\begin{table}%t3
\caption{\label{tab:h2}Neutral gas, metals, dust, and molecules in the VLT/UVES sample
  of {\it Swift}-era $z_{\rm abs}>1.8$ GRB hosts (as of June
  2008).}
%\centerline
 {
\begin{tabular}{llcclccccc}
\hline
\hline\\[-3.2mm]
GRB                      &
$z_{\rm abs}$            &
$\log N({\rm H}^0)$          &
[X/H]                    &
X                        &
${\rm [X/Fe]}^{c}$         &
$\log N({\rm Fe})_{\rm dust}$&
$\log N({\rm H}_2)^{f}$  &
$\log f^{h}$         &
$\log N({\rm C}^0)$          \\[-3.2mm]\\
\hline\\[-3.2mm]
050730  & 3.969 & $22.10\pm 0.10$ &     $-2.18\pm 0.11$        &            S          & $-0.06\pm 0.06^{d}$ & ${<}14.77^{d}$ & ${<}13.8^{g}$ & ${<}{-}8.0$ & ${<}13.05^{g}$\\
050820  & 2.615 & $21.05\pm 0.10$ &     $-0.39\pm 0.10$        &           Zn          & $+0.80\pm 0.03$        &           16.05 & ${<}14.1^{g}$ & ${<}{-}6.7$ & ${<}12.48^{g}$\\
050922C & 2.200 & $21.55\pm 0.10$ &     $-1.82\pm 0.11$        &            S          & $+0.76\pm 0.05^{e}$ & ${<}15.12^{e}$ & ${<}14.6^{g}$ & ${<}{-}6.7$ & ${<}12.45^{g}$\\%J=0 log N<14.4, J=1 log N<14.1
060607  & 3.075 & $16.95\pm 0.03$ &        ...$^{a}$        &          ...$^{a}$ &   ...$^{a}$         &    ...$^{a}$ & ${<}13.5^{g}$ & ${<}{-}3.1$ & ${<}12.03^{g}$\\%J=0 log N<13.1, J=1 log N<13.3
071031  & 2.692 & $22.15\pm 0.05$ &     $-1.73\pm 0.05$        &           Zn          & $+0.04\pm 0.02$        &           14.83 & ${<}14.1^{g}$ & ${<}{-}7.8$ & ${<}12.80^{g}$\\%J=0 log N<13.65, J=1 log N<13.9
080310  & 2.427 & $18.70\pm 0.10$ & ${\le}{-}1.91\pm 0.13^{b}$ &            O$^{b}$ &   ...$^{a}$         &    ...$^{a}$ & ${<}14.3^{g}$ & ${<}{-}4.2$ & ${<}12.55^{g}$\\%J=0 log N<13.9, J=1 log N<14.0
080413A & 2.435 & $21.85\pm 0.15$ &     $-1.60\pm 0.16$        &           Zn          & $+0.13\pm 0.07$        &           15.13 & ${<}15.8^{g}$ & ${<}{-}5.7$ & ${<}13.28^{g}$\\
\hline
\end{tabular}}
\medskip
\par
References to the observations: 050730: \citet{2007AetA...467..629D}; 050820:
\citet{2005GCN..3860....1L}; 050922C: \citet{2005GCN..4044....1D};
060607: \citet{2006GCN..5237....1L}; 071031: \citet{2007GCN..7023....1L};
080310: \citet{2008GCN..7391....1V}; 080413A: \citet{2008GCN..7601....1V}.\\
$^{a}$ The metallicity and/or depletion factor cannot be determined due
to ionization effects.\\
$^{b}$ The O~{\sc i}$\lambda$1302 line used for the fit is partially
blended with strong Al~{\sc ii}$\lambda$1670 absorption from an intervening
system at $z_{\rm abs}=1.671$.\\
$^{c}$ The total Fe$^+$~column density includes the contributions of
energy levels above the ground state up to $^4{\rm F}_{9/2}$ whenever transition
lines from these levels were detected.\\
$^{d}$ When [X/Fe] is negative, $N({\rm Fe})_{\rm dust}$ cannot be calculated
directly. In this case, we estimate a $3\sigma$ upper limit on
$N({\rm Fe})_{\rm dust}$ by considering the upper bound provided by the $3\sigma$
error on [X/Fe].\\
$^{e}$ Significant overabundance of sulphur compared to iron might
be present in this system (see Sect.~\ref{sec:statistics}). The column
density of Fe in dust derived from [S/Fe] (as all measurements based on S or
O) should thus be considered strictly as an upper limit.\\
$^{f}$ Sum of the contributions of the $J=0$ and 1 rotational levels.\\
$^{g}$ $3\sigma$ upper limit.\\
$^{h}$ $f\equiv 2N({\rm H}_2)/(2N({\rm H}_2)+N({\rm H}^0))$.
\end{table}