lands. He had, however, subsequently reversed the act and had restored
the lands, whose tenure had been thus altered, to their original
condition of rent-paying estate or "socage."

The very term "the new feoffment," which was applied to the knights'
fees created between the death of Henry I and the year in which the
account preserved in the _Black Book_ of the exchequer was taken, proves
that the process was going on for nearly a hundred years, and that the
form in which the knights' fees appear when called on by Henry II for
"scutage" was most probably the result of a series of compositions by
which the great vassals relieved their lands from a general burden by
carving out particular estates, the holders of which performed the
services due from the whole; it was a matter of convenience and not of
tyrannical pressure. The statement of Ordericus Vitalis that the
Conqueror "distributed lands to his knights in such fashion that the
kingdom of England should have forever sixty thousand knights, and
furnish them at the king's command according to the occasion," must be
regarded as one of the many numerical exaggerations of the early
historians. The officers of the exchequer in the twelfth century were
quite unable to fix the number of existing knights' fees.

It cannot even be granted that a definite area of land was necessary to
constitute a knight's fee; for although at a later period and in local
computations we may find four or five hides adopted as a basis of
calculation, where the extent of the particular knight's fee is given
exactly, it affords no ground for such a conclusion. In the _Liber
Niger_ we find knights' fees of two hides and a half, of two hides, of
four, five, and six hides. Geoffrey Ridel states that his father held
one hundred and eighty-four _carucates_ and a _virgate_, for which the
service of fifteen knights was due, but that no knights' fees had been