90 Ferguson and Merwin — The Ternary System. 



once formed, would dissolve but slowly, thus making 

 almost impossible the preparation of homogeneous com- 

 positions. Compositions lying within the magnesia field 

 showed a similar tendency to crystallize, and with these 

 also, difficulty was experienced in obtaining homogeneity. 



Once prepared, each composition was thoroughly 

 investigated by means of the quenching method. This 

 method consists in holding a small quantity of a given 

 composition (called a "charge") at a given temperature 

 long enough to insure the attainment of equilibrium and 

 then chilling it suddenly without disturbing this equili- 

 brium condition. The charge is wrapped in a small piece 

 of platinum foil 0-01 m.m. thick which is attached to a 

 thermoelement tube in such a manner as to be very near 

 the" junction, and is dropped into mercury by fusing the 

 supporting wire by means of an electric current. The 

 details of the method may be found in several of the pre- 

 ceding papers. 33 The apparatus is shown in fig. 7. 



A few compositions, thoroughly investigated, served 

 to locate approximately the various boundary lines, and 

 with this information the efficient selection of the subse- 

 quent compositions was an easy task. Most of the 

 charges required from 20 to 30 minutes to reach an 

 equilibrium condition, but some charges, notably those in 

 the 2CaO.Mg0.2Si0 2 field and those containing much 

 silica, required a much longer treatment. Charges in 

 which at equilibrium there was no glass, such as those 

 used in the study of the wollastonite solid solutions, were 

 heated for days before samples suitable for microscopic 

 examination could be obtained. The formation of 

 unstable pseudowollastonite crystals at temperatures 

 below but near the inversion temperature took place 

 readily when suitable glasses were crystallized at these 

 temperatures. The inversion of pseudowollastonite to 

 wollastonite does not take place readily and so when 

 charges of wollastonite free from pseudowollastonite 

 were desired, glasses were first crystallized at tempera- 

 tures ranging from 800 to 900 °C over long periods of 

 time (15 hours) and this material if free from pseudo- 

 wollastonite (in not more than one-half of the charges 

 was this true) and from glass was then reheated at 

 higher temperatures for some hours in order to let the 

 crystals grow and enable a final selection of material to 

 be made with certainty. 



23 See this Journal, 39, 1, 1915. 



