\begin{table}%t5
\caption{Structural parameters.}
\label{tab5}
\small%\centerline
{
\begin{tabular}{lccccc}
\hline\hline
Cluster&$\sigma_{\rm bg}$&$\sigma_{\rm 0K}$&\rc&\rl&\rt \\
       &$\rm(stars~pc^{-2})$&$\rm(stars~pc^{-2})$&(pc)&(pc)&(pc)\\
(1)&(2)&(3)&(4)&(5)&(6)\\
\hline
% & & & & & & \\[-8pt]
&\multicolumn{5}{c}{Before differential-reddening correction}\\
M~52&$3.2\pm0.2$&$49\pm13$&$0.80\pm0.14$&$8.0\pm1.0$&--\\
NGC~3960&$4.4\pm0.1$&$35\pm9$&$0.69\pm0.12$&$6.0\pm0.3$&-- \\
\cline{2-6}
% & & & & & & \\[-8pt]
&\multicolumn{5}{c}{After differential-reddening correction}\\
M~52&$3.7\pm0.2$&$45\pm10$&$0.91\pm0.14$&$8.0\pm1.0$&$13.1\pm2.2$\\
NGC~3960&$4.8\pm0.2$&$36\pm9$&$0.62\pm0.11$&$6.0\pm0.8$&$10.7\pm3.7$ \\
\hline
\end{tabular}}
\medskip
Notes. We express King's profile as $\sigma(R)=\sigma_{\rm bg}+\sigma_{\rm 0K}/(1+(R/R_{\rm core})^2)$. $\sigma_{\rm 0K}$ and~\rc\ were allowed to vary while~$\sigma_{\rm bg}$ was kept fixed.
\end{table}