In their 1993 paper, Mihos, Richstone and Bothun considered models of
NGC 4038/9 directly based on those used by Barnes in Encounters of
Disk/Halo Galaxies, but with star formation rates in mind.. They
tried to match the observed SFR with the model SFR by examining the
spatial distribution of H 
and CO, but discovered that the models predicted that the nuclear regions
should carry the majority of the star formation and ISM mass in the
system. Based on observations, the model should have demonstrated the
SFR and CO emission evenly divided between the nuclear and overlap
regions.