deletions | additions       diff --git a/Mode Visibility.tex b/Mode Visibility.tex  index f2a40d9..9832455 100644  --- a/Mode Visibility.tex  +++ b/Mode Visibility.tex 
...
Fig.~\ref{fig:moneyplot} compares our estimate for suppressed dipole mode visibility (equation \ref{eqn:vis}) with the observations of \cite{Mosser_2011}. Our estimate closely aligns with the branch of stars classified by \cite{Mosser_2011} as suppressed pulsators. The striking agreement holds over a large baseline in $\nu_{\rm max}$. The predicted visibility of equation \ref{eqn:vis} has no free parameters, although there is some uncertainty in the value of $\tau_0$. Additional scatter can be accounted for by a range of stellar masses, metallicities, and inclinations in the the observed sample.  We conclude that the cores of stars with suppressed dipole modes host a mechanism able to efficiently trap waves tunneling through the evanescent region. This is further supported by the normal $\ell=0$ mode visibility in suppressed pulsators  (since radial modes do not propagate within the core) and the lack (or perhaps smaller degree) of suppression observed in $\ell=2$ modes by \citet{Mosser_2011}, as quadrupole modes have a smaller transmission coefficient $T$. There are two additional consequences for mode visibility. First, the larger effective damping rate for suppressed modes will lead to larger line widths in the oscillation power spectrum. The linewidth of a non-suppressed mode is the mode damping rate $\gamma_{\alpha}$, which in general is not equal to $\tau_{0}^{-1}$ because non-radial mixed modes have some inertia in the core (in contrast to radial modes which are confined to the envelope). The linewidth of a suppressed mode is $\tau_{0}^{-1} + \Delta \nu T^2_{\ell}$ and is generally much larger. The suppressed modes in KIC 8561221 \citep{Garcia_2014} indeed have much larger linewidths.