diffusive reequilibration), and the disequilib-rium between226Ra and210Pb. Once210Pbexcess is formed, reestablishing secularequilibrium between226Ra and210Pb wouldrequire about 100 years; thus, the210Pb ex-cesses formed by accumulation of gases willpersist. Plagioclase grows during decompres-sion (under H2O-saturated conditions), thussealing off the melt inclusions shortly aftergas is lost. The available experimental datashow that Li diffuses more rapidly than allother trace elements in melt (9) and plagio-clase (10), which suggests that Li can bediffusively homogenized in both phases ontime scales of hours.In the context of Mount St. Helens, ourresults provide a detailed picture of magmaand gas movement during 1980. Specifically,the cryptodome followed a two-stage de-compression path interrupted at a depth of4- to 5-km where magma stalled and gasaccumulated before further ascent. Duringmagma ascent, continued gas fluxing througha semipermeable magmatic foam bufferedthe Li concentration of the melt and sup-plied Rn (Fig. 3). Evidence for breaching ofthe impermeable cap and gas flow comesfrom phreatomagmatic eruptions and steamventing in March and April (2). Magmaerupted during the Plinian phase of theeruption ascended rapidly from a depth ofmore than about 7 km without stalling (22).Before the post–18 May explosive eruptions,magmas stalled at 4- to 5-km before theimpermeable cap was breached and magmaascended rapidly. A dome sample from Octo-ber 1980 also has high Li melt inclusions atpH2O G 125 MPa, which suggests thatmagma ascended more slowly than duringthe preceding explosive eruptions and, aswith the cryptodome, Li concentrations inthe melt were buffered. The increase of(210Pb/226Ra) during the summer of 1980correlates with a decrease in magma ascentrate (4, 5). This suggests that the trend ofincreasing (210Pb/226Ra) with time reflectsprolonged stalling of magma at 4- to 5-kmdepth as the eruption intensity waned.Only recently has attention been turnedto the complexity of shallow conduit pro-cesses and their link to eruption style. Ourdata shed light on processes occurring ontime scales of years to hours before aneruption and, as such, may provide an aidin interpreting observations from establishedmonitoring techniques, for example, gas emis-sions and seismic surveys.210Pb excessesare coupled to210Pb deficits deeper in thesystem. Thus, magma erupting with210Pbexcess requires the presence of degassingmagma at depth.References and Notes1. P. W. Lipman, D. R. Mullineaux, Eds., U.S. Geol. Surv.Prof. Paper 1250, 93 (1981).2. K. V. Cashman, R. P. Hoblitt, Geology 32, 141 (2004).3. K. Cashman, Contrib. Mineral. Petrol. 109, 431 (1992).4. C.-H. Geschwind, M. J. Rutherford, Bull. Volcanol. 57,356 (1995).5. R. Scandone, S. D. Malone, J. Volcanol. Geotherm.Res. 23, 239 (1985).6. J. Blundy, K. Cashman, Contrib. Mineral. Petrol. 140,631 (2001).7. P.-J. Gauthier, M. Condomines, Earth Planet. Sci. Lett.172, 111 (1999).8. J. T. Bennett, S. Krishnaswami, K. K. Turekian, W. G.Melson, C. A. Hopson, Earth Planet. Sci. Lett. 60,61(1982).9. F. M. Richter, A. M. Davis, D. J. DePaolo, E. B. Watson,Geochim. Cosmochim. Acta 67, 3905 (2003).10. B. J. Giletti, T. M. Shanahan, Chem. Geol. 139,3(1997).11. J. D. Webster, J. R. Holloway, R. L. Hervig, Econ. Geol.84, 116 (1989).12. S. Newman, J. B. Lowenstern, Comput. Geosci. 28,597 (2002).13. J. Blundy, K. Cashman, in preparation.14. T. Casadevall et al., Science 221, 1383 (1983).15. A. C. Rust, K. V. Cashman, P. J. Wallace, Geology 32,349 (2004).16. D. S. Stevenson, S. Blake, Bull. Volcanol. 60, 307 (1998).17. B. E. Taylor, J. C. Eichelberger, H. R. Westrich, Nature306, 541 (1983).18. J. C. Eichelberger, C. R. Carrigan, H. R. Westrich, R. H.Price, Nature 323, 598 (1986).19. C. Klug, K. C. Cashman, Bull. Volcanol. 58, 87 (1996).20. S. Signorelli, M. R. Carroll, Geochim. Cosmochim.Acta 64, 2851 (2000).21. H. L. Barnes, Ed., Geochemistry of Hydrothermal OreDeposits (Wiley, New York, ed. 3, 1997).22. M. J. Rutherford, P. M. Hill, J. Geophys. Res. 98,19,667 (1993).23. Methods and data tables are available as supportingmaterial on Science Online.24. During this research, K.B. was supported by a Uni-versity of Bristol scholarship; J.B. and S.T. acknowl-edge support from the Royal Society. W. Melson at theSmithsonian Institution provided some of the samplesused for this study. Other samples were collected withthe generous support of J. Pallister and M. Clynneduring field work. We also thank S. Kasemann (NaturalEnvironment Research Council Ion Microprobe Facility,Edinburgh) for assistance with secondary ion massspectrometry analysis.Supporting Online Materialwww.sciencemag.org/cgi/content/full/1103869/DC1Materials and MethodsTables S1 to S310 August 2004; accepted 1 October 2004Published online 14 October 2004;10.1126/science.1103869Include this information when citing this paper.Ventilation of the Glacial DeepPacific OceanWallace Broecker,1*Stephen Barker,1Elizabeth Clark,1Irka Hajdas,2Georges Bonani,2Lowell Stott3Measurements of the age difference between coexisting benthic and plankticforaminifera from western equatorial Pacific deep-sea cores suggest thatduring peak glacial time the radiocarbon age of water at 2-kilometers depthwas no greater than that of today. These results make unlikely suggestionsthat a slowdown in deep-ocean ventilation was responsible for a sizablefraction of the increase of the ratio of carbon-14 (14C) to carbon in theatmosphere and surface ocean during glacial time. Comparison of14C ages forcoexisting wood and planktic foraminifera from the same site suggests thatthe atmosphere to surface ocean14CtoCratiodifferencewasnotsubstantially different from today’s.Hughen et al.(1) present a strong case thatduring the last glacial maximum (LGM) (i.e.,22,000 to 16,000 calendar years ago) the14Cto C ratio in the atmosphere and surfaceocean was 375 T 25° higher than that forpreindustrial time. The obvious explanationfor this increase is that Earth_s magneticfield was on the average weaker, allowingmore cosmic rays to reach our atmosphere.Hughen et al.(1) contend, however, thatalthough the field was weaker, the conse-quent increase in14C production was insuf-ficient to explain the entire observed14Cincrease. To explain the remainder, theseauthors call on a sizable reduction in deep-sea ventilation rate and a possible reductionin shallow marine carbonate deposition (i.e.,reef growth). Muscheler et al.(2) reach asimilar conclusion by another