DISCUSSION

This paper has briefly reviewed the results of Leg 173 with allusions to results of Leg 149, where appropriate and for completeness, that also relate to the development of the west Iberia continental margin before the onset of seafloor spreading. Further details can be found in the references that accompany the text. Other references refer to geophysical observations off west Iberia and to comparable work done in the rifted margin and transition zone fragments now exposed in the Alps. In spite of the importance of considering both members of a conjugate pair of margins when discussing their development, this has not been done here because of the lack of comparable ODP or other borehole data from beyond the shelf edge off the Newfoundland margin. Aspects of the postrift history, which was not a specific objective of Leg 173 drilling, are also omitted and are discussed elsewhere (Kuhnt and Urquhart, in press; Whitmarsh, Sawyer, Klaus, and Masson, 1996) and in this volume (Roessig and Wise, Chap. 4; Urquhart, Chap. 9; Wallrabe-Adams, Chap. 6; Zhao et al., Chap. 8).

One as-yet unresolved problem on the west Iberia margin is the apparently almost complete lack of synrift melt products in the ODP cores, although a synrift (122.1 ± 0.3 Ma) chlorite schist derived from a Fe-Ti-rich gabbro protolith has been sampled northwest of Galicia Bank (Schärer et al., 1995). It has also been suggested, from analysis of mostly deep-towed magnetometer profiles, that scattered gabbroic bodies may exist within the serpentinized peridotite basement of the OCT (Russell, 1999; Whitmarsh et al., in press). The lack of melt is a problem for melting models (Bown and White, 1995), given the relatively short minimum periods of continental lithosphere extension suggested by Wilson et al. (in press a) (5 m.y.) and Dean et al. (2000) (7 m.y.); for = 10 stretching of a 125-km lithosphere overlying a 1300°-1350°C potential temperature asthenosphere, Bown and White's model predicts 3-6 km of melt. Harry and Bowling (1999) and Bowling and Harry (in press) presented a finite element model of nonvolcanic rifted margins that predicts that melting begins only after lithospheric necking has become focused and results in a short period of magmatism confined to the end of the rift episode just before continental breakup. Unfortunately, the problem off west Iberia is a lack of evidence for significant melt, not only in the thinned continental crust prior to breakup but also in the broad OCT, for some time immediately after breakup. Minshull et al. (in press) considered and rejected the effects of lateral heat conduction, anomalously low mantle potential temperature, and depth-dependent stretching to explain totally the lack of melt products at the time of continental breakup and subsequently in the OCT. Instead, they proposed qualitative arguments that, even for OCT extension lasting only 10 m.y., melting was inhibited while asthenospheric upwelling remained relatively unfocused as the thermal structure of the lithosphere evolved from that of a rifting margin to that of a steady-state spreading ridge. In this context, it is salutary to recognize that radiometric dates from Site 1067 suggest that exhumation, and therefore extension, at the seaward edge of thinned continental crust lasted at least 24 m.y. (Table T2), far longer than suggested by Wilson et al. (in press a).

It is clear that the combined efforts of scientific drilling and geophysical observations off west Iberia, and particularly in the southern Iberia Abyssal Plain, have provided a valuable set of data with which to investigate the development of this nonvolcanic rifted margin. Even so, several problems remain. The significance and continental or oceanic origin of the peridotite ridge remains obscure. The OCT is 15 km wide off Galicia Bank and is up to 170 km wide in the southern Iberia Abyssal Plain, but there is insufficient evidence at present to estimate its width in the Tagus Abyssal Plain with confidence. How and why does the width of the OCT vary along the margin? Does the so-far unsampled deep OCT (Region C in Fig. F3) really consist of serpentinized peridotite, as inferred from indirect evidence? A deep drill hole in this region would test this inference and also provide a very valuable and potentially complete postrift sedimentary record (and subsidence history), which is currently lacking. Lastly, scientific drilling on the deep conjugate Newfoundland margin has the potential to make a very significant contribution to our understanding of the development and eventual lithospheric reconstruction of the west Iberia-Newfoundland conjugate pair of margins.