serie NOVA TERRA nº 49

75 rocks ( Corfu et al., 2003 ). The last two age groups are interpreted as inherited components ( Fig. 6 ) based on the variable morphology and CL features of the zircon grains. There is a cluster of six analyses at ca. 565 Ma; three single analyses between 625 and 650 Ma, and two at ca. 730 Ma. Five analyses reveal a Paleoproterozoic signature at ca. 2010 and 2080 Ma. All of these age groups match ages de fi ned by zircon age spectra for cycles of continental crust production ( Condie et al., 2009 ). 4.3. Trace element data ThemagmaticcharacteroftheAresdykezirconisfurthersupportedby theirrareearthelement (REE) composition. Overall,theirREEpatterns are typical of magmatic zircon ( Hoskin and Schaltegger, 2003 ), with low light (L) REE contents, a pronounced positive Ce anomaly, a variably negative Eu anomaly, and variable fractionation in the heavy (H) REE ( Fig. 7 ). Even though inter- and intra-grain compositional variation is usually large in zircons, and thus zircon REE studies should be based on both multiple zircon grains and multiple analyses per grain, some limited petrogenetic information can be suggested for the Ares dyke zircon based on their REE contents and certain elemental ratios. The following section is based on a number of general observations presented by Wooden et al. (2006) using the vast dataset obtained in their laboratory. According to these authors, Yb/Gd, which represents the steepness of the HREE pattern, appears to be an excellent monitor of magma evolution by fractional crystallization during zircon growth. For com- mon magmatic suites, it shows a starting ratio of about 10 and in- creases rapidly at relatively low temperatures (<750 °C). Conversely, Th/U tends todecreasewith zircon crystallizationtemperature,showing the strongest changes at higher temperature. The Ce/Sm ratio is preferred to that of Ce/Ce ⁎ by these authors because it varies more regularly when plotted against a fractionation index, typically showing an increase with increasing fractionation (e.g., Yb/Gd ratio). These features are illustrated for the Ares dyke zircons in plots of Yb/ Gd versus Th/U and Ce/Sm versus Yb/Gd ( Fig. 8 a and b, respectively). In both diagrams, the low and homogeneous Yb/Gd ratio in the zircon grains used to de fi ne the age of the dyke is apparent, consistent with crystallization from a poorly evolved magma such as a ma fi c dyke. Conversely, the remaining zircon grains usually show higher and more scattered Yb/Gd ratios, consistent with their inherited nature. The homogeneous character of the zircons used to determine the age of the dykeis alsoevident intheir lowCe/Smratioscomparedtotheremaining grains. Hence, given the CL features together and characteristics of the Ares dyke zircons, it can be argued that the grains selected for the age determination represent a homogeneous population of clear magmatic rather than being of metamorphic origin. This population is clearly different from the rest of the zircon grains, which show high variability in their chemical and CL features suggesting an inherited source. Based on the U – Pb and trace element data, a crystallization age of 510 Ma is proposed for the Ares dyke. However, the possibility that the whole zircon population is inherited cannot be completely discounted since this is a common situation in ma fi c rocks where the total number of zircon grains is often low. However, even if the zircon grains are inherited, this youngest zircon population would provide a maxi- mum age of intrusion in the same way that detrital zircons provide a maximum depositional age in sedimentary rocks. In this case, the maximum age of intrusion would likely be close to the actual crystallization age. The volcanic arc preserved in the upper units of NW Iberia is characterized by an early intra-arc high-temperature metamorphism dated in the interval 505 – 483 Ma ( Abati et al., 1999, 2007 ). Moreover, subsequent accretion of the arc to the southern margin of Laurussia at 410 – 390 Ma ( Ordoñez Casado et al., 2001; Fernández Suárez et al., 2007 ) resulted in high-pressure/high-temper- ature metamorphism in the lower parts of the upper units. The absence in the dyke of a signi fi cant zircon population younger than 510 Ma, and the complete absence of metamorphic zircon grains, consequently provide additional arguments favouring its intrusion at around 510 Ma, even in the case of a completely inherited zircon population. 5. Discussion Fuenlabrada et al. (2010) show that the sedimentary sequence in the uppermost part of the upper units has a maximum depositional age in the range 510 – 530 Ma (Middle Cambrian), which overlaps the Fig. 5. Tera-WasserburgplotshowingdistributionofSHRIMPzirconanalysesfromsample GCH-06-2. Grey ellipses correspond to analyses considered to date the crystallization age. Error ellipses are ±2 σ . Fig. 6. Diagrams illustrating the distribution of (a) 206 Pb/ 238 U ages, and (b) 207 Pb/ 206 Pb ages from the Ares dyke zircon grains. 358 F. Díaz García et al. / Gondwana Research 17 (2010) 352 – 362