deletions | additions       diff --git a/Discussion.tex b/Discussion.tex  index 1423a8b..cc8ea36 100644  --- a/Discussion.tex  +++ b/Discussion.tex 
...
For perfect wave trapping in the core, purely dipole modes only exist in the envelope, with part of their energy leaking into the core as running magneto-gravity waves. If some wave energy does escape the core, it may leave a signature in the form of mixed magneto-gravity acoustic modes, {\bf or by producing magnetic mode splitting}, which could be used to constrain the internal magnetic field geometry.  In principle, it is possible that another symmetry-breaking mechanism within the core could suppress dipole mode amplitudes. The only other plausible candidate is rapid core rotation. In order for rotation to strongly modify the incoming waves such that they will be trapped in the core, the core must rotate at a frequency comparable to $\nu_{\rm max}$, roughly two orders of magnitude faster than the values commonly {\bf measured} in red giant cores \cite{Beck_2011,Mosser_2012,deheuvels_2014}. The {\bf depressed dipole modes} mode}  star KIC8561221 \cite{Garcia_2014} does not exhibit rapid {\bf core} rotation and disfavors the rotation scenario. A magnetic field of amplitude $B \! > \! 10^4 \, {\rm G}$ (see Figure \ref{fig:Bc}) could be present in the core of a red giant if it was retained {\bf from previous phases of stellar formation/evolution (supplementary online text).} These strong fields may reside within the inner core with little external manifestation apart from the {\bf reduced} visibility of the dipole modes. However, fields of similar amplitude have been discussed in order to explain the suppression of thermohaline mixing in a small fraction of red giant stars, as inferred from the observations of their surface abundances \cite{Charbonnel_2007}. The {\bf inferred core field strength of $B_r \! \gtrsim \! 1.5 \! \times \! 10^7 \, {\rm G}$ in KIC8561221} shows very strong magnetic fields ($B \! \gg \! 10^6 \, {\rm G}$) can exist within the radiative cores of early RGB stars. Since these fields are likely inherited from previous stages of stellar evolution, slightly weaker ($B \! \gg \! 10^5 \,{\rm G})$ fields {\bf could exist in the cores of exceptional very highly magnetized main sequence stars.} A detailed analysis of a large {\bf sample} of red giants with {\bf depressed} dipole modes will put strong constraints on the amplitude and evolution of internal magnetic fields in {\bf different types of stars.}  
...