Another concern is that there are more dark halos in CDM simulations with circular velocity Vc 30 km s-1 than there are low-Vc galaxies in the Local Group [87, 88]. A natural solution to this problem was proposed by Bullock et al. , who pointed out that gas will not be able to cool in Vc 30 km s-1 dark matter halos that collapse after the epoch of reionization, which occured perhaps at redshift zreion 6 . When this is taken into account, the predicted number of small satellite galaxies in the Local Group is in good agreement with observations [89, 91]. It is important to develop and test this idea further, and this is being done by James Bullock and by Rachel Somerville and their collaborators; the results to date (e.g. [92, 93]) look rather promising. Other groups (e.g. [94, 95, 96]) now agree that astrophysical effects will keep most of the subhalos dark. As a result, theories such as warm dark matter (WDM), which solve the supposed problem of too many satellites by decreasing the amount of small scale power, may end up predicting too few satellites when reionization and other astrophysical effects are taken into account .
The fact that high-resolution CDM simulations of galaxy-mass halos are full of subhalos has also led to concerns that all this substructure could prevent the resulting astrophysical objects from looking like actual galaxies . In particular, it is known that interaction with massive satellites can thicken or damage the thin stellar disks that are characteristic of spiral galaxies, after the disks have formed by dissipative gas processes. However, detailed simulations [98, 99] have shown that simpler calculations  had overestimated the extent to which small satellites could damage galactic disks. Only interaction with large satellites like the Large Magellanic Cloud could do serious damage. But the number of LMC-size and larger satellites is in good agreement with the number of predicted halos , which suggests that preventing disk damage will not lead to a separate constraint on halo substructure.