, the vintage, the warm south. But in the
remote west, where the Celts held their savage own, no such traces
have ever been found.

Could this at last be one? The pavement, cleared with care, proved
of a disappointing size, measuring 8 feet by 4 at the widest. The
_tessellae_ were exceptionally beautiful and fresh in color; and each
separate design represented some scene in the story of Apollo. No
Bacchus with his panther-skin and Maenads, no Triton and Nymphs, no
loves of Mars and Venus, no Ganymede with the eagle, no Leda, no
Orpheus, no Danae, no Europa--but always and only Apollo! He was
guiding his car; he was singing among the Nine; he was drawing his
bow; he was flaying Marsyas; above all--the only repeated picture--he
was guiding the oxen of Admetus, goad in hand, with the glory yet
vivid about his hair. Could it (someone suggested) be the pavement
of a temple? And, if so, how came a temple of the sun-god upon this
unhomely coast?

The discovery gave rise to a small sensation and several ingenious
theories, one enthusiastic philologer going so far as to derive the
name Halzaphron from the Greek, interpreting it as "the salt of the
west winds" or "Zephyrs," and to assert roundly that the temple (he
assumed it to be a temple) dated far back beyond the Roman Invasion.
This contention, though perhaps no more foolish than a dozen others,
undoubtedly met with the most ridicule.

And yet in my wanderings along that coast I have come upon broken
echoes, whispers, fragments of a tale, which now and again, as I
tried to piece them together, wakened a suspicion that the derided
philologer, with his false derivation, was yet "hot," as children say
in the game of hide-and-seek.

For the stretch of sea overlooked by Halzaphron covers the lost land
of Lyonnesse. Take a boat upon a clear, calm day, and, drifting, peer
over the side through its shadow, and you will see the tops of tall
forests waving below you. Walk the shore at low water and you may fill
your pockets with beech-nuts, and sometimes--when a violent tide has
displaced the sand--stumble on the trunks of large trees. Geologists
dispute whether the Lyonnesse disappeared by sudden catastrophe
or gradual subsidence, but they agree in condemning the fables of
Florence and William of Worcester, that so late as November, 1099,
the sea broke in and covered the whole tract between Cornwall and the
Scillies, overwhelming on its way no less than a hundred and forty
chur