Dehydration in Laterite Formation — SHERMAN et d. 
439 
precipitation of the iron oxides in the zone 
above the fluctuating water table. This layer 
is made up of concretions of iron oxide which 
will harden on exposure to give the ironstone 
laterite, a material suitable for building mate- 
rial. Finally, there are soil scientists who hold 
that laterite has developed by residual accu- 
mulation as the result of leaching and removal 
of silica. Marbut (1932) and Fox (1933) were 
the leading advocates of this hypothesis of 
the mode of formation of laterite. 
In spite of the numerous hypotheses as to 
the origin of laterite, there are certain facts 
which are included in all proposals. One of 
these facts is that in the early stages the 
weathering surface is losing silica and bases. 
Desilication is taking place with the progres- 
sive development of soils having a sequence 
of minerals with a lower silica content. Sec- 
ondly, as desilication proceeds the sesqui- 
oxides accumulate in the soil. These weather- 
ing processes are accelerated under high 
rainfall and high temperatures which are com- 
mon to many tropical regions. Lastly, all 
ferruginous laterites harden on exposure. The 
hardened laterite is made up of lenses or con- 
cretions of iron oxides often cemented to- 
gether by a kaolinitic clay. 
A number of titaniferous ferruginous lat- 
erite crusts have been described in the Ha- 
waiian Islands by Sherman (1952). Each 
profile of these soils has a massive indurate 
surface horizon having both a high bulk den- 
sity and a high particle density. This hardened 
surface horizon is underlaid by a friable sub- 
soil. This subsoil horizon possesses the bulk 
density of an average soil. Likewise, the par- 
ticle density is lower than that of the surface 
horizon. Below the friable subsoil horizon is 
an impervious layer which may be the un- 
weathered country rock, a plastic clay, or an 
unconformity-like relationship which may 
have developed by pedogenic processes rather 
than being a geological unconformity. 
These laterite crusts were either barren or 
were covered by scattered bunches of grass 
or vines. However, some of the areas which 
exhibited the crust-like surface horizon had 
a fair amount of vegetation. These areas, also, 
exhibited a high bulk density and high par- 
ticle density relationship such as is found in 
the barren crust areas. From many observa- 
tions made during wet and dry weather, it 
was noted that the bulk density of the soil 
varied according to the soil moisture condi- 
tions. In a visit to a barren crust area on 
Kauai after a heavy rain, a marked difference 
was observed between the bulk density and 
hardness of the surface soil of the barren area 
and the adjacent identical soil under a vege- 
tative cover. It has been observed that the 
crust will develop on irrigation ditch banks 
where wetting and drying have occurred at 
very frequent intervals. The appearance of the 
hardened crust can be associated with dry 
periods, suggesting that its formation is due 
to dehydration. Thus, it is possible that vege- 
tation will protect the hydrated minerals from 
dehydration and prevent the crust formation. 
A study has been conducted to determine 
the role of dehydration of the hydrated ses- 
quioxides in the development of the laterite 
crust. The study also included the determina- 
tion of the effect of dehydration on the physi- 
cal and chemical properties of the soil. For 
this study two sets of soil samples from ad- 
jacent profiles were selected. The soil profiles 
were located on the slopes of the western rim 
of Waimea Canyon on the island of Kauai. 
A titaniferous ferruginous laterite crust covers 
a rather level bench on this slope. Profiles 
were selected at different areas of this laterite 
crust. The two main profiles used in this study 
were taken at the edge of the barren crust area 
so that the profile without a vegetative cover 
was taken only 3 feet from the site of the 
profile with a vegetative cover. The morphol- 
ogy of the two profiles was identical except 
for the obvious difference in the physical con- 
dition of the surface horizon. The characteris- 
tics of the two profiles are shown in Figure 1. 
METHODS OF ANALYSIS 
Soil samples were collected from the pro- 
