e site now occupied by Florence from very remote
times; but the form of the "ancient circle" carries us back to a Roman
camp and a military colony as the origin of the regular city. Beyond
this meagre basis the whole story of "Troy, and of Fiesole and Rome,"
in connection with Florence must be pronounced a myth. The notices of
Florence before the opening of the twelfth century are few and meagre,
but they suffice to prove that the story of its destruction by Totila,
and rebuilding by Charlemagne, is without foundation; and of all the
reported conquests of Fiesole that of 1125 is the first that we can
regard as historical.

The history of Florence is almost a blank until about 1115 A.D., the
date of the death of the Countess Matilda.

With respect to the second point, it is impossible to give so brief or
conclusive an answer. Villani is as valuable to the historian as he is
delightful to the general reader. He is a keen observer, and has a
quick eye for the salient and essential features of what he observes.
When dealing with his own times, and with events immediately connected
with Florence, he is a trustworthy witness, but minute accuracy is
never his strong point; and in dealing with distant times and places
he is hopelessly unreliable.

The English reader will readily detect his confusions in Book VII., Sec.
39, where at one time Richard of Cornwall, and at another Henry III.,
is called king of England; and Henry of Cornwall and Edward I. are
regarded indifferently as sons of Richard or sons of Henry III., but
are always said to be brothers instead of cousins.

Here there is little danger of the reader being misled, but it is
otherwise in such a case as that of Robert Guiscard and the house of
Tancred in Book IV., Sec. 19. By way of putting the reader on his guard,
we will go into this exceptionally bad, but by no means solitary,
instance of Villani's inaccuracies.

Tancred, of the castle of Hauteville (near Coutances, in Normandy),
had twelve sons, ten of whom sought their fortunes in southern Italy
and Sicily. Four of these were successively Counts of Apulia, the last
of the four being Robert Guiscard. He was followed by his son Roger,
and his grandson William, who died childless. Another of the sons of
Tancred was Roger, who became Count of Sicily. He was succeeded by his
son Roger II., who possessed himself of the Apulian domains of his
relative William, on the decease of the latter. Roger now had himself
p