424 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 13, No. 19 
The normal density 3.3 corresponds to a pyroxenite or a peridotite. 
Throughout this whole region the temperature must be very high, 
and it is difficult to avoid the conclusion that this layer is all at a 
temperature above its melting point, its high rigidity being maintained 
by pressure.22 Both the density and the earthquake velocity will 
probably be somewhat smaller in such a glassy material than in a 
crystalline layer of the same composition, but the difference can 
hardly be great enough to nullify the evidence in favor of an ultrabasie 
layer. 
It has been suggested that meteorites should have the same average 
composition as that of the Earth or of any other part of the solar 
system. Now this average composition?’ (due account being taken 
of the proportion in which stony and metallic meteorites are seen to 
fall) corresponds to: olivine, 35; aluminous pyroxene, 42; anorthite, 4; 
troilite, 5; nickel-iron, 18. The silicate portion is principally an 
olivine-pyroxene mixture and thus is essentially a peridotite, and 
should have nearly the same density and compressibility as that 
postulated for the basic layer. 
Pallasite layer. A remarkable feature of the earthquake velocity 
curves (Fig. 1) is the small amount of change beyond a depth of 
1600 km. From compressibility measurements the velocity of the 
longitudinal waves in iron at moderate pressures is 6.0 km/sec, 
whereas the velocity in peridotite is 7.2. At high pressures the 
difference will probably be greater. This circumstance immediately 
suggests that the nearly constant velocity below 1600 km may be 
due to a gradually increasing amount of metallic iron mixed with the 
siliceous rock. The normal tendency for the velocity to increase 
with depth is thus offset by the admixture, in gradually increasing 
amount, of iron (or nickel-iron). 
The material in this region may be thought of as resembling certain 
meteorites consisting of a heterogeneous mixture of silicates and 
metallic iron which is called pallasite. The lower limit of this zone 
of incomplete segregation is thought to lie at about 3000 km depth 
where the velocity shows distinct evidence of falling off. 
Central metallic core. The remaining part of the Earth consists, 
beyond reasonable doubt, mainly of iron or nickel-iron with density 
2 Cf. R. A. Daly. Igneous Rocks and their Origin. (New York, 1914.) p. 172. 
28 Cf. O. C. Farrington. Field Columbian Museum, Geol. Ser., Vol. 3, Publ. No. 120: 
211-13. 1911. 
W. D. Harkins. Journ. Am. Chem. Soc. 39: 864. 1917. 
