Dolomitization in Hawaiian Soils — Sherman et al. 
39 
In Hawaii, Lyman and Dean (1938) have 
reported soils containing twice as much mag¬ 
nesium as calcium. They found these soils 
in low-lying areas which had been under salt 
water in the past. The authors have found 
other Hawaiian soils rich in magnesium in 
areas which had never been under sea water. 
These soils in general occur only in the dry 
areas of the Hawaiian Islands. 
The dark gray soils of Lualualei Valley, 
Waianae Valley, and Makaha Valley on the 
west slope of the island of Oahu have a low 
calcium-to-magnesium ratio, as shown in the 
present study. Under a very limited annual 
rainfall of 16 to 20 inches per year, these 
soils were developed on alluvial materials 
brought down from the Waianae Range. 
These soils are intermingled with numerous 
outcrops of coral rock (95 per cent calcium 
carbonate) and a few outcrops of basaltic 
rocks. The soils developed on these mate¬ 
rials (coral and basaltic rock) can be readily 
distinguished from the alluvial soils by color. 
The native vegetation in the area is that 
characteristic of semi-arid regions, the domi¬ 
nant vegetation being algarroba (mesquite). 
Because of the highly dispersed condition of 
the soil and the consequent retarded penetra¬ 
tion of water, the "A” and "B” horizons of 
the soil profile have been very poorly de¬ 
veloped. Crystals of gypsum were found in 
some of the subsoils. In soil above the layer 
containing gypsum, the carbonates gave a 
very weak effervescence when treated with 
cold dilute hydrochloric acid and a violent 
effervescence with hot acid. This test would 
indicate the presence of dolomite in this soil. 
The object of this study is to determine 
whether dolomitization is taking place, and 
if so, to what extent it is occurring in these 
soils and its possible relationship to soil¬ 
forming processes. 
EXPERIMENTAL METHODS 
Soil samples from several areas of soils 
having a high percentage of their exchange¬ 
able cations as magnesium were collected, 
on the basis of profiles which showed evi¬ 
dence of dolomitization. These samples were 
analyzed for water-soluble salts, for ex¬ 
changeable cations, and for the composition 
of the carbonates. 
1. Water-soluble salts were leached from the soil 
by shaking 25 grams of soil with a liter of dis¬ 
tilled water and allowing the mixture to stand 
overnight. It was then filtered on a Buchner 
. funnel and the soil was washed with another 
liter of distilled water or until no test was given 
for sulfates in the leachate. The filtrate was 
evaporated to a small volume and the organic 
matter was destroyed by acid oxidations with a 
few drops of hydrogen peroxide. Calcium was 
determined by the standard volumetric method, 
using potassium permanganate as the oxidizing 
agent. Magnesium and sulfate were determined 
gravimetrically as magnesium pyrophosphate 
and barium sulfate, respectively. 
2. Exchangeable cations were extracted by neutral 
normal ammonium acetate solution containing 
70 per cent ethyl alcohol to reduce the solubil¬ 
ity of the calcium carbonate. Calcium and mag¬ 
nesium were determined by the same methods 
as were used for the water-soluble ions. Potas¬ 
sium was determined volumetrically by the 
cobaltinitrite method (Volk and Truog, 1934), 
and sodium was determined gravimetrically as 
sodium zinc uranyl acetate (Barber, 1928). 
3. The carbonate analysis was made by a method 
described for the analyses of dolomitic carbo¬ 
nates (Sherman, 1937). This method involved 
the determination of carbonate carbon dioxide 
by decomposing the carbonates with normal 
hydrochloric acid solution, followed by absorp¬ 
tion of the carbon dioxide in a 0.5 N sodium 
hydroxide solution in an absorption tower. Be¬ 
fore entering the tower, the carbon dioxide was 
passed through a silver sulfate-sulfuric acid 
solution to remove any hydrochloric acid fumes 
from the digestion flask. The digestion flasks 
were heated to boiling in order to insure the 
complete decomposition of the dolomite. After 
the decomposition of the carbonates, the so¬ 
dium hydroxide solution in the absorption 
tower was drained into a volumetric flask. The 
tower was washed with five portions of hot 
carbon-dioxide-free distilled water to remove 
all the sodium hydroxide. The carbonates were 
precipitated in the volumetric flask by the addi¬ 
tion of barium chloride and the solution was 
then made up to volume. The precipitate was 
allowed to settle and an aliquot was taken from 
the clear portion of the liquid. The excess so¬ 
dium hydroxide was determined in the aliquot 
