Charles Lyell, Historian of Time's Cycle - Time's Arrow, Time's Cycle: Myth and Metaphor in the Discovery of Geological Time - page 125

If, at certain periods of the past, rocks and peculiar mineral composition had
been precipitated simultaneously upon the floor of an "universal ocean," so
as to invest the whole earth in a succession of concentric coats, the
determination of relative dates in geology might have been a matter of the
greatest simplicity. To explain, indeed, the phenomenon would have been
difficult, or rather impossible, as such appearances would have implied a
former state of the globe, without any analogy to that now prevailing. (III, 37
—38)
In the most striking of all Lyellian statements, he sketches a series of tactics
to preserve uniformity of state in the face of almost any conceivable evidence
for directional change in the earth's history:
When we are unable to explain the monuments of past changes, it is always
more probable that the difficulty arises from our ignorance of all the existing
agents, or all their possible effects in an indefinite lapse of time, than that
some cause was formerly in operation which has ceased to act; and if in any
part of the globe the energy of a cause appears to have decreased, it is always
probable, that the diminution of intensity in its action is merely local, and that
its force is unimpaired, when the whole globe is considered. But should we
ever establish by unequivocal proofs, that certain agents have, at particular
periods of past time, been more potent instruments of change over the entire
surface of the earth than they now are, it will be more consistent with
philosophical caution to presume, that after an interval of quiescence they
will recover their pristine vigor, than to regard them as worn out. (I, 164-165)
I have quoted this passage for years, but it never ceases to astound me. It
begins with an acceptable statement of the second uniformity, then slides into
substantive claims about uniformity of state. Finally, Lyell actually holds that
we should reject directional change even if we have "unequivocal proofs . . .
over the entire surface of the