Polarization of photons emitted in weak decays occuring at distant star allows to determine whether this star is made from antimatter. Even more promissing is the observation of neutrinos (antineutrinos) produced at neutronization (antineutronization) reactions at the beginning of SN ($\bar{SN}$) explosion.

Since the Standard Model most probably cannot explain the large value of CP asymmetries recently observed in D-meson decays we propose the fourth quark-lepton generation explanation of it. As a byproduct weakly mixed leptons of the fourth generation make it possible to save the baryon number of the Universe from erasure by sphalerons. An impact of the 4th generation on BBN is briefly discussed.

Even tiny lepton flavor violation (LFV) due to some New Physics is able to alter the conditions inside a collapsing supernova core and probably to facilitate the explosion. LFV emerges naturally in a See-Saw type II model of neutrino mass generation. Experimentally LFV is constrained by rare lepton decay searches. In particular, strong bounds are imposed on the mu->eee branching ratio and on the mu-e conversion probability in muonic gold. Currently the mu->e gamma decay is under investigation in the MEG experiment which aims at dramatic increase of sensitivity in the next three years. We search for a See-Saw type II LFV pattern which fits all the experimental constraints, provides Br(mu->e gamma) not less than Br(mu->eee) and ensures a rate of LFV processes in supernova high enough to modify the supernova physics. These requirements are sufficient to eliminate almost all freedom in the model. In particular, they lead to a prediction 0.5*10^(-12) < Br(mu->e gamma)< 6*10^(-12), which is testable by MEG in the nearest future. The considered scenario also constrains neutrino mass-mixing pattern and provides lower and upper bounds on tau-lepton LFV decays. We also briefly discuss a model with a single bilepton in which the mu->eee decay is absent at the tree level.

Electrons and electron neutrinos in the inner core of the core-collapse supernova are highly degenerate and therefore numerous during a few seconds of explosion. In contrast, leptons of other flavors are non-degenerate and therefore relatively scarce. This is due to lepton flavor conservation. If this conservation law is broken by some non-standard interactions, electron neutrinos are converted to muon and tau-neutrinos, and electrons -- to muons. This affects the supernova dynamics and the supernova neutrino signal in several ways. In particular, the total neutrino luminosity in the first second of the collapse is increased due to the larger free path of the non-electron neutrinos. This effect may have important consequences as the increase of the neutrino luminosity is known to facilitate the explosion. We consider an extension of the Standard Model by scalar bileptons which mediate lepton flavor violation. It is shown that in case of TeV-mass bileptons the electron fermi gas is equilibrated with non-electron species inside the inner supernova core at a time-scale of order of (1-100) ms. In particular, a scalar triplet which generates neutrino masses through the see-saw type II mechanism is considered. Non-observation of rare decays and data on neutrino mixing and neutrino masses restrict possible lepton flavor violation effects in this case. However a region in the parameter space of the model exists which fits all the experimental constraints and provides lepton flavor violation sufficient for observable effects in supernova.

The photon polarization operator in superstrong magnetic fields induces the dynamical photon "mass" which leads to screening of Coulomb potential at small distances $z\ll 1/m$, $m$ is the mass of an electron. We demonstrate that this behaviour is qualitatively different from the case of D=2 QED, where the same formula for a polarization operator leads to screening at large distances as well. Because of screening the ground state energy of the hydrogen atom at the magnetic fields $B \gg m^2/e^3$ has the finite value $E_0 = -me^4/2 \ln^2(1/e^6)$.

Electrons and electron neutrinos in the inner core of the core-collapse supernova are highly degenerate and therefore numerous during a few seconds of explosion. In contrast, leptons of other flavors are non-degenerate and therefore relatively scarce. This is due to lepton flavor conservation. If this conservation law is broken by some non-standard interactions, electron neutrinos are converted to muon and tau-neutrinos, and electrons - to muons. This affects the supernova dynamics and the supernova neutrino signal. We consider lepton flavor violating interactions mediated by scalar bileptons, i.e. heavy scalars with lepton number 2. It is shown that in case of TeV-mass bileptons the electron fermi gas is equilibrated with non-electron species inside the inner supernova core at a time-scale of order of (1-100) ms. In particular, a scalar triplet which generates neutrino masses through the see-saw type II mechanism is considered. It is found that supernova core is sensitive to yet unprobed values of masses and couplings of the triplet.

We define the critical coordinate velocity v_c. A particle moving radially in Schwarzschild background with this velocity, v_c= c/sqrt 3, is neither accelerated, nor decelerated if gravitational field is weak, r_g << r, where r_g is the gravitational radius, while r is the current one. We find that the numerical coincidence of v_c with velocity of sound in ultrarelativistic plasma, u_s, is accidental, since two velocities are different if the number of spatial dimensions is not equal to 3