U–Pb SIMS dating of some granitoids from eastern Blekinge, southern Sweden

Abstract Zircons from seven granitoids in eastern Blekinge have been dated using secondary ion mass spectrometry. The analyzed rocks include one Småland granitoid from north of the Småland-Blekinge Deformation Zone (SBDZ), two samples of megacrystic “Filipstad-type” granite from south of that zone and one sample each of the “Småland-type” Rödeby, Almö, Tjurkö and Jämjö granites. The results yield a crystallization age of 1778 ± 5 Ma for the Småland granitoid, and crystallization ages between 1770 ± 4 and 1758 ± 5 Ma for the other granitoids, in most cases substantially older than previous TIMS ages. These data show that the “Småland-type” granitoids in eastern Blekinge are similar in age to the surrounding Tving granitoids, and the more felsic of them may represent late-stage differentiates belonging to the same magmatic suite. As the Tving granitoids show differences both in degree of deformation, in geochemistry and in age, compared to the Småland granitoids north of the SBDZ, it is suggested that these represent two separate but closely related igneous suites, which could both be included within a TIB-1 supersuite. The investigated zircons showed very limited signs of metamorphic overprinting, and no metamorphic ages could be determined. However, the combined evidence from field observations and earlier U–Pb geochronology would suggest the presence of two separate metamorphic episodes in Blekinge, one in close connection with the formation of these rocks at 1.76–1.75 Ga and one connected to the intrusion of the Karlshamn granitoid suite at around 1.45 Ga.


Introduction
The Proterozoic bedrock of eastern Blekinge in southern Sweden ( Fig. 1) is dominated by gneissic tonalites, granodiorites and granites belonging to the Tving granitoid suite, characterized by the presence of 1-2 cm large microcline augen (Kornfält 1999a(Kornfält , 1999b(Kornfält , 2007a(Kornfält , 2007bKornfält & Bruun 2002, which have been dated to 1.77-1.75 Ga both by conventional U-Pb TIMS geochronology on zircon (Johansson & Larsen 1989;Kornfält 1993) and by U-Pb spot dating of zircons using the Nordsim ion microprobe (Johansson et al. 2006). They thus overlap slightly in age with the 1.81-1.76 Ga Småland granitoids, which belong to the Transscandinavian Igneous Belt generation 1 (TIB-1; Högdahl et al. 2004), and outcrop north of the Småland-Blekinge Deformation Zone (SBDZ, Fig. 1 ;Wiklander 1974;Krauss et al. 1996;Lindh et al. 2001). The Tving granitoids can thus be considered a more deformed variety of the TIB-1 Småland granitoids, although they also differ in composition by being more mafic (granodioritic to adamellitic) compared to the more alkali-rich and felsic Småland granitoids (adamellitic to granitic; Kornfält 1999aKornfält , 1999bKornfält & Bruun 2002). The border zone between them, the SBDZ, although appearing as a sharp E-W trending fault zone on most geological maps, in fact is a wide and rather diffuse zone of increased E-W trending schistosity, along which the southern (Blekinge) deformed and metamorphosed block presumably has been uplifted relative to the northern (Småland) less deformed block (Krauss et al. 1996;Kornfält 1993Kornfält , 1999aLindh et al. 2001;Kornfält & Bruun 2002). This uplift occurred after the formation of the Tving and Småland granitoids, i.e. after 1.75 Ga, but before the intrusion of the younger Karlshamn granitoids (1.45 Ga Eringsboda pluton; Kornfält 1999a;Lindh et al. 2001;Kornfält & Bruun 2002). As pointed out by Kornfält (1999a) and Kornfält & Bruun (2002), the SBDZ is not the southern boundary of the Transscandinavian Igneous Belt, since the Tving granitoids and other rocks in Blekinge can be included within it, but it is the northern limit of subsequent penetrative deformation.
Within the area of Tving granitoids in eastern Blekinge (Fig. 1), smaller massifs with granitoids of somewhat deviating character have been distinguished during geological mapping, especially within the well-exposed coastal area and archipelago, and considered to be more direct equivalents of the Småland granitoids north of the SBDZ (Kornfält 1999a(Kornfält , 1999b. Some of these massifs consist of relatively coarse-grained granite with plagioclase-mantled K-feldspar megacrysts, resembling the TIB-1 Filipstad granite in Värmland, and have been referred to as "Småland granite of the Filipstad-type" in the map descriptions (Kornfält 1999a(Kornfält , 1999b. In other places, the granite is strongly deformed and/or more fine-grained, and has been given local names such as Tjurkö granite, Jämjö granite and Almö Karlshamn granitoids, or whether they simply reflect analytical scatter caused by the discordancy of the zircons.
As these granitoids have been given a special colour and code (88) on the latest geological overview bedrock map of Sweden from SGU (Bergman et al. 2012), and assigned a somewhat younger age (1.7 Ga) in the legend compared to the surrounding Tving granitoid suite (1.7-1.8 Ga), it became of interest to verify their real intrusion age. For that purpose, the original zircon separates used for conventional TIMS dating were recovered from the laboratories involved, zircons were selected by handpicking, mounted in epoxy and studied under the microscope in transmitted and reflected light and in cathodoluminescence (CL) using a Hitachi scanning electron microscope to see their internal structure, followed by U-Th-Pb isotope analyses of selected spots using the Nordsim Cameca 1280 ion microprobe (SIMS) with procedures as outlined by Whitehouse et al. (1999) and Whitehouse & Kamber (2005), and using the 91500 zircon standard (Wiedenbeck et al. 1995(Wiedenbeck et al. , 2004. Most of the selected spots were in magmatic-looking internal parts of the zircons, in order to determine the crystallization age of these rocks, but in some cases overgrowths of possible metamorphic origin were also analyzed in order to check for metamorphic influence. The results of the individual spot analyses are listed in Table 2, and granite (Kornfält 1999b). On the map (Kornfält 2007b;Fig. 1), these have been grouped together as "leucogranites" belonging to the Småland granitoid suite. A medium-grained variety occurring around Rödeby, and called Rödeby granite, also deviates from the surrounding Tving granitoids in texture and composition (Kornfält 1999a(Kornfält , 2007a. In connection with the mapping program by the Geological Survey of Sweden (SGU) in Blekinge, several of these smaller granitoid massifs were sampled and subjected to multigrain U-Pb TIMS (thermal ionization mass spectrometry) dating of zircon, either at the Laboratory for Isotope Geology (LIG) of the Swedish Museum of Natural History in Stockholm, or at the laboratory of the Geological Survey of Finland (GTK) in Espoo. The results, published by Kornfält (1993Kornfält ( , 1996, in part showed rather large uncertainties due to the high degree of discordancy of many zircon fractions, and scattered between 1.65 and 1.75 Ga (with one sample of Småland granite north of the SBDZ at 1.78 Ga; Table 1). The question thus arose whether the lower ages really reflect a prolonged period of magmatism up to 100 million years younger than the surrounding Tving granitoids and the Småland granitoids north of the SBDZ, whether they reflect partial metamorphic resetting, possibly caused by heating and deformation related to the intrusion of the ca. 1.45 Ga  Kornfält (2007aKornfält ( , 2007b, Kornfält & Bruun (2007), and SGU digital maps, with sample location and obtained weighted average 207 Pb/ 206 Pb age for each sample. SBDZ = Småland-Blekinge Deformation Zone. the obtained concordia and weighted average 207 Pb/ 206 Pb ages are listed in Table 1 together with relevant sample information. Images of all analyzed zircons in transmitted light, reflected light and CL, with analytical spots and obtained 207 Pb/ 206 Pb ages, is available as Supplementary material 1-7 (pdf-files). Sample locations are shown in Fig. 1.

Sample descriptions and results
Sample KK90:100: Småland granite, Falan Sample KK90:100 is derived from the large area of Småland granitoids north of the SBDZ, and was taken in a road cut ca 3 km north of the boundary between Tving and Småland granitoids as marked on the map sheet 3F Karlskrona NO (Kornfält 2007a). According to the description by Kornfält (1993), the sampled rock was medium-grained and reddish grey. The purpose of the original dating was to compare the age of the Småland granitoids immediately north of the SBDZ with the 1.77 Ga age of the Tving granitoids south of this zone. This dating gave a fairly precise upper intercept age of 1778 ± 11 Ma (Kornfält 1993), similar or slightly older than the Tving granitoids and within the typical 1.81-1.76 Ga range for the TIB-1 Småland granitoids.
Nevertheless, it was considered useful to verify this age by U-Pb SIMS dating, so that it could be used as a background value with which the alleged Småland granites south of the SBDZ could be compared. In spite of the non-metamorphic nature of this granite, the selected zircons had fairly complex interiors, sometimes with CL-dark overgrowths on CL-bright and zoned interior parts. Some of the analyzed zircons were homogeneously CL-dark (Supplementary material 1). Possibly, these complexities may reflect a complex multi-stage magmatic crystallization history, rather than metamorphic overprinting. In addition, the zircons were relatively fractured and somewhat turbid in transmitted light.
Seventeen zircons and spots were analyzed. Most of them represent the interior of grains, ranging from bright to dark grey in CL (Table 2). Twelve of these spots form a cluster of concordant or almost concordant points ( Fig. 2A). One of them (zr38) yields a slightly higher age ( 207 Pb/ 206 Pb age of 1805 Ma) and has a markedly higher Th/U-ratio of 0.92, and may contain an inherited component. The remaining 11 points yield a concordia age of 1776 ± 6 Ma (2σ, MSWD 0.58) and a weighted average 207 Pb/ 206 Pb age of 1778 ± 5 Ma (2σ, MSWD 0.74; Table 1, Fig. 2A), identical to the earlier obtained TIMS age. This age is thus taken as the crystallization age of the Småland granite at Falan.
The remaining five points are −16% to −35% discordant, and have evidently lost Pb during fairly recent times. Some of these analyses appear to have hit fractures. Most of them are relatively U-rich (>400 ppm), but they fall within the same range of Th/U-ratios as the concordant points, 0.20-0.60 ( Fig. 3A left), suggesting a similar magmatic origin. On a logarithmic plot of Th versus U all points fall on a linear trend with a correlation coefficient (r) of 0.81 (Fig. 3A Kornfält (1993). It was sampled in a road cut at Torstäva 2 km SW of Ramdala church east of Karlskrona, in a ca 5-km wide body distinguished as "Småland granite" on the geological map (3F Karlskrona NO;Kornfält 2007a), and whose central part consists of the porphyritic "Filipstad-like" variety. In the original TIMS dating, three out of four discordant zircon fractions gave a seemingly precise     at Torstäva, ca 6 km north of Torp at Senoren, but substantially older than the previous ca 1724 Ma TIMS ages of these samples. The zircons from the Torp sample also show a similar, rather narrow range in Th/U-ratios of 0.36-0.62 as the Torstäva zircons (0.35-0.59), following a well-defined linear trend with r = 0.97 (Table 2, Fig. 3C). The slightly deviating zircon 06 (with a 207 Pb/ 206 Pb age of 1735 Ma) stands out with its markedly lower Th and U concentrations (10 and 25 ppm, respectively), but has a similar Th/U-ratio (0.41) as the others. The similarity, not only in age but also in appearance and Th/U-ratio of the zircons of these two samples, together with the map pattern (Kornfält 2007a(Kornfält , 2007b, would suggest that they belong to the same intrusive complex.
Sample KK93:35 is from a road cut 2 km NW of the village of Rödeby. The conventional TIMS dating of four fractions of zircon from this sample yielded a discordia with an imprecise upper intercept age of 1751 +41/−33 Ma (Kornfält 1996), similar or slightly younger than the surrounding Tving granitoid suite.
The handpicked zircons are mostly relatively clear and of good quality, with magmatic appearance and interiors ranging from CL-bright through CL-zoned to CL-grey. In a few cases (e.g. zr10, 17, 18, 22, 37, 38; Supplementary material 4), CLdark or CL-zoned rims that could be of metamorphic origin were encountered. Those that were analyzed appeared, however, to have similar age as the underlying magmatic zircon, with the possible exception of point 18b. The latter point is highly U-rich (5590 ppm), with low Th/U (0.11) and is highly discordant plotting close to an age of 400 Ma on the concordia.
Thirty-two spots in 26 zircons were analyzed ( Table 2). Out of these, 22 spots form a tight cluster yielding a concordia age of 1768 ± 5 Ma (2σ, MSWD 1.5) and a weighted average 207 Pb/ 206 Pb age of 1765 ± 3 Ma (2σ, MSWD 0.90; Table 1, Fig. 2D), taken as the magmatic crystallization age. In addition to point 18b, nine points are −6% to −32% discordant, apparently due to recent Pb loss. Most of the points, whether concordant or discordant, have Th/U-ratios between 0.40 and 1.0 (Table 2, Fig. 3D). However, there are some outliers, and the overall correlation coefficient between Th and U is only 0.75 (spot 18b excluded). Zircon spot 45 has a Th/U-ratio of 1.17, while, apart from 18b, also 03b and 06 have low Th/U-ratios of 0.20 and 0.06, respectively. The latter two spots, however, give concordant ages in the same range as the other spots ( 207 Pb/ 206 Pb ages of 1768 and 1765 Ma, respectively). CL observations, however, show that these two spots are located in CL-dark and homogeneous parts of each crystal (Table 2; Supplementary material 4), suggesting that these areas may have undergone some recrystallization, apparently lowering their Th contents, but this recrystallization presumably occurred upper intercept age of 1723 ± 4 Ma (Kornfält 1993), somewhat younger than the typical age range for both Tving and Småland granitoids, but the geological significance of this age remained obscure. Kornfält (1993) suggested that the result may be due to deformation and migmatization of this rock.
The handpicked zircons from this sample were prismatic and relatively clear with a magmatic appearance, having CL-bright and/or CL-zoned interiors and no obvious signs of any metamorphic overgrowths (Supplementary material 2). Sixteen spots in 16 zircons were analyzed, out of which one (zr05) was −6% discordant while the remaining 15 form a concordant cluster with a concordia age of 1766 ± 6 Ma (2σ, MSWD 0.28) and a weighted average 207 Pb/ 206 Pb age of 1765 ± 5 Ma (2σ, MSWD 1.3; Table  1, Fig. 2B), taken to indicate magmatic crystallization. The data points show a relatively narrow range of Th/U-ratios between 0.35 and 0.59 (Table 2), with the Th and U concentrations falling on a well-defined linear trend with a correlation coefficient of 0.98, presumed to have originated by magmatic differentiation (Fig. 3B). In spite of the metamorphic character of this rock, no metamorphic overprinting in the zircons that could explain the lowering of the conventional TIMS age could be discerned.
The obtained 1765 Ma age is similar to U-Pb TIMS and SIMS ages of the surrounding Tving granitoids (Johansson & Larsen 1989;Kornfält 1993;Johansson et al. 2006), and within the lower end of the 1.81-1.76 Ga age range typical for TIB-1 Småland (and Värmland) granitoids. Compared to the age of the Filipstad granite proper in the type area close to Filipstad in Värmland, 1783 ± 10 Ma (U-Pb TIMS on zircon; Jarl & Johansson 1988), the presently obtained age is slightly younger.

KK93:36: "Filipstad-type granite", Torp, Senoren
Sample KK93:36 is a reddish grey, medium-grained, porphyritic, weakly migmatized granite similar to the TIB-1 Filipstad granite in Värmland, sampled in a road cut at Torp on the island of Senoren east of Karlskrona (Table 1, Fig. 1; Kornfält 1996). Both the K-feldspar megacrysts and the matrix minerals have been affected by deformation, and the former are sometimes elongated into cm-wide red bands, alternating with black bands of biotite and amphibole (Kornfält 1996). Four discordant zircon fractions yielded an upper intercept age of 1724 ± 39 Ma by conventional TIMS dating (Table 1; Kornfält 1996), almost identical to the TIMS result from the "Filipstad-type granite" at Torstäva (1723 ± 4 Ma; Kornfält 1993).
The zircons from sample KK93:36 are similar to those from the Torstäva sample (KK90:102): relatively clear and prismatic with a magmatic appearance, ranging from bright to grey and zoned in CL, without any obvious metamorphic rims (Supplementary material 3). Twenty-three spots from 23 zircons were analyzed, all of which are concordant within error (Table 2, Fig.  2C). Excluding the somewhat deviating point 6, a concordia age of 1774 ± 5 Ma (2σ, MSWD 11.7) and a weighted average 207 Pb/ 206 Pb age of 1766 ± 4 Ma (2σ, MSWD 1.3; Table 1, Fig.  2C) are obtained. The high MSWD value of the concordia age calculation is caused by a tendency for most points to plot with a slight reverse discordance, which will increase the 235 U/ 207 Pb and 238 U/ 206 Pb ages as well as the overall concordia age and cause the latter not to overlap with the concordia curve. In view of this, the weighted average 207 Pb/ 206 Pb age of 1766 ± 4 Ma is considered the best estimate of the crystallization age of this rock. This age is identical to the new SIMS age of the "Filipstad-type granite" The 1758 ± 5 Ma age is considered the magmatic crystallization age of the Almö granite. It is thus similar in age to the Tving and other older Blekinge granitoids, or only marginally younger. There are some signs of metamorphic overprinting, both in the rock itself and the zircons, but no metamorphic age could be obtained.

Sample KK93:37: Tjurkö granite, Tjurkö
Sample KK93:37 comes from a large abandoned quarry on the island of Tjurkö in the archipelago outside Karlskrona, and is a greyish red, medium-grained, foliated, leucocratic granite (Kornfält 1996). Similar reddish and felsic, leucocratic and strongly foliated granite occurs over much of the neighbouring islands of Tjurkö and Sturkö, and is considered a more deformed and foliated variety of the "Filipstad-type granite" on Senoren (Kornfält 1999b). However, there is also a compositional difference, with the "Filipstad-type granite" containing 67.9-70.7 wt% SiO 2 (n = 4) and 2-8 vol% biotite (n = 6), compared to 76.3-78.7 wt% SiO 2 (n = 5) and maximum 2 vol% biotite (n = 6) for the Tjurkö granite (Tables 7-10 in Kornfält 1999b). The foliation in this area may be related to the Karlskrona Deformation Zone, a ca 20 km wide NW-trending ductile deformation zone inferred from aeromagnetic maps (cf. Johansson et al. 2006). In the conventional TIMS analysis, the four zircon fractions analyzed plotted very discordantly due to high U contents (1100-2200 ppm), and yielded a very uncertain upper intercept age of 1658 ± 104 Ma (Table 1; Kornfält 1996).
The handpicked zircons from sample KK93:37 are mostly relatively turbid and fractured, but with good magmatic zonation. A few clear grains also occur, e.g. zr02, 04, 12, 13 (Table 2, Supplementary material 6). In CL, they range from bright to dominantly grey and zoned, with little or no signs of metamorphic overgrowths. During the SIMS analysis, 18 spots in 15 zircons were analyzed. Eleven of these spots form a concordant cluster with a concordia age of 1763 ± 6 Ma (2σ, MSWD 0.88) and a weighted average 207 Pb/ 206 Pb age of 1762 ± 5 Ma (2σ, MSWD 1.2; Table 1, Fig. 2F), which is considered the crystallization age of the undeformed protolith of the Tjurkö granite. The remaining points are mostly from more turbid zircon, and are −3% to −58% discordant, evidently due to fairly recent Pb loss, in part through cracks.
Th/U-ratios range between 0.24 and 0.78, with a substantial range both in Th (26-954 ppm) and U (41-2586 ppm), but still following a rough common linear trend with r = 0.96, and with the most U-and Th-rich points being discordant (Fig. 3F). This pattern suggest that they are late-stage magmatic differentiates, rather than metamorphic overgrowths which would be high in U but low in Th.

Sample KK93:38: Jämjö granite, Gisslevik, NW Torhamn
The Jämjö granite is a fine-to medium-grained, mostly reddish, strongly foliated and leucocratic granite occurring on the Torhamn peninsula in southeasternmost Blekinge (map sheet 3F Karlskrona SO; Kornfält 1999bKornfält , 2007b. Sample KK93:38, collected from a large abandoned quarry at Gisslevik 2.2 km NW of Torhamn, is a fine medium-grained, red granite with a weak structure outlined by thin streaks of mafic minerals (Kornfält 1996). Kornfält (1999b) regards the Jämjö granite as an even more intensely deformed variety of the "Filipstad-type granite" than the Tjurkö granite, with highly attenuated K-feldspar megacrysts. As the Tjurkö granite, it is highly silicic (75.8-shortly after these two zircons had formed and a separate age cannot be determined. Zircon 11 is a very clear CL-bright crystal which is the lowest in Th (21 ppm) and U (41 ppm), but with a "normal" Th/U-ratio of 0.50. Excluding the deviating points 03b, 06 and 45 (as well as 18b) improves the Th-U correlation to 0.93 (Fig. 3D).
The obtained weighted average 207 Pb/ 206 Pb age of 1765 ± 3 Ma is slightly older than the previously obtained upper intercept TIMS age of 1751 + 41/-33 Ma (Kornfält 1996), although within error the same, and indistinguishable from that of the surrounding Tving granitoids. This would suggest that the Rödeby granite formed in close connection with the Tving granitoids, although locally showing an intrusive relationship as exemplified by the Tving granitoid xenolith. A possible metamorphic overprinting is hinted on by point 18b, but no age could be determined from this overgrowth due to its extremely U-rich, metamict and highly discordant nature.

Sample 89071: Almö granite, Almö
The Almö granite forms a massif of red to reddish grey, fine-to medium-grained, leucocratic and gneissic or foliated granite on the island of Almö, which straddles the boundary between map sheets 3F NO and 3F SO west of Karlskrona (Kornfält 2007a(Kornfält , 2007b. It contains xenoliths of the so-called "coastal gneiss" (see Fig. 24 in Kornfält 1999b), a grey fine-grained rock that may be of volcanic or subvolcanic origin. Sample 89071 is from a small abandoned quarry on the northern part of Almö island and is finely medium-grained, red to greyish red and foliated (Kornfält 1996). Conventional TIMS dating of four zircon fractions yielded a discordia with an ill-defined upper intercept age at 1716 +105/−59 Ma (Kornfält 1996; Table 1). In addition, Kornfält (1996) reported the result from the U-Pb analysis of two titanite fractions, which yielded an upper intercept age of 1471 +153/−59 Ma, suggesting some metamorphic overprinting related to the intrusion of the large nearby ca 1.45 Ga Karlshamn granite massif.
Handpicked zircons from sample 89071 were relatively turbid and fractured and thus not of very good quality. Although dominated by CL-grey or CL-zoned interiors of magmatic appearance, many grains contained CL-dark outer rims that potentially could be of metamorphic origin (Supplementary material 5). Twenty-two spots in 19 zircon crystals were analyzed ( Table  2). Out of these, 11 points form a concordant cluster with a concordia age of 1758 ± 6 Ma (2σ, MSWD 0.005) and a weighted average 207 Pb/ 206 Pb age of 1758 ± 5 Ma (2σ, MSWD 1.04; Table  1, Fig. 2E), one point (zr22) is 7% reversely discordant and the remaining ten points are −4% to −66% discordant. Most of the points, whether concordant or discordant, show rather moderate U concentrations in a narrow range between 100 and 300 ppm, with only a few points at higher concentrations, and with relatively high Th/U-ratios between 0.48 and 1.00 (Table 2, Fig. 3E left) and a Th-U correlation coefficient of 0.94 (Fig. 3E right, point 20b excluded). Both texturally and from their Th/U-characteristics most of the discordant points appear to be normal magmatic zircon that have lost radiogenic Pb. This also includes the potential overgrowth spots 19b and 38b. Only the most discordant point, 20b, which texturally appears to be from a CL-dark overgrowth cutting across the older zircon, shows a very high U content (4930 ppm) and low Th/U-ratio (0.04) suggestive of a metamorphic origin, but unfortunately no reliable age can be obtained from that spot . GFF 138 (2016) Conventional TIMS dating of four zircon fractions from sample KK93:38 gave an uncertain upper intercept age of 1735 ± 56 Ma for the Jämjö granite (Kornfält 1996).
As in the Almö and Tjurkö granites, the zircons in the Jämjö granite sample are relatively turbid, fractured and rounded in their morphology, although some relatively clear crystals could be selected during handpicking. In CL, they range from bright to grey, and from zoned to having more irregular and patchy interiors, still without any visible metamorphic overgrowths. In several cases, the pattern goes from CL-grey and even in the centre, to CL-bright and zoned in the outer part (e.g. zr07, 17, 50, 52), i.e. a reverse magmatic zonation with apparent lowering of the Th and U concentrations, perhaps due to magma replenishment (Table 2, Supplementary material 7).
Twenty-five spots from 25 zircons were analyzed ( Table 2). The results are somewhat complex. Eighteen of these points form a cluster with a slight reverse discordancy, yielding a concordia age of 1779 ± 5 Ma (2σ, MSWD 16) and a weighted average 207 Pb/ 206 Pb age of 1770 ± 4 Ma (2σ, MSWD 0.73; Table  1, Fig. 2G). Two points (zr10 and 54) show a greater degree of reverse discordancy and were excluded from this calculation. Four points (zr06, 13, 14, 40) form an additional semi-concordant cluster with a slight normal discordancy, while a fifth point is −11.9% normal discordant. Considering the high MSWD value of the concordia age, caused by the slight reverse discordancy of the analyses within the main 18-point cluster, the weighted average 207 Pb/ 206 Pb age of 1770 ± 4 Ma is considered the best estimate for the crystallization age for the protolith of the Jämjö granite. The subsequent deformation, with development of foliation and subsequent folding, has not left any obvious imprint on the zircons.
The analyzed Jämjö granite zircons of sample KK93:38 show unusually high Th/U-ratios between 0.49 and 1.19, but still following a common magmatic differentiation trend with r = 0.91 (Fig. 3G). Given the limited amounts of discordancy, there are no systematic differences in Th and U contents between the "concordant" group of 18 points, and the points deviating in either direction in their discordancy. The differences in Th-U systematics between the zircons in the Jämjö and Tjurkö granite, compared to those in the "Filipstad-like granites" at Torstäva and Torp, however would suggest that the Jämjö and Tjurkö granites are not more foliated direct counterparts of the latter. The more silica-rich composition of the Tjurkö and Jämjö granites compared to the "Filipstad-like granites" would also argue against such an interpretation. Rather, the Jämjö and Tjurkö granites could be late-stage more felsic differentiates from the same magma as the "Filipstad-type granite", belonging to the same plutonic complex and within error having the same age, but being more susceptible to subsequent deformation.

Relation between the Småland and Tving granitoids
The new SIMS weighted average 207 Pb/ 206 Pb age of 1778 ± 5 Ma for Småland granite sample KK90:100, taken only 3 km north of the Småland-Blekinge Deformation Zone, confirms the previously obtained TIMS age. This age is somewhat older than SIMS ages obtained for Tving granitoids and other gneissic granitoids in Blekinge south of the SBDZ, which fall between 1763 and 1750 Ma (excluding the 1816 ± 9 Ma Nättraby gneissic 76.3 wt% SiO 2 , n = 4, Table 12 in Kornfält 1999b) and low in mafic minerals. The strong deformation may be attributed to its position within the ductile NW-trending Karlskrona Deformation Zone. The more coarse-grained and megacrystic "Filipstad-like" granitoids at Torstäva and Torp would then represent somewhat less felsic, more competent and well-preserved lenses within this deformation zone, which encompasses the whole coastal area in southeastern Blekinge. As pointed out by Kornfält (1999b), the foliation in the Jämjö granite has undergone subsequent ductile folding (see Fig. 20 in Kornfält 1999b). Fig. 4. Summary and comparison of weighted average 207 Pb/ 206 Pb SIMS ages of zircons from Blekinge rocks, illustrating the similarity in age of the "Filipstad-type granites" and leucogranites in eastern Blekinge dated here with the surrounding Tving granitoids and other Blekinge rocks (except for the older Nättraby gneissic granite) dated by Johansson et al. (2006), as well as the slightly older age of the Småland granite at Falan north of the Småland-Blekinge Deformation Zone. Grey error bars show 2σ uncertainty of each age. Shown on top is a compilation of 24 published U-Pb zircon TIMS and SIMS ages of Småland granitoids in the whole Småland lithotectonic unit north of the SBDZ, from the SGU website http://apps.sgu.se/kartvisare/kartvisare-bergets-alder-sv.html, most of which are significantly older than the Tving granitoids and the other Blekinge rock units.
Possibly, deformation and metamorphism occurred under relatively dry conditions, preventing growth of new metamorphic zircon. From this, it appears that the younger ages obtained by conventional TIMS in some samples were not caused by metamorphic overprinting, but rather by the scatter and uncertainty caused by the high degree of discordancy of many of the multi-grain fractions.
The age of the metamorphism and deformation in eastern Blekinge, and in particular along the Karlskrona Deformation Zone as well as the Småland-Blekinge Deformation Zone, thus remains somewhat enigmatic. The presence of a deformed xenolith of 'coastal gneiss' (dated to 1765 ± 6 Ma in western Blekinge; Johansson et al. 2006) within the 1758 ± 5 Ma Almö granite (Kornfält 1999b, Figure 24) would suggest that metamorphism and deformation of the former occurred shortly after its formation, prior to, or contemporaneous with, the intrusion of the Almö granite. A similar conclusion was reached by Johansson et al. (2006), based on the presence of 1754 ± 11 Ma red aplitic dykes crosscutting the deformed Tving granitoid at Hallarum in eastern Blekinge.
On the other hand, metamorphic rims on zircons from the "coastal gneiss" at Kullerön in western Blekinge and from a migmatite at Lindö, analyzed by SIMS by Johansson et al. (2006), both provided ages of 1441 ± 9 Ma. Titanite from several rocks analyzed by TIMS by the same authors yielded a cluster of 207 Pb/ 206 Pb ages at 1431 ± 6 Ma. Thus, the combined evidence seems to favour two separate metamorphic events, one encompassing pervasive deformation under relatively dry conditions in close connection with the formation of these rocks at 1.76-1.75 Ga, and a second episode, possibly of more static character, with heating and migmatization related to the intrusion of the Karlshamn granitoid suite at ca 1.45 Ga (Åberg 1988;Čečys & Benn 2007) and, more generally, to the so-called Danopolonian event of Bogdanova (2001). Movements along the Karlskrona Deformation Zone are envisaged to have occurred in close connection with the first event, while uplift along the SBDZ may have occurred somewhat later, but prior to the intrusion of the Karlshamn suite (Kornfält 1999a;Kornfält & Bruun 2002). granite; weighted average 207 Pb/ 206 Pb ages from Johansson et al. 2006), as well as the 1770-1758 Ma range of weighted average 207 Pb/ 206 Pb ages of "Småland-like" granitoids south of the SBDZ in the present study (Fig. 4). A compilation of published U-Pb zircon TIMS and SIMS ages of Småland granitoids within the entire Småland lithotectonic unit north of the SBDZ (top of Fig.  4) shows these to be systematically older (1.77-1.81 Ga, with most of them being around 1.80 Ga) than the Tving granitoids and other Blekinge rocks (1.75-1.77 Ga) Kornfält (1999a) and Kornfält & Bruun (2002) suggested the Tving granitoids in Blekinge to be more deformed counterparts of the Småland granitoids north of the SBDZ. They could represent a deeper crustal level uplifted by tectonic movements along the SBDZ, with their slightly lower age being caused by slower cooling (Kornfält 1999a). Considering the high closure temperature of zircon, making it record crystallization rather than cooling, the latter explanation appears unlikely and the age difference is probably real. Since there are also some differences in composition, with the Tving granitoids being somewhat more mafic, and the Småland granitoids more felsic and alkali-rich (Lindh et al. 2001;Kornfält & Bruun 2002), it may be useful to uphold a distinction between them as two separate but closely related magmatic suites. The Tving granitoid suite and the Småland granitoid suite could then be considered members of the TIB-1 supersuite. This would be similar to the conclusions reached by Lindh et al. (2001), when comparing Tving and Småland granitoids on either side of the SBDZ.

The age of the "Småland-type" granitoids in Blekinge and their relation to the Tving granitoids
The granitoids from eastern Blekinge dated here by SIMS have been considered by Kornfält (1993Kornfält ( , 1996Kornfält ( , 1999aKornfält ( , 1999b to be Småland granitoids enclosed within the large Tving granitoid area in eastern Blekinge south of the SBDZ, whether being megacrystic "Filipstad-type" granitoids or medium-to finegrained, more or less foliated leucogranites. The new age data show them to be coeval with the surrounding Tving granitoids (Fig. 4). Although they locally may show an intrusive relationship to the Tving granitoid (block of Tving granitoid within the Rödeby granite), the age difference is probably minor. These granitoids, and especially the more felsic and leucocratic ones, could possibly be late-stage differentiates from the Tving granitoid suite, which locally have undergone strong deformation within the Karlskrona Deformation Zone, although such a genetic relationship remains to be tested by geochemistry. In any case, they do not represent a series of separate younger magmatic events, as some of the older TIMS age data could be taken to indicate, if taken at face value.

Age of metamorphism and deformation
In spite of the pervasive deformation affecting several of these rocks, the zircons show only limited signs of metamorphic overprinting and metamorphic overgrowths. Only two examples of what appears to be metamorphic overgrowths could be seen, one each in sample 89071 (Almö granite, zr20b) and sample KK93:35 (Rödeby granite, zr18b), none of which could be dated.