Several molecular solids containing magnetic complexes (Fe_8, Mn_{12}) have now been seen to display spin tunneling. The magnetic complex has a large spin (~10), and an Ising type anisotropy, so the spin can tunnel from "up" to "down". The experimental evidence for this is that the tunnel splitting inferred from Landau-Zener-Stuckelberg (LZS) experiments shows tunnel splitting oscillations as a function of a static magnetic field along the hard axis, which can be understood in terms of interference of tunneling trajectories. This is a very strong quantum signature. Yet this scenario is problematic. A spin of 10 is nearly macroscopic, so its tunneling should be strongly suppressed by environmental decoherence, but experiments agree rather well with the LZS formula derived for an isolated spin. In Fe_8, dipole- dipole couplings are 10^6 times the putative tunnel splitting. We have developed a model to include the effect of dipole-dipole and nuclear spin couplings. The tunneling is found to be strongly incoherent, but the LZS formula still holds due to a cancellation between the incoherent single spin flip rate and the effective width of the LZS transition region. Departures from LZS are seen at slow sweeps.