\begin{table}%t4 \caption{\label{tab4}Results of Gaussian fits to the detected clouds in CO(1--0).} %\centerline \small \begin{tabular}{l cccc} \hline\hline Name & Velocity & Line width & $M_{\rm vir}$ & $M_{\rm lum}$ \\ & (km s$^{-1}$) & (km s$^{-1}$) & (1000 $M_\odot$) & (1000 $M_\odot$) \\ \hline M33\_2 & $-254.1 \pm 0.5$ & 5.0 $\pm$ 1.2 & 125 $\pm$ 60 & 6 \\ M33\_9 & $-256.2 \pm 0.1$ & 5.3 $\pm$ 0.2 & 140 $\pm$ 9 & 81 \\ M33\_11 & $-251.2 \pm 0.1$ & 8.5 $\pm$ 0.1 & 361 $\pm$ 10 & 312 \\ M33\_12 & $-248.2 \pm 0.0$ & 11.4 $\pm$ 0.1 & 643 $\pm$ 11 & 730 \\ M33\_3 & $-258.7 \pm 0.4$ & 5.5 $\pm$ 1.2 & 150 $\pm$ 66 & 20 \\ M33\_4 & $-269.8 \pm 0.1$ & 4.4 $\pm$ 0.2 & 95 $\pm$ 7 & 58 \\ M33\_5 & $-263.0 \pm 0.1$ & 6.7 $\pm$ 0.3 & 221 $\pm$ 23 & 55 \\ M33\_7 & $-263.9 \pm 0.1$ & 4.1 $\pm$ 0.2 & 82 $\pm$ 9 & 48 \\ M33\_8 & $-271.6 \pm 0.2$ & 10.7 $\pm$ 0.4 & 567 $\pm$ 40 & 44 \\ M33\_10 & $-267.1 \pm 0.1$ & 7.6 $\pm$ 0.2 & 284 $\pm$ 17 & 105 \\ M33\_18 & $-263.3 \pm 0.1$ & 2.5 $\pm$ 0.2 & 31 $\pm$ 4 & 19 \\ \hline \end{tabular} \tablefoot{Velocities and line widths, with the errors estimated by the Gaussian fitting algorithm in CLASS are provided in Cols.~2 and~3. Making the assumption that the cloud radius is equal to the beamsize, virial masses are calculated as in \citet{Braine_n4414b} and the errors using only the uncertainty in $\Delta V$. Masses from the CO(1--0) line intensity are given in Col.~5 using the line area from Table~\ref{tab2}a and a ``standard'' galactic conversion $\ratio = 2 \times 10^{20} \Xunit $. The first 4~clouds have Gaussian fits made using the 2.6~km~s$^{-1}$ channel width whereas the latter 7~fits have been made using data at 0.52~km~s$^{-1}$ channel width. Cloud 8 appears to have two velocity components, producing the broad line. Cloud~2 has a large error associated with the CO(1--0) observation; the Gaussian fit to the CO(2--1) line yields a width of $2.2\pm0.44$~km~s$^{-1}$ but the CO(2--1) beam is much smaller and may not include all the emission.} \end{table}