Ch. XXXV.] METAMOEPKIC THEORY. 599 



forgotten that, as a general rule, the less crystalline rocks do really occur 

 in the upper, and the more crystalline in the lower part of each meta 

 morphic series. 



Moreover, metamorphism must often begin to exert its force long after 

 the strata have assumed a vertical position, and it may then act locally 

 or -within limited areas, and will be as likely to affect the newer as the 

 older beds. As an illustration of such partial conversion into gneiss of 

 portions of a highly inclined set of beds, I may cite Sir R. Murchison's 

 memoir on the structure of the Alps. Slates provincially termed " flysch" 

 (see above, p. 230), overlying the nummulite limestone of Eocene date, 

 and comprising some arenaceous and some calcareous layers, are seen to 

 alternate several times with bands of granitoid rock, answering in charac- 

 ter to gneiss.* In this case heat, or vapor, or water at an intensely high 

 temperature, may have traversed the more permeable beds, and altered 

 them so far as to admit of an internal movement and re-arrangement of 

 the molecules, while the adjoining strata did not give passage to the same 

 heat, or if so, remained unchanged because they were composed of less 

 fusible materials. "Whatever hypothesis we adopt, the phenomena estab- 

 lish beyond a doubt the possibility of the development of the metamor- 

 phic structure in a tertiary deposit in planes parallel to those of stratifi- 

 cation. 



Whether such parallelism be the rule or the exception in gneiss, mica- 

 schist, and other formations of the same family, is a question which I 

 shall discuss at length in the next chapter. 



* GeoL Quart. Jiurn. vol. v. p. 211. 1848. 



