THE ARCHJÎAN FORMATION OF THE ABUKUMA PLATEAU. 215 



the same zone in the section is by no means nîiIikeJv. But, makinof 

 a small deduction for the breadth of the eruptive rock and a cor- 

 rection for the dip, we can estimate the probable thickness of the 

 formation, and this was found to be 5^ km. for the upper horizon 

 of the Takanuki series. 



At Sôri, wliere the road bifurcates, one leading to Tnkanuki, 

 the other to Kodaira, an important tectonic line was accidentally 

 discovered. Here an insignificant isoclinal valley comes from the east, 

 through the villages of Kodaira and Nakagura, along whicli tlie present 

 line of section was drawn. A precipitous wrdl on both sides of the 

 rivulet is wholly composed of alternate bands of the titanite-amphibole 

 schist and the two-mica schist ; all of them dip westwards at very 

 blüh angfles with the strike north and south. A ridg-e alono- the 



Co o o 



west side of the valley, forming a naked whitish precipice, at once 

 gives information of its being composed of a granitic material, con- 

 trasting greatly in colour and habitus with the black, schistose 

 rock of tlie opposite side (Fig. H, Fl. XXIII)'. 



If we adopt the view of the excessive effect of dislocation -met a- 

 morphism, and in accordance with it attribute the formation of banded 

 gneis.-es and amphibolites to the crushing of dioritic and graniti«" rock 

 by strong pressure, it is not yet easy to conceive of such sudden changes 

 of structure at so short a distance. Therefore, I examined very 

 closely the spot where possibly the line of junction of the schists and 

 granite might occur high up on the cliff ; and in doing so, I found 

 actually the place where two geological bodies of entirely different 

 origin come in direct contact, side by side, having nothing in 

 common either in appearance or structure. The granite is the 

 hornblende-bearing variety with an imperfectly schistose structure and 

 its plane of schistosity inclining towards the west ; and the streaky, 



1 In Fig. 11, PI. XXIII, X:ikagura valley is erroneously spelled as Naqura. 



