The Milagro experiment has announced the discovery of an excess flux of TeV cosmic rays from the general direction of the heliotail, also close to the Galactic anticenter. We investigate the hypothesis that the excess cosmic rays were produced in the
SN explosion that gave birth to the Geminga pulsar. The assumptions underlying our proposed scenario are that the Geminga supernova occurred about 3.4 10^5 years ago (as indicated by the spin down timescale), that a burst of cosmic rays was injected with total energy 10^49 erg (i.e., about 1% of a typical SN output), and that the Geminga pulsar was born with a positive radial velocity of 100--200 km s^-1. We find that our hypothesis is consistent with the available information. In a first variant (likely oversimplified), the cosmic rays have diffused according to the Bohm prescription (i.e., with a diffusion coefficient on the order of c times r_L, with c the speed of light and r_L the Larmor radius). An alternative scheme assumes that diffusion only occurred initially, and the final propagation to the Sun was a free streaming in a diverging magnetic field. If the observed cosmic ray excess does indeed arise from the Geminga SN explosion, the long--sought smoking gun connecting cosmic rays with supernovae would finally be at hand. It could be said that, while looking for the smoking gun, we were hit by the bullets themselves.
We investigate the X-ray properties of the most luminous radio sources in the 3CR catalogue, in order to assess if they are similar to the most luminous radio quiet quasars, for instance in the X-ray normalization with respect to the optical luminosi
ty, or in the distribution of the absorption column density. We have selected the (optically identified) 3CR radio sources whose 178-MHz monochromatic luminosity lies in the highest factor-of-three bin. The 4 most luminous objects had already been observed in X rays. Of the remaining 16, we observed with XMM-Newton 4 randomly chosen, optical type 1s, and 4 type 2s. All targets have been detected. The optical-to-Xray spectral index, alphaox, can be computed only for the type 1s and, in agreement with previous studies, is found to be flatter than in radio quiet quasars of similar luminosity. However, the Compton thin type 2s have an absorption corrected X-ray luminosity systematically lower than the type 1s, by a factor which makes them consistent with the radio quiet alphaox. Within the limited statistics, the Compton thick objects seem to have a reflected component more luminous than the Compton thin ones. The extra X-ray component observed in type 1 radio loud quasars is beamed for intrinsic causes, and is not collimated by the absorbing torus as is the case for the (intrinsically isotropic) disk emission. The extra component can be associated with a relativistic outflow, provided that the flow opening angle and the Doppler beaming factor are 1/5 - 1/7 radians.
