316 
PACIFIC SCIENCE, VoL IX, July, 1955 
the geomorphic contrasts. In fact, the areas of 
greatest rainfall, 300 inches a year, are in the 
less deeply eroded northwestern part. 
After eruption of the K-2 series of lavas, 
there came another erosional episode, during 
which streams not only cut into this new 
ridge but also continued their work in the 
older part. Vigorous wave erosion may have 
commenced at this time or later, but the 
northwest end of the new ridge has been cut 
back by sea cliffs, which now survive as some- 
what subdued cliffs from a little west of 
Kahuku nearly to Haleiwa (Fig. 6d). Wave 
action has also truncated various spurs on the 
windward side, especially from Kualoa to 
Hauula (Fig. lc, d). 
Next in our story, the last Koolau eruptive 
episode or episodes occurred. For one of these 
we have the "Honolulu Volcanic Series" of 
Stearns (1935) which built various pyroclastic 
cones and poured out several intra-valley lava 
flows in the Honolulu area, or the south- 
eastern end of the Koolau Range. That the 
Honolulu Volcanic Series postdates the build- 
ing of the southeastern part of the range is 
very clear from the unconformable positions 
and from the topographic youth of the cones, 
and flows. 
The other late episode or sub-episode was 
the effusion of a large amount of very fluid 
lavas from the leeward side of the north- 
western (K-2) ridge of the Koolau Range to 
build the smooth Wahiawa Plateau with gen- 
tle slopes leading not only toward the Waia- 
nae Range but also both ways, toward Pearl 
Harbor and toward Haleiwa, from the low 
divide between Waikakalaua and Kaukona- 
hua streams. Small eruptions from transverse 
dikes or rifts built the various lateral lava 
shields. The lavas of the plateau and of the 
small shields constitute the K-3 series of the 
tentative map. There is no evidence in the 
structures by which to determine the relative 
age of the K-3 series and the Honolulu Vol- 
canic Series, as they do not overlap, or no 
overlap of one on the other is exposed, nor is 
any recorded in well logs. 
Perhaps the fact that pyroclastics are abun- 
dant in one and lacking in the other indicates 
that the two are of different age. Another 
difference is that the Honolulu Volcanic Se- 
ries affected both the windward and leeward 
sides of the range, whereas the K-3 series is 
restricted to the leeward side. 
An attempt to map the extent of the erup- 
tive products of the several episodes must of 
necessity be tentative, because subsequent 
weathering and erosion have masked the 
former boundaries. To such extent as bound- 
aries exist, they are those marked by topo- 
graphic differences. 
Separation of the K-3 and the K-2 lots of 
lava from one another will explain several 
topographic contrasts. The upper profile in 
Figure 10b is in the region of the K»2 Series, 
and has none of the original upland surviving 
as there is in the upper profile of Figure 10a, 
in the K-3 region. This, however, may be 
partly, or largely, a matter of differences in 
rainfall and in distance between stream courses. 
The profiles A — A'* and B — B / in Figure 11 
have a change of slope at the boundary be- 
tween K-2 and K-3. And in Figure 12, the 
wide separation of the lower contour lines 
marks the area of the fluid K-3 lavas. Finally, 
the smoothness of the Wahiawa Plateau as a 
whole and the rather sharp incision of the 
gulches indicate the youth of the K-3 area 
as compared to the K-2 area. 
The story concludes with the erosional epi- 
sode that is going on at present. It continues 
the earlier erosional events on the older areas 
but began on the Honolulu Volcanic Series 
and on the K-3 Series when their eruptive 
activity ended. This episode is much com- 
plicated by the several glacially controlled 
oscillations of sea level, which are omitted as 
this paper is concerned primarily with the 
volcanic constructional events. 
REFERENCES 
Dana, James Dwight. 1890. Characteristics 
of volcanoes . xvi + 399 pp. 5 16 pis., figs. 
Dodd, Mead & Co., New York. 
