3^5 
MAGNETIC OXIDE OF IRON IN HAWAIIAN SOILS. 
W. P. Kelley, Chemist Hawaii Agricultural Experiment Station. 
For many years it has been a matter of common knowledge 
ai::ong surveyors on the Hawaiian Islands that readings of the 
mrgnetic compass are often very misleading. In the Coast and 
Geodetic Survey of the islands abnormal deflections of the nag- 
ne tic needle occasion much comment by those in charge, and nu- 
merous discrepancies concerning boundary lines in the interior 
hc.ve been traced in a large measure to inaccuracies of the com- 
pass. By reference to old land maps it is apparent that conflicting 
land claims exist ; and so well established is the fact of magnetic 
deflection here that surveyors no longer rely on the compass in 
survey work. It is further stated that such magnetic disturbance 
is by no means confined to land, as pilots on board ships fre- 
quently observe peculiar deviations of the compass when sailing 
close to these shores. 
The cause of this phenomenon has generally been attributed to 
the presence of metallic iron in the enormous rock beds and lava 
flows of the islands. This view would seem to have more or less 
basis since Hawaiian lavas in general contain much iron, and since 
magnetic deflection is usually greatest in the vicinity of large 
masses of lava or near hill and mountain sides. In passing around 
certain masses of lava on Hawaii, for instance, the magnetic 
needle is said to be deflected through an arc of 360 degrees. 
In the course of some investigations at the U. S. Experiment 
Station it was found that the large deposits of black sand on 
the reservation possess magnetic property to a marked degree and 
subsequently it has been shown that the soil not only on the 
station grounds but throughout the islands is magnetic A small 
horseshoe magnet almost universally attracts particles from both 
surface soil and sub-soil. Experiments have been made to deter- 
mine what substance in the soil is thus attracted by the magnet 
and as an outcome of this investigation it has been found to be one 
of the oxides of iron, namely: Fe 3 0 4 , the so-called magnetic 
oxide of iron or magnetite, and not metallic iron in the uncom- 
bined state. This compound of iron has long been known to be 
highly magnetic and is a constituent of most volcanic rocks. It 
is reported by Dr. Hilgard as being generally distributed in the 
black sands and alluvial soils of the Pacific Coast, also as occurring 
in Hawaiian lavas, though no mention is made of its occurrence 
in the soils of Hawaii. Magnetite occurs scattered throughout the 
Rocky Mountains and is found in enormous deposits in Northern 
Pennsylvania, New Jersey and New York where it is mined as 
one of the richest iron ores in America. 
From preliminary investigations magnetite appears to be uni- 
formly distributed throughout a large part of the rocks of Hawaii, 
