Geomorphic Contrasts in Koolaus — Palmer 
311 
Fig. 11. Lengthwise profiles of three ridge crests. 
Vertical exaggeration about 5.2 times. See Figure 12 
for the location of the profiles. 
The profile along St. Louis Heights is what 
one would expect to find on a long eroded 
lava shield built as a single unit and with 
greater rainfall and greater erosion at the 
higher levels. Projection, or extrapolation, of 
the slope of the lower part of the profile, as 
shown by the broken line, runs distinctly 
above the much more eroded, much rainier, 
upper part of the profile. 
In contrast, the profiles for the middle and 
northwestern parts of the range, one upslope 
from Wahiawa and one along the ridge on 
the south side of Helemano Gulch (Fig. 6d) 
are very different. Extrapolation of the slope 
of the lower part of each profile, instead of 
going above the rainier, more intricately dis- 
sected summit region, goes well below the 
Crestline. This discrepancy leads to several 
hypotheses, which involve different construc- 
tional histories. The profiles are in a sense 
concave upward, at least for part of their ex- 
tent. One thinks of lava shields as being con- 
vex upward in general, and it is the concavity 
that needs explanation. 
Dr. Gordon A. Macdonald has pointed out 
to me in a letter that concave profiles are 
found on Hualalai, which might be due to 
maximum precipitation and erosion in the 
middle altitudes of that 8,231-foot mountain. 
This seems unlikely for the degree of dissec- 
tion is rather uniform in all parts of Hualalai. 
He also points out concavities on the lower 
slopes of Mauna Loa, where lavas ponded 
against Mauna Kea, and on the lower slopes 
of Mauna Kea with similar ponding against 
the Kohala Mountains at the Waimea Saddle. 
In addition to these examples from the island 
of Hawaii, he wrote of similar concavity of 
the lower slopes of East Maui at the Isthmus 
and of East Molokai against West Molokai. 
An east-west profile of Mauna Kea shows 
concavity upward, but this may be due to the 
more viscous, later lavas coming to rest with 
steeper slopes than the older, more fluid lavas. 
Ponding of lavas against the Waianae 
Mountains certainly has been a factor in 
limiting the extent of the gentle slopes of the 
lower parts of these profiles, as well as of the 
Wahiawa Plateau in general. However, there 
seems to be a fairly definite or sharp break in 
the slopes of these two profiles at about 1,200 
feet altitude. Farther northwest, near Waimea 
Bay (Fig. 13), there is a fault with upthrow 
on the north side, which makes distinct 
breaks in profiles drawn at right angles across 
the ridge crests (Palmer, 1947, fig. 6). 
The hypothesis is offered that the upper 
parts of these profiles (A — A' and B — B') 
represent an older long, narrow constructional 
land form, which was considerably eroded, 
and that subsequently younger volcanic ac- 
tivity discharged very fluid lavas that built 
the gentler slopes of the lower parts of the 
profiles. The upper parts are in regions of 
rather heavy rainfall and intricate dissection 
by streams, whereas the lower, drier parts are 
as yet but little dissected. These are the groups 
of lavas designated as K-2 and K-3 on the 
tentative map, Figure 19- 
No petrographic evidence is known to me 
to indicate any marked difference in viscosity 
of lavas in these areas. 
LATE VOLCANIC ACTIVITY 
Three main volcanic episodes have long 
been recognized for Oahu. The first, which 
Stearns (1935) divided into three subepisodes, 
built the lava shield from which the Waianae 
Range has been carved. The second was 
