esistant to failure, despite potentially severe local
damage. These systems would suffer serious local outages, particularly
in the first several days after the event, but would resume service
over a few weeks to months. The principal difficulty would be the
greatly increased need for these systems in the first few days after
the event, when lifesaving activities would be paramount.

Portions of the San Francisco Bay Area and of the Los Angeles Area
contain substantial concentrations of manufacturing capacity for
guided missiles and space vehicles, semiconductors, aircraft parts,
electronic computing equipment, and airframes. Their specific
vulnerability to the postulated earthquakes was not analyzed. In the
event of major damage, however, the long-term impacts may be mitigated
somewhat by such measures as the use of underutilized capacity located
elsewhere, substitution of capacity from other industries, imports,
use of other products, and drawing-down of inventories.

Since we have not recently experienced a catastrophic earthquake in
the United States, there are many unknowns which must be estimated
with best judgment. This is true particularly for the response of
individuals as well as governmental and other institutions. Popular
assumptions of post-disaster behavior include antisocial behavior and
the need for martial law, the breakdown of government institutions,
and the requirement for the quick assertion of outside leadership and
control. Practical experience and field studies of disasters, however,
indicate that these assumptions are not necessarily correct. On the
contrary, the impacts of the disaster commonly produce a sense of
solidarity and cooperativeness among the survivors. Nonetheless, the
perception remains among emergency response officials that there will
be an increased need for law enforcement following the event.

Another major unknown involves whether a medium or short-term warning
of the event would be possible and how such a warning could be
utilized most effectively. The technology for earthquake prediction is
in an early stage of development and, therefore it is problematical
that researchers will succeed in issuing a short-term warning before a
catastrophic earthquake, should the event occur in the next few
years. Yet as research progresses, scientifically-based,
intermediate-term warnings are possible, but subject to a high degree
of uncertainty. Consequently, response preparations must b