400 Scientific Proceedings, Royal Dublin Society. 
indication of deformability, we have direct petrological evidence to: 
show that such an hypothesis of soft crystals deforming one 
another is untenable. Olivine is a mineral melting at 1371°C. 
(therefore below quartz) and exhibiting (according to Mr. Ralph 
Cusack)! a far longer period of viscosity during fusion than any 
other mineral. ‘The melting period of olivine as observed on the 
meldometer was 84° (138238°-1407°), that of quartz only 34° 
(1406°-1440°). If then the slow melting of quartz be regarded as 
accounting for its appearing in allotriomorphic shapes then a@ fortiort 
olivine should always be allotriomorphic. But this is exactly what 
we never find. Olivine is always idiomorphic with regard to the 
other silicates in the rocks in which it occurs.” 
A possible factor in the problem may perhaps be the “ melting 
hysteresis,” or want of agreement between the melting and freezing 
temperature of a substance. This has often been noticed in the 
case of certain organic substances, but has been fully demonstrated 
by Dr. C. Barus,* and is illustrated by a series of curves. These 
illustrate the ‘‘ volume lag” in isothermal fusion, and from these 
curves the isopiestic volume lag could readily be deduced. J. g. 
under a pressure of about 600 atmospheres naphthalene melts at 
100° and solidifies at 90°. Barus also points out that there is a 
perfectly definite amount of lag. It is conceivable that quartz 
might crystallize at a lower temperature than its melting-point. 
There are then several more purely chemical aspects of the 
subject which probably all enter to a greater or less extent in the 
building up of a complete theory. Indeed, a complete theory 
could only be developed from a consideration of all the connected 
phenomena included under the Theory of Solution. All molten 
rock magmas were probably once true solutions. Thus, for instance, 
Bunsen‘ dwelt on the important influence on the melting-points of 
both solvent and dissolved substance exerted on each by the presence 
of the other. It is well known that the order of crystallization of 
1 Loc. cit., p. 406. 
2 Cf. J. J. H. Teall, ‘‘ British Petrography,”’ p. 53. 
3C. Barus, ‘‘The Continuity of Solid and Liquid,’? Am. J. of Sc., m.° v. 42 
(1891), p. 125. 
4R. Bunsen, ‘‘ Ueber die Bildung des Granites,’’ Zeitschrift d. deut. geol. Ges. 
Bd. x11. (1861), p. 61. 
Cf. H. C. Sorby, ‘‘On some Peculiarities in the Arrangement of the Minerals in 
Tgneous Rocks,’’ Brit. Assoc. Report, 1858 (pt. 2), p. 107. 
