\relax 
\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces (Left): Our major new result based on several cycles of HST data showing the averaged Hubble diagram (SNe within $z < 0.01$ of each other have been combined) for all SNe from our low-extinction subsample. The solid curve overlaid on the data represents our best-fit flat-universe model, $(\Omega _M,\Omega _\Lambda = (0.25, 0.75)$. Two other cosmological models are shown for comparison. (Center): Our latest joint measurements of $\Omega _M$ and $w$ assuming $\Omega _M + \Omega _\Lambda = 1$ and that $w$ is not time-varying. Confidence regions (68\% and 90\%) are shown for the SCP SN\nobreakspace  {}Ia data overlaid with LSS measurements and combined CMB measurements. (Right): 68\% statistical confidence intervals on the dark energy equation of state possible from SNLS when it is completed 5 years from now, assuming a flat universe (from CMB measurements), and with (solid) and without (dotted) a prior on $\Omega _M$ from LSS. If a small systematic error is introduced, the statistical confidence interval will miss the correct simulated $\Lambda $ ($w=-1$) solution by $2\sigma $ (long-dash). }}{4}}
\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Preliminary SNLS $B$-band Hubble diagram. The restframe $I$ observations we propose would allow uniformly small uncertainties out to $z\sim 0.9$, whereas currently poor color measurements limit high-quality measurements to $z\mathrel {\raise 0.27ex\hbox {$>$}\kern -0.70em \lower 0.71ex\hbox {{$\scriptstyle \sim $}}}0.6$. In addition, we will generate $I$-band Hubble diagram which will be largely independent of the $B$-band Hubble diagram.}}{5}}
\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces Actual SNLS lightcurves, showing the ``rolling'' nature of the search. Shaded vertical regions indicate example times when HST could be pre-scheduled. The bold lightcurves then indicate the example SNe\nobreakspace  {}Ia that we could follow using the observing strategy we propose here. Note that suitable SNe are always available in this pre-scheduled mode, so no disruptive ToO is required.}}{5}}