We use new FUSE data to determine the column densities of interstellar D I , N I , O I , Fe II , and H _ { 2 } along the HD 41161 and HD 53975 sightlines .
Together with N ( H I ) from the literature ( derived from Copernicus and IUE data ) we derive D/H , N/H , and O/H ratios .
These high column density sightlines have both log N ( H I ) > 21.00 and allow us to probe gas up to \sim 1300 pc .
In particular these sightlines allow us to determine the gas phase D/H ratio in a hydrogen column density range , log N ( H ) > 20.70 , where the only five measurements available in the literature yield a weighted average of D/H = ( 0.86 \pm 0.08 ) \times 10 ^ { -5 } .
We find D/H = ( 2.14 \pm~ { } ^ { 0.51 } _ { 0.43 } ) \times 10 ^ { -5 } along the HD 41161 sightline .
This ratio is \sim 3 \sigma higher than the weighted mean D/H ratio quoted above , for sightlines with log N ( H ) > 20.70 , while the D/H ratio for the HD 53975 line of sight , D/H = ( 1.02 \pm~ { } ^ { 0.23 } _ { 0.20 } ) \times 10 ^ { -5 } , agrees within the 1 \sigma uncertainties .
Our D/H measurement along the HD 41161 sightline presents the first evidence of variations of D/H at high N ( H ) .
Our result seems to indicate that either the long sightlines that according to the deuterium depletion model are dominated by cold undisturbed gas where deuterium would be depleted onto carbonaceous grains occur at higher N ( H ) than previously thought or that the clumping of low D/H values in the literature for the long sightlines has another explanation .
Another possibility to explain the high D/H toward HD 41161 is local production of D I which can not be ruled out .
Both of the O/H ratios derived here , ( 9.12 \pm~ { } ^ { 2.15 } _ { 1.83 } ) \times 10 ^ { -4 } and ( 5.37 \pm~ { } ^ { 1.35 } _ { 1.14 } ) \times 10 ^ { -4 } ( for HD 41161 and HD 53975 , respectively ) are higher than what has been found by other authors in studies containing many sightlines , although not inconsistent , in the case of HD 53975 , with individual measurements along some sightlines .
However , O/H along HD 41161 is extremely high , and together with other high O/H ratios based on FUSE data might indicate the presence of unknown systematic effects when deriving N ( O I ) along high column density sightlines .
Finally , we derive ( N/H ) \times 10 ^ { 5 } = 8.32 \pm~ { } ^ { 2.09 } _ { 1.76 } and 5.07 \pm~ { } ^ { 1.45 } _ { 1.21 } and ( D/O ) \times 10 ^ { 2 } = 2.29 \pm~ { } ^ { 0.40 } _ { 0.35 } and 1.91 \pm~ { } ^ { 0.51 } _ { 0.43 } , for HD 41161 and HD 53975 , respectively .
In addition , the relatively high signal-to-noise ratio of the HD 41161 data allows us to place constraints on the f -values of some neutral chlorine transitions , present in the FUSE bandpass , for which only theoretical values are available .