In the above discussion we have shown that a generalized form of Wheeler-Feynman absorber theory can be made consistent with an open Big Bang model of the universe by imposing a plausible four-vector reflection boundary condition at the T=0 "origin" of the universe. This then provides the solutions for the Weak Absorption and Time-Arrow problems presented in the introduction and in AT1, for it provides for "open-ended" emission of weakly absorbed radiation (radio waves and neutrinos) and also gives an explanation for the observed dominance of retarded radiation. Effectively, in this model the T=0 boundary condition becomes "the absorber of last resort", confirming emitted waves which would otherwise have no corresponding absorption.
We note, however, that this boundary condition model only provides
solutions to these problems if the universe is not closed, i.e., the
density of the universe is less than or equal to the critical density so that
k=0 or -1 in the Friedmann model. If the density of the universe exceeds
the critical density so that k=+1 and the universe is bounded in both time
directions by 
=0
conditions at terminating singularities, then neither
advanced nor retarded radiation would be allowed, so that no emission
would be permitted.
Thus, the present work in essence connects the observed time asymmetry in radiation processes with the intrinsic time asymmetry in non-closed Friedmann models of the universe and is inconsistent with closed and cyclic models. This connection has been established by replacing the absorption boundary condition which previous authors have used to describe the early universe with a four-vector reflection boundary condition. This boundary condition model leads us to the insight that the Arrow of electromagnetic time points in the "future" time direction for the same reason that the light from a spotlight points in a particular spatial direction: both have a reflector "behind" them which reflects all rays going the wrong way.
Thus the electromagnetic arrow of time, according to the arguments presented above, has a direct connection to the cosmological arrow of time, i.e., the time direction in which the universe is observed to expand. This leads to the question of how these two "arrows" are connected to the other arrows of time, particularly the thermodynamic arrow of time, i.e. the time direction of entropy increase as specified by the Second Law, and the CP arrow of time, which is related to the CP violation in the decay of the K20 meson. Recent progress on grand unified theories has pointed to a connection between the CP violation and the observed dominance of matter over antimatter in the universe, but as yet no connection has been made between the CP arrow and the other time arrows. It would seem that this problem should be carefully reconsidered in the framework of advanced and retarded waves as they pertain to the decay of the K20 meson and other similar CP violating processes.
Several authors[, 9, 5, 11, 12, 14] have asserted that the thermodynamic arrow of time is a direct consequence of the electromagnetic arrow. On the other hand, Davies[22] has argued that a better connection can be made between the thermodynamic and cosmological arrows. We do not wish to enter deeply into this controversy in the present paper, but we feel that a case can be made for the former point of view. In particular, Boltzmann's famous H-Theorem, which "demonstrates" the thermodynamic arrow's direction seemingly from first principles, is based on the apparently harmless assumption that the motions of particles in a system of particles are uncorrelated before collision. This leads to the conclusion that the entropy of the system is constant or increasing. If, on the other hand, one had made the assumption that the particles' motions were uncorrelated after the collision, one would have been led to an entropy which was constant or decreasing. It can be argued that the correlation or lack thereof arises from the propagation of information by the medium of retarded electromagnetic fields, and that the dominance of the retarded solution of Maxwell's equations insures that there will be no correlations of electromagnetically interacting particles before the interaction has occurred. This would lead to a direct connection between the electromagnetic and thermodynamic arrows of time.