Two new low galactic D/H measurements from the Far Ultraviolet Spectroscopic Explorer

We analyze interstellar absorption observed toward two subdwarf O stars, JL 9 and LS 1274, using spectra taken by the Far Ultraviolet Spectroscopic Explorer (FUSE). Column densities are measured for many atomic and molecular species (H I, D I, C I, N I, O I, P II, Ar I, Fe II, and H-2), but our main focus is on measuring the D/H ratios for these extended lines of sight, as D/H is an important diagnostic for both cosmology and Galactic chemical evolution. We find D/H = ( 1.00 +/- 0.37) x 10(-5) toward JL 9 and D/H = (0.76 +/- 0.36) x 10(-5) toward LS 1274 (2sigma uncertainties). With distances of 590 +/- 160 and 580 +/- 100 pc, respectively, these two lines of sight are currently among the longest Galactic lines of sight with measured D/ H. With the addition of these measurements, we see a significant tendency for longer Galactic lines of sight to yield low D/ H values, consistent with previous inferences about the deuterium abundance from D/ O and D/ N measurements. Short lines of sight with H I column densities of log N(H I) < 19.2 suggest that the gas-phase D/H value within the Local Bubble is (D/H)(LBg) = (1.56 +/- 0.04) x 10(-5). However, the four longest Galactic lines of sight with measured D/H, which have d > 500 pc and log N(H I) > 20.5, suggest a significantly lower value for the true local disk gas-phase D/H value, (D/H)(LDg) = (0.85 +/- 0.09) x 10(-5). One interpretation of these results is that D is preferentially depleted onto dust grains relative to H and that longer lines of sight that extend beyond the Local Bubble sample more depleted material. In this scenario, the higher Local Bubble D/ H ratio is actually a better estimate than (D/H)(LDg) for the true local disk D/H, (D/H)(LD). However, if (D/H)(LDg) is different from (D/H)(LBg) simply because of variable astration and incomplete interstellar medium mixing, then (D/H)(LD) = (D/H)(LDg).