Tensile strength: From 1 to 4 tons per sq. in. (Trautwine). ” ⅓ to 1¼ ” ” ” (Henrivaux). ” 2 to 5½ ” ” ” (Winkelmann and Schott). ” 5 to 6 ” ” ” (Kowalski). Crushing strength: From 9 to 16 tons per sq. in. (Trautwine). ” 3 to 8 ” ” ” (Winkelmann and Schott). ” 20 to 27 ” ” ” (Kowalski). Of the above figures the experiments of Winkelmann and Schott are probably by far the most reliable, but these refer to a series of special Jena glasses, selected with a view to determining the influence of chemical composition on mechanical properties, and, unfortunately, this series does not include glasses at all closely resembling those ordinarily used for practical purposes. The attempt to connect tensile and crushing strength with chemical composition was also only very partially successful; but the results serve to show that the chemical composition has a profound influence on the mechanical strength of glass, so that by systematic research it would probably be possible to produce glasses of considerably greater mechanical strength than those at present known. It must be noted in this connection that the mechanical properties of glass depend to a very considerable extent upon the rate of cooling which the specimen in question has undergone. It is well known that by rapid cooling, or quenching, the hardness of glass can be considerably increased; such treatment also increases the strength both as against tension and compression, and numerous processes have been put forward for the purpose of utilising these effects in practice. Unfortunately the “hardened” glass thus obtained is extremely sensitive to minute scratches, and flies to pieces as soon as the surface is broken, and the great internal stress which always exists in such glass is thereby relieved. All these peculiarities are, of course, dependent as to their degree upon the rapidity with which the glass has been cooled, and the aim of inventors in this field has been to devise a rapid cooling process which should strike the happy mean between the increased strength and the induced brittleness resulting from quenching. Thus processes for “tempering” glass by cooling it in a blast of steam or in a bath of hot oil or grease have been brought forward; but, although some such glass is manufactured, no very extensive practical application has resulted.