2. Geological setting of Charnwood Forest samples

The rocks that have been sampled (Fig. 1b) are located within the Charnwood Forest inlier of basement rocks lying immediately to the east of the Midlands Microcraton, the tectonic boundary of which is defined by the Thringstone Fault (Smith, Reference Smith, Brooks and Glennie1987; Lee, Pharaoh & Soper, Reference Lee, Pharaoh and Soper1990). The Charnwood cleavage is slightly sinusoidal, but with a predominant ESE strike (100° to 120°) that is for the most part oblique with respect to the SE axial trend of the main Charnwood anticline. It is imprinted on volcanic and volcaniclastic rocks of the Charnian Supergroup (Moseley & Ford, Reference Moseley and Ford1985), of latest Neoproterozoic age, as well as on the sequence of silty mudrocks that comprises the overlying Swithland Formation (Brand Group). It had long been assumed that all of the Charnwood Forest basement rocks were Neoproterozoic in age; however, it is now thought that the Swithland Formation is significantly younger than this. In many local churchyards, headstones carved from the Swithland Formation have yielded trace fossils (Teichichnus) indicative of a Phanerozoic, possibly Early Cambrian age (Bland & Goldring, Reference Bland and Goldring1995; McIlroy, Brasier & Moseley, Reference McIlroy, Brasier and Moseley1998).

All of the samples are characterized by an intense, millimetre- or micron-scale cleavage, as described in Table 1. It is therefore anticipated that the isotope determinations will include the age of the fabric-forming event in both the Swithland Formation and the underlying Charnian Supergroup. The in situ samples (Fig. 1b) are from three different structural settings. Samples JNC 649 and JNC 659 are silty mudstones with abundant detrital quartz and albitized feldspar grains. They were collected from disused quarries in the Swithland Formation at geographically separated localities, respectively on the eastern and western limbs of the Charnwood anticline, in order to minimize bias caused by local and possibly aberrant lithological and/or tectonic conditions. A further in situ sample (JNC 542A, Table 1) is a silvery, strongly foliated and weakly lineated, highly micaceous rock (phyllonite) from a composite brittle/ductile fault zone cutting a Charnian porphyritic dacite in Whitwick Quarry. The ESE orientation of this fault zone is parallel to the regional cleavage (Carney, Reference Carney2005), suggesting that faulting and cleavage formation were contemporary, and that strain during the cleavage-forming event may have been heterogeneously distributed. The value of this sample is that, because its protolith was a primary igneous rock, it will be devoid of detrital mica contaminant grains.

Table 1. Summary petrographical features of the samples

Mica types: a – fabric-forming; b – detrital; c – replacive.

Two further Swithland Formation samples (JNC 662, 663; Table 1) are not in situ; they were collected as small, fresh cleavage flakes from the crumbled corners of headstones in Ratby churchyard, to the south of the Charnwood study area shown in Figure 1b. They were included in this study because the headstones have been quarried from the best quality local slate resource, which we have found to be particularly pure in terms of their high original mud content. In many quarries, such material occurred in narrow, subvertical ‘seams’, less than 5 m wide, which are now inaccessible due to flooding of the quarries. The headstones are also of major significance to the stratigraphy of Charnwood Forest, as noted previously.

3. Metamorphic conditions

X-ray diffraction (XRD) techniques were used to determine the Kübler indices of white mica (illite) ‘crystallinity’. The Kübler index (KI in Δ°2θ) measures small reductions in the half-height width of the dioctahedral mica ~ 10 Å XRD peak resulting from crystallite thickening during reaction progress in the smectite–illite–muscovite series (Merriman & Peacor, Reference Merriman, Peacor, Frey and Robinson1999). White mica crystallite thickness generally increases with grade, a process accelerated during slaty cleavage development, so that the mean thickness of epizonal mica would be expected to exceed 520 Å, whereas anchizonal white micas have mean thicknesses in the range 230–520 Å (Merriman & Peacor, Reference Merriman, Peacor, Frey and Robinson1999). These crystallographic changes can be used to measure the progress of advancing metamorphic grade (Merriman, Roberts & Peacor, Reference Merriman, Roberts and Peacor1990; Merriman et al. Reference Merriman, Roberts, Peacor and Hirons1995a; Warr & Rice, Reference Warr and Rice1994). In pelitic rocks, KI values are used to define the limits of a series of metapelitic zones of very low- and low-grade metamorphism: diagenetic zone (KI > 0.42), anchizone (KI 0.42–0.25) and epizone (KI < 0.25) (Kisch, Reference Kisch1990).

For the northern part of the Charnwood Forest area shown in Figure 1b, a metamorphic zonation was previously determined by Merriman & Kemp (Reference Merriman and Kemp1997), based on the distribution of mean values and the range of KI data for 71 samples. The near-ubiquitous development of epizonal KI values, which are seen in the data for the Swithland Formation samples analysed for this study (Table 1), reflects the well-developed tectonic fabric of mica-rich rocks in the Charnwood basement, and suggests that white mica recrystallization is strain-enhanced (note that due to the small size of the samples, KI determinations could not be carried out for the headstone rocks JNC 662 and 663). The highest grades indicate temperatures of 300–350 °C, corresponding to burial depths of up to 10 km. The Charnwood shear zone rock (JNC 542A; Table 1) shows slightly lower grades than the in situ slates and this may reflect retrogression or comminution of white mica crystallites caused by high strain rates and sluggish recovery following ductile deformation (Kemp & Merriman, Reference Kemp and Merriman1995).

4. Petrography and mica fabrics

Three petrographic types of white mica have been distinguished and their estimated proportions, along with other summary data for the samples, are shown in Table 1. Type (a) fabric-forming micas (Fig. 2a–c) are found as intergrowths with chlorite in the orientated microstructural P-domains that define the slaty cleavage (Knipe, Reference Knipe1981). Estimates of their maximum length (Table 1) are based on measurements made on apparently optically continuous grains. The uncertainties of resolving crystallographic continuity in mica–chlorite intergrowths on this scale by optical methods, however, underlines the approximate nature of the measurements. Type (b) detrital micas (Fig. 2c) include the white mica within chlorite–mica stacks. Although these stacks are of detrital origin, both their white mica and chlorite minerals probably developed during late diagenesis and burial (e.g. Li et al. Reference Li, Peacor, Merriman and Roberts1994). Type (c) replacive white mica was generated by secondary processes pre-dating the fabric-forming event, and does not show a preferred orientation. It includes mica replacing crystal, vitric and lithic volcanic grains, and grain coatings representing recrystallized matrix clay mica.

Figure 2. Photomicrographs of dated samples of Charnian slates and phyllonite. (a) Sample JNC 542A: continuous, non-domainal slaty cleavage in phyllonite. Note penetration of fabric into fractures in quartz phenocryst; crossed polarizers. (b) Sample JNC 542A: continuous, anastomosing slaty cleavage in phyllonite matrix; crossed polarizers. (c) Sample JNC 649: domainal cleavage in siltstone sample JNC 649; many of the larger, detrital clasts show pressure-solution textures; crossed polarizers.

In terms of their relative ages, type (a) white micas that define the slaty cleavage microfabric are the most recently formed and would be expected to give ages that approximate to the fabric-forming event. Type (b) detrital micas are assumed to have been derived from contemporary basement outcrop at the time of sedimentary accumulation and would be expected to give the oldest ages. As the degree of tectonic fabric formation increases with grade, however, more detrital micas are likely to have recrystallized into the cleavage domains (e.g. Ho, Peacor & van der Pluijm, Reference Ho, Peacor and Van Der Pluijm1999; Van de Pluijm et al. Reference Van de Pluijm, Ho, Peacor and Merriman1998) and therefore in epizonal grade samples such as those from Charnwood Forest, a high proportion of the detrital mica population is likely to have isotopically equilibrated under the metamorphic conditions that generated the tectonic fabric-forming micas (e.g. Dong et al. Reference Dong, Hall, Halliday, Peacor, Merriman and Roberts1997). It is noted that the muscovite closure temperature for argon is generally about 500 °C (Villa, Reference Villa1998); however, given that the grains measured for this study are relatively small (typically less than 100 μm; Fig. 2 and Table 1), the closure temperature for the studied micas would be somewhat lower, for instance, 400 to 450 °C, or even lower (300 to 350 °C) if the calculations of Parry et al. (Reference Parry, Bunds, Bruhn, Hall and Murphy2001) are to be followed. These estimates for the closure temperature of the dated muscovite are therefore similar or to higher than the metamorphic temperatures. The detrital population may not be completely reset, however, as indicated by previously determined age ranges for Welsh Basin epizonal slates reviewed by Sherlock et al. (Reference Sherlock, Kelley, Zalasiewicz, Schofield, Evans, Merriman and Kemp2003). As already noted, detrital mica will be absent from the ductile shear zone sample (JNC 542A), which cuts dacitic rocks of primary igneous origin. Type (c) replacive white mica, because of its very fine grain size, has almost certainly equilibrated to the metamorphic conditions prevailing during fabric formation. In terms of ⁴⁰Ar–³⁹Ar isotope systematics, this type of mica would have formed closed systems at the same time as the fabric-forming mica.

All samples possess an intense slaty cleavage microfabric comprising orientated intergrowths of white mica and chlorite. In the phyllonite (ductile shear) sample JNC 542A, a continuous non-domainal fabric is developed, which may penetrate fractures in some granulated, relict quartz phenocrysts (Fig. 2a). Although the cleavage is domainal in character in the siltstone sample JNC 649, pressure-solution has elongated most detrital grains into the penetrative fabric (Fig. 2c). The penetrative development of the microfabric is reflected in the epizonal grade of the Charnwood samples, and in the relatively low proportions of type (b) and type (c) micas (less than 10% and 20%, respectively; Table 1). Nevertheless, a few detrital micas have survived in the slates (Fig. 2c).

5. Analytical methods

Thin slices of about 5 × 10 × 0.2 mm were cut from the various lithologies, polished on one side, and cleaned in acetone in an ultrasonic bath. They were then wrapped individually in aluminium foil, and irradiated for 120 hours in the Oregon State University TRIGA reactor. The neutron flux was measured with mineral standards wrapped in aluminium foil and surrounding and intercalated between the sample packets. Its variability is known down to approximately ± 0.3% relative (1σ) at any point on the rock slices, and is mainly due to the horizontal flux gradient across the sample rock slice. The uncertainty of the measured neutron flux is propagated into the final weighted average ages (for spot fusion) and plateau ages (for incremental heating) and typically contributes to less than 30% of the age uncertainty. Absolute ages are reported relative to the biotite standard GA1550 (98.79±0.96 Ma; McDougall & Roksandic, Reference McDougall and Roksandic1974; Renne et al. Reference Renne, Swisher, Deino, Karner, Owens and Depaolo1998).

Argon was extracted from the samples by both laser spot fusion and incremental heating techniques at the Scottish Universities Environmental Research Centre. For the spot fusion analyses, argon was extracted from up to 12 spots per sample, chosen in areas of the rock slices with the finest grain sizes (to avoid detrital micas) and highest mica contents (to avoid other mineral phases). Heating was accomplished with a 65W CW Nd-YAG laser beam focused through a modified petrographic microscope and shuttered externally in 5 to 10 bursts of 0.1 second to minimize heating of the surrounding sample. The resulting fusion pits were generally less than 200 μm in diameter and went through the thickness of the rock slice. For the incremental heating analyses, carefully selected chips of the rock slices measuring about 0.5 mm² were heated for 120 seconds for each step with a defocused 25W CO₂ laser beam of about 1.5 mm in diameter. Power levels for each step ranged from a nominal 1.3 to 8.0% of full power, corresponding to 0.3 to 2 W.

The released gas was purified after heating for 7 to 8 minutes with two SAES Zr–Al GP50 getters, one operated at about 400 °C and the other at room temperature, as well as with an off-line stainless steel finger cooled with an acetone and dry ice slush. For the spot fusion analyses, the cleaned gas was measured with a MAP-215 mass spectrometer using a fixed slit giving a resolution of about 300, and an electron multiplier operated at a gain of about 4000. For the incremental heating analyses, the cleaned gas was measured with a MAP-215/50 mass spectrometer using an adjustable slit set to a resolution of about 600 and a Johnson MM-1 electron multiplier operated at a gain of about 15000. The raw argon data were corrected for total system blank, reactor-induced mass interferences, mass fractionation and ³⁹Ar and ³⁷Ar decay. Individual ⁴⁰Ar–³⁹Ar age and error calculations follow Dalrymple et al. (Reference Dalrymple, Alexander, Lanphere and Kraker1981), and all analytical uncertainty margins are reported at the 1σ level.

6. ⁴⁰Ar–³⁹Ar analytical data

The full analytical data are presented in Table 2 and are separated into two groups as a function of the extraction method: 52 spot fusion analyses (summarized in Fig. 3) and six step heating analyses (Fig. 4). The ages (at the 1σ confidence level) are hereafter presented, treated, and compared on the basis of a separation into: Swithland Formation in situ rocks (JNC 649 and JNC 659), Swithland Formation headstones (JNC 662 and JNC 663), and the Charnian ductile shear zone phyllonite (JNC 542A). The statistical treatment used for the spot fusion results was the cumulated probability analysis (e.g. Alexandre et al. Reference Alexandre, Chalot-Prat, Saintot, Wijbrans, Stephenson, Wilson, Kitchka and Stovba2004); it was used in order to obtain age group distributions and therefore age information. One of the advantages of this method is that there is no choice of bin width, as in classical histograms, but of how often the probability is calculated, the only effect of which is the smoothness of the final curve. In our case, probability was calculated every 1 Ma, ensuring that the overall shape of the curve is not affected at all by any calculation bias, as the error margin of each age used for the calculation is significantly higher than 1 Ma (Table 2). The individual age probabilities (assuming Gaussian distribution) were integrated over the entire time span chosen, between 400 and 460 Ma. This method thus avoids the risk of creating non-meaningful peaks, or missing important peaks out. The treatment relies on the hypothesis that the dated phases contain several sites with distinct isotopic composition, each of them therefore giving different, but geologically meaningful, age information. In the case of the Charnwood Forest samples analysed here, the largest reservoir (microscopically estimated at 70 to 92%; Table 1) corresponds to micas that define the slaty cleavage microfabric, followed by replacive (3 to 20%) and by detrital micas (5 to 10%; Table 1). The expected ages of the deformation event would be given by the cleavage microfabric fine-grained white micas, as well as large proportions of the detrital and replacive micas, as discussed above. The fabric ages would correspond to the major peak in the cumulative probability diagrams (Fig. 3).

Table 2. Results of the ⁴⁰Ar/³⁹Ar analyses of the samples from the studied Charnwood samples

Ages in bold are those used for the calculation of a plateau age.

Figure 3. Cumulative probability diagrams for all ⁴⁰Ar–³⁹Ar spot analyses given in Table 2, organized by groups of dated samples, as indicated in the text. The age and error margin of each individual peak are indicated, together with the number of individual ages defining the relevant peak.

Figure 4. ⁴⁰Ar–³⁹Ar step heating age spectra of individual rock fragment slices. Note that the Charnwood samples yielded good age plateaus (see text for discussion).

Ages determined by spot analyses range from 404.1±1.9 to 452.7±5.1 Ma, with two major clusters at c. 435 and c. 415 Ma (Table 2). For the Swithland Formation in situ samples (JNC 649, 659; Fig. 3a), the major peak is broad, with an age of 432.0±7.2 Ma (10 spot analyses; MSWD = 1.01). For the Swithland Formation headstone samples, which are expected to be the most pure in terms of original mud content, one well-defined peak is present at 415.8±4.0 Ma (n=16; MSWD=0.22), with a minor one on its left flank (Fig. 3b). The Charnian ductile shear sample (JNC 542A) shows one single and well-defined peak at 419.4±3.8 Ma (n=10; MSWD=1.60; Fig. 3c); this sample is significant because it is a tectonized igneous rock and so cannot have contained any detrital sedimentary components. All peaks defined in the probability density diagrams satisfy the goodness of fit criterion of MSWD lower than 2.5, as the calculated MSWD values range from 0.22 to 1.60.

Step-heating experiments were conducted on three samples from Charnwood Forest, each being analysed twice, and the results are presented in Figure 4. The spectra obtained are mostly flat, even though the first two to three steps of sample JNC 663 indicate significant superficial radiogenic argon loss; the higher ages of the last one or two steps in sample JNC 649 probably indicate the presence of inherited ⁴⁰Ar, and the first step of the duplicate analysis of sample JCN542A might suggest limited ³⁹Ar_K recoil (Fig. 4). These perturbations of the age spectra are to be expected with such fine-grained material as used here, where loss of both ⁴⁰Ar* and ³⁹Ar_K are to be expected. However, the obtained spectra permit the definition of plateau ages (based on at least three consecutive steps corresponding to more than 50% of the gas) or the calculation of gas proportion-weighted average ages of relatively flat portions of the spectrum, ranging from 413.7±1.2 Ma (sample JNC 542A, steps 2 to 5, 58% of the gas) to 429.9±0.5 Ma (sample JNC 649, steps 1 to 3, 51% of the gas). The ages obtained by double analyses are consistent, even though they do not overlap at the 1σ uncertainty level (Fig. 4).

In summary, the step-heating results closely reflect the spot analyses in revealing two age groupings: the main one at c. 425 to 416 Ma (three plateau ages and an average age from samples JNC 542A and JNC 663) and a further one at c. 430 Ma (sample JNC 649). It is important to note that for sample JNC 649 the step-heating analysis (Fig. 4) and the majority of single spot analyses (Fig. 3a) give very similar ages of c. 430 Ma and 432 Ma, respectively.

7. Discussion and interpretation of the age data

The spread of the ⁴⁰Ar–³⁹Ar age data gives a Silurian to earliest Devonian age range, between 432 and 416 Ma, for Charnwood samples. Of the two age groupings represented within this cluster, the younger one, of c. 425 to 416 Ma, which is both well defined (Fig. 3) and present in all dated samples (see also Table 2), is interpreted to constrain the age of the main deformational event, which included metamorphism, folding, ductile shearing and the development of the penetrative cleavage. This age grouping was obtained from single-spot and step-heating analyses, and it indicates a Silurian (Wenlock) to earliest Devonian (Lochkovian) age on the timescale of Gradstein et al. (Reference Gradstein, Ogg and Smith2005). It is important to stress that the three purest samples, in terms of tectonic mica content, are represented within this younger age grouping; they were collected from Swithland Formation headstones (two samples; JNC 662, 663) and a ductile shear zone cutting igneous rocks (one sample; JNC 542A) and they give well-defined age peaks at 415.8±4.0 and 419.4±3.8 Ma, respectively (Fig. 3b, c).

The deformational event at c. 425–416 Ma that is constrained by the new ⁴⁰Ar–³⁹Ar determinations does not contradict the possibility of an Early Cambrian age recently proposed for the Swithland Formation (Bland & Goldring, Reference Bland and Goldring1995; McIlroy, Brasier & Moseley, Reference McIlroy, Brasier and Moseley1998). The new age estimate does, however, rule out a recent proposal by Compston, Wright & Toghill (Reference Compston, Wright and Toghill2002) that the cleavage deformation could have arisen during earliest Cambrian time. A metamorphic hiatus, implying weak deformation, probably did originally exist between the Neoproterozoic Charnian Supergroup and ?Early Cambrian strata of the Swithland Formation, by analogy with the situation in the Midland Microcraton's interior exposed at Nuneaton (Fig. 1a). There, deep diagenetic zone sedimentary rocks of Early Cambrian age unconformably overlie relatively higher grade (lower anchizonal) rocks of the Caldecote Volcanic Formation, equated with the Charnian Supergroup (Bridge et al. Reference Bridge, Carney, Lawley and Rushton1998; Merriman et al. Reference Merriman, Pharaoh, Woodcock and Daly1993). In Charnwood Forest, however, the evidence for this metamorphic discontinuity would have been obliterated by the subsequent epizonal recrystallization of the Neoproterozoic and Early Cambrian rocks documented here.

The older (Silurian) age grouping at c. 432–430 Ma for the in situ Swithland Formation may reflect the imprint of an earlier thermal event. The much lower age values, of c. 404 Ma and c. 408 Ma, most probably correspond to the effects of argon loss due to perturbations in the isotopic system, as they are defined by a very limited number of analyses.

8. Regional implications

Given the limitations of the whole-rock dating methods employed, these data have provided a Silurian to Early Devonian estimate for the age of Charnwood deformation, which is significantly older than Acadian deformational ages recorded elsewhere in southern mainland Britain. For example, within the paratectonic Caledonides belts to the west and north of the Midlands Microcraton (Fig. 1a), Sherlock et al. (Reference Sherlock, Kelley, Zalasiewicz, Schofield, Evans, Merriman and Kemp2003) used ⁴⁰Ar–³⁹Ar laser microprobe analysis to date pressure fringe white mica, giving a mean age of 396.1±1.4 Ma for the Acadian cleavage. Farther north, a similar cleavage age of 397±7 Ma was obtained from a Silurian mid-anchizonal grade metabentonite in the Ribblesdale inlier (Fig. 1a) of northern England (Merriman et al. Reference Merriman, Rex, Soper and Peacor1995b). There, however, an Rb–Sr isochron age of 465±10 Ma has also been determined, suggesting the possibility that earlier, diagenetic or metamorphic events may survive penetrative cleavage deformation (Dodson & Robinson, Reference Dodson and Robinson2006). The c. 397 Ma age determinations constrain the peak of Acadian metamorphism and deformation to latest Emsian time, according to the U–Pb chronology for this part of the Devonian suggested by Kaufmann et al. (Reference Kaufmann, Trapp, Mezger and Weddige2005).

Although more work clearly needs to be carried out on the Charnwood rocks, our preliminary findings suggest that their argon reservoirs could have been influenced by the events that accompanied closure of the Iapetus Ocean, towards the end of the Silurian Period. According to Soper & Woodcock (Reference Soper and Woodcock2003), a ‘sizeable time gap’ is present between Iapetus closure and the onset of Acadian deformation. Within this pre-Acadian tectonic regime (Fig. 5), a changeover from sinistral convergence to sinistral transtension occurred at about 420 Ma in the Welsh Basin and Lake District regions of southern Britain (Soper & Woodcock, Reference Soper and Woodcock2003), roughly coinciding with the younger estimated age range for the Charnwood deformation. Further afield, the Charnwood event may also have parallels in a c. 425–414 Ma ‘Neo-Acadian’ tectonic imprint recognized in the Avalonian terranes of the USA (Robinson et al. Reference Robinson, Tucker, Bradley, Berry and Osberg1998). The main Acadian deformational event in southern Britain has been attributed to transpression caused by the subsequent collision of Armorica with the terranes of Laurentia, Avalonia (the Early Palaeozoic microcontinent) and Baltica that had previously been docked along the Iapetus closure line (Soper, Webb & Woodcock, Reference Soper, Webb and Woodcock1987; Dewey & Strachan, Reference Dewey and Strachan2003).

Figure 5. Proposed tectonic setting during the c. 425–416 Ma deformation in Charnwood Forest. Major basement terranes are: EC – Eastern Caledonides; BM – Brabant Massif. Structural lineaments are: NC/FH – North Craven/Flamborough Head fault system; DSHFZ – Dowsing–South Hewett Fault Zone. Other information is given in Figure 1a. Data sources are from compilations by Pharaoh (Reference Pharaoh1999), Soper & Woodcock (Reference Soper and Woodcock2003) and Van Grootel et al. (Reference Van Grootel, Verniers, Geerkens, Laduron, Verhaeren, Hertogen and De Voss1997).

An attempt has been made to clarify the mechanism that caused the Charnwood deformation, in order to compare it with Acadian structural styles in the Welsh Basin and northern England. The Charnwood cleavage orientation is slightly sinuous, bearing ESE (100° to 120°) and cutting obliquely across the SE axial trend of the main Charnwood anticline (Fig. 1b). This was described by J. Moseley (unpub. Ph.D. thesis, Univ. Leicester, 1979) as an anticlockwise cleavage transection, and in order to constrain it further, detailed measurements of fold/cleavage relationships in the Charnian Supergroup were made by TCP and JNC. It was found that the fold-girdle representing the best fit to bedding poles is offset from the mean cleavage pole (and the regional fold-axis predicted from the fold-girdle likewise offset from the observed bedding on cleavage intersections) by an angle of 8°, confirming anticlockwise transection (Fig. 1c). These relationships suggest that the Charnwood deformation arose within a prevailing regime of dextral transpression. It is noteworthy that in the paratectonic Caledonides domains to the west and north of the Midlands Microcraton, the Acadian event produced a transecting cleavage generated within a sinistral transpressional regime. These younger, Acadian structures were attributed to the northwards movement of the Midlands Microcraton as a rigid indenter (Soper, Webb & Woodcock, Reference Soper, Webb and Woodcock1987; Woodcock, Reference Woodcock1990).

To account for these different ages and styles of transection geometry on either side of the Midlands Microcraton, it is suggested that the ⁴⁰Ar–³⁹Ar age estimate of 425–416 Ma in Charnwood Forest may include the record of early deformation in the assemblage of terranes that, south of the Scottish Highlands, were converging with the Iapetus Ocean suture line (Fig. 1a). It is possible that, as sinistrally oblique accretion in the Scottish Southern Uplands was slowing down (Soper & Woodcock, Reference Soper and Woodcock2003), dextral transpressive deformation in Charnwood Forest was commencing along the NE margin of the Midlands Microcraton indenter, perhaps documenting the latter's early relative northward movement against the Eastern Caledonides terrane. This involvement of the microcraton is broadly compatible with the concept (Soper & Woodcock, Reference Soper and Woodcock2003) of sinistral transtension prevailing across Wales and NW England at that time (Fig. 5). Uplift of the microcraton eastern margin, perhaps commensurate with the Charnwood deformation, would also be in keeping with an upwards shallowing of depositional facies, reddening of strata and possible cessation of deposition of sediments within the East Anglia Basin, in which the youngest biostratigraphically dated strata are of Silurian, late Přídolí age (Woodcock & Pharaoh, Reference Woodcock and Pharaoh1993), or about 416 Ma on the timescale of Gradstein et al. (Reference Gradstein, Ogg and Smith2005). This transition from marine to non-marine deposition is also seen across the whole of the western edge of the microcraton, into the Welsh Basin and as far north as the Lake District, as reviewed in Soper & Woodcock (Reference Soper and Woodcock2003); it is the evidence formerly used to suggest that the ‘Caledonian Orogeny’ was end-Silurian in age.

It is noteworthy that in the paratectonic Eastern Caledonides domain immediately to the east of Charnwood Forest (Fig. 1a), Cambro-Ordovician basement rocks intersected in boreholes differ from the Charnwood rocks in possessing generally weaker tectonic fabrics and lower, anchizonal KI values, corresponding to a maximum burial of approximately 8 km depth (Kemp, Reference Kemp1997; Pharaoh et al. Reference Pharaoh, Merriman, Webb and Beckinsale1987; Merriman et al. Reference Merriman, Pharaoh, Woodcock and Daly1993; Van Grootel et al. Reference Van Grootel, Verniers, Geerkens, Laduron, Verhaeren, Hertogen and De Voss1997). Thus the style of Charnwood deformation, and its epizonal metamorphism, could represent a narrow zone of relatively deeper burial and stronger compression that was localized along the tectonic boundary between the Eastern Caledonides and Midlands Microcraton.

Within the continuation of the Eastern Caledonides belt in Belgium, dextral transpressive deformation has been tentatively suggested (Sintubin, Brodkom & Laduron, Reference Sintubin, Brodkom and Laduron1998), and the Charnwood deformation reported here thus fulfils their prediction that such a tectonic regime should be widespread within the ‘Anglo-Brabant fold-belt’, which encompasses the Eastern Caledonides and Brabant Massif terranes (Fig. 5). The Brabant deformation appears to have been particularly protracted, compared with the areas of southern Britain affected by the Acadian event (Debacker et al. Reference Debacker, Dewaele, Sintubin, Verniers, Muchez and Boven2005; Woodcock, Soper & Strachan, Reference Woodcock, Soper and Strachan2007). On the basis of unconformities, a relatively young ‘Acadian’ Brabantian event has been suggested, between earliest Devonian (Late Lochkovian) and Givetian times (Van Grootel et al. Reference Van Grootel, Verniers, Geerkens, Laduron, Verhaeren, Hertogen and De Voss1997). However, radiometric age determinations (⁴⁰Ar–³⁹Ar) on cleavage-parallel micas (Dewaele, Boven & Muchez, Reference Dewaele, Boven and Muchez2002; Debacker et al. Reference Debacker, Dewaele, Sintubin, Verniers, Muchez and Boven2005) indicate an earlier phase of Brabantian deformation, with one age cluster at 419–412 Ma (Přídolí and Lochkovian), overlapping with that seen in Charnwood Forest, and another at 407–401 Ma (Emsian).