274 
PACIFIC SCIENCE, Vol. VI, October, 1952 
Rood crocks ^ 
Bonk caving 
Stone wall dovjn H 
South ■ 60 
Fig. 3. Diagram showing the frequency distribution 
of three of the principal types of earthquake damage 
along the main highway. The arrows indicate the 
position on the highway of some villages and other 
features. Note the centering of damage close to the 
Kealakekua fault. 
were dressed back to any approximation to 
an equilibrium slope. The earthquake of 
August 21 greatly exceeded in size any previ- 
ous quake in the affected area since the road 
cuts were made, and shaking during the 
earthquake merely dislodged much of the 
loose material and allowed it to roll down 
onto the road. 
The distribution of abundance of rock 
slides in road cuts is shown in Figure 3, in 
which it is represented by the portion of the 
columns labeled "bank caving.” Like the 
other damage shown in the graph, it was 
greatest in the immediate vicinity of the 
Kealakekua fault, inland from and a little 
south of Kealakekua Bay. 
Many large slides took place on the fault 
scarp at the northern edge of Kealakekua Bay. 
The slides caused a disturbance of the water 
of the bay just after the earthquake, and many 
residents of the coastal village of Napoopoo 
fled inland, fearing a big tsunami. Slides 
continued on the Kealakekua cliff for several 
days after the earthquake, sending up clouds 
of yellowish-brown dust, leaving fresh scars 
on the cliff face, and building talus fans at 
the foot of the cliff. 
Less numerous and smaller slides also oc- 
curred along the cliff just inland from the 
village of Hookena Beach. The cliff at Hoo- 
kena is believed to be an ancient fault scarp, 
mantled by lava flows from the upper slopes 
of Mauna Loa during prehistoric times. Many 
fragments of the lava veneer were shaken 
down during the earthquake. 
Tsunami 
Despite early reports to the contrary, there 
is no doubt that the earthquake was accom- 
panied by a small tsunami, or "tidal wave.” 
At Napoopoo wharf the water was observed 
to withdraw from shore. The tide was low 
at the time. Withdrawal of the water lowered 
the level to about 4 feet below normal low- 
tide level. Immediately afterward the water 
returned shoreward, and the level rose about 
2 feet above low-tide level. 
At Milolii, Eugene Kaupiko reported that 
a few minutes after the earthquake, which 
he felt while in a canoe anchored offshore, 
the water receded from shore, revealing the 
sea bottom as far out as the edge of the 
wharf. This represents a lowering of the water 
level of about 3 feet. After the withdrawal the 
water returned shoreward, causing a rise of 
the water level 3 or 4 feet above normal low 
water and floating away a canoe that had 
been drawn up on the beach about 2.5 feet 
above high-tide level. One large fall and rise 
of the water level appears to have been fol- 
lowed by many small oscillations. 
At Honaunau, between Napoopoo and 
Milolii, Eli Cooper, caretaker of the City of 
Refuge, went down to the water’s edge a few 
minutes after the earthquake. At that time 
he could see no signs of disturbance of the 
water, but a small tsunami could have oc- 
curred between the time of the earthquake 
and his arrival at the strand. At Hookena no 
tsunami was observed, and there was none 
large enough to flood the floor of the dock, 
about 4 feet above normal water level. How- 
ever, it cannot be said definitely that no small 
tsunami occurred there. 
The Honolulu tide gauge record shows a 
distinct oscillatory disturbance of the water 
starting at approximately 01:35, about 38 
minutes after the earthquake. Seven or more 
oscillations are detectable, with an average 
