We measured waveform peak amplitudes for the various samples,
as discussed in the procedural section.
The ¶ and § amplitudes attenuated by a factor of about 2.5 from
the source-receiver delay time measurement, to the stainless-steel plug
measurement. This decay is probably due mostly to poor coupling, and some
relatively small attenuation due to intrinsic *Q*, since the plug was only
1.187 inches long and made of the same material as the source-receiver
transducer clamping apparatus.

We made similar measurements on the
wet and dry Massillon Sandstone rock samples. The ¶-wave amplitude attenuated
about a factor of 2.5 from dry to wet. In light of the stainless-steel
result, this implies that the attenuation due to coupling and *Q* are not
that different from that of stainless steel. However, the §-wave amplitude
attenuated a factor of 15 from dry to wet, indicating a significant
intrinsic *Q*-attenuation component. Intuitively, this might be explained
as shear stresses being more efficient at ``squirting'' fluid through the
available pore spaces, hence losing energy to friction heat-loss and the
work required to permanently displace the fluid to a non-equilibrium position.
This result could also be explained
as destructive interference from ¶-§ conversions and diffractions
attenuating the direct § arrival. Finally, these amplitude measurements
were made on one peak of the waveform. Attenuation would be more accurately
studied by examining the amplitude frequency response by Fourier Transform
spectral techniques over a more complete direct arrival waveform time series.

11/18/1997