Compressional Wave Velocities in Basic Rocks 
Nikolas I. Christensen 1 
ABSTRACT: Compressional wave velocities determined by measurement of travel 
times of pulses at pressures to 10 kilobars are given for specimens of basalt. Varia- 
tions of velocity with propagation direction are related to feldspar orientation and 
inhomogeneity in alteration of the specimens. Velocity differences reported for 
diabase, gabbro, eclogite, and basalt can be explained in terms of variation of 
density and mean atomic weight. The basalts have the lowest compressional wave 
velocities of basic rocks. The low velocities are a consequence of slight alteration, 
high mean atomic weight, and relatively low density. 
Ultrasonic measurements of the elastic 
properties of rocks are required for the interpre- 
tation of seismic velocities. Comparisons of seis- 
mic velocities with laboratory-measured veloc- 
ities provide the simplest and most direct evi- 
dence concerning the constitution of the earth’s 
interior. It is surprising that with the abundant 
velocity data now available very little attention 
has been given to basaltic rocks at high pres- 
sures. In this paper compressional wave veloc- 
ities are reported at pressures to 10 kb for three 
specimens of basalt. In addition to presenting 
new velocity data which may be important for 
oceanic crustal areas, this note is part of a 
continuing effort to understand the factors which 
influence the elastic properties of rocks. 
MEASUREMENT TECHNIQUE 
The technique for measuring the velocities 
was similar to that described by Birch (I960) 
and Christensen (1965) ; therefore it is described 
only briefly here. The specimens were cylindri- 
cal cores, % inch in diameter and 2 inches 
in length, jacketed with a thin copper tube. 
Barium titanate transducers with natural fre- 
quencies of 1 Mc/sec were placed on the ends 
of the specimens and then backed by aluminum 
electrodes. Rubber tubing was used to seal the 
pressure fluid from the spaces between the 
sample, electrodes, and transducers. Rectangular 
electrical pulses of about 50 volts were applied 
to one transducer. The resulting mechanical 
1 Department of Geology, University of Wash- 
ington, Seattle, Washington. Manuscript received 
December 12, 1966. 
pulse in the sample was received by an identical 
transducer and converted to an electrical signal 
which was amplified and displayed on a dual- 
trace oscilloscope. Transit times were measured 
by comparing the signal from the rock specimen 
with that through a variable mercury delay line 
displayed simultaneously on the oscilloscope. 
Kerosene was used as the pressure fluid. 
Pressure was generated by the advance of a 
piston driven by a 6-inch ram into a cylinder 
with an outside diameter of 6 inches and an 
inside diameter of 1.5 inches. Pressure was 
measured by determining the change in resis- 
tance of a calibrated manganin wire gage. All 
measurements were made at temperatures be- 
tween 20° and 30°C 
DATA 
Compressional wave velocities and densities 
are given in Table 1. Velocities are recorded 
for each specimen from three cores cut in 
mutually perpendicular directions. The veloci- 
ties are considered accurate to 1%. 
The basalts were collected from the Triassic 
Hampden basalt near Hartford, Connecticut. 
Petrologically they are fine-grained tholeiitic 
basalts. Average grain size is about 0.2 mm. 
Modal analyses are given in Table 2. 
DISCUSSION 
Anisotropy in the basalts is related to varia- 
tions in composition of the three cores from 
each sample and a subparallel orientation of 
plagioclase laths. Directions of low velocity in 
41 
