The structure of southwest Dyfed November 1979 The Editor, Dear Sir, Although our contribution to the discussion of Hurst, Hancock & McKerrow (1978) comes more than a year after the publication of their work we would like to add our reservations to those of Allen & Williams (1979) who question some of their conclusions. Whereas Allen & William s' arguments are stratigraphical, ours are structural. Hurst et al. (ibid., figs. 5-11) present maps showing their interpretations of the palaeogeography of Wales and the Welsh Borderland for seven intervals of Silurian time . Southwest Dyfed (formerly Pembrokeshire) is depicted as comprising four thrust sheets and one autochthon, each thru st sheet and the autochthon being separated from its nearest neighbour by an arbitrary distance of 5 km . Sanzen-Baker (1972 , 140) in her anal ysis ofSW Dyfed ,on which Hurstetal. base their arguments, was careful to refer to the form erly separated units as allochthonous blocks; implying that although they had undergone substantial horizontal translation, they were not necessaril y thrust sheets. The age of the faulting she believed to be post-Carboniferous; a conclusion in broad agreement with Hancock's (1973) proposal that the major strike faults developed during the Variscan orogeny , a late Carboniferous-early Permian event in SW Dyfed . We differ from Hurst et al. in that we do not recognise the existence of four allochthonous blocks each bounded by a thrust. Figure 1(a) shows the locations of the faults presumed to bound the blocks, and fig. 1(b) shows the outcrops of those faults as determined by our mapping. The Ritec Fault. This structure, the only strike fault in the region which may extend from the west coast to the east coast, is commonly regarded as a steep, southdipping reverse fault. Near Tenby the downthrow to the north is about 500 m. This increases to at least 1000 m where the fault enters Milford Haven, but then decreases westwards until in the extre me west it is less than 100 rn, and the direction of down throw is to the south (Hancock et al. , in press). Because the Ritec Fault achieves such a small displacement in the west it is unlikel y to be a boundary to an allochthonous block. The Musselwick Fault. At the western end of its outcrop , where the fault plane is inclined steeply to the south, stratigraphical relationships and an oblique slickenside lineation indicate that movement was a combination of sinistral strike-slip and reverse dip-slip. Detailed mapping of the well-exposed shore section where the eastern extension of the fault is presumed by SanzenBaker (1972 ) and Hurst et al. to enter Milford Haven has not revealed a major strike fault. The limited lateral extent of the fault makes it an improbable boundary to an allochthonous block. The Benton Fault. Between St. Brides Bay and about 1 km ea st of the northern arm of Milford Haven the outcrop of this south dipping fault is clear. Despite being interpreted as reverse by most workers, rocks exposed in its hanging wall are everywhere younger than those in its footwall ; a normal displacement in excess of 1500 m is Fig. 1. Locations of major Variscan strike faults in southwest Dyfed. (a) According to Sanzen-Baker (1972, fig. 1). (b) According to Hancock (1973, fig. 1) and the present authors. 237 238 CORRESPONDEN CE indicated (Hancock et al., ibid .]. Stratigraphic contrasts across the fault suggest that it was active during Palaeozoic sedimentation, some of these earlier displacements may have been reverse. East of the outcrop of the Old Red Sandstone the fault disappears beneath the limestones of the Lower Carboniferous. There is also no sign of a major strike fault cutting the well-exposed Coal Measure rocks on Carmarthen Bay through which the projected fault should pass . Again it is necessary to reject the idea that the Benton Fault acted as a thrust bounding the northern margin of an allochthonous block during Variscan deformation. The Johnston Thrust. The sinuous outcrop of this gently south-dipping fault zone is well-established in the west; to the east of the northern arm of Milford Haven it splits into subordinate faults until disappearing before the Carmarthen Bay coast. The well-known complex of folds and faults in the Coal Measures immediately south of Amroth (7 km NNE of Tenby) has been shown by Tringham (1979) to be unrelated to a major fault zone despite the structures being on-line with an eastward projection of the Johnston Thrust. Tringham's analysis of structures in the western part of the Johnston Thrust zone demonstrates that overthrusting to the north was followed by dextral str ike-slip. The absence of the Johnston Thrust as a Variscan structure in the east makes it impossible to interpret the thrust as a structure bounding allochthonous blocks. Thus we conclude that Hurst et al. have selected faults which do not intersect and which, excluding the Ritec Fault, do not extend along the entire length of the fold belt. Because there are unfaulted boundaries between regions which have been referred to as the Haverfordwest autochthonous area, and the Rosemarket, Winsle and Marloes blocks, it is impossible to interpret them as allochthonous fault blocks. It is tru e that any palin sp astic reconstruction should allow for some substantial N-S shortening related to folding and flattening, but the connections between blocks would not be eliminated. Our field observations also demonstrate that displacements on the faults were not uniformly reverse, and that some achieved translations which are much less than the arbitrary 5 km of separation employed by Hurst et al. References ALLEN, J. R. L. & B. P. J . WILLIAMS, 1979. Old Red Sandstone facies and stratigraphy and palaeogeography in Wales and the Welsh Borderland. Proc. Geol. Ass. 90, 229-232. HANCOCK, P. L., 1973. Structural zones in Variscan Pembrokeshire. Proc. Ussher Soc. 2, 509-520. HANCOCK, P. L., W. M. DUNNE, & M. E. TRINGHAM, in press. Variscan deformation in southwest Wales. In (Ed: Hancock, P. L.) The Variscan Fold Belt in the British Isles. Adam Hilger, Bristol. HURST, J. M., N. J . HANCOCK, & W. S. McKERROW, 1978. Wenlock stratigraphy and palaeogeography of Wales and the Welsh Borderland. Proc. Geol. Ass. 89, 197-226. SANZEN-BAKER, I., 1972. Stratigraphical relationshipsand sedimentary environments of the Silurian-early Old Red Sandstone of Pembrokeshire. Proc. Ceol. Ass., 83, 139164. TRINGHAM, M. E., 1979. Structures in Upper Carboniferous rocks in the Pembrokeshire Coalfield, Dyfed, Wales. Unpublished Ph.D. thesis, University of Bristol. W. M. Dunne M. E. Tringham Department of Geology University of Bristol Queen's Building University Walk Bristol BS8 lTR P. L. Hancock Esso Exploration and Production U.K. Block 5, The Centre Walton-on-Thames Surrey KT12 IQN