SCATSTOSITY AND SLATY CLEAVAGE 437 
experimental side of the question will be touched upon later. 
Flow must cause molecular rearrangement even in chemically 
stable bodies. One may imagine flow to consist in the rolling 
over of molecules in alternate very thin layers and the effect 
being something lke the lamellar twinning of feldspars. In 
such cases it would seem inevitable that the twinning planes 
should be planes of weakness. The energy of the strain is con- 
verted into heat and this heat is developed exclusively along the 
flow surfaces. In chemically unstable bodies this heat will man- 
ifest itself in the production of secondary minerals such as mica, 
and the new minerals will arrange themselves along the lines of 
flow. This action appears to me to constitute dynamo-meta- 
morphism so far as such metamorphism attends direct pressure. 
Taking it for granted that flow surfaces are surfaces of weak- 
ness rather than of increased strength, two distinguishable cases 
may arise in the progress of shear. It may happen (with some 
substances and under some relations between the active forces) 
that, although the direction of greatest tangential load passes 
away from the direction of initial flow, the diminished tangential 
load will still produce more motion on the weakened surfaces 
than in the unweakened but more heavily loaded surfaces. In 
such cases rupture will ensue at nearly 45°, the distortion will 
5) 
be insensible and the substance is a ‘brittle’? one; or in other 
words, the difference between the elastic limit and the ultimate 
strength is exceedingly small. In the opposite case flow will 
cease on the initial planes and commence anew on the planes 
where the tangential load has risen to a maximum. Experi- 
mentally no absolutely brittle substances are known. There is 
always so far as known a range of pressure (usually a small one ) 
within which flow occurs along successive surfaces so that con- 
siderable deformation without rupture can be affected. 
If distortion by pure shear is carried very far without true 
rupture, the mass will be more or less cleavable in a number of 
directions separable into two systems. All the cleavages of 
either system will lie within a few degrees of one another. The 
consequence will be a somewhat confused foliation. It will be 
