The Spite Lithium Plateau: Ultrathin but Postprimordial

We have studied 23 very metal-poor field turnoff stars, specifically chosento enable a precise measurement of the dispersion in the lithium abundance ofthe Spite Li plateau. We concentrated on stars having a narrow range ofeffective temperature and very low metallicities ([Fe/H] < -2.5) to reduce theeffects of systematic errors, and have made particular efforts to minimizerandom errors in equivalent width and effective temperature. A typical formalerror for our abundances is 0.033 dex (1-sigma), which represents a factor oftwo improvement on most previous studies. Our observed spread about the LiSpite plateau is a mere 0.031 dex (1-sigma), hence we conclude the intrinsicspread is effectively zero at the very metal-poor halo turnoff. We recover arobust dependence of lithium abundance with metallicity, dA(Li)/d[Fe/H] = 0.118+/- 0.023 dex per dex. Previous claims for a lack of dependence of A(Li) on[Fe/H] are shown to have arisen from the use of noisier estimates of effectivetemperatures and metallicities. Although we cannot rule out a uniform depletionmechanism, the lack of spread in the A(Li) abundances limits permissibledepletion by current rotationally- induced mixing models to < 0.1 dex.Correcting for the GCE contribution to both 6Li and 7Li, we infer a primordiallithium abundance of A(Li) = 2.0 dex. We predict that observations of lithiumin extremely low-metallicity stars, having [Fe/H] < -3.0, will yield smallerA(Li) values than the bulk of the stars in this sample, consistent with a lowprimordial abundance.Comment: 42 Pages, 10 figures, 5 tables Submitted to the Astrophysical Journa