Where are the missing galactic satellites?

According to the hierarchical clustering scenario, galaxies are assembled b y merging and accretion of numerous satellites of different sizes and masse s. This ongoing process is not 100% efficient in destroying all of the accr eted satellites, as evidenced by the satellites of our Galaxy and of M31. U sing published data, we have compiled the circular velocity (V-circ) distri bution function (VDF) of galaxy satellites in the Local Group. We find that within the volumes of radius of 570 kpc (400 h(-1) kpc assuming the Hubble constant(1) h = 0.7) centered on the Milky Way and Andromeda, the average VDF is roughly approximated as n(> V-circ) approximate to 55 +/- 11(V-circ/ 10 km s(-1))(-1.4+/-0.4) h(3) Mpc(-3) for V-circ in the range approximate t o 10-70 km s(-1) The observed VDF is compared with results of high-resoluti on cosmological simulations. We find that the VDF in models is very differe nt from the observed one: n(> V-circ) approximate to 1200(V-circ/10 km s(-1 ))(-2.75) h(3) Mpc(-3). Cosmological models thus predict that a halo the si ze of our Galaxy should have about 50 dark matter satellites with circular velocity greater than 20 km s(-1) and mass greater than 3 x 10(8) M., withi n a 570 kpc radius. This number is significantly higher than the approximat ely dozen satellites actually observed around our Galaxy. The difference is even larger if we consider the abundance of satellites in simulated galaxy groups similar to the Local Group. The models predict similar to 300 satel lites inside a 1.5 Mpc radius, while only similar to 40 satellites are obse rved in the Local Group. The observed and predicted VDFs cross at approxima te to 50 km s(-1), indicating that the predicted abundance of satellites wi th V-circ greater than or similar to 50 km s(-1) is in reasonably good agre ement with observations. We conclude, therefore, that unless a large fracti on of the Local Group satellites has been missed in observations, there is a dramatic discrepancy between observations and hierarchical models, regard less of the model parameters. We discuss several possible explanations for this discrepancy including identification of some satellites with the high- velocity clouds observed in the Local Group and the existence of dark satel lites that failed to accrete gas and form stars either because of the expul sion of gas in the supernovae-driven winds or because of gas heating by the intergalactic ionizing background.