R EPORTS and 32 for Kurinwas and 34, 34, 35, 36, 37, 42, 44, Loma de Mico, and Fonseca, respectively. In 1996, the 26. M. C. Bove, J. B. Elsner, C. W. Landsea, X. Niu, J. J. and 47 for the other sites. corresponding figures were 0.09, 0.14, 0.15, and 0.16, O’Brien, Bull. Am. Meteorol. Soc. 79, 2477 (1998). 17. M. R. Guariguata, R. L. Chazdon, J. S. Denslow, J. M. respectively. Note that the order of sites ranging 27. We thank students from Universidad de las Regiones Dupuy, L. Anderson, Plant Ecol. 132, 107 (1997). from farthest to nearest to the edge of the nondam- Auto«nomas de la Costa Cariben˜a de Nicaragua, Blue- 18. R. L. Chazdon, R. K. Colwell, J. S. Denslow, M. R. aged forest (i.e., a potential source of pioneer seeds) fields Indian Caribbean University, University of Cen- Guariguata, in Forest Biodiversity Research, Monitor- is also Bodega, La Unio«n, Loma de Mico, Fonseca. tral America, Hood College, University of Guelph, and ing and Modeling: Conceptual Background and Old 22. N. C. Garwood, D. P. Janos, N. Brokaw, Science 205, University of Michigan for help in data collection; D. World Case Studies, F. Dallmeier and J. Comiskey, Eds. 997 (1979). Goldberg for guidance on data analysis; and L. Curran, (Parthenon, Paris, 1998), pp. 285Ð309. 23. P. J. Bellingham, E. V. J. Tanner, J. R. Healey, J. Ecol. R. Burnham, and S. Levin for advice on the manu- 19. M. Begon, J. L. Harper, C. R. Townsend, Ecology 82, 747 (1994). script. The Centro de Investigaciones y Documenta- (Blackwell Science, Oxford, ed. 3, 1996). 24. T. C. Whitmore and D. F. R. P. Burslem, in Dynamics of cio«n de la Costa Atla«ntica (CIDCA) contributed logis- 20. J. H. Vandermeer, M. A. Mallona, D. Boucher, I. Per- Tropical Communities, D. M. Newbery, H. H. T. Prins, tic and intellectual support. Supported by NSF and fecto, K. Yih, J. Trop. Ecol. 11, 465 (1995). N. Brown, Eds. (Blackwell Science, Oxford, 1998), pp. the University of Michigan. 21. In 1991, the proportions of stems that were pioneers 549Ð565. were 0.05, 0.13, 0.17, and 0.19 for Bodega, La Unio«n, 25. K. E. Trenberth, Current 15, 12 (1998). 3 April 2000; accepted 22 August 2000 single domain (PSD) crystals (alternating- A Low Temperature Transfer of field demagnetization is more effective at removing an IRM than an ARM). X-ray maps ALH84001 from Mars to Earth obtained by electron microprobe analysis de- tected Fe-sulfide crystals dispersed through Benjamin P. Weiss,1* Joseph L. Kirschvink,1 the pyroxene matrix, suggesting that pyrrho- Franz J. Baudenbacher,2 Hojatollah Vali,3 Nick T. Peters,2 tite may also be present outside the carbonate. Francis A. Macdonald,1 John P. Wikswo2 We have thus detected two major magnetic minerals in ALH84001, located in the car- The ejection of material from Mars is thought to be caused by large impacts bonate blebs and also probably in the pyrox- that would heat much of the ejecta to high temperatures. Images of the ene: magnetite and pyrrhotite ranging in size magnetic field of martian meteorite ALH84001 reveal a spatially heterogeneous between SP, SD, and PSD. This confirms a pattern of magnetization associated with fractures and rock fragments. Heating previous identification of these magnetic the meteorite to 40¡C reduces the intensity of some magnetic features, indi- minerals (13) and argues against the presence cating that the interior of the rock has not been above this temperature since of titanomagnetite (14). before its ejection from the surface of Mars. Because this temperature cannot Kirschvink et al. (13) suggested that the sterilize most bacteria or eukarya, these data support the hypothesis that interior of ALH84001 has been cooler than meteorites could transfer life between planets in the solar system. 110°C since before the formation of the car- bonate. To obtain more precise thermal con- Large-body impacts are the only known nat- wandering through space, landed in Antarc- straints, we imaged the perpendicular (east/ ural processes capable of ejecting a rock from tica at about 11 thousand years ago (ka) (10). west in the meteorite orientation system) Mars. It has been suggested that some rocks Transmission electron microscopy imag- component of the magnetic field of eight could be ejected without being shocked and ing of the rims of the carbonate blebs in an oriented slices of ALH84001 (15) using the heated (1, 2), and laboratory shock experi- ultrathin section prepared by focused ion Ultrahigh Resolution Scanning SQUID Mi- ments have spalled lightly shocked material beam detected single domain (SD) and super- croscope (UHRSSM). This magnetometer moving at about 20% of Mars’ escape veloc- paramagnetic (SP) (11) magnetite (Fe3O4) has a sensitivity of better than 0.1 nT and is ity (3). Thermal conductivity calculations (4) and monoclinic pyrrhotite (Fe7S8) with char- capable of making two-dimensional images demonstrate that passage through Earth’s at- acteristic lattice fringes. Because magnetic of the magnetic field of materials at room mosphere will not heat the interior of mete- minerals have not been positively identified temperature with a resolution of 500 ␮m orites larger than ϳ0.3 cm above 100°C. outside the carbonate bleb’s rims, the blebs (16). ALH84001 is a meteorite composed of probably carry most of the magnetization in The fusion crust on the top-south surface ϳ95% orthopyroxene that accumulated in a ALH84001. Other studies (12) have shown of ALH84001,228b formed during its high- magma chamber on Mars ϳ4.5 billion years that this magnetite is stoichiometric (impuri- temperature passage through Earth’s atmo- ago (Ga) (5). Carbonate blebs, which may ties Ͻ0.1%). After exposure to a 5-T field at sphere and magnetic field, and is associated contain evidence for ancient life on Mars (6), room temperature, a 20-mg pyroxenite grain with an intense magnetic anomaly (Fig. 1). formed in its fractures at about 4 Ga (7). from ALH84001,236 containing multiple Moment magnetometry measurements of this During its first few billion years, ALH84001 carbonate blebs exhibited a remanence tran- and two additional samples with fusion crusts experienced several shocks, probably from sition at 112 K and a possible weaker rema- indicate that the meteorite initially came to minor planet impacts (8). It was launched nence change at ϳ35 K, diagnostic of low-Ti rest in the ice with its east-southeast axis ϳ Ͻϳ ϳ from the surface of Mars at 15 million magnetite (Fe3–zTizO4 with z 0.01) and pointing up. Much weaker ( 1% intensity) years ago by another impact (9) and after pyrrhotite, respectively. The grain’s magneti- positive and negative magnetic features are zation increased during cooling and then re- present at distances of Ͻ5 mm in from the covered ϳ90% of its original magnetization surface, implying that the heat pulse from 1 Division of Geological and Planetary Sciences, 170- upon warming to room temperature, indicat- atmospheric deceleration did not travel fur- 25, California Institute of Technology, Pasadena, CA 91125, USA. 2Department of Physics and Astronomy, ing the presence of SP and SD crystals and a ther than this into the meteorite. This shallow Vanderbilt University, 6301 Stevenson Center, Nash- lack of multidomain (MD) crystals. Anhys- depth of heating is typical of most meteorites ville, TN 37235, USA. 3Electron Microscopy Centre, teretic remanent magnetization (ARM) and of this size (4). This suggests that the heter- Department of Anatomy and Cell Biology and Depart- isothermal remanent magnetization (IRM) ogeneous magnetization in the interior (Fig. ment of Earth and Planetary Sciences, McGill Univer- sity, 3640 University Street, Montreal QC H3A 2B2, acquisition and demagnetization experiments 2) predates arrival at Earth. Canada. on this grain also provide evidence of pyr- Using the same technique, we made mul- *To whom correspondence should be addressed. E- rhotite (it acquires an IRM up to and beyond tiple magnetic images of ALH84001,232e, mail: [email protected] 1000 mT) and a small fraction of pseudo– which was extracted from the interior of the www.sciencemag.org SCIENCE VOL 290 27 OCTOBER 2000 791 R EPORTS meteorite. The first image of the natural moving trapped air under vacuum before magnetization (TRM) during our heating remanent magnetization (NRM) at room tem- each heating cycle. Following the heating experiment and throughout its 4.5–billion perature (Fig. 2A) reveals a spatially hetero- to 40°C, the magnitude of many magnetic year history (19). This conclusion is sup- geneous pattern of magnetization like that features decreased (including the dipole ported by two experiments. In the first, we observed in the interior portion of 228b. We centered on the carbonate) (Fig. 2, B and C) subjected the large pyroxene grain studied also observed a similar, stable (17) pattern in [first frame of Web fig. 1 (18)], completely by Kirschvink et al. (13) to three 100-mT UHRSSM images of four other cm-sized slic- erasing some that were present in the NRM magnetic pulses at room temperature to es taken from the interior of the meteorite. scan (Fig. 2A). These features never re- give the grain an IRM, which we measured Several positive (eastward) and negative gained the intensities present in the NRM after each pulse. After the first and second (westward) features are present (black ar- image (Fig. 2C and Web fig. 1). At each pulses, the grain was rotated 90° along the rows), some of which are dipolar in character. temperature step above 40°C, different ran- north and east directions, respectively, The strongest of these is centered on a car- domly oriented features weakened or while keeping the field direction constant.