Abstract The fundamental principles underlying the annealing of glass and the present state of knowledge concerning changes in the nature of glass during heating and cooling are briefly reviewed. Emphasis is placed on the fact that a quantitative explanation of the formation and removal of strain in glass, and a precise determination of the requisite conditions for any desired degree of annealing, may be obtained on a purely mechanical basis, without consideration of the internal constitution of materials in the vitreous state. Annealing Practice. —An improvement in the definition of annealing time (made feasible by the relatively great speed with which large strains are released) allows a considerable simplification in the determination of the proper procedure for annealing. As an illustration, the procedure for annealing a disk of Pyrex glass, 200 inches in diameter, is given. The best conditions for annealing, when for any reason the temperature must be above or below the most favorable temperature, are also discussed. Effect of Internal Stresses on Density. —It is demonstrated that internal stresses can have no effect on the total volume or on the average density of a block of glass. The stresses may cause a deficiency in density of the interior portion and an excess near the outside, but the mean density of any large or small piece is exactly the same as if the glass of any large or small piece is exactly the same as if the glass were free from stress. Although unannealed glass may have an abnormally low density, the stresses themselves have nothing to do with this effect. The Vitreous State. —The anomalies of glass in the annealing range, the relation between viscosity and rigidity, the physical basis of the rate of release of strain, and the possibility that glass at low temperatures may have a finite strength, are discussed. Although a lack of data prevents the drawing of any positive conclusions, nevertheless the state of aggregation of glass is a problem of very great interest. It appears that more attention should be given to the physico-chemical side of the problem; but it is essential not to confuse phenomena that involve the internal constitution of glass with those that do not.