The morphology and characteristics of the so-called GeV gamma-ray excess detected in the Milky Way lead us to speculate about a possible common origin with the 511 keV line mapped by the SPI experiment about ten years ago. In the previous version of
our paper, we assumed 30 GeV dark matter particles annihilating into $b bar{b}$ and obtained both a morphology and a 511 keV flux (phi_{511 keV} ~ 10^{-3} ph/cm^2/s) in agreement with SPI observation. However our estimates assumed a negligible number density of electrons in the bulge which lead to an artificial increase in the flux (mostly due to negligible Coulomb losses in this configuration). Assuming a number density greater than $n_e > 10^{-3} cm^{-3}$, we now obtain a flux of 511 keV photons that is smaller than phi_{511 keV} ~ 10^{-6} ph/cm^2/s and is essentially in agreement with the 511 keV flux that one can infer from the total number of positrons injected by dark matter annihilations into $b bar{b}$. We thus conclude that -- even if 30 GeV dark matter particles were to exist-- it is impossible to establish a connexion between the two types of signals, even though they are located within the same 10 deg region in the galactic centre.
(abridged) This comment is intended to show that simulations by Smith et al. (S12) support the Dark Star (DS) scenario and even remove some potential obstacles. Our previous work illustrated that the initial hydrogen densities of the first equilibriu
m DSs are high, ~10^{17}/cm^3 for the case of 100 GeV WIMPs, with a stellar radius of ~2-3 AU. Subsequent authors have somehow missed the fact that equilibrium DSs have the high densities they do. S12 have numerically simulated the effect of dark matter annihilation on the contraction of a protostellar gas cloud en route to forming the first stars. They show results at a density ~5 10^{14}/cm^3, slightly higher than the value at which annihilation heating prevails over cooling. However, they are apparently unable to reach the ~10^{17}/cm^3 density of our hydrostatic DS solutions. We are in complete agreement with their physical result that the gas keeps collapsing to densities > 5 10^{14}/cm^3, as it must before equilibrium DSs can form. However we are in disagreement with some of the words in their paper which imply that DSs never come to exist. It seems to us that S12 supports the DS scenario. They use the sink particle approach to treat the gas that collapses to scales smaller than their resolution limit. We argue that their sink is effectively a DS, or contains one. An accretion disk forms as more mass falls onto the sink, and the DS grows. S12 not only confirm our predictions about DS in the range where the simulations apply, but also solve a potential obstruction to DS formation by showing that dark matter annihilation prevents the fragmentation of the collapsing gas. Whereas fragmentation might perturb the dark matter away from the DS and remove its power source, instead S12 show that further sinks, if any, form only far enough away as to leave the DS undisturbed in the comfort of its dark matter surroundings.
