UNBREAKABLE OR TOUGHENED GLASS. 
229 
throwing them about indiscriminately on the hare floor. Glass 
plates, dishes, coloured lantern-glasses, and the like, have been 
similarly thrown about by the handful, stood upon, and other- 
wise maltreated, but without the slightest injury accruing to 
them, except perhaps when a solitary specimen which had been 
imperfectly tempered got in with the rest. Experiments have 
also been carried out to ascertain the comparative strength of 
toughened and untoughened glass when submitted to bending 
stress. Here a number of pieces of glass, each measuring 6 
inches in length by 5 inches in breadth, and having a thickness 
of about ^ of an inch, were tried. Each sample in its turn was 
supported at the ends, and a stirrup-piece was hung upon the 
■centre of the glass, a weight rod hanging vertically from the 
underside of the stirrup. With this arrangement applied to a 
piece of ordinary glass, the weight rod was gradually loaded until 
a weight of 279 lbs. was reached, when the glass broke. A 
piece of toughened glass of similar dimensions, similarly treated, 
did not give way until a strain of 1,348 lbs. had been reached, 
and before it yielded a considerable deflection was produced in 
it, showing its elasticity. Had its strength been due to rigidity 
or inflexibility alone, it would not have assumed a curve before 
yielding to the pressure brought upon it. 
Satisfactory as the above results may appear at the first glance, 
they will be seen upon reflection most inadequately to represent 
the relative strength of toughened and untoughened glass. It 
will be observed that the test applied was that of long-sustained 
and gradually-increasing pressure, which could rarely occur to 
glass articles in everyday use. Glass is subject to sudden, 
sharp blows, either from articles falling down on other sub- 
stances or from extraneous bodies falling upon or being brought 
into contact with them. Hence it is clear that to obtain a true 
estimate of the new process glass must be subjected to tests 
which fairly represent the conditions of the accidents to which it 
is ordinarily exposed. This estimate has been arrived at repeat- 
edly by placing pieces of plate glass in a frame and allowing 
weights to fall on them from given heights. One experiment 
from a number — and which was made publicly — will illustrate 
this test. A piece of ordinary glass 6 inches long by 5 inches wide 
and \ inch thick was placed in a small frame which supported 
the glass around its edges, and kept its underside about an inch 
from the floor. A 4-oz. weight was dropped on it from a height of 
1 foot, and the glass was broken. A piece of toughened glass of 
corresponding dimensions was then placed in the frame and the 
same weight dropped on it several times from a height of 10 feet, 
but without fracturing the glass. An 8-oz. weight was then sub- 
stituted, and repeatedly dropped upon the glass from the same 
height as before, and with the same result, no impression what- 
