system to say that the derivation lines show the course of
an absolute feudal tenure, and they are not meant for that
purpose, but simply to indicate the succession of the
inequalities of rank.

Turning now to the territorial divisions of the country at this
period, we find them practically unchanged. The _civitas_ still stands
as the sectional unit; the territory with its city still represents
the administrative division of the state. It is fundamental to a
correct understanding of the early development of communal
institutions that we should have a thorough knowledge of the meaning
of this term _civitas_; of the extent of its application and of its
limitations. I used the words "territory with its city" in defining
the administrative division of the state, and perhaps this term
describes the _civitas_ better than any single word would do. In the
Roman municipal system we have the city with its surrounding
territory, over which extends the jurisdiction of the _curia_; in the
Lombard system we have the territory, the land, in some part of which
is located a city, a fortified place.

This is to my mind the important point which settles satisfactorily
the vexed question of the dominance or the disappearance of Roman
influences. The institutions of the Lombards were similar in character
to those of the other Germanic races, and the continuance of any
overruling municipal influence among them would have done violence
alike to their traditions and to the nature of their race. The old
municipal predominance as a system disappeared, the old municipal
divisions and many of the minor forms and offices as a fact remained.
It is these latter which give some color to the arguments of writers
like Savigny,[12] who endeavor to maintain the continuance of the old