Journal of Petroleum Geology, vol.27 (1), January 2004, pp 95-104
Copyright © 2004 by Scientific Press Ltd. All rights reserved.

*Hans Ramberg Tectonic Laboratory, Dept of Earth Sciences, Uppsala University, Villavägen 16, SE-752 36 Uppsala, Sweden.
email: [email protected]

WILSON (2003) has renewed his longstanding challenge to the consensus view that most of the salt at high stratigraphic levels in the Gulf of Mexico is remobilised Jurassic salt repeatedly extruded through clastic sediments prograding basinward. Science advances by continually reassessing multiple hypotheses and, as it is notoriously difficult to date salt bodies directly, we must accept Wilson’s plea to consider all reliable and pertinent geological observations — but that includes learning lessons from subaerial salt extrusions. While unable to deny most of Wilson’s reinterpretations, I show here that most are unnecessarily complicated and thus vulnerable to Occam’s Razor. With the aim of correcting several misunderstandings about salt tectonics revealed in Wilson’s onslaught, I use new observations to show how younger sediments and microfossils can be incorporated within emergent older salt and use classical arguments to explain how such younger shales are later extended into planar layers concordant with the salt layering.

Wilson raises thirteen arguments against the shallow salt in the Gulf of Mexico being allochthonous Louann salt. I abbreviate and number these here (in a different order) and comment after first dismissing three illogical arguments:

1. Claimed equivalents south of the Cretaceous shelf edge are not as pure as definite Louann salt in the interior basins to the north The petrography of any regional stratigraphic unit is likely to change in primary facies space. Thus, salt accumulated in clastic-starved onshore basins is likely to be purer than contemporaneous salt deposited in connected but relatively open basins beyond the shelf. 2. On- and offshore intercalations of shale in salt are so similar to that they must be the same age However, similarities between some of the shale intercalations in the salt sheets of Iran and those figured by Wilson does not make them the same age! 3. The argument that mixed Tertiary fauna found within and below salt sheets looks so much like sedimentary reworking that they must be due to sedimentary reworking needs no comment (but see later).

DEGRADATION OF EMERGENT SALT

4. Wilson is able to argue that Tertiary microfauna and shales within salt sheets challenge the Gulf of Mexico paradigm because there has been little discussion about how contemporaneous material is incorporated into emergent allochthonous salt sheets. In this section, I report new observations that indicate some of the processes involved.

The smooth rounded profiles of viscous fountains or droplets of salt (Talbot 1993, 1998; Talbot and Alavi, 1996;Talbot and Jarvis, 1984) on scales of >1 km, whether they are extruding at m per year (Talbot et al., 2000) or cm/year (Talbot and Aftabi, in press), imply that they degrade mainly by viscous gravity spreading and that any submarine or subaerial weathering is relatively minor.

However, on scales below a few hundred metres, the relief of salt extrusions increases radially down-slope from minor (c.5 m) on their crests to maxima (e.g. 100 m) around their termini. Seismic dip and azimuth maps of the top of Mickey and an adjoining buried salt sheet in the Gulf of Mexico (Fletcher et al., 1990, figs. 22-23) indicate concentric steps c.1 km apart and radial gullies c.200 m apart. These authors (op. cit.) speculate that the steps are due to onlap of individual beds and that the radial relief records radial faults. However, subaerial salt extrusions in Iran have equivalent relief, and the concentric steps on their top surface relate to irregularities in the bottom boundaries while the radial gullies relate to surface drainage (Talbot and Aftabi, in press). Although subaerial salt extrusions in Iran have an outer carapace of brittle and broken dilated salt (Talbot, 1998), none of the 50 or so examples I have visited in Iran have any of the radial faults conjectured by Fletcher et al. (1990).

Because active salt extrusions are proud of their surroundings, rain or brines dense with dissolved salt are likely to drain radially downslope through, over and off the salt via surface and subsurface conduits. However, none of the internal anisotropies represented by layering, grain shape fabrics and fractures have any significant influence on the topography of Iranian salt extrusions on scales less than about 100 m (see Fig. 1a, and photographs in Talbot, 1998).

As so little is known about the thin veneers covering submarine salt extrusions in the Gulf of Mexico, it is worth considering subaerial equivalents in Iran as analogues. There, insoluble components dispersed in the extruding sequence accumulate by salt dissolution as weak amorphous residual cap soils that are joined by air-borne dust. Such loose soils are magenta or buff in colour and rich in anhydrite and clay minerals beneath a puffy gypsum crust. They can reach thicknesses of hundreds of m and can slump or be washed off steep slopes, or be conveyed by the flowing salt and spilled over its terminus (thereby probably accounting for Wilson¥s “sedimentary reworking” of mixed Tertiary microfossils above and below salt sheets in the Gulf of Mexico).

It is unusual to be able to see how such soil veneers relate to the dissolved salt beneath. However, a 1999 “salt-rush” in Iran led to 35 new quarries being opened in bodies of white salt in the Eivanekey plateau west of Garmsar, the hills north of Garmsar, and in dome 22 in the Great Kavir (Jackson et al., 1990). The bedding and foliation in most of the 20 x 10 km Eivanekey plateau is subhorizontal as a result of the advance of the Alborz Mountains extruding Eocene-Oligocene salt southward over the Great Kavir (Fig. 2a). A strong “gneissose” foliation is axial planar to the younger of two suites of recumbent folds verging generally southward. Shallow bedding and foliation steepen with the top of the Eivanekey salt sheet in NNE trending welts with cuspate-upward sections along thin-skinned transfer faults in the local roof (similar to welts in the top of the tabular body of Hormuz salt in the Salt Range of Pakistan: Butler et al., 1987). On visits to many of these quarries in 2000, I found that the salt can be stripped of its residual soils (by bulldozers, picks, shovels, brooms and paint brushes), so that the salt is clean for the (culinary?) market when blasted using black powder in shot-holes made by hand-held pneumatic percussion drills.

Smooth planar salt surfaces recently stripped of soils have a negative relief of c.5 m as a result of dissolution by the <400 mm/y winter snow melt and rainfall having filtered through the soil (Figs. 1b-e). Such surfaces are punctuated by dolina (cylindrical or funnel-shaped downward-closing openings wth subvertical axes dissolved in the salt between dm-wide salt septa and normally full of soil). Exposed in steep quarry faces, downward-merging dolina have smooth walls completely independent of either bedding or grain-shape fabrics (Fig. 1f). At depths between 1 and 4 m, the rounded bases of these dolina connect to numerous small diameter (cm-mm) horizontal channels (Figs. 1 c, e, f). These presumably allowed brines laden with dissolved salt to diffuse back to the open hillside via several dolina. The small channels anastomose through dilated but sound salt above slope-parallel relief fractures, but are close to horizontal in the confined salt beneath (right-hand side of Fig. 1c). The small-scale horizontal channels survive as gentle corrugations in the walls of deepened dolina (Fig. 1f).

It takes time for vertical anisotropy- and relief-related joints to develop with W-E and N-S strikes in exposed salt. Even in quarries km long and decades old, vertical joints 7 to 14 m apart occur in groups of 4 or 5 that are hundreds of metres apart. Such joints first appear near the top free surface within c.4 months after definition of a new face. It then takes 3 to 5 years for slightly en-Èchelon (<5 mm wide) joint traces to propagate c.15 m downward to within a few m short of the quarry floors (Fig. 2a). Plumose markings on an example exposed in a N-S quarry wall suggest it dilated in only one or two (slow!) events (Fig. 2b). In the oldest quarries, the top few m of the initially vertical fractures curve downslope as a result of ductile flow of surficial salt. Ptygmatic veins of halite or anhydrite have axial planes that parallel the subhorizontal grain-shape and orientation fabric in their host rocks (Fig.3d) and are interpreted as infilled vertical joints shortened by subsequent subhorizontal extending flow.

Horizontal planar openings cross bedding and grain-shape fabrics to partially truncate the top 10 m of NNE-SSW trending welts in the top of the Eivanekey salt sheet (Fig. 4a-d). The ends deepest into the hillsides tend to resemble fractures (vertical arrows in Figs 4a,b), whereas the ends near the free slope appear to have been infilled by amorphous or bedded soils washed from upstream. Such horizontal openings can occur alone (Fig. 4d) or in swarms (Figs. 4c). Most of the intervening salt screens are remarkably thin and

Fig. 1. (a) Salt surface ribbed by decades of rain are unaffected by old fractures (western quarry, Qum Kuh, see Talbot and Aftabi, in press). (b-e) photos of new quarries in Eivanekey Plateau, Iran. (b) Surface of unfractured salt recently stripped of residual soils, Qaem quarry, and (c and d) Immature (1 to 3m deep) dolina in salt not thoroughly cleaned, Morvarid quarry. (e and f) close-up of cleaned dolina 4m deep at Qaem quarry. Area shown in (f) is outlined in (e). Macroscopic dissolution surfaces are unaffected by layering, grain shape fabric or fractures.

Fig. 2. Symptoms of extending flow in extrusive sheet of allochthonous salt in Derakhshan quarry, Eivanekey Plateau. (a) A 22 x 15m face of subhorizontal halitic “gneisses”. Man arrowed for scale. (b) Boudinage of horizontal halite pegmatite. (c) Vertical ptygmatic halite vein. (d) “Gneissose” foliation crossing an older recumbent fold is (e) axial planar to a younger fold.

planar over tens of metres but water-worn channels are stacked locally (oblique arrows in Fig. 4c). Loss of the roof and septa in many examples would result in the salt being topped by the smooth horizontal “salt mirror” often found beneath the cap rocks of onshore salt bodies.

Some of the roofs and internal salt septa appear to have arched upward (Figs 4a, c, d). It is not clear whether this flexure is due to lateral forces or due to lift by the swelling of soil infills as anhydrite hydrates to gypsum. Nor is it clear whether these planar horizontal features are due to lateral erosion by water, the closure of caves, or the opening of de-stressing joints. The startling 200 x 100 x 5 m aspect ratio of an open cave in vertical Hormuz salt (Fig. 4f & g, by Jacob Smid, pers. comm. 1999) suggests they are opened by water flow. However, local distortion in the roof (estimated to be c.100 m thick) of a former cave exposed in a sidewall of the same cave (Fig. 4g) suggests that the propagation of horizontal fractures is involved. Although open caves in salt can have complex cross-sections (see photos in Frumkin, 1994, and Talbot, 1998), most have simple planar floors and arched roofs (Figs 4a, d, e) before they are infilled by younger sediments (Fig. 4g & h) and rotated (Fig. 4i) and extended to concordance with the salt layering and fabric by further salt flow.

5. Wilson mistakes the relative significance of subaerial and submarine dissolution The openings, dissolved and infilled, in an allochthonous salt sheet described above are subaerial. Although freshwater is more aggressive to salt than seawater, we can expect similar phenomena in submarine salt extrusions to introduce younger sediments and fossils into allochthonous salt, and this may account for many of Wilson¥s observations.

Wilson’s assumption that extruding salt is dissolved as fast as it reaches the sea bed does not recognise numerical models by Fletcher et al. (1990), who used sophisticated diffusion laws to estimate that veneers of overlying sediments only a few 10s of metres thick will significantly retard submarine salt dissolution. These authors estimated that only 10 vol % of an expanding salt extrusion dissolves at a rate that would result in a cap rock 26m thick after 3Ma and 48m after 5Ma.

Past extrusions of salt in the Zagros have indeed dissolved away, but most are onshore and located in a shortening fold-thrust belt without much likelihood of burial. Submarine salt extrusion is more likely during times of relatively slow clastic deposition and less likely when clastic deposition is rapid. There are plenty of salt extrusions hundreds of metres proud of the current floor of the Gulf of Mexico; and the salt sheets at Eocene, Oligocene, Mio-Pliocene and Plio-Pleistocene levels in the Gulf of Mexico which suggest episodes of evaporation to Wilson indicated episodes of relatively slow deposition to Wu et al. (1990).

DEFORMATION OF SALT

6. Wilson is incorrect in assuming that large bodies of allochthonous salt are only common in the Gulf of Mexico Similar examples, past or present, are reported in the Red Sea (Heaton et al., 1995), the Pricaspian Basin (Volozh et al.,2003), and Iberia (Maestro et al.,in press) and are shown in seismic profiles of offshore Morocco (Tari et al.,2000), if not in the

Fig. 3. Development of stress relief fractures in quarried salt. (a) Vertical N-S joint, obvious in face 3 to 5years old in Rah Raha quarry, Garmsar hills, is just appearing in 4 month old face below bench (arrowed). (b) Oblique view of a 30 x 30m west-facing joint in Derakhshan quarry, Eivanekey Plateau.

Neoproterozoic Congo Basin (Jackson et al., 2003). Most of the field photographs shown here come from another 20 x 10 km sheet previously undocumented in northern Iran.

7. Wilson draws insufficient distinction between canopies and sheets of allochthonous salt Salt canopies characteristically form as shallow bulbs spread by salt diapirs that partially suture (Jackson et al., 1990). Remote sensing shows this characteristic in the type canopy of the Semnan diapirs in the Great Kavir of central Iran without any need for subsurface control and whatever the emplacement concept (e.g. Jackson et al. 1990). Although the Semnan diapirs share similarities with the salt structures in the Gulf of Mexico, they are not ideal analogues. This is because the overburden through which the Semnan diapirs actively rose (upbuilt) was so weak that shallow salt in the Great Kavir mushrooms largely beneath the weak desert surface rather than fountains above it (Talbot, 1990). The emergent bulbs of the Semnan diapirs float with the low freeboards of ice floes rather than stand proud like glaciers. As a result, salt emergent in the Semnan diapirs is like that in the Belle Isle Salt Mine and is characterised by steep rather than gently-dipping foliations (Fig. 5).

Salt extrusions elsewhere in Iran are more appropriate analogues for the sheets of allochthonous salt in the Gulf of Mexico, and the type locality for fountains of diapiric salt are in the Zagros Mountains, further south in Iran (Lees, 1927). Here, the carbonate overburdens provides strong nozzles to reactive salt diapirs that fountain 900m in the sky before flowing over their surroundings.

8. Wilson overemphasises the limitations of analogue models…

Dynamically-scaled models deformed in a centrifuge (e.g. Ramberg, 1982) and at normal gravity can certainly simulate most of the internal structures in salt structures while constraining the times for the salt prototype to reach equivalent development (see e.g. Fig. 6 and Talbot and Aftabi, in press). Indeed, centrifuge models have simulated the whole paradigm of sedimentary progradation in the Gulf of Mexico driving diapirs and sheets of “Jurassic salt” basinward, through, over and incorporating younger “clastic sediments” (Talbot, 1992a).

9. …and field observations of subaerial salt sheets It is worth learning some lessons from subaerial salt extrusions in Iran before going further. These vary in shape from simple thick axisymmetric flows spreading

Fig. 4. Horizontal stress relief fractures or caves crossing layering and foliation in salt (a-c) partially truncating welt in top salt in Drakhsharnamak quarry, NW Eivanekey Plateau, showing (a) closed horizontal “fractures” (vertical arrows), arched roof (dashed) and (c) internal septa with local indications of water flow (oblique arrows). (d) An arched roof closing to left, Morvarid quarry. (e) Arch of salt in dome 28, Great Kavir. (f) Open (200 x 100 x 5m) cave beneath c.100 m of vertical Hormuz salt in the Namakdam diapir on the island of Qeshm, S. Iran. (g) Close-up of older cave infill in sidewall of (f); notice lateral flow of downward extruded roof block. Photographs (g & f) by Jakub Smid. (h) Cave with bedded infill rotated in welt, Qaem quarry. (i) Infilled cave, Rah Raha quarry, Garmsar hills.

Fig. 5. Steep foliations characterise emergent Semnan diapirs of the Great Kavir (Jackson et al. , 1990). (a) Vertical layering and megacrystic grain shape fabric characteristic of mushroom margin and core in dome 22. (b) Nearly vertical crenulation cleavage superposed on subhorizontal sheet dip of layering and foliation between margin and core of dome 28.

Fig. 6. Flow folds in passive layers in extrusions of transparent silicone (Talbot & Aftabi, in press) simulate: (a) the basic kinematic tank-track fold in a slow steady extrusion over gravel plain away from a country rock collar. (b) As above with additional dynamic folds indicating pulsed drive. (c) Complex internal folds in a thin fast flow down the flank of a Zagros anticline (cf. this model by Rosemary Talbot with prototype described in Talbot, 1979).

slowly (dm/y) over gravel plains (cf. Figs. 6a & b) to thin flows that develop complex internal structures, as markers on them are carried comparatively rapidly (e.g. 10 m/y in Talbot et al., 2000) down the irregular flanks of Zagros anticlines (cf. Fig. 6c). If the former are steady over planar substrates, they exhibit isotropic grain-shape fabrics and planar layering (parallel to the top or bottom boundary, whichever is closer) and only one internal fold. This is the tank-track fold of Talbot (1998) that probably accounts for the most of the complexities encountered in the lower levels of salt sheets drilled in the Gulf of Mexico (and Wilson¥s supposed sedimentary reworking). Like lava flows and ice sheets, the top surfaces of extrusive salt sheets advance by flowing faster under air or water than their bases that attach to their substrates. The basic tank-track fold forms kinematically as the top surface of an advancing sheet (and internal marker layers) rolls over the front and sheds any superposed cover which the salt then overrides but seldom entrains.

Even in the most internally complex sheets (cf. Fig. 6c), the colour layers pick up flow folds (± mylonite zones) only where the flowing salt decelerates in zones tens to hundreds of metres long upstream of fixed obstructions in the country rock channel. Such folds become inconspicuous as they tighten to isoclinal in the zones a few hundred metres long where the salt accelerates and extends downstream of the obstruction (Talbot, 1979, 1981).

The grain boundaries of clean halite are so mobile that grain-shape fabrics in salt have remarkably short strain memories compared to larger strain markers like the layering. As a result, salt can maintain steady-state isotropic (“undeformed”) or gneissose (<1.5:1) grain-shape fabrics (Talbot and Jackson, 1987) by dynamic recrystalisation during supershear (Talbot, 1992b). Most spectacular curtain folds in intrusive diapirs have isotropic grain fabric despite having risen a few km. Even in the most complex extrusions in Iran, most of the salt has isotropic grain-shape fabrics and megacrysts shredded from “halitic pegmatites” are spherical. Only in local fold-generation zones is the salt fabric “gneissose” with ellipsoidal megacrysts. The most extreme grain-shape fabrics, with grain aspect ratios of up to 5:1, are confined to mylonite zones only dm thick in ductile detachments in the lower limbs of kinematic (Talbot, 1981) or dynamic flow folds (Talbot and Aftabi, in press).

Thus, the furthest-travelled Iranian salt looks undeformed to a casual first inspection. Closer study reveals a few isoclinal fold closures and dm thick mylonite zones among the generally horizontal planar layering and almost isotropic grain-shape fabrics. However, salt in the Zagros has likely travelled only a few tens of km along its source layer, up its diapiric feeder, and downslope over the surface, and it has done this only once. By contrast, salt in the some of the youngest submarine extrusions in the Gulf of Mexico may have flowed hundreds of km from its place of origin and been through more than one cycle of burial and extrusion. Nevertheless, like ice that has flowed hundreds of km from the interior to the coasts of Antarctica, salt that has flowed hudreds of km need not look very different from how it began, and may well resemble the salt and shales shown in the Gulf of Mexico cores Wilson refers to.

10. We now come to Wilson¥s main complaint, that the offshore salt in the Gulf of Mexico is not sufficiently deformed in seismic profiles and drill cores to have travelled the vertical and horizontal distances claimed for it. 11. To bolster this argument, he makes much of the simplicity of the fabrics and the concordance between planar layering in the salt and the entrained shale bodies (with their Tertiary microfauna).

Three markers can be used for assessing the degree of deformation in a salt body: (a) its external shape, (b) the shape of its internal layers, and (c) its grain-shape fabrics (b and c are taken together below).

(a) Seismic profiles in the Gulf of Mexico show every stage in the transition from almost planar sheets of shallow salt being buried by clastic sediments that “sink” in minibasins and “downbuild” passive intervening diapirs. These can mature to extrude domes on the seafloor that gravity spread downslope, while clastic sediments continue to downbuild the upslope contacts of many. Salt is so weak that, as already mentioned, the tops of extruded salt bodies >1 km have the smooth profiles of a flowing viscous fluid. Also because it is so weak, salt can be lost (at seismic resolution) from source layers that close to primary welds, or from wall- or stock-like diapiric stems that close to secondary welds (and, as noted by Wilson, are thus difficult to find).

(b + c) The internal structures and fabrics in flowing salt depend on the external boundaries guiding the flow. Salt layers that form planar and horizontal are likely to remain so where salt flows steadily along a buried source channel. Convergent flow towards the stem of a diapiric stock amplifies initial irregularities in the layering to upright folds with radial axes. Both the layering and axes of these folds steepen up the diapirs and turn back to horizontal in the crest of a salt extrusion. The curtain folds in the cores of diapirs indicate such convergent flow, not extreme deformation. Contrary to Wilson’s expectations of “wavy” layering in allochthonous salt sheets, we should therefore expect simple subhorizontal (top-parallel) layering and fabrics, both above any extrusive vents not closed to secondary welds, and beneath the smooth tops of the allochthonous sheets in which the salt has flowed long downslope distances.

Gravity spreading of extruding salt over planar substrates results in extensional internal strain. Extensional flow along the layering unfolds any irregularities or folds in the salt and extends any clastic sediment bodies to simple planar layers (Fig. 2a), which it then thins by horizontal extension (and boudinage of any competent layers: Fig. 2b) with vertical shortening (and folding of any steep competent layers: Fig. 2c). This accounts for the simplicity of the seismic reflectors inside the salt sheets in the Gulf of Mexico. Just as younger dikes initially perpendicular to gneissose fabrics can be thinned and rotated to concordance with the foliations in major shear zones, so beds of younger shales deposited with contemporary or reworked microfauna on or in allochthonous salt (Figs. 1 & 4) can be thinned and rotated to concordance with the salt layering and foliation. They may be “primary layers” but are intensely deformed and much straighter and thinner than they began. The symptoms to be expected of long-distance salt flow over a simple deposition surface are precisely those Wilson figures from Belle Isle Salt mine (complete with a gneissose grain-shape fabric parallel to the layering) -- but the horizontal of extrusive salt rather than the vertical of intrusive salt.

12. Wilson re-interprets angular contacts at the base of salt sheets as sub-salt unconformities rather than shear or thrust contacts Fluids (whether melts like water or lava, or crystalline like ice or salt) flow over irregular surfaces by gravity flattening their top surface. While gravity spreads the fluid downward beneath its top boundary, it may meanwhile drive the fluid horizontally, up, or downward along the bottom boundary (or all three in different locations, see Fig. 6c). Consequently, the basal contact can be conformable or cut either up- or down-sequence with respect to its surroundings.

More like the contacts of igneous bodies than faults, the layering and grain shape fabrics of deformed salt bodies parallel their contacts whatever their attitude relative to any bedding in the country rocks. Angular sub-salt contacts to allochthonous salt bodies are as much unconformities as the angular contacts along the base of lava flows or glaciers.

13. Wilson also re-interprets exotic blocks in piercement domes as plucked from base-salt unconformities As illustrated earlier, it is easy to imagine younger sediments and microfauna being incorporated into the tops or sides of salt extrusions or shallow intrusions. Sediments that onlap or even bury downbuilt passive diapirs are readily incorporated into the salt when the surrounding sediments are further downbuilt. Thus Frio fossils occur in the screens of shales coiled inside at least one of the five island salt domes (Talbot and Jackson, 1987).

However, it is more difficult to advocate the entrainment (rather than drag) of solid blocks into the bottom contacts of salt structures. This is because salt flow is likely to be slowest along rigid substrates. Aware of this problem, Kent (1979) explained blocks of basement in emergent Hormuz salt in Iran as having fallen onto the top surface of still-autochthonous salt from steep overlooking fault scarps (like the submarine cliffs Wilson interprets as fault scarps in his figs. 7 & 8). However, pebbles derived from underlying gravels occur in the lowest few dm of subaerial sheets of Iranian salt (Talbot, 1998), and I have since seen pebbles incorporated high in a salt sheet just downstream of a pocket of gravel. Perhaps like ice, salt can ruck-up and entrain irregular soft substrates.

Wilson concludes that salt sheets at Eocene, Oligocene, Mio-Pliocene and Plio-Pleistocene levels in the Gulf of Mexico are autochthonous largely on the basis of interpretations of observation in the Belle isle Salt Mine. I hope to have shown how his interpretations are unnecessary if not incorrect. To argue his alternative model, Wilson needs to demonstrate more convincingly repeated closure of the straights leading into the Gulf of Mexico. Most of the Messinian salts were deposited on the deep floor of the Mediterranean, well below the level of the sill at Gibraltar.

ACKNOWLEDGEMENTS

I thank Dr Korehei, the General Director of the Geological Survey of Iran and his colleagues for their encouragement and logistic support for my research in Iran. GSI geologist Mr Mahmoud Rahmati accompanied me on all my visits to the Eivanekey plateau, together with Pedram Aftabi on most days and Dr Safei on one day. The Swedish Natural Science foundation pays for my travel to and from Iran.

REFERENCES

BUTLER, R.W.H., COWARD, M.P., HARWOOD, G.M., and

KNIPE, R.J., 1987. Salt control of the thrust geometry, structural style and gravitational collapse along the Himalayan mountain front in the Salt Range of northern Pakistan. In: LERCHE, I. and O’BRIEN, J. J., (Eds) Dynamical Geology of salt and related structures, Academic Press, Orlando, US, 339-418 FRUMKIN, A., 1994. Hydrology and denudation rates of halite karst. Journ. Hydrology, 162, 171-189.

FLETCHER, R. C., HUDEC, M. and WATSON, A. A., 1995. Salt glacier and composite sediment-salt glacier model for the emplacement and early burial of allochthonous salt sheet, In: JACKSON , M.P. A., ROBERTS, D. J. and SNELSON, S. (Eds), 1995. Salt tectonics, a global perspective. AAPG Mem. 65, 77-108.

HEATON R. C., JACKSON, M.P.A., BAHAMOUD, M. and NANI, A.S.O., 1995. Superposed Neogene extension, contraction and salt canopy emplacement in the Yemini Red Sea. In: JACKSON , M. P. A., ROBERTS, D. J. and SNELSON, S. (Eds), Salt tectonics, a global perspective. AAPG Mem. 65, 333-351. JACKSON, M. P. A., CORNELIUS, R. R., CRAIG, C. R.,

GANSSER, A., ST÷CKLIN, J. and TALBOT, C. J., 1990. Salt diapirs of the Great Kavir, Central Iran. Geol. Soc. America Memoir 177, 139pp.

JACKSON, M. P. A., WARIN, O. N., WOAD, G. M. and HUDEC, M. R., 2003. Neoproterozoic allochthonous salt tectonics during the Lufilian orogeny in the Katangan Copperbelt, central Africa. GSA Bull, 115, 314-330.

KENT, P.E., 1979. The Emergent Hormuz Salt Plugs of Southern Iran. Journ. Petrol. Geol., 2, 117-144.

LEES, G. M., 1927. Saltzgletscher in Persien. Mitteilungen Geologische Gesellschaft Wien, 20, 29-34.

MAESTRO, A., SOMOZA, L., MEDIALDEA, T., TALBOT, C. J., LOWRIE, A., V¡ZQUEZ, J. T. and DÕAZ-DEL-RÕO, V., 2004. Large-scale slope-failure involving Triassic and Middle Miocene salt and shale in the Gulf of C·diz (Atlantic Iberian Margin). Terra Nova, in press.

TALBOT, C. J., 1979. Fold trains in a glacier of salt in southern Iran. Journ. Struct Geol., 1, 5-18.

TALBOT, C. J.,1981. Sliding and other deformation mechanisms in a salt glacier, Iran. Geol. Soc. London, Spec. Publ., 9, 173-183.

TALBOT, C. J., 1990. Allochthonous salt spreading. Proceedings of Symposium on Diapirism, Bandarabbas, Geological Survey of Tehran, 2, 508-510.

TALBOT, C. J.,1992a. Quo vadis tectonophysics? With a pinch of salt!. Journ. Geodynamics, 16, 1-20.

TALBOT, C. J.,1992b. Centrifuged models of Gulf of Mexico profiles. Mar. Petrol. Geol., 9, 412-432.

TALBOT, C. J., 1993. Spreading of salt structures in the Gulf of Mexico. Tectonophysics, 228, 151- 166.

TALBOT, C. J., 1995. Molding of salt diapirs by stiff overburdens. in JACKSON , M. P. A., ROBERTS, D. J. and SNELSON, S. (Eds) Salt tectonics, a global perspective. AAPG Memoir 65, 61-75.

TALBOT, C. J., 1998, Extrusions of Hormuz salt in Iran, In: BLUNDELL, D. J. and SCOTT, A. C. (Eds), Lyell: the Past is the Key to the Present. Geol. Soc. London, Spec.Publ., 143, 315-334.

TALBOT, C. J. and JARVIS, R. J., 1984. Age, budget and dynamics of an active salt extrusion in Iran. Journ. Struct..Geol., 6, 521-533.

TALBOT, C. J. and JACKSON, M. P. A., 1987. Internal kinematics of salt diapirs: Reply AAPG. Bull., 73, 946-950.

TALBOT, C. J. and ALAVI, M., 1996. The past of a future syntaxis across the Zagros. Geol. Soc. London, Spec. Publ. 100, 89-109.

TALBOT, C. J., MEDVEDEV, S., ALAVI, M., SHAHRIVAR, H. and HEIDARI, E., 2000. Salt extrusion rates at Kuh-e-Jahani, Iran: June 1994 to November 1997. Geol. Soc. London, Spec. Publ., 174, 93-110.

TALBOT, C. J., and AFTABI, P., 2004. Geology and models of salt extrusion at Qum Kuh, central Iran. Journ. Geol. Soc. London, 161, 1-14, in press.

TARI, G., MOLNAR, J., ASHTON, P. and HEDLEY, R., 2000. Salt tectonics in the Atlantic margin of Morocco. The Leading edge Ocober 2000, 1074-1080.

WILSON, H. H. 2003. Extensional evolution of the Gulf of Mexico and the deposition of Tertiary evaporites. Journ. Petrol. Geol. 26 (4), 403-428.

WU, S., BALLY, A. W. and CRAMEZ, C., 1990. Allocthonous salt, structure and stratigraphyof the northeastern Gulf of Mexico. Part II, structure. Mar. Petrol. Geol. 7, 334-370.

VOLOZH, V., TALBOT, C. J. and ISMAIL-ZADEH, A., 2003. Salt structures and oil in the Pricaspian basin. AAPG Bull., 87, 313-334.

EXTENSIONAL EVOLUTION OF THE GULF OF MEXICO BASIN AND THE DEPOSITION OF TERTIARY EVAPORITES: Discussion