There is growing evidence that gas clouds in galaxy halos are a common phenomenon and a natural part of the galaxy evolution process. This halo gas is part of accreting satellites in some cases, but some of the clouds are consistent with a condensing cloud model. The hot material the clouds condense out of at z = 0 most likely consists of a mixture of materials that has been gradually enriched over time. Since the medium is thought to be responsible for much of the stripping of dwarf galaxies (Mayer et al. 2006), and extends out to at least 55 kpc as evident from the observations of clouds in the Magellanic Stream (Stanimirovic et al. 2002; Putman et al. 2003; Sembach et al. 2003), it is unlikely to all originate from feedback from the Galaxy itself. In fountain scenarios the hot gas generally rises to only ~ 10 kpc (Booth & Theuns 2007; de Avillez 2000). One plausible scenario is the hot halo is created with the initial collapse of baryons into the dark matter potential well and is supplemented through time with fountain material and stripped gas from satellite galaxies.
In terms of the Local Group galaxies, the three largest galaxies all show evidence for HI halo clouds with a total mass in clouds of ~ 107-8.5 M. This is not a large reservoir of gas in the context of the chemical evolution models, but with a continual new supply through incoming gas-rich satellites and condensing clouds it may be sufficient at z = 0. Presumably both of these sources were more abundant at higher redshift. The future may be bright for both the Milky Way and M31 in the context of future star formation fuel with the accretion of gas from the Magellanic System for the Milky Way and possibly M33's gas for M31. There may also be additional Leo T's sitting at the outskirts of the Local Group that will serve as future Galactic and M31 morsels. The role of the diffuse halo of these galaxies will be further probed with observations and simulations of the head-tail clouds and new results from the Cosmic Origins Spectrograph.
Acknowledgments. M.E.P. thanks F. Fraternali, R. Sancisi, and J. Ostriker for useful discussions. Many thanks to the Turn On GALFA Survey team, Snezana Stanimirovic, Kevin Douglas, Steven Gibson, Carl Heiles, and Eric Korpela for their help with the GALFA data presented here.