64 
PHYSICS: C. BARUS 
Proc. N. a. S. 
the results for T and X in the lapse of time. To determine the latter, the 
needle was deflected from rest to the maximum elongation, by placing the 
lead balls M in either of the active positions; and with the needle so 
deflected for an instant, to quickly put the weights in the neutral position 
again; i.e., in the vertical plane through the needle. It is thus possible 
to determine both the period T of the needle and the logarithmic decre- 
ment : the former by timing successive passages (first and second) through 
the equilibrium position with a stopwatch, and the latter by reading the 
elongations. The zero point was nearly constant. The radiation is thus 
taken symmetrically from both sides. So far as the method goes, the 
values of T and X are quite accurately found in this way, the difficulty 
being in the interpretation. The plenum value of X log e, so far as deter- 
minable in a heated room was about 1, the needle being now practically 
aperiodic. This has decreased to about .27 (ratio of arcs 1.9) at 2.6 X 10 ~^ 
mm. of pressure and the curve suggests a more rapid decrease beyond. 
With the pressure values added the curve will afford data for the vis- 
cosity of highly rarified air. 
Periods. — Unlike X, the values for the period T (figure) show no marked 
improvement as the vacuum increases. In spite of the steady zero point, 
they rather vary with the time of day, being smallest in the morning and 
largest toward the end of the day and they drop off at night (A denoting 
forenoon, P afternoon, N night). The latitude of values is enormous, 
ranging from about 10 minutes to over 13 minutes. The former cases 
(low T) are usually encountered in the morning, and probably indicate 
repulsive radiant forces acting from the plates toward the middle plane 
of the case, like an elastic buffer, whereby the period is shortened. More- 
over the effect of the changes of X on 7 are nearly negligible in comparison 
with the enormous fluctuations of T resulting from modified environmental 
radiation. 
Whereas the deflections are largest in the morning and least at night, 
the T values pursue a nearly opposite course, increasing from morning to 
afternoon, except that the night periods again drop off. When the en- 
vironmental temperature is gradually increasing (morning) and when it is 
gradually decreasing (night), small values of T, the period of the needle, 
occur. This was also shown strikingly during a violent rain storm (chart 
11-15 hours), in which the atmospheric temperature dropped about 20° F. 
