364 PEor. J. w. jrDD on the tertiaky and 



display this structure in its most complete development, others are 

 quite destitute of it, and others, again, exhibit it in irregular and 

 interrupted portions of their mass. 



At one time the presence of lamellar twinning was regarded by 

 petrographers as an essential characteristic of the plagioclase felspars, 

 and when untwinned or simply-twinned felspar- crystals were met 

 with, even in the most basic rocks, they were referred to ortho- 

 clase. But this view has now been generally abandoned, and the 

 late Dr. G. Hawes * has demonstrated that the most typical oligo- 

 clases, andesines, labradorites, and anorthites sometimes present no 

 trace whatever of lamellar twinning. 



Much new light has been thrown by recent researches on the 

 manner in which twin lamellae may be produced in crystals ; and the 

 results of these investigations are worthy of the most careful con- 

 sideration by the petrographer and geologist. 



As long ago as 1859, it was shown by Pfatf that remarkable 

 changes are induced in the optical properties of calcite when its 

 crystals are subjected to pressure t ; and in 1867 Prof. E. Reusch 

 demonstrated that by pressure applied in certain directions to these 

 crystals a definite lamellar twinning could be developed in them J. 

 In 1879 Baumhauer showed that the pressure necessary to bring 

 about this twin structure could be applied by simply forcing the 

 edge of a knife-blade in a certain direction into the crystal §, and 

 this conclusion was confirmed by the observations of Brezina ||. 



l^ow calcite crystals, when freely developed, may attain the 

 largest dimensions without betraying any trace of this lamellar 

 twinning, as is seen in the case of ordinary Iceland-spar. But the 

 great majority of rock-forming calcite-crystals, even when of the 

 smallest size, like those occurring in crystalline limestones and 

 dolomites, for instance, exhibit this lamellar twinning in the most 

 striking manner. 



Similar lamellar twinning is exhibited by many rock-forming 

 minerals, such as leucite, garnet, boracite, &c., and has led to many 

 speculations concerning the possibility of these crystals being re- 

 garded, not as the simple forms to which they were first referred, 

 but as examples of very complicated structures built up by polysyn- 

 thetic twinning and belonging to a lower order of symmetry. 



. But certain recent investigations have thrown a new and unex- 

 pected light on this most interesting question. Mallard has shown 

 that boracite loses its twin lamellae when the crystals are heated to 

 the temperature at which they were originally formed ^. Merian 

 has proved that tridymite becomes optically hexagonal at high 



* Proc. National Mus. 1881, pp. 134-136. See Smithson. Miscall. Collect. 

 vol. xxii. 



t Pogg. Ann. vol. cvii. (1859) pp. 333-538. 



X Pogg. Ann. Tol. cxxxii. (1867) ; ibid. cxli. (1872) ; also Proc. Rov. See. 

 Edinb. vol. vi. (1869) p. 136. 



§ Zeitschr. Kryst. vol. iii. (1879) p. 588. 



li Ibid. vol. iv. (1880) p. 518. 



^ Bull. Soc. Min. de France, vol. v. (1882) p. 144. 



