14 W. T. BORN AND J. E. OWEN 
lying in air at room temperature for several months. The observed 
natural frequency was go1 cycles per second and the weight 6,450 
grams. The rod was then placed in an oven at 95°C. for a period of 20 
hours, and after it was allowed to cool to room temperature, the weight 
was found to be 6,440.7 grams and the natural frequency 931 cycles per 
second (Point 2). The weight of the bar after this baking process was 
taken to be the “dry” weight, that is, the assumption was made that 
only a negligible amount of water remained after this baking period. 
The rod was then moistened and placed in a closed tube at room 
temperature for a period of 36 hours in order to allow the moisture 
to distribute itself throughout the bar. The weight and natural 
on 
900 
ee ee DE BARE ee een ee ee 

6400 6500 6600 6700 66 6900 
WEIGHT IN GMS. 
Fic. 4.—Variation in natural frequency of Amherst sandstone bar with 
weight as water content is changed. 
frequency were then determined again, the former being 6,831.3 
grams and the latter 538 cycles per second. These values are those 
corresponding with Point 3 on the curve. The rod was next allowed 
to dry in air for several hours and then placed in a closed tube for a 
day to reéstablish moisture equilibrium. Point 4 indicates the weight 
and natural frequency after this procedure. Subsequent determina- 
tions, corresponding to points 5, 6, 7 and 8, were made in similar 
fashion. To obtain Point g, the rod was again baked for about 5 hours 
and the weight and natural frequency determined. 
Several features of this curve are of interest. Perhaps the most 
striking is the extremely rapid reduction of velocity which accom- 
panies the addition of the first increments of moisture and the slower 
reduction as a greater amount of water is added. A second interesting 
point is the fact that the curve indicates that the addition and re- 
moval of the water caused little, if any, permanent change in the 
674 
