evolution. The old form of society founded on sexual relations
is abolished in the clash with the recently developed social classes. A
new society steps into being, crystallized into the state. The units of
the latter are no longer sexual, but local groups; a society in which
family relations are entirely subordinated to property relations,
thereby freely developing those class antagonisms and class struggles
that make up the contents of all written history up to the present time.

Morgan deserves great credit for rediscovering and re-establishing in
its main outlines this foundation of our written history, and of finding
in the sexual organizations of the North American Indians the key that
opens all the unfathomable riddles of most ancient Greek, Roman and
German history. His book is not the work of a short day. For more than
forty years he grappled with the subject, until he mastered it fully.
Therefore his work is one of the few epochal publications of our time.

In the following demonstrations, the reader will, on the whole, easily
distinguish what originated with Morgan and what was added by myself. In
the historical sections on Greece and Rome, I have not limited myself to
Morgan's material, but have added as much as I could supply. The
sections on Celts and Germans essentially belong to me. Morgan had only
sources of minor quality at his disposal, and for German
conditions--aside from Tacitus--only the worthless, unbridled
falsifications of Freeman. The economic deductions, sufficient for
Morgan's purpose, but wholly inadequate for mine, were treated anew by
myself. And lastly I am, of course, responsible for all final
conclusions, unless Morgan is expressly quoted.

FREDERICK ENGELS.

AUTHOR'S PREFACE TO THE FOURTH EDITION, 1891.

The first large editions of this work have been out of print for nearly
six months, and the publisher has for some time requested of me the
arrangement of a new edition. Urgent duties have hitherto prevented me.
Seven years have passed, since the first edition made its appearance;
during this time, the study of primeval forms of the family has made
considerable progress. Hence it became necessary to apply diligently the
improving and supplementing hand, more especially, as the proposed
stereotyping of the present text will make further changes impossible
for some time.

Consequently, I have subjected the whole text to a thorough revision and
made a number of additions wh