pressure-volume @V) work. The pressure-induced amorphization of R-A1,Li3Cu supports a suggestion of a struc- tural relation between quasi-crystals and amorphous metals. Nelson and Spaepen have suggested that the ordering in quasi- crystals and metallic glasses is based on polytetrahedral packings like the icosahe- dron (1 7). This relation can be shown by examining the close structural relation u among crystalline R-A1,Li3Cu, quasi-crys- talline i-A16Li3Cu, and the pressure-in- duced amorphous phase of R-A1,Li3Cu (24). Extended x-ray absorption fine struc- ture (EXAFS) measurements show that the icosahedral clusters in the auasi-crvstal i-A16Li,Cu are similar to those'in its crys- talline counterpart R-A1,Li3Cu except that they are more locally ordered (25). Single- crystal structural refinements of i-A16Li3Cu show the same atomic shells as in the R-phase (26). The measurements reported here suggest that the amorphous metallic phase R-Al5Li3Cu created by compression has the topology of the crystalline R-phase, with more nearly perfect local icosahedral order. In addition, i-A16Li3Cu shows be- havior similar to that of the crystalline R-phase when compressed; it undergoes a phase transition by way of a disordered state (27). These observations provide support for a close structural relation between the quasi-crystalline i-phase and the pressure- amorohized R-vhase. Our results show that we can produce an amorphous metal, at ambient temperature, by compression. Because the change is largely isoconfigurational, the amorphous state arises from the disordering of the disclination lines of a Frank-Kasoer uhase. * . Thus, the order present in the amorphous state can be described in a curved or higher - dimensional space. The pressure-amor- phized material is not necessarily a glass in the traditional sense: A glass is formed by the continuous solidification from the melt and exhibits a glass transition (10, 28). How is this pressure-amorphized material related to an amorphous metal produced by quenching from the melt? We speculate that it is an "ideal" glassv state and that - , melt-quenched metals are, owing to kinetic constraints, defective forms of amorphous materials formed by projections from curved space (29). REFERENCES AND NOTES 1. S. R. Elliott, Nature 354, 445 (1991). 2. P. H. Gaskell, M. C. Eckersley, A. C. Barnes, P. Chieux, ibid. 350, 675 (1991) 3. 1. Amato, Science 252, 1377 (1991). 4. Y. Fujii, M. Kowaka, A. Onodera, J Phys. C 18, 789 (1 985). 5. R. J. Hemley, A. P. Jephcoat, H. K. Mao, L. C. Ming, M. H. Manghnani, Nature 334, 52 (1988). 6. M. B. Kruger, Q. Williams, R. Jeanloz, J. Chem. Phys. 91, 5910 (1989). 7. G. C. Serghiou, R. R. Winters, W. S. Hammack, Phys. Rev Lett. 68, 331 1 (1 992). 8. We have suggested elsewhere [R. R. Winters, G. C. Serghiou, W. S. Hammack, Phys. Rev B. 46, 2792 (1992)l that topological frustration based on polytetrahedral packings is crucial in understand- ing the pressure-induced amorphization of nondi- rectionally bonded solids. The results presented here are an implementation of this idea. 9. R. R. Winters, A. Garg, W. S. Hammack, Phys. Rev Lett. 69, 3751 (1 992). 10. P. Chaudhari and D. Turnbull, Science 199, 11 (1 978) 11. D. R. Nelson, Phys. Rev 6 2 8 , 5515 (1983). 12. J. F. Sadoc and R. Mosseri, Philos. Mag. 6 45, 467 (1982). 13. J. F. Sadoc and N. Rivier, ibid 55, 537 (1 987). 14. D. P. Shoemaker and C. B. Shoemaker, in Aperi- odicity and Order: Introduction to Quasicrystais, M. V. Jaric, Ed. (Academic Press, New York, 1988), p. 1. 15. S. Samson, in Structural Chemistry and Molecular Biology, A, Rich and N. Davidson, Eds. (Freeman, San Francisco, 1968), p. 687. 16. M. R. Hoare and P. Pal, Adv Phys. 20, 161 (1971). 17. D. R. Nelson and F. Spaepen, Solid State Phys. 42, 1 (1989). 18. F. C. Frank and J. C. Kasper, Acta Crystallogr. 11, 184 (1 958) 19. , ibid. 12, 483 (1959). 20. M:Widom, in Aperiodicity and Order: Introduction to Quasicrystals, M. V. Jaric, Ed. (Academic Press, New York, 1988), p. 59. 21. The sample of R-AI,Li,Cu was obtained from F. Gayle of the National Institute of Standards and Technology. A modified Merrill-Bassett style dia- mond anvil cell (DAC) was used to generate high pressures [L. Merrill and W. A. Bassett, Rev. Sci. Instrum. 45, 290 (1974)l; both a 4 : l methanol- ethanol mixture and argon were used as pres- sure-transmitting media. The same result was obtained in both media. Observing the same effect for both media implies that the amorphiza- tion is not principally due to shear [H. K. Mao, J. Xu, P. M. Bell, J. Geophys. Res. 91, 4673 (1986)l. The x-ray diffraction patterns were measured by energy-dispersive x-ray diffraction (EDXD) at the Cornell High Energy Synchrotron Source (CHESS) with a germanium solid-state detector [M. A. Baublitz, V. Arnold, A. L. Ruoff, Rev Sci. Instrum. 52, 1616 (1981); K. Brister, Y. K. Vohra, A. L. Ruoff, b i d 57, 2560 (1986)l. For all measure- ments Edwas 88.18 ke'4.A. (Ed = 6.1993 ke~.& sin 0, where 0 is the acute angle between the incoming x-ray radiation and the detector). We monitored the pressure by mixing gold powder (Alpha, 99.95% pure; grain size, 1 to 3 pm) with the sample. The change in the (111) diffraction line of gold was monitored, and the pressure was then calculated with an equation of state [D. L. Heinz and R. Jeanloz, J Appl. Phys. 55, 885 (1984)l. The gold also served to indicate that we had truly lost the diffraction patterns of the sam- ples and had not erred in aligning the DAC with the synchrotron beam. 22. M. Audier et al., Physica B 153, 136 (1 988). 23. C. A. Guryan, P. W. Stephens, A. I. Goldman, F. W. Gayle, Phys. Rev. B 37, 8495 (1988). 24. B. Dubost, J.-M. Lang, M. Tanaka, P. Sainfort, M. Audier, Nature 324, 48 (1 986) 25. Y. Ma, E. A. Stern, F. W. Gayle, Phys. Rev. Lett. 58, 1956 (1987). 26. M, de Boissieu, C. Janot, J. M. Dubios, M. Audier, B. Dubost, J. Phys. Condensed Matter 3, 1 (1991). 27. Y. Akahama, et a/., J:Phys. Soc. Jpn. 58, 2231 (1989) 28. J. Zarzycki, Glasses and the Vitreous State (Cam- bridge Univ. Press, Cambridge, 1991). 29. For example, J. C. Phillips has pointed out that the icosahedral ordering models ignore the influence of quenching kinetics and mechanical stability that are seminal in the formation of a rapidly quenched metal [J. Mater Res. 1, 1 (1986)l. The preparatory methods used here bypass these considerations. 30. This work was supported by the Division of Mate- rials Sciences, Office of Basic Energy Sciences, Department of Energy, under contract DE-FGO2- 92ER45474. We thank K. Brister and the staff at CHESS for their assistance with the high-pressure EDXD experiments, M. Widom and J. Robeson for critical comments, and F. Gayle for providing the sample of R-AI,Li,Cu. 3 November 1992; accepted 21 January 1993 Velocity Structure of a Gas Hydrate Reflector Satish C. Singh, Timothy A. Minshull, George D. Spence Seismic reflection profiles across many continental margins have imaged bottom-simu- lating reflectors (BSRs) parallel to the seabed; these are often interpreted as the base of a zone in which methane hydrate "ice" is stable. Waveform inversion of seismic reflection data can be used to estimate from seismic data worldwide the velocity structure of a BSR and its thickness. A test of this method at a drill site of the Ocean Drilling Program predicts that sediment pores beneath the BSR contain free methane for approximately 30 meters. The hydrate and underlying gas represent a large global reservoir of methane, which may have economic importance and may influence global climate. Bottom-simulating reflectors (BSRs) are many continental margins. They are most found in the upper few hundred meters of often found in accretionary sediment prisms ocean bottom sediments on and adjacent to at convergent continental margins. BSRs have been widelv interoreted as marking - S. C. Singh, British Institutions Reflection Profiling the base of the Zone in which methane Syndicate, Bullard Laboratories, University of Cam- hydrate is stable (I); methane hydrate sta- bridge, Madingley Road, Cambridge CB3 OEZ, United biliN is temperature-controlled ,,:--AA- nlrlyuulrl. T. A. Minshull, Bullard Laboratories, University of Cam- under thise conditions -and therefore the bridge, Madingley Road, Cambridge CB3 OEZ, United base of this zone follows local isotherms. As Kingdom. G. D. Spence, School of Ocean and Earth Sciences, such, BSRs have been used to estimate University of Victoria, Victoria, British Columbia V8W thermal gradient and hence heat flow (2, 2YZ, Canada. 3). However, BSRs have wider significance SCIENCE VOL. 260 9 APRIL 1993