serie NOVA TERRA nº 49

73 affected in their lower part by the general presence of the regional S 2 foliation, the development of which culminates in the formation of the Ponte Carreira detachment. The whole package is then overprinted by upright D 3 Variscan folding. 3. Relative timing of dyke emplacement From a regional point of view, ma fi c dykes are almost exclusive to the low-grade part of the upper units and are likely linked to the major igneousbodies in the underlyingsection.Their absence in the remaining part of the orogenic section, however, suggests that theywere emplaced prior to the stacking of the allochthonous units. The sills and cross- cutting dykes show a rectilinear geometry and range from 20 – 30 cm to 20 – 30 m in thickness, both in map view and in vertical exposures. Individual dyke segments are sub-parallel (from N30°E to N15°W) and perpendicular (N110°E) to the regional foliation (S1 or S2). The dykes are mainly composed of medium-grained gabbro and diabase showing chilled margins, and consist of plagioclase, hornblende, epidote and sphene with minor amounts of quartz and relict pyroxene. The similarity of the macroscopic appearance, phase assemblages, and textures of the gabbroic and diabasic dykes and sills precludes fi eld and petrographic distinction of different magma series. Ma fi c dykes in the lower part of the study area show clear cross-cutting relationships with the S 2 foliation, but some are boudinaged and sheared at their margins. Syntectonic garnet growth in these sheared margins points to a syn- to post-tectonic intrusion with respect to the regional S 2 foliation. In the uppermost part of the unit, which is devoid of S 2 , the ma fi c dykes are undeformed and transect D 1 folds. Conversely, a number of thin, rectilinear diabase dykes are folded by upright D 3 Variscan folds. In addition, a number of quartz veins were formed at high angles to the S 1 cleavage during D 2 . The presence of these veins together with the diabase dyke swarm suggests the existence of a dilational event that may be linked to brittle behaviour during the fi nal stages of D 2 . 4. U – Pb dating of the dyke network 4.1. Analytical methods Zircon crystals from the Ares dyke were separated at Universidad Complutense (Madrid) using conventional gravimetric and magnetic techniques. At the Stanford-US Geological Survey Micro-analytical Center (SUMAC), 39 zircon grains were handpicked under a binocular microscope and mounted on double-sided adhesive on glass slides in 1×6 mmparallel rows together with chips of the zircon standard R33 ( Black et al., 2004 ). After being set in epoxy resin, the zircon grains were abraded to expose their central portions by using 1500grit wet sandpaper, and then polished with 6 μmand 1 μm diamond abrasive on a lap wheel. Prior to isotopic analysis, the internal structure, inclusions, fractures and physical defects were identi fi ed with transmitted and re fl ected light on a petrographic microscope, and with cathodolumines- cence (CL) on a JEOL JSM 5600 electron microscope ( Fig. 4 ). Following analysis, secondary electron images were taken to locate the exact position of the spots. U – Th – Pb analyses of zircon were conducted at the Bay SHRIMP-RG (sensitive high-resolution ion microprobe-reverse geometry) facility operated by SUMAC in a single analytical session in July 2008. Analytical procedures for zircon dating followed methods described in Williams ( Williams, 1997 ). The primary oxygen ion beamoperated at 6 – 7 nA and produced a spot with a diameter of ~25 μ m and a depth of 1 – 2 μ m for an analysis time of 12 – 13 min. Data for each spot were collected in sets of fi vescans throughthemassrange. Theconcentrationof Uwascalibrated using zircon standard CZ3 (550 ppm U; ( Pidgeon et al., 1995 )), and isotopic compositions were calibrated against R33 ( 206 Pb ⁎ / 238 U=0.06716, equivalent to an age of 419 Ma; ( Black et al., 2004 )), which was analyzed every fourth scan. Data reduction follows the methods described by Williams (1997) and Ireland andWilliams (2003) , and used SQUID and ISOPLOT software Fig. 4. Cathodoluminescence images of selected zircon crystals from the Ares dyke: (a) Variscan zircons, (b) crystallization age zircons, and (c) inherited zircons. 356 F. Díaz García et al. / Gondwana Research 17 (2010) 352 – 362