pressure from the atomic bombs differed from that of
ordinary high explosive bombs in three main ways:

A. Downward thrust. Because the explosions were well up in the air,
much of the damage resulted from a downward pressure. This pressure of
course most largely effected flat roofs. Some telegraph and other
poles immediately below the explosion remained upright while those at
greater distances from the center of damage, being more largely exposed
to a horizontal thrust from the blast pressure waves, were overturned
or tilted. Trees underneath the explosion remained upright but had
their branches broken downward.

B. Mass distortion of buildings. An ordinary bomb can damage only a
part of a large building, which may then collapse further under the
action of gravity. But the blast wave from an atomic bomb is so large
that it can engulf whole buildings, no matter how great their size,
pushing them over as though a giant hand had given them a shove.

C. Long duration of the positive pressure pulse and consequent small
effect of the negative pressure, or suction, phase. In any explosion,
the positive pressure exerted by the blast lasts for a definite period
of time (usually a small fraction of a second) and is then followed by
a somewhat longer period of negative pressure, or suction. The
negative pressure is always much weaker than the positive, but in
ordinary explosions the short duration of the positive pulse results in
many structures not having time to fail in that phase, while they are
able to fail under the more extended, though weaker, negative pressure.
But the duration of the positive pulse is approximately proportional to
the 1/3 power of the size of the explosive charge. Thus, if the
relation held true throughout the range in question, a 10-ton T.N.T.
explosion would have a positive pulse only about 1/14th as long as that
of a 20,000-ton explosion. Consequently, the atomic explosions had
positive pulses so much longer then those of ordinary explosives that
nearly all failures probably occurred during this phase, and very
little damage could be attributed to the suction which followed.

One other interesting feature was the combination of flash ignition and
comparative slow pressure wave. Some objects, such as thin, dry wooden
slats, were ignited by the radiated flash heat, and then their fires
were blown out some time later (depending on their distance from X) by
the pressure blast which follow