It has been previously reported that a general electric field solution
and its initial condition, E(g)(z, t) and E(g)(z, 0), respectively, a
re not causal when formed by a superposition of time-harmonic waves in
an attenuating medium. However, this is not the case. Further, the re
lationship between attenuation and phase velocity as well as their dep
endence on frequency arise simply from the form chosen for the time ha
rmonic particular solutions. Even though causality is not introduced d
uring the solution to the wave equation, the general solution can subs
equently be shown to be a time convolution of a causal boundary condit
ion (time history of the electric field as it crosses the z = 0 plane,
E(g)(0, t)), and the medium's impulse response g(z, t), which can be
shown to be causal. Hence, the general solution is also causal. The in
itial condition occurs at the instant, t = 0, when the electric field
arrives at the z = 0 plane, and it has been previously reported that t
he initial condition depends on the boundary condition for times after
the initial time thereby violating causality. A re-examination shows
that the initial condition does not depend on times after the initial
time. Hence, the initial condition obeys causality, and it can also be
shown to be properly determined (E(g)(z, 0) = 0 for z > 0) even when
the boundary condition is not zero. It has also been reported that lim
iting expressions for the boundary and initial conditions, E(g)(0, t -
-> 0) and E(g)(z --> 0,0), respectively, are not equal. However, a re-
examination reveals that E(g)(0, t --> 0) = E(g)(z --> 0, 0).