ey 
LXII. 
A MECHANICAL CAUSE OF HOMOGENEITY OF STRUCTURE 
AND SYMMETRY GHOMETRICALLY INVESTIGATED ; 
WITH SPECIAL APPLICATION TO CRYSTALS AND TO 
CHEMICAL COMBINATION. By WILLIAM BARLOW. 
[Read Junz 16; Received for Publication Junz 18; Published Decumzzr 20, 1897.] 
From the early days of crystallographic study, attempts have 
from time to time been made to find some clue to the nature of 
the ultimate structure of crystals by means of artificial devices 
which imitate as closely as possible the various kinds of 
symmetry displayed by these bodies; these attempts have mainly 
consisted in packing together spheres, ellipsoids, and other regular 
bodies in a symmetrical manner, the methods employed being 
generally such that the packing is the closest possible, and the 
bodies packed together being all alike. 
In the meantime, however, evidence has been accumulating, 
notably in connexion with stereo-chemical investigations, that a 
regular repetition in space which portrays the homogeneity of 
structure of crystals should be that of groups composed of two or 
more individuals rather than that of single bodies as generally 
hitherto represented, and further that the bodies forming a group 
need not be similar. Thus, if we adopt the simile used by Lord 
_ Kelvin, and compare the homogeneous structure of a crystal to a 
regiment of soldiers in battle array,? we shall in many cases take 
a troop of cavalry rather than one of infantry for the comparison, 
1 Comp. ‘‘ Molecular Constitution of Matter,’’ by Sir William Thomson, in Proc. 
Roy. Soc. of Edinburgh, vol. xvi. pp. 712-715. 
2“«The Molecular Tactics of a Crystal.’”? Second Robert Boyle Lecture, 1893, 
p. 26. 
SCIEN. PROC. R.D.S., VOL. VIII., PART VI. 2R 
