A correlation between dark matter and neutral hydrogen in spiral galaxies

We test the hypothesis put forward by Bosma (1981) that the surface density of dark matter is proportional to the surface density of HI, using decompositions of the rotation curves of a number of galaxies according to the THINGS, along with data for the galaxy NGC 6822. The rotation curves of these galaxies can be explained by assuming the existence of a massive gaseous disk in the absence of a dark halo, although the proportionality factor ??_dark/??_HI between the surface densities of dark matter and HI is different for different galaxies. However, there emerges the problem of the gravitational stability of galaxies whose stellar-velocity dispersions have been estimated, if the thickness of the dark-matter disk is similar to or less than the thickness of the stellar disk. The proportionality between ?? _dark and ??_HI is probably due to the fact that the radial profiles of ??_HI for galaxies with flat rotational curves are close to the critical density of a gravitationally stable gaseous layer (??_HI ?? R ^?1), and ??_dark(R) for a pseudo-isothermal halo obeys the same law.