10
0.35 0.35
0.86 0.29
0.95 0.35
1.34 0.47
1.34 0.44
1.01 0.21
P2O5
5 190 52 26 57 65
15 16
50 20
51 36
33 23
49 26
81 57
Cr
La/Zr 0.15(0.02) 0.06 0.1 0.09 0.15
2.12 2.00 3.10 3.30 2.60 3.00
LOI
Ba/La 21(6) 10 36 17 28
7 92 45 90 52 34
33 43
26 10
34 28
60 24
46 22
52 27
Ni
344 269 431 343 448 277
249 248
281 15
424 40
362 52
370 82
325 60
V
Zr/Y 5.5(1.9) 2.5 2.8 4.5 4.9
17 24 24 20 19 17
13 10
20 6
21 4
18 2
20 4
20 2
Sc
524 297 384 162 548 679
470 358
592 110
399 115
468 162
465 292
398 121
Ba
Ti/Zr 110(43) 116 112 61 67
54 5 15 22 17 47
71 83
60 25
43 57
35 24
33 31
35 25
Rb
K/Zr 70(41) 22 52 29 58
551 549 463 538 536 675
381 47
637 108
525 150
526 66
477 104
556 34
Sr
48 24 48 38 49 35
28 27
31 11
48 3
41 7
40 5
35 11
Nb
Ba/Nb 14(13) 10 64 16 52
177 74 162 77 166 269
177 20
260 17
168 12
132 26
159 93
112 38
Zr
The average and standard deviations (SD) are given for the six sampled localities. n = number of samples in each locality. The full chemical analysis is given only for representative samples. Mg no. = 100 mol. prop. [Mg2+ / (Mg2+ + Fe 2+)].
Average incompatible elemental ratios of youngest mafic dyke suite compared to relevant reservoirs K/Rb K/Ba Y/Nb Zr/Nb Youngest mafic dyke suite (SD) 433(367) 22(9) 0.8(0.08) 4.6(3.9) Primitive mantle (Sun & McDonough, 1989) 394 35.4 0.03 16 Arabian Lower Crust (McGuire & Stern, 1993) 1527 15.5 0.003 19.2 Upper crust (Kempton et al. 1991) 316 18.3 0.06 10 Lower crust (Kempton et al. 1991) 329 13.3 0.04 12
Nd 19.60 37.80 23.10
Er 2.60 3.87 1.90
48.05 1.89
Yutum SD, n = 11
2.53 0.54
15.94 1.28
Fe2O3
Rare Earth Elements (ppm) La Ce SH-5 12.33 30.86 SJ-4 27.60 63.43 SJ-9 16.90 39.70
48.97 4.77
Quweira SD, n = 15
2.20 0.82
Al2O3
7.19 8.77 8.37 9.90 7.22 6.79
48.85 1.32
Shurairyh SD, n = 8
TiO2
Analyses of representative samples (oxides in wt % & trace elements in ppm) SH-1 48.8 3.37 14.21 3.55 9.58 0.22 4.26 SH-5 49.6 1.50 16.85 2.66 7.17 0.15 7.03 SJ-4 47.4 3.15 13.75 3.62 9.86 0.32 4.52 SJ-9 47.2 2.23 15.74 3.12 8.43 0.19 6.03 Y-1 46.7 3.49 14.16 3.78 10.20 0.20 5.27 R-2 52.8 1.90 15.09 2.44 6.57 0.14 4.51
SiO2
Locality
Table 2. Summary of geochemical data for the investigated youngest mafic dyke suite
32 25 25 17 31 27
23 8
30 5
29 1
26 6
30 8
25 4
Y
Sr/Y 20(6) 4.6 33 7 23
32 12.3 27.6 16.9 22 49
20 4
36 12
22 1
18 10
23 12
16 9
La
314 G . JA R R A R
Neoproterozoic mafic dyke suite, Jordan
315
Figure 3. Geochemical classification diagrams for the investigated dolerites. (a) Total alkalis vs. silica plot of Le Maitre (1989). II & I are the dividers between alkaline and subalkaline after Macdonald & Katsura (1964) and Irvine & Baragar (1971), respectively. One sample plotting in the phonolite field is not shown. (b) Zr/TiO2 vs. Nb/Y diagram (Winchester & Floyd, 1977). (c) P2O5 vs. Zr discrimination diagram (Winchester & Floyd 1977). (d) Ti vs. (Ca + Na) discrimination diagram for clinopyroxenes in the investigated dykes (Leterrier et al. 1982). Clinopyroxene compositions are expressed in cations per six oxygens.
The following trends are prominent: Al2O3 and CaO decrease, while TiO2, total iron, Nb, Y, Zr and V increase with decreasing Mg no. Nonetheless, the total iron, TiO2, Nb and V increase until Mg no. of about 40 (which corresponds to MgO of 5 wt %) and then decline. The trace element variation patterns are consistent with those of the major elements (e.g. Nb, Y, and V compared to TiO2 and total iron). The major element variations of the investigated dykes suggest that fractional crystallization may have played an important role in producing the observed chemical features. In this regard an Mg phase is one of the possibilities, although only pseudomorphs after olivine are present in some samples of the Shuraiyh locality (Fig. 1). The compositional trends furthermore call for fractionation of a calcic mineral. Since the investigated dykes contain occasional plagioclase phenocrysts, this mineral is a candidate. Observed decrease in Al2O3 supports fractionation of plagioclase. The Al2O3 trends flatten at Mg no. 40–30 and start to increase at lower Mg no.
A noticeable decrease in Al2O3 contents is expected after 40–50 % of plagioclase removal (Albaréde, 1992). The role of plagioclase fractionation is also supported by the smooth decreasing trends of Sr/La vs. SiO2 (see Section 7; Fig. 9). The fractionation of olivine and plagioclase alone cannot explain the observed trends. CaO/Al2O3 vs. SiO2 variations (see Fig. 9) show that the fractionation of a combination of phases with CaO > Al2O3 is required. Clinopyroxene is an obvious good choice. Titanaugite is present in the phenocrystal assemblage as well as an abundant phase in the groundmass. The steady decrease in Ni with differentiation is consistent with the removal of olivine, whereas the decreasing Cr content can be accounted for by removal of clinopyroxene. The Nb, Zr, V, and Y are relatively well constrained and show almost two-fold increases with decreasing Mg no. Chondrite-normalized REE patterns for three samples are presented in Figure 5 and the REE data are listed in Table 2. The REE patterns of the three sam-
316
G . JA R R A R
Figure 4. Binary plots of major and trace elements vs. Mg no. Symbols as in Figure 3.
Figure 5. Chondrite-normalized REE plot for the investigated dykes. Normalizing values from Sun & McDonough (1989).
ples display an increasing slope with differentiation (La/Yb)n = 4.8 and 8.1 for Mg no. of 57 and 38, respectively. Consequently, the original magma with Mg no. > 57 should have had (La/Yb)n < 4.8. This suggests the absence of residual garnet in the melting source. Furthermore, the three patterns are parallel and show a progressive increase in the ∑REE with differentiation, that is, ∑REE is 87, 102 and 169 ppm for Mg no. of 57, 48 and 38, respectively. Furthermore, the patterns show a slightly positive Eu anomaly (Eu/Eu* = 1.05–1.15) that is indicative of very minor plagioclase accumulation in the most primitive sample (Mg no. 57). Multi-elemental ‘spider-diagrams’ are particularly useful for comparing the geochemistry of the investigated dykes with other dyke suites from adjacent countries and from the same tectonic setting. A primitive mantle-normalized spider-diagram for the range and average of the studied dykes is shown in Figure 6a.
317
Neoproterozoic mafic dyke suite, Jordan
1995) are plotted in Figure 6d. The similarity between the patterns displayed by the studied dykes and the Kenya rift alkali and transitional basalts is striking. All plots lack the Nb trough, which is otherwise present in the older generations of dykes in southern Jordan and the adjacent areas. The lack of Nb depression indicates negligible crustal contamination (Wilson, 1989). 6. Tectonic setting Assuming that each tectonic environment invokes a diagnostic geochemical signature (e.g. Pearce & Norry, 1979; Mullen, 1983; Meschede, 1986; Wilson & Versfeld, 1994), the compositions of the studied dykes are plotted on a variety of tectonic discrimination diagrams designed to classify basaltic rocks on the basis of their minor and trace element contents. On the triangular diagrams as well as binary plots (Fig. 7) involving the HFSC, which are considered immobile during low temperature alteration, the investigated dolerites show the geochemical signature of withinplate basalts (Fig. 7d), within plate alkali basalts (Fig. 7a), and continental flood basalts (Fig. 7b,c). The overwhelming evidence for the tectonomagmatic environment in the Arabian–Nubian Shield during the intrusion of these dykes indicates strong extensional tectonic regime and the absence of subduction related features. This may be related to the development of a passive margin on the northern flank of Greater Gondwanaland at this time (Stern, 1994).
Figure 6. Primitive mantle-normalized (Sun & McDonough, 1989) elemental compatibility diagrams. (a) Range and average patterns of the studied dolerites and the average of data in (c). (b) Three samples from the investigated suite with complete REE and trace element data. (c) Equivalent dykes from the Arabian–Nubian Shield. For the locations see Figure 1. (d) Compositionally equivalent volcanics from the Main East African Rift and from Jordan.
Slight negative anomalies at Zr and K are evident, although when the average of the whole suite is plotted (Fig. 6a) a slight depression at Zr and prominent P spike can be observed. Similar plots for presumably equivalent dykes in the Wadi Feiran, southern Sinai (Friz-Töpfer, 1991), Mt Amram (Kessel, Stein & Navon, 1998) and Mt Timna (Beyth et al. 1995) (Fig. 1a) areas are also drawn for comparison (Fig. 6c). These dykes share similar patterns with the studied suite. Alkali basalts, ferrobasalts from the southern Gregory Kenya Rift (Baker et al. 1977), transitional basalts from the bimodal Naivasha volcanic field, Kenya rift (Macdonald et al.1987) and Quaternary alkali basalts from Jordan (Al-Malabeh, 1994; AlDajani, unpub M.Sc. thesis, Univ. Jordan (Amman),
7. Petrogenesis The following arguments are used to put constraints on the source of partial melting. First, the flat patterns of the REE, as pointed out above, indicate the absence of a residual garnet from the source area of the magmas. Second, investigations on mantle xenoliths from Cenozoic basalts associated with Red Sea rifting help select the suitable mineralogy of the mantle source (El-Sharkawi, 1982; Henjes-Kunst, Altherr, & Baumann, 1990; Stein, Garfunkel, & Jagoutz, 1993). A metasomatized lithospheric mantle source of phlogopite-bearing spinel lherzolite is selected. The composition of the modelled mantle source is taken as olivine : orthopyroxene : clinopyroxene : spinel : phlogopite = 65 : 20 : 10 : 4 : 1. The concentrations of Cr, Ni, Ti, Y, Zr, P and REE are similar to the primitive mantle (Sun & McDonough, 1989) compositions whereas concentration of the large ion lithophile elements (LILE) and Nb are taken as twice the primitive mantle compositions. The input data and results of the modal batch partial melting are shown in Table 3. The 1, 5 and 10 % partial melts along with the least evolved sample (SH-5) and the mantle source are shown on the multi-elemental compatibility plot and chondrite-normalized plot (Fig. 8). The figure shows
318
G . JA R R A R
Figure 8. (a) Primitive mantle-normalized (Sun & McDonough, 1989) elemental spider diagram for the modelled partial melts along with the least-evolved sample of the investigated suite and the source rock of the suite. (b) Chondrite-normalized REE (Sun & McDonough, 1989) plot for the modelled partial melts, the least-evolved sample and the source rock.
Figure 7. Tectonic setting discrimination diagrams for the investigated dolerites. Symbols as in Figure 3. (a) Zr–Nb–Y diagram (Meschede, 1986); (b) K2O–TiO2–P2O5 diagram (Pearce, Gorman, & Birkett, 1975); (c) Ti/Zr–Al2O3/TiO2 diagram (Wilson & Versfeld, 1994); (d) Zr/Y–Zr diagram (Pearce & Norry, 1979).
that the plot of least differentiated samples fits that of the 5 % partial melt. The deviations are attributed to uncertainties in the selection of partition coefficients and the composition of the source. The Ni and Cr
contents of the modelled 5 % partial melt are higher than the observed concentrations of 92 and 190 ppm, respectively. Accepted criteria for primary basaltic magma according to Clague & Frey (1982) are Mg no. greater than 65 and high nickel contents > 235 ppm. None of the investigated dykes fulfils this requirement. However, the samples with Mg no. of 57 and Cr and Ni concentrations of 190 and 92, respectively, can be considered as the most primitive and best approximating the primary magma of this suite. A minor amount of olivine and clinopyroxene fractionation of the primary magma (not encountered in the suite) could have brought about the observed drop in Ni and Cr from 178 to 92 and 364 to 190, respectively. Adding about 8 % forsterite-rich olivine to the most primitive sample (Mg no. = 57) would raise the Mg no. of this sample to 65. To summarize, the primitive melts that gave rise to the investigated suite were derived from possibly a phlogopite-bearing spinel lherzolite mantle source by a low 5 % modal batch partial melting. The intra-suite elemental variations can be best explained by fractional crystallization for reasons presented below.
319
Neoproterozoic mafic dyke suite, Jordan Table 3. The input data and the results of the modal batch partial melting of a phlogopite-bearing spinel lherzolite Mineral–melt distribution coefficients Element K Rb Ba Sr U Th Ti Zr Nb Y La Ce Nd Sm Eu Gd Dy Er Tm Yb Lu Ni Cr Sc V P
Ol 0.0068 0.0098 0.0099 0.014 0.00002 0.000052 0.02 0.012 0.01 0.01 0.0067 0.0069 0.0066 0.0066 0.0068 0.0077 0.0096 0.011 0.009 0.014 0.016 16.3 1.92 0.17 0.06 0.043
Opx
Cpx
0.014 0.022 0.013 0.04 0.015 0.0056 0.1 0.18 0.15 0.18 0.02 0.02 0.03 0.05 0.05 0.09 0.15 0.23 0.3 0.34 0.42 5 10 1.2 0.6 0.03
0.038 0.031 0.001 0.06 0.013 0.014 0.4 0.12 0.005 0.9 0.056 0.15 0.31 0.50 0.51 0.61 0.68 0.65 0.54 0.62 0.56 2.75 11.5 3.2 1.2 0.0044
Sp 0.01 0.01 0.01 0.048 0.14 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.02 77 0.57
Calculated and observed melt compositions (Cl)
Phlog
D
Co
F=1
F=5
F = 10
SH-5
1.5 1.7 1.5 0.081 0.011 0.12 0.98 0.6 0.14 0.03 0.034 0.034 0.032 0.031 0.03 0.03 0.026 0.03
0.026 0.031 0.025 0.024 0.004 0.004 0.085 0.067 0.039 0.133 0.015 0.024 0.042 0.065 0.066 0.085 0.105 0.119 0.120 0.140 0.152 11.883 7.532 0.776 0.279 0.034
516.4 1.27 13.978 42.2 0.042 0.17 1205 11.2 1.42 4.55 0.687 1.775 1.354 0.444 0.168 0.596 0.737 0.48 0.074 0.493 0.074 2080 2625 29.76 164 180
14441 31.0 408 1239 2.9 12.4 12837 146 29 32 28 52 26 6 2.2 6.4 6.5 3.8 0.6 3.3 0.5 177 352 38 573 4087
6911 15.9 191 577 0.8 3.2 9235 98 16.3 26 10.7 24.3 15.1 4.0 1.5 4.6 4.9 2.9 0.5 2.7 0.4 183 364 38 521 2177
4184 9.9 115 346 0.4 1.6 6837 70 10.5 21 6.1 14.6 9.8 2.8 1.1 3.4 3.8 2.3 0.4 2.2 0.3 193 382 37 467 1375
5562 21 297 549 2.1* 3.29* 8993 74 24 25 12.33 30.86 19.61 6.13 1.87 3.93 4.87 2.6 0.27 2.53 0.4 92 190 24 269 2941
0.042 0.03 1.3 5.4 8.3
D is the bulk distribution coefficient calculated for the source: phlogopite-bearing spinel lherzolite. Minerals used in the calculations are in the following proportions: olivine : orthopyroxene : clinopyroxene : spinel : phlogopite = 65 : 20 : 10 : 4 : 1; *U and Th values are from Beyth et al. (1995) on the dykes of Timna. Distribution coefficients are from the compilation of Rollinson (1993) and from the Geochemical Earth Reference Model (GERM). Co is composition of the source. Cr, Ni, Ti, Y, Zr, P and REE are similar to the primitive mantle compositions (Sun & McDonough, 1989) whereas the LILE and the Nb are taken as twice the primitive mantle.
On the Al2O3/TiO2 vs. TiO2 (Fig. 9) the samples define a curved continuum with most primitive sample (Mg no. = 57) having the highest ratio and lowest TiO2 content. According to Sun & Nesbitt (1978) this indicates that the plotted samples belong to one magma type. Sr/La forms a smooth decreasing trend with decreasing Mg no. (Fig. 9), indicating plagioclase controls the fractionation of these elements. The plot of TiO2 vs. FeO/MgO (Fig. 9) shows an increasing trend with an inflection at a TiO2 content of about 3.4 and Mg no. of 40 (FeO/MgO of 2) indicating the onset of titanomagnetite fractionation. To quantify the effects of fractional crystallization, major element leastsquares calculations using the XLFRAC software of Stormer & Nicholls (1978) were carried out. The fractionation was computed from sample SH-5, the most primitive, to sample SJ-3, the most evolved, which have Mg no. of 57 and 27, respectively. The results are shown in Table 4, and a relatively good fit between calculated and observed major element compositions is evident (∑r2=0.69). The composition of the cumulate rock, which was produced from fractional crystallization, is also shown and it is compared to the modelled weighted average of the lower crust of the Arabian– Nubian Shield (McGuire & Stern, 1993). The observed differences are accounted for by the lower crust of the Arabian–Nubian Shield having been
formed due to fractionation of mantle-derived magmas during the whole period of the Pan-African Orogeny. Least-squares modelling of major elements shows that the magma having the composition of sample (SH5) must have crystallized about 64 % Ol + Cpx + Plag + Titmag to produce the most evolved composition (SJ3), assuming that fractionation was the only operating differentiation mechanism. In order to test the validity of the major element modelling, trace and rare earth element modelling were also attempted. The fractionated phases were used in the proportions deduced from major element modelling, that is, the 64 % solids recalculated to 100 would give the ratio olivine : clinopyroxene : plagioclase : titanomagnetite = 18 : 11 : 64 : 7. These ratios were used for the calculation of bulk distribution coefficients of the modelled elements. The Rayleigh fractional crystallization equation (Arth, 1976) has been used and F was taken as 0.36, C0 and Cl the elemental concentration in the most primitive sample SH-5 and most differentiated sample (SJ-3), respectively. There is good agreement between observed and modelled concentrations (Table 5; Fig. 10). Deviations are also present but these are within the analytical error of trace elements. Zr is as much as 30 % lower than the observed value. Other accessories could have also played a minor role in the geochemical evolution of the investigated dykes.
320
G . JA R R A R
Table 4. Least squares modelling of major elements Oxide
Parent SH-5
Daughter SJ-3
SiO2 50.87 58.04 TiO2 1.58 2.04 Al2O3 17.39 14.26 FeO* 9.63 10.52 MgO 7.51 2.42 CaO 7.94 4.46 Na2O 2.99 3.10 K2O 0.95 3.66 P2O5 1.14 1.50 Residual sum of squares (r2) = 0.69 Phase wt % relative to the initial magma
Cumulate
LCANS
Ol
Cpx
Plag
Titm
46.56 1.69 19.16 8.91 10.48 9.97 3.04 0.18
49.3 0.9 16.0 9.6 7.9 10.5 3.0 0.3 0.1
40.24 0.00 0.00 12.30 47.02 0.44 0.00 0.00 0.00
49.30 1.67 4.60 8.55 14.69 20.72 0.45 0.02 0.0
53.29 0.04 29.17 0.25 0.07 12.16 4.74 0.28 0.00
0.10 20.59 2.83 76.40 0.08 0.00 0.00 0.00 0.00
11.93
6.66
40
4.57
Compositions of the magmas and the phases are normalized to 100 %. Parent is the most primitive sample of the suite (SH-5, Mg no. 57); daughter is the most evolved sample of the suite (SJ-3, Mg no. 27). Ol – olivine; Cpx – clinopyroxene; Plag – plagioclase; Titm – Titanomagnetite. The cumulate is calculated from the fractionated minerals and their percentages relative to the initial magma. LCANS – Weighted average of the lower crust of the Arabian–Nubian Shield (McGuire & Stern, 1993). FeO* – total iron. Chemistry of minerals is from Table 1 and the olivine values from literature.
Figure 9. Binary plots showing petrogenetic aspects of the investigated dolerites. Symbols as in Figure 3. Al2O3/TiO2 vs. TiO2 plot after Sun & Nesbitt (1978); Sr/La vs. Mg no.; TiO2 vs. FeO/MgO, Mg no. are indicated on some of the samples; (Ni + Cr) vs. MgO; CaO/Al2O3 vs. SiO2; K/Rb vs. Rb, Mg no. are indicated on some of the samples.
The scarcity of primitive magmas and the strong evidence for large-scale fractional crystallization suggest that large volumes of mafic and ultramafic cumulates must have been produced during the evolution of this suite. McGuire & Stern (1993) calculated a mafic lower continental crust for the late Precambrian Arabian–
Nubian Shield on the basis of geochemical and isotopic investigation of granulite xenoliths from western Saudi Arabia. Furthermore, they suggested that the formation of this crust was caused by fractionation of mantle-derived magmas to produce the felsic and intermediate upper crust and mafic cumulates lower crust.
321
Neoproterozoic mafic dyke suite, Jordan Table 5. Modelling of trace elements assuming simple Rayleigh fractionation KD values
Rb Ba Sr Ti Zr Nb Y La Ce Nd Sm Eu Gd Yb Ni Cr Sc V
Daughter (SJ-4)
Plag
Titm
D
Parent (SH-5)
Observed
Calculated
0.071 0.23 1.83 0.04 0.012 0.025 0.03 0.19 0.12 0.081 0.067 1.12 0.063 0.067 0.05 0.06 0.03 0.1
0.01 0.01 0.01 7.5 0.10 1.8 0.2 1.5 1.3 1.0 1.1 0.6 1.0 0.9 10 40 3 7.4
0.05 0.15 1.18 0.59 0.03 0.15 0.13 0.24 0.19 0.16 0.18 0.82 0.18 0.18 3.93 4.45 0.61 0.72
37 234 499 1.21 79 25 21 12.33 31 19.6 6.13 1.87 3.93 2.53 92 190 20 256
90 1271 463 1.96 371 39 51 27.6 63.43 37.8 11.53 2.2 8.72 3.2 3 3 13 371
97.5 558 414.7 1.82 213 59.4 50.9 27 71.2 46.2 14.22 2.26 9.09 5.87 4.4 3.4 29.7 339.1
Distribution coefficients (KD) for olivine and clinopyroxene are similar to those in Table 3. Distribution coefficients are taken from the compilation of Rollinson (1993), from GERM, and McBirney (1984). D (bulk distribution coefficient) was calculated for the fractionated mineral assemblage in Table 4. The observed REE concentrations are for sample SJ-4. Plag – plagioclase; Titm – titanomagnetite.
8. Conclusions The constancy of incompatible element ratios in a given suite is strong evidence that fractional crystallization has been the predominant process in the evolution of a particular suite. The ratios of the LILE and HFSC for the studied dykes are listed in Table 2 along with those for the upper continental crust, lower continental crust, and the lower crust of the Arabian–Nubian Shield. The majority (75 %) of these ratios cluster around the average and most of these ratios approximate upper crustal values. However, some ratios increase with decreasing Mg no. (e.g. Zr/Y); other ratios (e.g. Y/Nb; Zr/Nb, La/Nb; Ba/Nb) increase just when fractionation of titanomagnetite starts (at Mg no. 40), due to the removal of Nb. Ti/Zr starts to decrease at the same Mg no. also due to the separation of titanomagnetite. K/Rb values, however, show an erratic increase. Figure 9 shows that the K/Rb increases at different Mg values. This can be explained by the assimilation of lower crustal material, which has high K/Rb ratios (Table 2).
Figure 10. Primitive mantle-normalized plot of the modelled trace elements (Table 5). Normalizing values are from Sun & McDonough (1989).
The last stage of Pan-African evolution of the Arabian–Nubian Shield (~ 600–540 Ma) was marked by extension and emplacement of several generations of dykes. Three of these are clearly identifiable in Jordan, and up to six phases elsewhere in the shield according to Eyal & Eyal (1987). The youngest generation defined here as the Youngest Mafic Dyke Suite has been studied over almost the whole basement complex of south Jordan. The following conclusions can be drawn from this investigation: (1) The investigated dyke suite represents the youngest mafic Neoproterozoic dyke activity in the Arabian Shield dated before the onset of Cambrian time at 545 Ma. (2) This dyke generation has a dominant trend of NE to ENE and attains a thickness of up to 25 metres. It is, moreover, characterized by the presence of pyrite and by being the only dyke generation cross-cutting the alkali feldspar and syenogranites of the Humrat suite (dated at 568 Ma) throughout the whole basement complex. (3) The crystallization temperatures of this suite range mainly between 1050 and 800 °C, as deduced from single pyroxene thermometry. (4) Geochemically, the studied suite is mildly alkaline in nature, enriched in LILE, shows moderate enrichment of REE, and strong enrichment of HFSC, and lacks Nb anomaly. These features are consistent with a predominantly extensional continental tectonic setting without any sign of subduction-related features. (5) The geochemical modelling suggests the derivation of this suite from a metasomatized lithospheric mantle by 5 % modal batch melting of phlogopitebearing spinel lherzolite.
322 (6) The intra-suite geochemical variations can be best explained by Rayleigh fractional crystallization of about 64 % of a combination of olivine, clinopyroxene, plagioclase and titanomagnetite. (7) The fractionated mineral assemblage gave rise to a cumulate in the lower crust and/or crust mantle boundary, which contributed to the formation of the mafic lower crust of Arabian–Nubian Shield as it has been modelled by McGuire & Stern (1993). (8) Elemental ratios and petrographic evidence indicate crustal contamination, although not significant, of the investigated dykes. (9) The investigated dykes show striking similarity to the time-equivalent dykes from the Elat and Mt Timna areas and the S-2 dyke generation of the Wadi Feiran of south Sinai, the transitional to mildly alkaline basalt of the East African Rift system, and the Quaternary Jordanian basalts. Acknowledgements. The sabbatical leave offered by the University of Jordan for the author is highly appreciated. This work was carried out while the author was on a research visit to Germany supported by the DAAD and completed while spending sabbatical leave at the Geosciences Department, University of Texas at Dallas, as a Fulbright Fellow. The author wishes to express his sincere thanks to D. Zachmann, TU Braunschweig, Germany and Jürgen Koepke, Institut für Physikalische Chemie der Universität Hannover, Germany, for their help during lab work. The manuscript benefited from the critical comments made by Robert Stern and an anonymous referee.
References ALBARÉDE, F. 1992. How deep do common basaltic magmas form and differentiate? Journal of Geophysical Research 97(B7), 10997–11009. AL-MALABEH, A. 1994. Geochemistry of two volcanic cones from the intra-continental plateau basalt of Harra ElJabban, NE-Jordan. Geochemical Journal 28, 517–40. ARTH, J. G. 1976. Behaviour of trace elements during magmatic processes – a summary of theoretical models and their applications. Journal of Research of the U.S. Geological Survey 4, 41–7. BAKER, B. H., GOLES, G. G., LEEMAN, W. P. & LINDSTROM, M. M. 1977. Geochemistry and petrogenesis of a basalt–benmoreite–trachyte suite from the southern part of the Gregory Rift, Kenya. Contributions to Mineralogy and Petrology 64, 303–32. BENTOR, Y. K. 1985. The crustal evolution of the Arabian–Nubian massif with special reference to the Sinai Peninsula. Precambrian Research 28, 1–74. BEYTH, M. & HEIMANN, A. 1999. The youngest igneous event in the crystalline basement of the Arabian–Nubian Shield, Timna igneous complex. Israel Journal of Earth Science 48, 113–20. BEYTH, M., STERN, R. J., ALTHERR, R., PELTZ, S. & HEIMANN, A. 1995. Petrochemistry of dolerite dikes from mount Timna, southern Israel: Implications for plate tectonic setting. Geological Survey of Israel Current Research 9, 24–6. BROOK, M., IBRAHIM, KH. & MCCOURT, W. J. 1990. New geochronological data from the Arabian Shield area of
G . JA R R A R
southwest Jordan. Proceedings of the 3rd Jordanian Geological Conference, Amman, 361–94. BUDDINGTON, A. F. & LINDSLEY, D. H. 1964. Iron-titanium oxide minerals and synthetic equivalents. Journal of Petrology 5, 310–57. CLAGUE, D. A. & FREY, F. A. 1982. Petrology and trace elements geochemistry of the Honolulu volcanics, Oahu: implications for the oceanic mantle beneath Hawaii. Journal of Petrology 23, 447–504. EL-SHARKAWI, M. A. 1982. Lherzolite xenoliths from the Shihan basalts, Jordan. Journal of the University of Kuwait 9, 287–300. EYAL, Y. & EYAL, M. 1987. Mafic dyke swarms in the Arabian–Nubian Shield. Israel Journal of Earth Science 36, 195–211. FRIZ-TÖPFER, A. 1991. Geochemical characterisation of Pan-African dyke swarms in southern Sinai: From continental margin to intraplate magmatism. Precambrian Research 49, 281–300. GREENWOOD, W. R. 1981. Geology of the Abha quadrangle, sheet 18F, Kingdom of Saudi Arabia, 1:100,000. Saudi Arabia Ministry of Petroleum and Mineral Resources, 49 pp. HENJES-KUNST, F., ALTHERR, R. & BAUMANN, A. 1990. Evolution and composition of the lithospheric mantle underneath the western Arabian Peninsula: Constraints from Sr–Nd isotope systematics of mantle xenoliths. Contributions to Mineralogy and Petrology 105, 460–72. IRVINE, T. N. & BARAGAR, W. R. A. 1971. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Science 8, 523–48. JARRAR, G. 1985. Late Proterozoic crustal evolution of the Arabian–Nubian shield in the Wadi Araba area, SWJordan. Geologisches Jahrbuch B 61, 3–87. JARRAR, G. 1992. Geochemistry and petrogenesis of an alkali feldspar rhyolite suite from Wadi Museimir, Central Wadi Araba, Jordan. Chemie der Erde 52, 301–12. JARRAR, G., WACHENDORF, H. & SAFFARINI, G. 1992. A late Proterozoic bimodal volcanic/sub volcanic suite from Wadi Araba, Southwest Jordan. Precambrian Research 56, 51–72. JARRAR, G., WACHENDORF, H. & ZACHMANN, D. 1993. A Pan-African alkaline pluton intruding the Saramuj Conglomerate, south-west Jordan. Geologische Rundschau 82, 121–35. JARRAR, G., WACHENDORF, H. & ZELLMER, H. 1991. The Saramuj conglomerate: Evolution of a pan-African molasse sequence from southwest Jordan. Neues Jahrbuch für Geologie und Paläontologie, Monatsheft 3, 335–56. KEMPTON, P. D., HARMON, R. S., HAWKESWORTH, C. J. & MOORBATH, S. 1991. Petrology and geochemistry of lower crustal granulites from the Geronimo volcanic field, southeastern Arizona. Geochimica et Cosmochimica Acta 54, 3401–26. KESSEL, R., STEIN, M. & NAVON, O. 1998. Petrogenesis of late Neoproterozoic dikes in the northern Arabian–Nubian shield: Implications for the origin of A-type granites. Precambrian Research 92, 195–213. LANDING, E., BROWNING, S. A., DAVIDEK, K. L., WESTROP, S. R., GEYER, G., & HELDMAIER, W. 1998. Duration of the early Camrian:U–Pb ages of volcanic ashes from Avalon and Gondwana. Canadian Journal of Earth Sciences 35, 329–38. LE BAS, M. J. 1962. The role of aluminium in igneous clinopyroxenes with relation to their parentage. American Journal of Science 260, 267–88.
Neoproterozoic mafic dyke suite, Jordan LE MAITRE, R. W. 1989. A classification of Igneous rocks and glossary of terms. Recommendations of the IUGS Subcommission on the Systematics of Igneous Rocks. Oxford: Blackwell. LETERRIER, J., MAURY, R., THONON, P, GIRARD, D. & MARCHAL, M. 1982. Clinopyroxene composition as a method of identification of magmatic affinities of paleo-volcanic series. Earth and Planetary Sciences Letters 59, 139–54. LINDSLEY, D. H. 1983. Pyroxene thermometry. American Mineralogist 64, 477–93. MACDONALD, G. A. & KATSURA, T. 1964. Chemical composition of Hawaiian lavas. Journal of Petrology 5, 83–133. MACDONALD, R., DAVIES, G. R. , BLISS, C. M., LEAT, P. T., BAILEY, D. K. & SMITH, R. L. 1987. Geochemistry of high-silica peralkaline rhyolites, Naivasha, Kenya rift Valley. Journal of Petrology 28, 979–1008. MCBIRNEY, A. R. 1984. Igneous Petrology. San Francisco: Freeman Cooper, 504 pp. MCGUIRE, A. V. & STERN, R. J. 1993. Granulite xenoliths from western Saudi Arabia: the lower crust of the late Precambrian Arabian–Nubian Shield. Contributions to Mineralogy and Petrology 114, 395–408. MESCHEDE, M. 1986. A model of discriminating between different types of mid-ocean ridge basalts and continental tholeiites with the N–Zr–Y diagram. Chemical Geology 56, 217–18. MORIMOTO, N. 1988. Nomenclature of pyroxenes. Mineralogical Magazine 52, 535–550. MULLEN, E. D. 1983. MnO/TiO2/P2O5. A minor element discriminant for basaltic rocks of oceanic environments and its implications for petrogenesis. Earth and Planetary Science Letters 62, 53–62. PEARCE, J. A. & NORRY, M. J. 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology 69, 33–47. PEARCE, T. H., GORMAN, B. E. & BIRKETT, T. C. 1975. The TiO2–K2O–P2O5 diagram: A method of discriminating between oceanic and non-oceanic basalts. Earth and Planetary Sciences Letters 24, 419–26. PUDLO, D. & FRANZ, G. 1994. Dike rock generation and magma interactions in the Bir Safsaf igneous complex, SW Egypt: Implications for the Pan-African evolution in North East Africa. Geologische Rundschau 83, 523–36. ROLLINSON, H. 1993. Using geochemical data: Evaluation, presentation, interpretation. Longman Group UK, 352 pp. SAFFARINI, G. & JARRAR, G. 1991. Geochemical and petrogenetic aspects of Late Precambrian diabase in Wadi Um Salab, SW Jordan: Implications for Mantle below Jordan. Chemie der Erde 51, 261–74.
323 SAFFARINI, G. & JARRAR, G. 1992. Ocelli formation in Late Proterozoic Ocellar diabase dikes of Wadi Um Salab, SW Jordan: Petrographic Inferences. Dirasat 19B, 63–85. STEIN, M., GARFUNKEL, Z. & JAGOUTZ, E. 1993. Chronothermometry of peridotitic and pyroxenitic xenoliths: Implications for the thermal evolution of the Arabian lithosphere. Geochimica et Cosmochimica Acta 57, 1325–37. STERN, R. J. 1985. The Najd-fault system, Saudi Arabia and Egypt: A late Precambrian rift-related transform system? Tectonics 4, 497–511. STERN, R. J. 1994. Arc assembly and continental collision in the Neoproterozoic east African Orogen: implications for the consolidation of the Gondwanaland. Annual Review Earth Planetary Sciences 22, 319–51. STOESER, D. B. & CAMP, V. E. 1985. Pan-African microplate accretion of the Arabian Shield. Geological Society of America Bulletin 96, 817–26. STORMER, J. C. & NICHOLLS, J. 1978. XLFRAC: A program for the interactive testing of magmatic differentiation models. Computers & Geosciences 4, 143–59. SUN, S. S. & MCDONOUGH, W. F. 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In Magmatism in the Ocean Basins (eds A. D. Saunders and M. J. Norry), pp. 313–45. Geological Society of London, Special Publication no. 42. SUN, S.-S. & NESBITT, R. W. 1978. Petrogenesis of Archean ultrabasic and basic volcanic rocks: evidence from rare earth elements. Contributions to Mineralogy and Petrology 65, 301–25. TUCKER, R. D. & MCKERROW, W. S. 1995. Early Paleozoic chronology: a review in light of new U–Pb zircon ages from Newfoundland and Britain. Canadian Journal of Earth Sciences 32, 368–79. WILSON, M. 1989. Igneous Petrogenesis. London: Unwin Hyman, 466 pp. WILSON, A. H. & VERSFELD, J. A. 1994. The early Archean Nondweni greenstone belt, southern Kaapvaal craton, South Africa; part II, characteristics of volcanic rocks and constraints on magma genesis. Precambrian Research 67, 277–320. WINCHESTER, J. A. & FLOYD, P. A. 1977. Geochemical discrimination of different magma series and their differentiation products using immobile elements. Chemical Geology 20, 325–43. ZACHMANN, D. W. 1988. Matrix effects in separation of Rare-Earth Elements, scandium, and yttrium and their determination by inductively coupled plasma optical emission spectroscopy. Analytical Chemistry 60, 420–7.
