https://geosci.uchicago.edu/~rowley/Rowley/Publications_files/Nature%201998%20Spray.pdf


Nature © Macmillan Publishers Ltd 1998 8 letters to nature NATURE | VOL 392 | 12 MARCH 1998 171 (1 3 1 km) images of visible reflectance and infrared brightness temperature from a companion instrument ATSR-2, to estimate cloud fraction, cloud-top height and other cloud and surface parameters. M Received 29 November 1996; accepted 9 December 1997. 1. Houghton, J. T. et al. (eds) Climate Change 1994, Part 1, Radiative Forcing of Climate Change (Cambridge Univ. Press, 1995). 2. Houghton, J. T. et al. (eds) Climate Change 1995 (Cambridge Univ. Press, 1996). 3. Hollandsworth, S. M. et al. Ozone trends deduced from combined NIMBUS-7 SBUV and NOAA-11 SBUV/2 data. Geophys. Res. Lett. 22, 905–908 (1995). 4. McPeters, R. D., Hollandsworth, S. M., Flynn, L. E., Herman, J. R. & Seftor, C. J. Long-term ozone trends derived for the 16-year combined Nimbus 7 Meteor 3 TOMS Version 7 record. Geophys. Res. Lett. 23, 3699–3702 (1996). 5. WMO, Report of the International Ozone Trends Panel 1988 Vol. I, 17–52 (Rep. No. 18, Global Ozone Research and Monitoring Project, World Meteorological Organization, Geneva, 1988). 6. Fishman, J. Watson, C. E., Larsen, J. C. & Logan, J. A. Distribution of tropospheric ozone determined from satellite data. J. Geophys. Res. 95(D4), 3599–3617 (1990). 7. Jiang, Y. B. & Yung, Y. L. Concentration of tropospheric ozone from 1979 to 1992 over tropical pacific South-America from TOMS data. Science 272, 714–716 (1996). 8. Hahne, A., Lefebvre, A., Callies, J. & Christensen, B. GOME—The development of a new instrument. ESA Bull. 83, 41–46 (1995). 9. Francis, C. R. et al. The ERS-2 spacecraft and its payload. ESA Bull. 83, 13–31 (1995). 10. Rosanov, V., Diebel, D., Spurr, R. J. D. & Burrows, J. P. GOMETRAN: A radiative transfer model for the satellite project GOME—the plane-parallel version. J. Geophys. Res. (in the press). 11. Rodgers, C. D. Retrieval of atmospheric temperature and composition from remote measurements of thermal radiation. Rev. Geophys. Space Phys. 14, 609–624 (1976). 12. Rodgers, C. D. Characterization and error analysis of profiles retrieved from remote sounding measurements. J. Geophys. Res. 95(D5), 5587–5595 (1990). 13. Collins, W. J., Stevens, D. S., Johnson, C. E. & Derwent, R. G. Tropospheric ozone in a global-scale three-dimensional Lagrangian Model and its response to NOx emission controls. J. Atmos. Chem. (in the press). 14. Burrows, J. et al. A Study of Methods for the Retrieval of Atmospheric Constituents (Final Rep. ESA Contract 9687/91/NL/BI, European Space Agency, Noordwijk, The Netherlands, 1994). 15. Diebel, D. et al. Detailed Analysis of the Retrieval Algorithms Selected for Level 1-2 Processing of GOME Data (Final Rep. ESA Contract 10728/94/NL/CN, European Space Agency, Noordwijk, The Netherlands, 1995). 16. Chance, K. V., Burrows, J. P., Perner, D. & Schneider, W. Satellite measurements of atmospheric ozone profiles, including tropospheric ozone, from ultraviolet/visible measurements in the nadir geometry: a potential method to retrieve tropospheric ozone. J. Quant. Spectrosc. Radiat. Transfer 57(4), 467– 476 (1997). 17. Grainger, J. F. & Ring, J. Anomalous Fraunhofer line profiles. Nature 193, 762 (1962). Acknowledgements. Permission to use GOME data before its public release was granted by ESA, and GOME Level-1 Products were provided by DLR. The GOMETRAN radiative transfer model and the highresolution solar reference spectrum used in this study were supplied by J. Burrows (Univ. Bremen) and K. Chance (Harvard-Smithsonian Astrophysical Observatory), respectively. Correspondence and requests for materials should be addressed to R.M. (e-mail: r.munro@ecmwf.int). Evidence for a late Triassic multiple impact event on Earth John G. Spray*, Simon P. Kelley† & David B. Rowley‡ * Department of Geology, University of New Brunswick, Fredericton E3B 5A3, Canada † Department of Earth Sciences, The Open University, Milton Keynes MK7 6AA, UK ‡ Department of Geophysical Sciences, University of Chicago, Chicago, Illinois 60637, USA ......................................................................................................................... Evidence for the collision of fragmented comets or asteroids with some of the larger (jovian) planets and their moons is now well established following the dramatic impact of the disrupted comet Shoemaker–Levy 9 with Jupiter in 1994 (ref. 1). Collisions by fragmented objects result in multiple impacts that can lead to the formation of linear crater chains, or catenae, on planetary surfaces2 . Here we present evidence for a multiple impact event that occurred on Earth. Five terrestrial impact structures have been found to possess comparable ages (,214 Myr), coincident with the Norian stage of the Triassic period. These craters are Rochechouart (France), Manicouagan and Saint Martin (Canada), Obolon’ (Ukraine) and Red Wing (USA). When these impact structures are plotted on a tectonic reconstruction of the North American and Eurasian plates for 214 Myr before present, the three largest structures (Rochechouart, Manicouagan and Saint Martin) are co-latitudinal at 22.88 (within 1.28, ,110 km), and span 43.58 of palaeolongitude. These structures may thus represent the remains of a crater chain at least 4,462 km long. The Obolon’ and Red Wing craters, on the other hand, lie on great circles of identical declination with Rochechouart and Saint Martin, respectively. We therefore suggest that the five impact structures were formed at the same time (within hours) during a multiple impact event caused by a fragmented comet or asteroid colliding with Earth. Approximately 150 impact structures are now known on Earth, most of which are ,200 Myr old3 . They represent a small portion of what would have been a much larger number, most structures having been buried or destroyed via the tectonic and erosional activities of our dynamic planet. The discoveries of an iridiumenriched clay layer4 and the 65-Myr-old Chicxulub impact structure in Yucata´n, Mexico5 , both coincident with the Cretaceous/Tertiary (K/T) boundary, have drawn attention to the global environmental damage such impacts can cause and, hence, their association with mass extinction6 . Geologists have tended to seek single giant impact structures as potential sources of catastrophe on Earth, but the temporal and spatial association of five impact structures described here indicates that the effects of synchronous multiple impact events produced by fragmented comets or asteroids should not be discounted. The Rochechouart impact structure in France occurs within Hercynian target rocks of the Massif Central. The structure is partly eroded and no impact-related topography remains. A thin (,10 m) impact melt sheet is overlain by a fallback sequence 20– 50 m thick, and underlain by allochthonous and autochthonous basal breccias. A shock zoning study and the distribution of various impact lithologies indicate that the structure is ,25 km in diameter7 . Previous radiometric work has yielded ages of 154– 173 Myr (K–Ar)8 and 186 Myr (Rb–Sr)9 , but the recent 40Ar/39Ar laser spot fusion dating of pseudotachylyte generated in impactrelated basement faults has yielded10 214 6 8 Myr (2j). This revised age is in keeping with the regional geological setting of the structure, whereas the younger age determinations, which implied a Jurassic origin, are now considered to indicate the timing of subsequent Ar loss and Rb and Sr mobilization. The Manicouagan impact structure in eastern Canada is ,100 km in diameter. The target consists of amphibolite to granulite facies metamorphic rocks and anorthosites of Grenville age (,1 Gyr) with overlying Ordovician carbonates. The anorthositic rocks have been locally shocked to yield diaplectic plagioclase glass (maskelynite)11. High precision U–Pb 260 250 240 230 220 210 200 190 180 170 Triassic Jurassic A Norian H S P T L C M O R SM Time (Myr before present) RW Figure 1 Radiometric and biostratigraphic age data for the five impact structures plotted for the late Triassic to early Jurassic using the timescale of Gradstein et al.22. Abbreviations for stages as follows: A, Anisian; L, Ladinian; C, Carnian; H, Hettangian; S, Sinemurian; P, Pliensbachian; T, Toarcian. (Rhaetian substage included in the Norian.) Shaded columns show errors for the Carnian/Norian (64.4 Myr) and Norian/Hettangian (64 Myr) boundaries22. Letter abbreviations for the impact structures (given in circles) as follows: R, Rochechouart; M, Manicouagan; SM, Saint Martin; O, Obolon’; RW, Red Wing. Nature © Macmillan Publishers Ltd 1998 8 letters to nature 172 NATURE | VOL 392 | 12 MARCH 1998 dating of zircons from the .200-m-thick impact melt sheet yields12 an age of 214 6 1 Myr (2j). The Saint Martin impact structure in western Canada is not well exposed, but gravity and magnetic surveys, coupled with the results of a drilling programme in the late 1960s, have revealed its essential structure. It is ,40 km in diameter13. The target rocks comprise Superior Province (Archaean) granitic gneisses and overlying Palaeozoic sedimentary cover. The youngest pre-impact material is Devonian, while Jurassic red beds and evaporites constitute the oldest post-impact cover. Rb/ Sr dating of the impact melt sheet yields14 an age of 219 6 32 Myr (2j). The Obolon’ impact structure of the Ukraine is ,15 km in diameter and buried by 150–300 m of post-impact Middle Jurassic to Quaternary sedimentary rocks15. Geophysical and bore-hole data indicate that the target lithologies are early Proterozoic migmatitic gneisses and granitoids, overlain by a mid-Carboniferous to Lower Triassic sedimentary cover. Shatter cones, melt-bonded breccias, diaplectic glass and the presence of coesite16, microdiamonds17 and planar deformation features in quartz15 have been identified from drill core. No radiometric constraints are available for Obolon’, but stratigraphic evidence supports a formation age18 of 215 6 25 Myr. The Red Wing impact structure of the western USA is ,9 km in diameter and buried by ,1.5 km of Jurassic to Neogene post-impact sediments. The impact structure is formed in Silurian to Triassic carbonates, minor sandstones and evaporites, and provides the structural trap to one of the most prolific oil fields in the United States19. It has a stratigraphically constrained age20 of 200 6 25 Myr. Each of the five structures shows features characteristic of hypervelocity impact, including most of the following: impactgenerated melt sheets of crustal bulk composition, shatter cones, planar deformation features in minerals, high-pressure polymorphs and diaplectic glasses. All are complex craters and are probably central peak basins21, though Manicouagan may be large enough to be a peak-ring or multi-ring basin. Based on the timescale of Gradstein et al.22, four of the five impact structures have ages that fall within the Norian stage of the late Triassic, while Red Wing overlaps with the Norian within error (Fig. 1). However, given the possibility of the impacts coinciding to within a few hours, temporal constraints provided by radiometric dating alone cannot prove the synchroneity of the impacts due to experimental uncertainties. The critical test for synchronous impact remains a spatial one. Using the two best-dated structures as an age constraint (Rochechouart and Manicouagan), the five sites have been plotted on a map showing the reconstructed positions of the North American and Eurasian plates 214 Myr ago23 (Fig. 2). The three largest impact structures (Rochechouart, Manicouagan and Saint Martin) plot as co-latitudinal at a mean palaeolatitude of 22.88, and a latitudinal width of ,1.28. This is a remarkably good fit to a small-circle path about the Earth’s spin axis. The spread in palaeolongitude is 43.58 (4,462 km). The three aligned structuresthus form a crater chain, or catena, like the catenae described on Jupiter’s moons2 . The probability of the three co-latitudinal structures representing a random event was assessed using Monte Carlo simulation. The probability ranges in a nonlinear fashion from P < 0:0003 to P < 0:083 for N ¼ 3 to N ¼ 13, where N is the total number of coeval impactors. These probabilities decrease by a factor of about 10 if the longitudinal range of the co-latitudinal impacts is restricted to <608, which is the case here.It is thus highly unlikely that these represent a random array. The two smallest impact structures, Obolon’ (15 km) and Red Wing (9 km), have essentially identical trajectories with respect to the latitude-parallel trajectory of the other three. Obolon’ and Rochechouart (the easternmost pair) define a great circle that has a declination of 37.58, while Red Wing and Saint Martin (the westernmost pair) define a great circle that has a declination of 43.48. Thus, they have the same sense and essentially the same magnitude of rotation with respect to the small-circle trajectory. If the longitudinal offset of 43.58 is removed for Red Wing and Saint Martin, while maintaining their latitudes, a best-fit great circle with a delineation of 37.28 is obtained for the four ‘end’ craters (Red Wing, Saint Martin, Rochechouart and Obolon’) (Fig. 2). Deviations of these data from the best-fit great circle are very small (,0.48). From the temporal and spatial constraints, we conclude that Saint Martin, Manicouagan and Rochechouart were generated by projectiles that were essentially coaxial with respect to each other (like Shoemaker–Levy 9; refs 24, 25). The spatial relations between the projectiles that generated Obolon’ and Red Wing and those that generated the three larger impacts are not clear. Red Wing, at only 9 km diameter, could have been produced by a fragment of the same projectile that generated the larger Saint Martin structure. It is probable that there were more than five impact structures generated by the fragmented bolide. Those craters generated by fragments that hit the Tethys or Panthalassa oceans rather than Pangaea, however, would have been subsequently destroyed by subduction. It is of interest to note that two other impact structures show possible spatial associations with the five already discussed. Wells Creek, Tennessee, USA (12 km diameter, 200 6 100 Myr in age)3 , lies on a great circle that intersects Manicouagan, with a declination identical to that of the two great circles shown in Fig. 2 (37.28). Newporte, North Dakota, USA (3 km diameter, ,500 Myr in age)3 , lies on the same great circle as, and between, Red Wing and Saint Martin. However, the age constraints for these two structures will need refining before any association with the five principal Palaeolongitude (degrees) Palaeolatitude (degrees) 0 10 20 30 40 50 0 10 20 30 40 50 -48 -24 0 24 48 -48 -24 0 24 48 SM Eurasia North America Africa 214 Myr M R O RW Present Reconstructed Greenland Crater name Rochechouart (R) Manicouagan (M) Saint Martin (SM) Obolon, (O) Red Wing (RW) Lat. 23.5 22.1 22.7 38.6 19.1 Long. 16.6 -7.0 -26.9 34.9 -31.1 Long. 0.93 -68.70 -98.53 32.92 -103.55 Lat. 45.83 51.38 51.78 49.50 47.60 Great circles with declination = 37.2° 43.5° 22.8° Figure 2 Reconstruction of the North American (Laurentian) and Eurasian plate positions in the Northern Hemisphere of Earth 214 Myr ago (Mercator projection), showing the locations of the five impact structures. Nature © Macmillan Publishers Ltd 1998 8 letters to nature NATURE | VOL 392 | 12 MARCH 1998 173 structures can be considered further. If the five impact structures are contemporaneous then they would have been generated by projectiles of common affinity. Attempts at identifying the body type of the impactor at Manicouagan and Saint Martin have been unsuccessful; melt sheet analyses yield low to undetectable iridium values26. Initial work on Rochechouart melt rocks implied contamination by a IIA iron meteorite27, but subsequent analysis of footwall metal veinsrevealed an elevated Cr content and Ni/Co ratios more compatible with a chondritic source28. Small amounts of kamacite and taenite have been reported in the heavy mineral fractions of allogenic breccia from Obolon’ 15, but their presence does not decisively distinguish an iron from a chondritic source. No appraisal has yet been made of Red Wing. Melt sheets may be poor indicators of projectile compositions owing to the effects of crustal dilution, and depending on the velocity of impact and subsequent mode of projectile dissemination29. It should also be cautioned that the non-volatile components of comets have chemical compositions close to those of primitive carbonaceous chondrites30. Furthermore, because comets comprise roughly equal portions of silicates, hydrocarbons and volatile ices, their expected bulk siderophile contents are expected to be less than one-third that of an undifferentiated carbonaceous asteroid. This may make detection of a cometary projectile difficult. More work is needed to identify the projectile for the five impact structures. Until then, it is not possible to determine whether the source was a comet or an asteroid on the basis of impact structure geochemistry alone. The Norian stage of the Triassic is marked by one or more major mass extinction events. One of these occurred at the Triassic/Jurassic boundary at ,205 Myr (refs 12, 22, 31) and is associated with impact-generated shocked-quartz-bearing shale beds in Italy32. This is significantly later than the 214-Myr multiple impact event described here. However, Benton33 has alluded to an earlier Carnian-Norian mass extinction event at ,220 Myr, which affected tetrapods and plants on land and certain marine groups34. Unlike the end-Triassic, the age of the Carnian/Norian boundary is less well constrained and it is possible that a 214-Myr multiple impact could be associated with an extinction event at that time. As the late Triassic period is now known to contain evidence for a multiple impact event, a renewed effort should be made to study Carnian/Norian boundary and Norian sections with the aim of locating and characterizing the impact ejecta layer, as well as refining biostratigraphic-radiometric correlations for that time period. The calculated probability of a multiple impact occurring on Earth, comparable to that of Shoemaker–Levy 9 colliding with Jupiter, appears to be low35. This is because, relative to Earth, Jupiter’s huge mass presents a formidable gravitational attraction and disruptive force for asteroids and comets. Pairs of craters formed by near-simultaneous impact of binary asteroids are expected for ,10% of the impact structures on Earth and Venus36. However, a terrestrial multiple impact event resulting from the collision of several projectiles would appear to require a special situation, possibly involving the close fly-by, orbital capture and decay, and consequent clustered strike of a fragmented or weakly coherent comet or asteroid35. The mechanisms by which such a particular circumstance may arise now need to be further explored. Critical to identifying any additional crater chains on Earth will be the continued precision dating of proven impact structures, followed by their placing on accurate plate reconstructions for the time of impact. M Received 30 January; accepted 19 February 1998. 1. Orton, G. et al. Collision of comet Shoemaker-Levy 9 with Jupiter observed by the NASA infrared telescope facility. Science 267, 1277–1282 (1995). 2. Melosh, H. J. & Schenk, P. Split comets and the origin of crater chains on Ganymede and Callisto. Nature 365, 731–733 (1993). 3. Grieve, R. A. F., Rupert, J., Smith, J. & Therriault, A. The record of terrestrial impact cratering. GSA Today 5, 189–196 (1995). 4. Alvarez, L. W., Alvarez, W., Asaro, F. & Michel, H. V. Extraterrestrial cause for the Cretaceous-Tertiary extinction. Science 208, 1095–1108 (1980). 5. Hildebrand, A. R. et al. Chicxulub crater: a possible Cretaceous/Tertiary boundary impact crater on the Yukata´n Peninsula. Geology 19, 867–871 (1991). 6. Sharpton, V. L. & Ward, P. E. (eds) Global catastrophes in earth history: an interdisciplinary conference on impacts, volcanism and mass mortality. Spec. Pap. Geol. Soc. Am. 247 (1990). 7. Lambert, P. The Rochechouart crater: shock zoning study. Earth Planet. Sci. Lett. 35, 258–268 (1977). 8. Lambert, P. La structure d’impact de me´te´orite ge´ante de Rochechouart. Thesis, Univ. Paris-Sud (1974). 9. Reimold, W. U. & Oskierski, W. in Research in Terrestrial Impact Structures (ed. Pohl, J.) 94–114 (Vieweg, Braunschweig/Weisbaden, Germany, 1987). 10. Kelley, S. P. & Spray, J. G. A late Triassic age for the Rochechouart impact structure, France. Meteorit. Planet. Sci. 32, 629–636 (1997). 11. Dworak, U. Stoßwellenmetamorphose des Anorthosits vom Manicouagan Krater, Que´bec, Canada. Contrib. Mineral. Petrol. 24, 306–347 (1969). 12. Hodych, J. P. & Dunning, G. R. Did the Manicouagan impact trigger end-of-Triassic mass extinction? Geology 20, 51–54 (1992). 13. McCabe, H. R. & Bannatyne, B. B. Lake St. Martin cryptoexplosion crater and geology of surrounding area. Geol. Surv. Manitoba Geol. Pap. 3/70 (1970). 14. Reimold, W. U., Barr, J. M., Grieve, R. A. F. & Durrheim, R. J. Geochemistry of the melt and country rocks of the Lake St. Martin impact structure, Manitoba, Canada. Geochim. Cosmochim. Acta 54, 2093–2111 (1990). 15. Masaitis, V. L., Danilin, A. N., Karpov, G. M. & Raykhlin, A. I. Karla, Obolon’ and Rotmisrovka astroblemes in the European part of the U.S.S.R. Dokl. Akad. Nauk SSSR 230, 174–177 (1976). 16. Gurov, Y. P., Val’ter, A. A. & Rakitskaya, R. B. Coesite in rocks of meteorite explosion craters on the Ukrainian shield. Int. Geol. Rev. 22, 329–332 (1978). 17. Gurov, Y. P., Gurova, E. P. & Rakitskaya, R. B. Impact diamonds in the craters of the Ukrainian shield. Meteoritics 30, 515–516 (1995). 18. Masaitis, V. L. et al. The Geology of Astroblemes (Nedra Press, Leningrad, 1980). 19. Grieve, R. A. F. & Masaitis, V. L. The economic potential of terrestrial impact craters. Econ. Geol. 36, 105–151 (1994). 20. Gerhard, L. C., Anderson, S. B., Lefever, J. A. & Carlson, C. G. Geological development, origin, and energy mineralresources of Williston Basin, North Dakota. Am. Assoc. Petrol. Geol. Bull. 66, 989–1020 (1982). 21. Melosh, H. J. Impact Cratering: A Geologic Process (Oxford Univ. Press, 1989). 22. Gradstein, F. M. et al. A Mesozoic time scale. J. Geophys. Res. 99, 24051–24074 (1994). 23. Ziegler, A. M. et al. in The Tectonic evolution of Asia (eds Yin, A. & Harrison, T. M.) 371–400 (Cambridge Univ. Press, 1996). 24. Hammel, H. B. et al. HST imaging of atmospheric phenomena created by the impact of comet Shoemaker-Levy 9. Science 267, 1288–1296 (1995). 25. Weaver, H. A. et al. The Hubble Space Telescope (HST) observing campaign on comet ShoemakerLevy 9. Science 267, 1282–1288 (1995). 26. Palme, H. Identification of projectiles of large terrestrial impact craters and some implications for the interpretation of Ir-rich Cretaceous/Tertiary boundary layers. Geol. Soc. Am. Spec. Pap. 190, 223–233 (1982). 27. Jannsens, M. J., Hertogen, J., Takahiashi, H., Anders, E. & Lambert, P. Rochechouart impact crater: identification of projectile. J. Geophys. Res. 82, 750–758 (1977). 28. Horn, N. P. & Goresy, A. E. The Rochechouart crater in France: stony and not iron meteorite? Lunar Planet. Sci. XI, 468–470 (1980). 29. McClaren, D. J. & Goodfellow, W. D. Geological and biological consequences of giant impacts. Annu. Rev. Earth Planet. Sci. 18, 123–171 (1990). 30. Hut, P. et al. Comet showers as a cause of mass extinctions. Nature 329, 118–126 (1987). 31. Hallam, A. in Global Events and Event Stratigraphy (ed. Walliser, O. H.) 265–283 (Springer, New York, 1996). 32. Bice, D. M., Newton, C. R., McCauley, S., Reiners, P. W. & McRoberts, C. A. Shocked quartz at the Triassic/Jurassic boundary in Italy. Science 259, 443–446 (1992). 33. Benton, M. J. More than one event in the late Triassic mass extinction. Nature 321, 857–861 (1986). 34. Hallam, A. & Wignall, P. B. Mass Extinctions and their Aftermath (Oxford Univ. Press, 1997). 35. Love, S. G., Bottke, W. F. & Richardson, D. C. Alternative formation mechanisms for terrestrial crater chains. Lunar. Planet. Sci. XXVIII, 837–838 (1997). 36. Bottke, W. F. & Melosh, H. J. Formation of asteroid satellites and doublet craters by planetary tidal forces. Nature 381, 51–53 (1996). Acknowledgements. We thank M. Benton and J. Melosh for discussions and reviews. This work was supported by NSERC (Canada) and the Open University (UK) grants to J.G.S. and S.P.K., respectively. D.B.R. acknowledges industrial support for the Palaeogeographic Atlas Project. Correspondence and requests for materials should be addressed to J.G.S. (e-mail: jgs@unb.ca). Fission-track ages of stone tools and fossils on the east Indonesian island of Flores M. J. Morwood*, P. B. O’Sullivan†, F. Aziz‡ & A. Raza† * Department of Archaeology and Palaeoanthropology, University of New England, New South Wales 2351, Australia † School of Earth Sciences, La Trobe University, Victoria 3083, Australia ‡Geological Research and Development Centre, Bandung 4011, Indonesia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The islands of Wallacea, located between the Southeast Asian (Sunda) and Australian (Sahul) continental areas, offer unique potential for the study of evolution and cultural change. Located east of Java and Bali, which were periodically connected to the Asian mainland, the Wallacean islands could only be reached by