Vol.. 8, 1922 
PHYSICS: C. BARUS 
63 
The values of the radiation potentials for strontium and barium have 
not been directly determined, but Mohler, Foote and Megers {Sci. Paper, 
Bureau of Standards, No. 403, 1920) give estimates from the wave-lengths 
of the tail lines of the proper spectral series for these elements. Hydrogen 
is added to the table, although not given by Eve. An inspection of the 
table brings out the following: 
For Group I, the degree of constancy of the products is nearly the same, 
being slightly in favor of the use of I rather than (I — R) . Group II A, 
shows, on the other hand, that (I — R) gives a better agreement than I. 
This is still more marked in Group II B. Group V A is considerably 
more favorable to (I — R). In case of the inert gases the product using 
I alone does not hold at all, while the product (I — R) gives nearly a 
constant. The product using (I — R) places hydrogen in good agreement 
with the first Group. Mercury, which has some of the physical qualities 
of the inert gases, has a product about equal to that for those gases. 
In case of the whole table, going from Group to Group, the products using 
(I — R) are much more constant than is the case with I alone. 
PERIODS AND LOGARITHMIC DECREMENT OF THE GRAVITA- 
TION NEEDLE UNDER HIGH EXHAUSTION'' 
By C. Barus 
Department of Physics, Brown University 
Communicated, February 4, 1922 
The Deflections. — After a search for a finer quartz fibre than was used 
in my last paper, one was found giving a double deflection of l^y = 13.42 
=±= .03 cm. as compared with the former 2.67 cm., for the same case and 
needle. In spite of this astonishing sensitivity and otherwise admirable 
behavior of the apparatus, the new result for Ay from 30 successive night 
observations came out relatively less accurate than the former. At the 
same time, the observations on any single night (details must be omitted 
here) rarely differed by more than .1%. It is a case therefore in which the 
increasing importance of the radiation forces renders further finessing with 
the fibre of doubtful use in the given environment. 
Logarithmic Decrements X log e. — In the endeavor to cope with this 
formidable difficulty, I began a study of the vibration of the needle in 
high vacua of a few thousandths of a millimeter ; but the summer vanished 
before I completed it. The results so far obtained are interesting, however, 
and are given in the attached chart. The case was exhausted in the be- 
ginning to about 2.6 X 10 ~^ mm. on the McLeod gauge, and then sealed off. 
Air, however, in the lapse of 1 1 days very slowly leaked in through the glass 
cock and rubber tubes; and as the installment prevented me from finding 
the successive vacua (always beyond the U-gauge limit) I have expressed 
