deletions | additions       diff --git a/summary.tex b/summary.tex  index f0edd73..7c2fbbc 100644  --- a/summary.tex  +++ b/summary.tex 
...
In this study we investigated the effect effects  of including accounting for  spin when considering PE for estimating the parameters of  BNS binaries sources  with aLIGO. We expect neutron stars NSs  to be only slowly spinning, and hence that their spins only have a small effect of the GW signature of a BNS merger. However, introducing the component accounting for  spins into PE has \emph{does} have  a significant effect. effect on parameter constraints.  First, this requires the use of more computationally expensive waveforms (that include more physics) and so physics), increasing  the latency of PE is increased. analyses.  Second, strong degeneracies are present in  the correlation of spin with other parameters affects model; not only are  the recovered PDFs. spins themselves poorly constrained, these degeneracies result in weaker constraints on other parameters, particularly masses.  The spins themselves are poorly constrained. There It  is a large uncertainty in most cases important to note that these weaker constraints are the result of accounting for our prior assumptions on NS spins. We tested various choices for conservative prior assumptions about NS spins and found them to have little effect on mass estimates. Only very strong prior assumptions, such  as indicated by some numbers. say $\chi_{1,~2}\leq .05$ (consistent with the simulated population, and NSs found in binaries to date) are likely to significantly affect mass constraints. Such prior assumptions will be hard to justify, however, based on the small number of observed systems and unknown selection effects.  Themass--spin degeneracy means posterior PDFs that are broader for the chirp mass and the mass ratio when spin is included. These posterior PDFs better represent what we believe about the parameters describing the source systems and so a broader posterior must not be viewed as degraded PE performance, rather more realistic.\footnote{We have now reliably measured that the glass is $50\%$ full.} Including spins broadens the chirp-mass distribution by a factor of two \citep{Poisson_1995,Berry_2014}, but the chirp-mass still remains precisely measured, the median something is $42$.  The  sky-location, which is central to performing EM follow-up, is not correlated with the spin; it is unaffected by the inclusion (or exclusion) of spin in PE. parameter estimation.  The luminosity distance is similarly unaffected for this population of slowly spinning neutron stars. NSs.  This means that the lower-latency results that could be supplied in time for EM observatories to search for a counter-part are already as good as we can hope to achieve with GW measurements.