278 Day and Shepherd — Lime-Siliea Series of Minerals. 



not a difficult matter to secure a preliminary survey of a field 

 of this kind. The mineral wollastonite is known, and more 

 than that, is known to possess a melting temperature lower than 

 either lime or silica, There is therefore immediate reason for 

 anticipating eutectic relations somewhere in the series. If 

 wollastonite forms a eutectic with components on one or both 

 sides of it, mixtures containing slightly more lime or slightly 

 more silica than wollastonite will have lower melting tempera- 

 tures than it. A simple and effective mode of preliminary pro- 

 cedure is therefore to take a tiny pinch of a number of the 

 percentage mixtures adjacent to wollastonite, place them in 

 order upon a narrow platinum ribbon which can be heated 

 electrically to uniform brightness, and observe the order in 

 which they melt. No temperature measurement is worth 

 while ; the information obtained can serve only for orientation 

 and must be verified by more reliable pyrometric methods. 



If a eutectic is present on either side of the compound, it 

 will be the first to melt, and the compound last ; the interme- 

 diate mixtures are not important. If the materials are not too 

 viscous the melting will be sharp and the material will crystal- 

 lize again on slow cooling. A few repetitions, or the intro- 

 duction of intermediate compositions in doubtful cases, will 

 usually enable a preliminary curve to be drawn in which the 

 compounds and eutectics which are within reach will be cor- 

 rectly located. In fact, for many substances they can be very 

 exactly located in this way. Intermediate compositions, on the 

 other hand, may be very misleading, depending upon the 

 behavior of the eutectic present after the melting temperature 

 of the, latter has been passed. (In applying this method, very 

 small particles (0'2 miu ) must be used in order to obtain com- 

 parable results.) 



Proceeding in this way, a eutectic w T ill be readily located 

 between silica and wollastonite at the composition 63 per cent 

 Si0 2 , 37 per cent CaO, and another on the other side of wollas- 

 tonite at the composition 46 per cent Si0 2 , 54 per cent CaO. 

 We will allow the other component of this second eutectic to 

 remain unidentified for the moment, as no stable lime-silica min- 

 eral richer in lime than wollastonite is known. If we continue 

 our platinum ribbon experiment with continually increasing 

 percentages of lime, we shall find that after one or two steps 

 beyond this second eutectic the platinum ribbon will burn out 

 without melting the little grains. In other words, the melting 

 temperatures of lime-silica mixtures richer in lime than 60 per 

 cent are all higher than that of the platinum. To meet this diffi- 

 culty we built a small but very serviceable piece of apparatus 

 the essential portion of which is a thin ribbon of pure iridium 

 about 2 mm wide and 10 cm long, stretched between electrodes 



