Parallel and Intersecting Joints. — Crosby. 373 
instead of continuing- to develop slowly and reaching the 
breaking intensity by insensible degrees, culminates, under 
the influence of the shock, instantaneously, accomplishing its 
work, at last, swiftly instead of sluggishly. The vibrations 
are merely the last straw, or the finger that pulls the trigger 
and releases the accumulated energy. 
I have repeated Daubree's experiments with sheets of 
glass, obtaining similar results; and have also modified 
them, to note the effect of a sudden shock. Strips of glass 
24 inches long and from 3 to 5 inches wide were tested in a 
horizontal position on an ordinary table, one end being held 
by a stationary clamp, while the other was attached by a 
hinge in the axis of torsion to a fixed support or fulcrum. 
The lever was long enough so that it could be weighted and 
the torsional strain in the glass thus carefully regulated. 
The shock was applied, at first, by striking the table, to 
which the apparatus was firmly fastened, with a hammer, the 
glass being entirely free from the table except at its extrem- 
ities. Before applying the shock, however, the lever was 
first weighted until it seemed that the strain on the glass 
must be near the breaking point, and then the strain was 
materially diminished, to make sure that the glass was not 
dangerously near collapsing before the passage of the vibra- 
tions. A smart blow with the hammer then precipitates the 
breaking; and the result is similar to that obtained by tor- 
sion alone, except that the fractures are likely to be fewer, if 
the torsional strain is not near its maximum. The directions 
df the vibrations make, apparently, but little difference; 
and in every trial two rectangular systems of fractures, 
making equal .ingles with the axis of torsion, were obtained. 
The breaking Or jointing of the glass was most regular and 
perfect when a sharp blow with a light hammer directly upon 
end of the glass itself, sent vibrations through it bearing 
some resemblance loan actual earthquake. When the tor 
sional strain is slight, strong vibrations are required to pre- 
cipitate the breaking and the fractures are irregular and 
mainly transverse, tending to cul the axis of torsion al ninety 
degrees instead of forty-five degrees. Hut if the torsional 
strain is great a very >light shock suffices and the result is. 
with glass, indistinguishable from that obtained by torsion 
