ory will give the rest. Sometimes a date can
be obtained by examining some public proclamation or some legal
document; in fact in the times of which we are speaking the difficulty
is easily surmounted.

The matter is by no means so simple, however, when we come to deal
with periods much earlier than this--with a scene from early Egypt,
Chaldaea, or China, or to go further back still, from Atlantis itself
or any of its numerous colonies. A date can still be obtained easily
enough from the mind of any educated man, but there is no longer any
means of relating it to our own system of dates, since the man will be
reckoning by eras of which we know nothing, or by the reigns of kings
whose history is lost in the night of time.

Our methods, nevertheless, are not yet exhausted. It must be
remembered that it is possible for the investigator to pass the
records before him at any speed that he may desire--at the rate of a
year in a second if he will, or even very much faster still. Now there
are one or two events in ancient history whose dates have already been
accurately fixed--as, for example, the sinking of Poseidonis in the
year 9564 B.C. It is therefore obvious that if from the general
appearance of the surroundings it seems probable that a picture seen
is within measurable distance of one of these events, it can be
related to that event by the simple process of running through the
record rapidly, and counting the years between the two as they pass.

Still, if those years ran into thousands, as they might sometimes do,
this plan would be insufferably tedious. In that case we are driven
back upon the astronomical method. In consequence of the movement
which is commonly called the precession of the equinoxes, though it
might more accurately be described as a kind of second rotation of
the earth, the angle between the equator and the ecliptic steadily but
very slowly varies. Thus, after long intervals of time we find the
pole of the earth no longer pointing towards the same spot in the
apparent sphere of the heavens, or in other words, our pole-star is
not, as at present, [Greek: a] Ursae Minoris, but some other celestial
body; and from this position of the pole of the earth, which can
easily be ascertained by careful observation of the night-sky of the
picture under consideration, an approximate date can be calculated
without difficulty.

In estimating the date of occurrences which took place millions of
years ago in earl