13 
Again, a friend communicated to me the experience of 
slaters, who find that when by yielding of the timbers the 
surface of an old roof has become bowed, the tiles are dis- 
torted to such an extent that they will not lie flat upon a 
new roof. Old stone tiles (I do not know whether the same 
is true of slates and brick tiles) are often rendered perfectly 
useless in this way, however sound they may be. It is per- 
haps unnecessary to cite other similar cases. Every observant 
architect and engineer can give from his own experience 
facts of interest in reference to the influence of long- 
continued pressure upon an unsupported edge of stone. 
The magnificent instances of contortion sometimes dis- 
played in coast sections are certainly more impressive, but 
perhaps less wonderful in reality, than the cases on record of 
distorted pebbles. The unlimited effects of long pressure are 
nowhere so clearly demonstrated as in the bending of round 
or oval masses of small size. Instances of pebbles elongated 
in the direction of planes of cleavage occur at Llyn Padarn, 
near Llanberis, in the Lake District, and elsewhere. But the 
most remarkable cases of alteration of figure effected by 
pressure are those described by Dr. Hitchcock, Mr. Gr. L. 
Vose and others, as occurring in New England. In Vermont, 
Maine, Massachusetts and Ehode Island are found con- 
glomerates where sometimes for hundreds of square feet 
every pebble, whether of granite, sandstone, schist or quartz, 
has been flattened. Occasionally one pebble has been driven 
into another, so as to indent it or squeeze it into a semi- 
circular form, yet without fracture. Some of the examples 
figured resemble soft cakes jammed together into one mass 
with unyielding stones, so freely do they curve round in 
layers and adapt their shape to the various lines of force. 
Yet plasticity in any ordinary sense of the word is out of the 
question. These very pebbles are water-worn and some of 
them cleaved. Not a few are rolled fragments of plutonic 
rock. (See Plate, Figs. 2, 3.) 
