RIGIDITY OF SEDENIUM CRYSTAES 
111 
termination of five constants. This is a big reduction from the 
twenty-one constants required for the triclinic form of crystal. 
Voigt ^ has shown how one can determine these five constants, but 
to do so requires that one have sections of the crystal cut transverse- 
ly to the principal axis, and at 45° to the axis, as well as the sec- 
tions possessed in the present work. Voigt actually determined 
these five constants for quartz and for beryl. Such sections, with 
the crystals discussed here, are out of the question, and so one 
must content himself with the gross constants, such as those which 
we have here, reported upon. 
A few determinations of the logarithmic decrement led to re- 
markable constancy for results. For example, in crystals Nos. 
3 and 12, which were quite different in size, the logarithmic dec- 
rement proved to be respectively 1.08 and 1.07. In these tests 
errors of measurement of the sides, or errors of assuming the 
sides equal do not enter, and these results go towards bearing out 
our belief that if we could obtain regular crystals, the elastic 
constants, experimentally determined, would show the same uni- 
formity. 
State University oe Iowa. 
4Voigtj Ann. d. Phys., 16, 398, 1882; 29, 604, 1886; 31, 474, 1887. 
