508 GEOTECTONIC (STRUCTURAL) GEOLOGY. [Boox IV. 
sometimes in one direction sometimes in the other.’ The remarkable 
spheroids which appear in many weathered igneous rocks besides 
basalts, where they are not the result of weathering, may probably be 
due to some of the conditions under which the original contractions 
took place. They are quite untraceable on a fresh fracture of the 
rock. It is only after some exposure to the weather that they begin 
to appear, and then they gradually crumble away by the successive 
formation and disappearance of external weathered crusts or coats, 
- which fall off into sand and clay. Almost all augitic or hornblendic 
rocks, with many granites and porphyries, exhibit the tendency to 
decompose into rounded spheroidal blocks. The columnar:structure, 
though abundant among modern volcanic rocks, is by no means 
confined to these. It is as well displayed among the felsites of the 
Lower Old Red Sandstone, and the basalts of the Carboniferous 
Limestone in central Scotland, as among the Tertiary lavas of 
- Auvergne or the Vivarais. 
MN 

A a , 
IM ie We 
a iM 
NUN iis VND) 
Fic. 223. — OrDINARY Fic. 224.—Bat.- Fig. 225.— Moprrica- 
CoLUMNAR STRUCTURE AND - SOCKET TION OF BALL-AND- 
oF LAVA. JOINTING OF Socket STRUCTURE. 
CoLUMNS. 
3. In Foliated (Schistose) Rocks.—The schists likewise possess 
their joints, which approximate in character to those among the 
massive igneous rocks, but they are on the whole less distinct and 
continuous, while their effect in dividing the rocks into oblong masses 
is considerably modified by the transverse lines of foliation. These 
lines play somewhat the same part as those of stratification among 
the stratified rocks, though with less definiteness and precision. The 
jointing of the more massive foliated rocks, such as the coarser 
varieties of gneiss, approaches most closely to that of granite; in the 
finely fissile schists, on the other hand, it is rather linked with that 
of sedimentary formations. Upon these differences much of the 
characteristic variety of outline presented by cliffs and crests of 
foliated rocks depends. 
1 Mr. Scrope pointed this out (Geol. Mag. September, 1875), though Mr. Mallet 
(Ibid. November, 1875) replied that in such cases the articulations must be formed just 
about the dividing surface, between the part of the rock which cooled from above and 
that which cooled from below. 

bi x Se Bee 
