Nature © Macmillan Publishers Ltd 1998

8

letters to nature

146 NATURE | VOL 395 | 10 SEPTEMBER 1998

Unalteredcosmicspherulesina1.4-Gyr-oldsandstonefromFinlandAlexander Deutsch*, Ansgar Greshake*†, Lauri J. Pesonen‡& Pekka Pihlaja‡

* Institut fur Planetologie, Westfalische Wilhelms-Universitat Munster,Wilhelm-Klemm-Strasse 10, D-48149 Munster, Germany‡ Geological Survey of Finland, PO Box 96, FIN-02151 Espoo, Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Micrometeorites—submillimetre-sized particles derived fromasteroids and comets1–5—occur in significant quantities in deepsea sediments1,2,4, and the ice sheets of Greenland6,7 andAntarctica8,9. The most abundant micrometeorites are cosmicspherules3, which contain nickel-rich spinels10 that were crystal-lized and oxidized during atmospheric entry, therefore recordingthe oxygen content in the uppermost atmosphere10–12. But the useof micrometeorites for detecting past changes in the flux ofincoming extraterrestrial matter, and as probes of the evolutionof the atmosphere, has been hampered by the fact that mostobjects with depositional ages higher than 0.5 Myr show severechemical alteration2. Here we report the discovery of unalteredcosmic spherules in a 1.4-Gyr-old13–15 sandstone16,17 (red bed) fromFinland. From this we infer that red beds, a common lithology inthe Earth’s history, may contain substantial unbiased populationsof fossil micrometeorites. The study of such populations wouldallow systematic research on variations in the micrometeorite fluxfrom the early Proterozoic era to recent times9 ( a time span ofabout 2.5 Gyr), and could help to better constrain the time whenthe atmospheric oxygen content was raised to its present level18–20.

Micrometeorites represent the whole spectrum of Earth-crossingbodies more completely than do meteorites2,5,6,8,9. Therefore, asystematic study of micrometeorite populations collected in differ-ent geological horizons could reveal whether time-related variationsin the abundance and type of accreted extraterrestrial matter exist.Such an approach, however, has failed so far, as micrometeoriteswith high terrestrial residence ages are rarely known. Uniqueexceptions are cosmic spherules in Eocene2 and Jurassic sediments21

(about 40 and 190 Myr old, respectively), and the material22 whichwe describe from the Satakunta sandstone.

The Satakunta sediments (up to 1,800 m thick) fill a grabenstriking northwest-southeast in southwestern Finland14,16,17. Theunit correlates with the Jotnian sandstone17, overlying the Sveco-fennian rocks (,1.9–1.8 Gyr old) and rapakivi granites (,1.6 Gyrold) of the Fennoscandian shield. Deposition of the Satakuntasandstone may have already begun during emplacement of rapakivigranites; in contrast the 1;260 6 10-Myr-old post-Jotnian diabasedykes transecting the sandstone set a stringent age limit for cessationof sedimentation13–15. Sedimentological features of the formation17

point to rapid deposition in an alluvial environment with only shortdistances of material transport. Following earlier descriptions16,22,we have sampled the Satakunta sediments at the Murronmakioutcrop (lat. 618 59 43.510 N, long. 228 179 29.360 E). There, themoderately sorted, medium-grained sandstone is bedded on thedecimetre scale, and occasionally interlayered by thin siltstone beds.The purplish red bed containing the cosmic spherules is an arkosewith a fine-grained matrix of authigenic quartz, clay minerals andchlorite, in which detrital, slightly rounded quartz and feldspargrains (up to 3 mm sized), as well as a few micas, are embedded.Standard procedures for mineral separation were employed torecover the spherules. From the 60–125 mm heavy mineral fractions

† Present address: Museum fur Naturkunde, Humboldt-Univesitat zu Berlin, Invalidenstrasse 43, D-10115 Berlin, Germany.

of the processed 5 kg sample, 18 strongly magnetic spherules werehand-picked, of which four were sectioned for microanalysis. All the18 objects are opaque with a metallic shine under reflected light.Most have a spherical morphology (Fig. 1a), yet a few show anelongated shape. Signs of abrasion are totally lacking. We ascribe theexcellent morphological preservation to the extraterrestrial spher-ules having quickly settled in a fluvial environment. Immediatecovering by other particles, such as quartz and feldspar, preventedfurther transportation or re-deposition. Despite the high deposi-tional age of the host lithology, around 1.4 Gyr, the surface of mostspherules appears surprisingly unaltered (Fig. 1a, b). Post-deposi-tional processes which resulted in formation of a haematitic pig-ment in the sandstone matrix did not significantly alter either theextraterrestrial objects or the detrital magnetites which occur inhigh abundance. Electron microscopy, however, reveals that nearlyall the objects suffered light weathering in the sediment, causingsculptured surfaces (Figs 1a, b). The cosmic spherules so farinvestigated from Murronmaki belong to the S (stony) type3. Asshown in Fig. 1, they are ‘‘melted micrometeorites’’ of either (1) theporphyritic (Fig. 1a, c), or (2) the fine-grained barred-olivine type

Figure 1 Secondary (a, b), and backscatter (c, d) electron micrographs of cosmic

spherules from Murronmaki with (a, c) a porphyritic, and (b, d) a barred-olivine

texture. Slight weathering of the spherules in the host sandstone caused

preferential removal of the mesostasis yielding a sculptured surface with

accentuation of octahedral (a) or skeletal (b) magnetite crystals, which are

connected in a typical chain-like fashion (d). The equant olivines in the centre of c

are relics with tiny FeNi metal inclusions (bright). a, b, Uncoated samples; field

emission SEM JEOL 6300F; operating conditions: 1 kVacceleration voltage, 60pA

beam current. c, d, Polished sections, SEM JEOL 840A; operating conditions:

15 kV acceleration voltage, 3 mA beam current.

Table 1 Composition of olivines and matrix in the Satakunta micrometeorites

wt% 1 2 3 4 5 6.............................................................................................................................................................................SiO2 39.8 42.2 42.2 40.5 39.9 41.6TiO2 0.32 n.d. 0.05 n.d. 0.33 0.10Al2O3 5.3 n.d. n.d. n.d. 8.9 2.85Cr2O3 0.12 0.48 0.29 0.05 0.06 0.72FeO 41.5 4.0 2.21 18.6 35.7 21.4MnO 0.27 0.46 0.41 0.53 0.46 0.25MgO 6.9 53.0 53.4 40.5 1.83 31.5NiO 0.16 n.d. n.d. n.d. n.d. n.d.CaO 5.2 0.21 0.13 0.03 7.1 0.77Total 99.57 100.35 98.69 100.21 94.28 99.19.............................................................................................................................................................................Fayalite (mol%) 4.5 2.9 20.5.............................................................................................................................................................................Representative electron microprobe (EMPA) analyses on polished sectioned spherules.Porphyritic textured spherules—no. 1, matrix; olivines, nos. 2, 3, forsteritic cores, and no. 4,rim; barred-olivine spherules—no. 5, matrix (the low total reflects the glassy nature of thismaterial), no. 6,Mg-rich olivine (analysis, taken at the centre of the grain, contains significantcontributions of the matrix due to the small grain size of the olivine). n.d., below detectionlimits of 0.05wt% for TiO2, 0.02wt% for Al2O3, and 0.07wt% for NiO. ‘Matrix’ is defined hereas an area which at the resolution of the SEM is free of visible crystal phases. Microprobeused was an JEOL JXA-8600 S, operating at 15 keV and a sample current of 15 nA, at theInstitut fur Planetologie. Standards: natural silicates and oxides; B and A correction29.

Nature © Macmillan Publishers Ltd 1998

8

letters to nature

NATURE | VOL 395 | 10 SEPTEMBER 1998 147

(Fig. 1b, d)2,3,5. These textures originate during atmospheric entry ofthe precursor material at hypervelocities, causing melting andquenching at different cooling rates1–8,10,11,23–25.

The porphyritic spherules are characterized by olivines, up to20 mm across (Fig. 1c), and small spinels embedded in an apparentlyglassy, Fe-rich matrix (Table 1). Some olivines contain Mg-richcores surrounded by an Fe-rich rim (Table 1). In the Mg-rich cores,there is up to 0.2 wt% CaO. In addition, one grain contains severalsmall FeNi metal inclusions (Fig. 1c). These features led us tointerpret the olivine cores as relics of the precursor material23,24,which escaped melting during atmospheric entry. The barred-olivine type spherules are also composed of olivines and spinelsembedded in an apparently glassy matrix (Fig. 1d), consistingmainly of silica and iron (Table 1). Olivines typically form anhedralto subhedral, 2–5 mm sized crystals with Mg-rich cores (Table 1)rimmed by more Fe-rich compositions; yet relic grains are absent.

The spinels vary widely in grain sizes (maximum, 5 mm) withindifferent regions of one specimen, suggesting a high thermalgradient during cooling. The crystals exhibit skeletal, often cruci-form, or octahedral morphologies. In the barred-olivine spherules,the spinels form chain-like, orientated dendritic growths (Fig. 1b,d). Chemically (Table 2), they are magnetites with up to 24 mol%magnesioferrite (MgFe3+

2 O3). Compositional variation is larger inspinels of the porphyritic spherules, which also contain Al and Cr-rich grains that may represent partial relics of spinels in theprecursor material (Table 2).

The textural features, mineral compositions, and bulk chemis-try (Table 3) of the Satakunta spherules compare excellently withmelted micrometeorites with young terrestrial residence ages2–7,23,24,thus confirming the assessment16,22 that they represent microme-teorites. In the porphyritic spherules, some minerals survived theatmospheric entry heating26, indicating that this material experi-enced lower maximum temperatures compared to the barred-olivine spherules25. The presence of relic MgO-rich olivine cores,high in Ca, and with FeNi metal inclusions, provides the most solidargument for the extraterrestrial origin of the Satakunta spherules.Such olivine compositions are typical of carbonaceous chondrites,relating the precursor material of the porphyritic spherules to thisparticular class of stony meteorites3,23,24.

Spinel morphology and chemistry further substantiate the extra-terrestrial origin. In cosmic spherules, spinels show a restrictedrange in their Fe3+/(total Fe) ratio (60–75 at.%), and contain asignificant trevorite (NiFe2O4) component reflecting the high Niconcentration in the chondritic precursor material10,11. The Sata-kunta spinels with 1.05–2.65 wt% NiO match spinel compositionsin other melted micrometeorites7,10,24; their Fe3+/(total Fe) ratios fall

exactly in the above given range, with exception of the relic crystals,high in chromium (Table 2).

According to experiments11, the Fe3+/(total Fe) ratio of spinelswhich crystallized from melted micrometeorites depends criticallyon the oxygen fugacity, fO2

, and hence, the altitude of deceleration inthe Earth’s atmosphere; whereas temperature only plays a subordi-nate role. In the Satakunta spherules, the Fe3+/(total Fe) ratios ofspinels (Table 2) indicate that fO2

was in the range of 10−5 to7 3 10 2 4 bar during melting and crystallization10,11. The estimateis corroborated by the composition of the olivine rims (,20 mol%fayalite; Table 1) also serving as experimentally calibrated11 fO2

indicator in melted micrometeorites. In the present atmosphere,such an fO2

corresponds to an altitude of 70–40 km. According tocurrent understanding, a substantial increase in the atmosphericoxygen content occurred during the interval 2.5 to ,1.85 Gyr (refs18–20). In the following late Proterozoic era, the atmosphericoxygen content apparently remained at a value intermediatebetween modern levels and those of the early Proterozoic18,19, yethigher atmospheric O2 levels would be consistent with geologicaldata19. The spinel composition in cosmic spherules of the ,1.4-Gyr-old Satakunta sandstone may favour the latter interpretation, giventhat the density of the upper atmosphere was similar to the presentvalue.

The depositional age of these sediments exceeds by far that of thehitherto oldest known fossil extraterrestrial material—meteoritesin Ordovician sediments27 (,460 Myr old). Despite their excep-tionally long residence in sediments, the Satakunta micrometeoritesstill display primary features just like micrometeorites deposited inthe Quaternary. The occurrence of cosmic spherules, whose altera-tion effects are restricted to the outer parts of the rims, in such oldsediments opens a new research perspective: a systematic analysis ofabundance and type of all spherules in the Satakunta sandstone, andin similar lithologies (red beds) of distinctly different age, couldyield a better knowledge on whether the flux of extraterrestrialmaterial has varied with time. For example, ,1.4-Gyr-old red bedsoccur not only in Finland and Sweden, but are known from theNorth American craton, the Ukraine, India and Siberia (see, forexample, ref. 28). In addition, the presence of unaltered spinels infossil spherules of different size could provide otherwiseunobtainable19 information on the evolution of the oxygen fugacityin the atmosphere. But if the fossil micrometeorites are found to berestricted to a few layers in the sedimentary rocks, their occurrencemay define marker horizons, which are of specific value in fossil-barren red beds. M

Table 2 Composition of spinels in the Satakunta micrometeorites

wt% 1 2 3 4 5 6.............................................................................................................................................................................TiO2 0.25 — — 0.64 0.14 0.57Al2O3 1.89 2.11 2.36 5.7 1.36 9.0Cr2O3 1.78 1.80 1.74 1.03 0.87 24.5Fe2O3* 64.4 64.8 64.7 55.5 65.0 27.7FeO* 23.8 23.3 23.1 35.3 27.7 31.4MnO — 0.05 0.11 0.06 0.18 0.21MgO 6.1 6.3 6.1 0.15 3.7 5.9NiO 1.78 1.64 1.89 1.62 1.05 0.72Total 100† 100† 100† 100† 100† 100†.............................................................................................................................................................................Fe3+/(tot. Fe) (at.%) 73.0 73.6 73.7 61.1 70.1 46.8.............................................................................................................................................................................Representative EMPA analyses on polished sectioned spherules; for analytical techniques,see footnotes to Table 1. Analyses 1 to 3——barred olivine; 4 to 6—porphyritic spherules.Grains nos 1 to 3 show a rather uniform composition with ,64mol% magnetite, 20mol%magnesioferrite, and 4.7mol% trevorite, and small yet varying amounts of Ti, Cr, Al spinelcomponents. The compositional variation is larger in the porphyritic spherules, ranging frommagnetite-magnesioferrite (grain no. 5) to grains rich in either the spinel-hercynite, or thechromite component (grain no. 6); in addition, the ulvospinel component makes up to3.2mol%. The Al and Cr-rich grain no. 6 may represent a partial relics of spinels in theprecursor material. Contributions of the host matrix to spinel raw analyses were correctedassuming that all SiO2 (,10wt%) comes from thematrix; corrections forotherelementsweredone using elemental ratios as obtained in the matrix.* Calculated. Molar components of the spinels and atomic Fe3+/(total Fe) were calculatedaccording to ref.11).† Normalized to 100%.

Table 3 Bulk composition of the Satakunta micrometeorites

wt% Porphyritic spherule A Barred-olivine spherule B

Mean 6 1 jD Range Mean 6 1 jD Range.............................................................................................................................................................................SiO2 33:47 6 3:41 27.5–35.9 33:32 6 0:13 33.3–33.4TiO2 0:18 6 0:09 0.06–0.30 0:20 6 0:09 0.09–0.27Al2O3 4:74 6 0:34 4.23–5.08 3:06 6 0:16 2.96–3.35Cr2O3 0:54 6 0:25 0.21–0.90 0:49 6 0:13 0.35–0.69FeOtot 42:54 6 4:89 37.8–50.8 39:34 6 1:12 38.0–41.0MnO n.d. ,0.10 n.d. ,0.05MgO 15:40 6 2:50 13.1–19.5 19:35 6 1:54 17.9–21.8NiO Irregular ,0.27 0:52 6 0:23 0.24–0.86CaO 2:82 6 0:58 1.98–3.52 3:02 6 0:19 2.82–3.31Na2O Irregular ,0.36 Irregular ,0.72K2O Irregular ,0.27 Irregular ,0.03–0.23SO3 0:13 6 0:10 ,0.24 Irregular ,0.16.............................................................................................................................................................................Total 99.09–101.2 98.48–101.5.............................................................................................................................................................................Energy dispersive analyses (JEOL 840A scanningelectron microscope (SEM) with EDS LinkAN1000; 20 keVacceleration voltage, 9 nA samplecurrent) on polishedsectioned spherules.Standards: natural oxides and silicates; ZAF correction. N ¼ 5; scanned area is 20 3 20 mm(A); 80 3 80 mm (B). Relative to the composition of CI chondrites30, the analysed samples aresignificantly depleted in Ni, Cr, Mn and sulphur, a feature known also from deep seaspherules2. The atomic ratios of the spherules normalized to CI30 cluster at 1 for Ca/Si, rangebetween 0.4 and 0.6 for Fe/Si, and between 0.7 and 1.1 for Mg/Si; the Al/Si ratio, however,exceeds that of CI. As expected, the spherules lack sodium and potassium due to loss byvolatilization during atmospheric entry. Minor K2O and Na2O contents in some areas mayreflect slight weathering in the host sediment.

Nature © Macmillan Publishers Ltd 1998

8

letters to nature

148 NATURE | VOL 395 | 10 SEPTEMBER 1998

Received 3 March; accepted 22 June 1998.

1. Blanchard, M. B., Brownlee, D. E., Bunch, T. E., Hodge, P. W. & Kyte, F. T. Meteoroid ablation spheresfrom deep-sea sediments. Earth Planet. Sci. Lett. 46, 178–190 (1980).

2. Taylor, S. & Brownlee, D. E. Cosmic spherules in the geologic record. Meteoritics 26, 203–211 (1991).3. Brownleee, D. E., Bates, B. & Schramm, L. The elemental composition of stony cosmic spherules.

Meteoritics Planet. Sci. 32, 157–175 (1997).4. Brownlee, D. E. in The Sea Vol. 7 (ed. Emilinani, C.) 733–762 (Wiley, New York, 1981).5. Brownlee, D. E. Cosmic dust: collection and research. Annu. Rev. Earth Planet. Sci. 13, 147–173 (1985).6. Maurette, M., Jehanno, C., Robin, E. & Hammer, C. Characteristics and mass distribution of

extraterrestrial dust from the greenland ice cap. Nature 328, 699–702 (1987).7. Robin, E., Christophe Michel-Levy, N., Bourot-Denise, M. & Jehanno, C. Crystalline micrometeorites

from Greenland blue lakes: their chemical composition, mineralogy and possible origin. Earth Planet.Sci. Lett. 97, 162–176 (1990).

8. Maurette, M. et al. A collection of diverse micrometeorites recovered from 100 tonnes of Antarcticblue ice. Nature 351, 44–46 (1991).

9. Taylor, S., Lever, J. H. & Harvey, R. P. Accretion rate of cosmic spherules measured at the South Pole.Nature 392, 899–902 (1998).

10. Robin, E., Bonte, Ph., Froget, L., Jehanno, C. & Rocchia, R. Formation of spinels in cosmic objectsduring atmospheric entry: A clue to the Cretacious-Tertiary boundary event. Earth Planet. Sci. Lett.108, 181–190 (1992).

11. Gayraud, J., Robin, E., Rocchia, R. & Froget, L. Formation conditions of oxidized Ni-rich spinel andtheir relevance to the K/T boundary event. Geol. Soc. Am. Spec. Pap. 307, 425–443 (1996).

12. Brownlee, D. E., Bates, B. A. & Wheelock, M. M. Extraterrestrial platinum group nuggets in deep-seasediments. Nature 309, 693–695 (1984).

13. Suominen, V. The chronostratigraphy of southwestern Finland, with special reference to Postjotnianand Subjotnian diabases. Geol. Surv. Fin. Bull. 304 (1991).

14. Amatov, A., Laitakari, I. & Prooshin, Ye. in Explanation to the Map of Precambrian Basement of the Gulfof Finland and Surrounding Area 1:1 mil (ed. Koistinen, T.) Geol. Surv. Finl. Spec. Pap. 21, 99–113 (1996).

15. Neuvonen, K. J. Remanent magnetization of the Jotnian sandstone in Satakunta, SW-Finland. Bull.Geol. Soc. Fin. 45, 23–27 (1973).

16. Marttila, E. Satakunnan hiekkakiven sedimentaatio-olosuhteista. (On the sedimentary environmentof the Satakunta sandstone.) Thesis, Univ. Turku (1969).

17. Kohonen, J., Pihlaja, P., Kujala, H. & Marmo, J. Sedimentation of the Jotnian Satakunta sandstone,western Finland. Geol. Surv. Fin. Bull. 369 (1993).

18. Des Marais, D. J., Strauss, H., Summons, R. E. & Hayes, J. M. Carbon isotope evidence for the stepwiseoxidation of the Proterozoic environment. Nature 359, 605–609 (1992).

19. Kasting, J. F. Earth’s early atmosphere. Science 259, 920–926 (1993).20. Ohmoto, H. When did the Earth’s atmosphere become oxic? Geochem. News 93, 12–27 (1997).21. Jehanno, C., Boclet, D., Bonte, Ph., Castellarin, A. & Rocchia, R. Identification of two populations of

extraterrestrial particles in a Jurassic hardground of the Southern Alps. Proc. Lunar Planet. Sci. Conf.18, 623–630 (1988).

22. Marttila, E. Mikrometeorlitteja Satakunnan kiekkakivessa. (Micrometeorites in the Satakuntasandstone.) Geologi 20, 135–136 (1968).

23. Steele, I. M. Olivine in Antarctic micrometeorites: Comparison with other extraterrestrial olivine.Geochim. Cosmochim. Acta 56, 2923–2929 (1992).

24. Beckerling, W. & Bischoff, A. Occurrence and composition of relict minerals in micrometeorites fromGreenland and Antarctica—Implication for their origjns. Planet. Space Sci. 43, 435–449 (1995).

25. Brownlee, D. E., Bates, B. & Beauchamp, R. H. in Chondrules and their Origin (ed. King, E. A.) 122–133 (Lunar and Planetary Inst., Houston, 1983).

26. Love, S. G. & Brownlee, D. E. Heating and thermal transformation of micrometeoroids entering theEarth’s atmosphere. Icarus 89, 26–43.

27. Thorslund, P. & Wickman, F. E. Middle Ordovician chondrite in fossiliferous limestone from Brunflo,central Sweden. Nature 289, 285–286 (1981).

28. Goodwin, A. M. Precambrian Geology; the Dynamic Evolution of the Continental Crust (Academic, SanDiego, 1991).

29. Bence, A. E. & Albee, A. L. Empirical correction factors for the electron microanalysis of silicates andoxides. J. Geol. 76, 382–403 (1968).

30. Kerridge, J. F. & Matthews, M. S. (eds) Meteorites and the Early Solar System Appendix 3 (Univ.Arizona Press, Tucson, 1988).

Acknowledgements. We thank T. Grund, F. Bartschat, U. Heitmann, M. Flucks and D. Kettrup fortechnical assistance, and A. Bischoff, E. Marttila, A. Putnis, U. Scharer, and H. Strauss for discussions.Field work was supported by a DAAD-Finish Academy of Science exchange program (A.D. and L.J.P.) andwe acknowledge additional support by DFG.

Correspondence and requests for materials should be addressed to A.D. (e-mail: deutsca@ uni-muenster.de).

Magnetic trappingof calciummonohydridemoleculesatmillikelvin temperaturesJonathan D. Weinstein*, Robert deCarvalho*,Thierry Guillet*, Bretislav Friedrich*† & John M. Doyle*

* Department of Physics, † Department of Chemistry, Harvard University,Cambridge, Massachusetts 02138, USA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Recent advances1–5 in the magnetic trapping and evaporativecooling of atoms to nanokelvin temperatures have opened impor-tant areas of research, such as Bose–Einstein condensation andultracold atomic collisions. Similarly, the ability to trap and coolmolecules should facilitate the study of ultracold molecularphysics and collisions6; improvements in molecular spectroscopy

could be anticipated. Also, ultracold molecules could aid thesearch for electric dipole moments of elementary particles7. Butalthough laser cooling (in the case of alkali metals1,8,9) andcryogenic surface thermalization (in the case of hydrogen10,11)are currently used to cool some atoms sufficiently to permit theirloading into magnetic traps, such techniques are not applicable tomolecules, because of the latter’s complex internal energy-levelstructure. (Indeed, most atoms have resisted trapping by thesetechniques.) We have reported a more general loading technique12

based on elastic collisions with a cold buffer gas, and have used itto trap atomic chromium and europium13,14. Here we apply thistechnique to magnetically trap a molecular species—calciummonohydride (CaH). We use Zeeman spectroscopy to determinethe number of trapped molecules and their temperature, and setupper bounds on the cross-sectional areas of collisional relaxationprocesses. The technique should be applicable to many paramag-netic molecules and atoms.

Evaporative cooling has proved to be a powerful technique forproducing ultracold atoms9,15,16. To employ evaporative cooling,atoms are first loaded into a magnetic trap. Although laser cooling iscommonly used to load Li, Na, Rb and Cs, extension of this methodto molecules is precluded by their complex internal level structure.(A technique to use multiline Raman cell radiation to optically coolalkali-metal dimers has been proposed, but not yet implemented17.)Hydrogen has also been trapped but is the only atoms that can beloaded by surface thermalization, as its very low binding energy toliquid helium (1 K) is unique. A molecule would be stronglyadsorbed by such a cold surface. A different approach to producecold molecules for trapping is photoassociation of laser-cooledalkali-metal atoms18,19. This technique has recently beendemonstrated20, producing Cs2 molecules at a translational tem-perature of 300 mK. This is sufficiently cold that it should be possibleto trap these molecules with far-off-resonance optical trapping

Window

Detection laserAblation laser

Magnet current

Confined molecules

Copper cell

Mirror

Solid lump of CaH2

Figure 1 Cutaway diagram of the experimental apparatus. The copper cell is

anchored to the mixing chamber of a dilution refrigerator. The magnet is

immersed in liquid helium. A vacuum space isolates the relatively warm (4K)

magnet from the cold (300mK) cell. The superconducting magnet coils are

arranged in the anti-Helmholtz configuration (currents travel in opposite

directions) and generate a spherical quadrupole magnetic trap up to 3T deep.

For the 1mB magnetic moment of the ground state of CaH, this results in a trap

depth of 2K. Detection and ablation lasers enter through three borosilicate

windows (not shown) at 300K, 77K and 4K before passing through the fused

silica window at the bottom of the cell. Fluorescence from the trapped molecules

(induced by the detection laser) is collected outside the 300K window by a

photomultiplier tube.

0.15 M NaCl, 3 mM EDTA, 0.005% Surfactant P20 and 1 mM DTT, were flown overimmobilized RNAs at a rate of 30 ml min21, and dissociation monitored by flowing theblank buffer over a subsequent 3-min period. The residual bound protein was removed byinjecting 20 ml of 2 M NaCl at 20 ml min21. Experiments at each concentration of proteinswere repeated at least twice. The apparent K d values were derived from the global fit of theassociation and disassociation curves to a simple 1:1 Langmuir interaction model with acorrection for mass transport effects using BIAevaluation 3.0 software.

Received 27 January; accepted 29 March 2004; doi:10.1038/nature02519.

1. Denli, A. M. & Hannon, G. J. RNAi: an ever-growing puzzle. Trends Biochem. Sci. 28, 196–201 (2003).

2. Bartel, D. P. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297 (2004).

3. Voinnet, O. RNA silencing as a plant immune system against viruses. Trends Genet. 17, 449–459 (2001).

4. Volpe, T. A. et al. Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by

RNAi. Science 297, 1833–1837 (2002).

5. Hall, I. M. et al. Establishment and maintenance of a heterochromatin domain. Science 297,

2232–2237 (2002).

6. Bernstein, E., Caudy, A. A., Hammond, S. M. & Hannon, G. J. Role for a bidentate ribonuclease in the

initiation step of RNA interference. Nature 409, 363–366 (2001).

7. Elbashir, S. M., Lendeckel, W. & Tuschl, T. RNA interference is mediated by 21- and 22-nucleotide

RNAs. Genes Dev. 15, 188–200 (2001).

8. Hammond, S. M., Bernstein, E., Beach, D. & Hannon, G. J. An RNA-directed nuclease mediates post-

transcriptional gene silencing in Drosophila cells. Nature 404, 293–296 (2000).

9. Hammond, S. M., Boettcher, S., Caudy, A. A., Kobayashi, R. & Hannon, G. J. Argonaute2, a link

between genetic and biochemical analyses of RNAi. Science 293, 1146–1150 (2001).

10. Nykanen, A., Haley, B. & Zamore, P. D. ATP requirements and small interfering RNA structure in the

RNA interference pathway. Cell 107, 309–321 (2001).

11. Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R. & Tuschl, T. Single-stranded antisense

siRNAs guide target RNA cleavage in RNAi. Cell 110, 563–574 (2002).

12. Yan, K. S. et al. Structure and conserved RNA binding of the PAZ domain. Nature 426, 468–474 (2003).

13. Lingel, A., Simon, B., Izaurralde, E. & Sattler, M. Structure and nucleic-acid binding of the Drosophila

Argonaute 2 PAZ domain. Nature 426, 465–469 (2003).

14. Song, J. J. et al. The crystal structure of the Argonaute2 PAZ domain reveals an RNA binding motif in

RNAi effector complexes. Nature Struct. Biol. 10, 1026–1032 (2003).

15. Elbashir, S. M., Martinez, J., Patkaniowska, A., Lendeckel, W. & Tuschl, T. Functional anatomy of

siRNAs for mediating efficient RNAi in Drosophila melanogaster embryo lysate. EMBO J. 20,

6877–6888 (2001).

16. Fire, A. et al. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis

elegans. Nature 391, 806–811 (1998).

17. Chiu, Y. L. & Rana, T. M. siRNA function in RNAi: a chemical modification analysis. RNA 9,

1034–1048 (2003).

18. Theobald, D. L., Mitton-Fry, R. M. & Wuttke, D. S. Nucleic acid recognition by OB-fold proteins.

Annu. Rev. Biophys. Biomol. Struct. 32, 115–133 (2003).

19. Liu, Q. et al. R2D2, a bridge between the initiation and effector steps of the Drosophila RNAi pathway.

Science 301, 1921–1925 (2003).

20. Hohjoh, H. RNA interference (RNAi) induction with various types of synthetic oligonucleotide

duplexes in cultured human cells. FEBS Lett. 521, 195–199 (2002).

21. Harborth, J. et al. Sequence, chemical, and structural variation of small interfering RNAs and short

hairpin RNAs and the effect on mammalian gene silencing. Antisense Nucleic Acid Drug Dev. 13,

83–105 (2003).

22. Chiu, Y. L. & Rana, T. M. RNAi in human cells: basic structural and functional features of small

interfering RNA. Mol. Cell 10, 549–561 (2002).

23. Holen, T., Amarzguioui, M., Wiiger, M. T., Babaie, E. & Prydz, H. Positional effects of short interfering

RNAs targeting the human coagulation trigger Tissue Factor. Nucleic Acids Res. 30, 1757–1766 (2002).

24. Amarzguioui, M., Holen, T., Babaie, E. & Prydz, H. Tolerance for mutations and chemical

modifications in a siRNA. Nucleic Acids Res. 31, 589–595 (2003).

25. Czauderna, F. et al. Structural variations and stabilising modifications of synthetic siRNAs in

mammalian cells. Nucleic Acids Res. 31, 2705–2716 (2003).

26. Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in oscillation mode.

Methods Enzymol. 276, 307–326 (1997).

27. Brunger, A. T. et al. Crystallography & NMR system: A new software suite for macromolecular

structure determination. Acta Crystallogr. D 54, 905–921 (1998).

28. Jones, T. & Kjeldgaard, M. Electron-density map interpretation. Methods Enzymol. 227, 174–208

(1997).

29. Nicholls, A., Sharp, K. A. & Honig, B. Protein folding and association: insights from the interfacial and

thermodynamic properties of hydrocarbons. Proteins 11, 281–296 (1991).

30. Katsamba, P. S., Park, S. & Laird-Offringa, I. A. Kinetic studies of RNA-protein interactions using

surface plasmon resonance. Methods 26, 95–104 (2002).

Supplementary Information accompanies the paper on www.nature.com/nature.

Acknowledgements We thank K. Saigo for providing us with the eIF2C1 complementary

DNA clone. This research was supported by the NIH. We thank Y. Cheng and personnel at the

Advanced Photon Source (APS) beamlines 19BM and 14IDB for help in collecting the X-ray

diffraction data. Use of the APS beamline was supported by the US Department of Energy, Basic

Energy Sciences, Office of Science.

Competing interests statement The authors declare that they have no competing financial

interests.

Correspondence and requests for materials should be addressed to D.J.P. (pateld@mskcc.org).

Coordinates for the PAZ–siRNA complexes containing 2-nt ribo- and deoxyribonucleotide 3 0

overhangs have been deposited in the Protein Data Bank under accession codes 1SI3 and 1SI2,

respectively.

..............................................................

retraction

Unaltered cosmic spherules in a1.4-Gyr-old sandstone from FinlandAlexander Deutsch, Ansgar Greshake, Lauri J. Pesonen & Pekka Pihlaja

Nature 395, 146–148 (1998)..............................................................................................................................................................................

We earlier reported finding unaltered cosmic spherules in samplesfrom the Satakunta sandstone, Finland1. At that time, we had found18 spherules in 5 kg of material, and an additional 100 spheruleswere recovered from the 60–125 m heavy mineral separates of thesamples. Rocks collected later from the same and nearby sites had nospherules in them2. We have subsequently concluded that thecosmic spherules1 were not part of the Satakunta sandstone samples.The total number of spherules we found during processing in theMuenster laboratory (about 120) is in itself a compelling argumentfor an exceptional contamination. A

1. Deutsch, A., Greshake, A., Pesonen, L. & Pihlaja, P. Unaltered cosmic spherules in a 1.4-Gyr-old

sandstone from Finland. Nature 395, 146–148 (1998).

2. Kettrup, D. Mikrometeorite in terrestrischen Sedimenten—Eine systematische Analyse potentieller

Wirtsgesteine unter besonderer Berucksichtigung des mesoproterozoischen Satakunta-Sandsteins

(SW-Finnland). Inaug.-Diss., Univ. Munster (2002).

letters to nature

NATURE | VOL 429 | 20 MAY 2004 | www.nature.com/nature322 © 2004 Nature Publishing Group