Gamma-ray spectra from cosmic-ray proton and electron interactions with dense gas clouds have been calculated using a Monte Carlo event simulation code , GEANT4 .
Such clouds are postulated as a possible form of baryonic dark matter in the Universe .
The simulation fully tracks the cascade and transport processes which are important in a dense medium , and the resulting gamma-ray spectra are computed as a function of cloud column-density .
These calculations are used for predicting the Galactic diffuse gamma-ray spectrum which may be contributed by baryonic dark matter ; the results are compared with data from the EGRET instrument , and used to constrain the fraction of Galactic dark matter which may be in the form of dense gas clouds .
In agreement with previous authors , we find useful constraints on the fraction of Galactic dark matter which may be in the form of low column-density clouds ( \Sigma \lesssim 10 { g cm ^ { -2 } } ) .
However , this fraction rises steeply in the region \Sigma \sim 10 ^ { 2 } { g cm ^ { -2 } } , and for \Sigma \gtrsim 200 { g cm ^ { -2 } } we find that baryonic dark matter models are virtually unconstrained by the existing gamma-ray data .