THE MALAYAN ARCHIPELAGO. 97 



ingenious interpretation of the account, the phenomena of earth- 

 ([uake and explosion are so hopelessly interwoven, that, from 

 reading his paper, I am almost convinced that all were here 

 dealing no more than a series of violent earthquake shocks. 

 Our experience here in Japan shows that destructive earthquakes 

 and explosions do not fall simultaneously. That Wawani is of 

 volcanic origin and once a seat of activity can, however, be 

 affirmed from the occurrence of pyroxene-dacite which moreover 

 contains cordierite^"^ as an accessory. It is very desirable to see 

 the question of Wawani settled, for, upon it largely depends 

 ^vhether the chain of volcanoes of the Moluccas and of the Banda 

 Sea have a claim to separate existence, or are only parts of 

 one system. 



From the foregoing, it is evident that the curved chain of 

 islands from Kisser northwards, is only the shattered relic of an 

 okl mountain-range, composed mainly of crystalline schists and 

 old massive rocks, beset with the Tertiary limestone and modern 

 coral reefs, with the exception of a part of Moa, the island Tjoor, 

 and possibly also Manawoka, which are said to be volcanic. 



Wichmann pointed out the existence of a third, external zone, 

 which may be indicated by a line, going through the Sula group, 

 Great Obi, Misool, and then a part of New Guinea and the 

 Aru group. 



It has been repeatedly asserted'^'^ that there exists a close 

 resemblance between the Banda islands and the Lesser Antilles. 



32) Vide Schroeder von der Kolk, Neues Jahrb. für Mineralogie, etc. 1896. I. Bd. S. 154. 

 Japanese cordierite occurs in tlie volcanoes of Asama, Iwaté, and a hill near Nagano (Prov. 

 Nagano), only in tlie form of white ejected blocks with conchoidal fracture, but, so far as 

 I know, has never found in tlie main body of lava-flows. 



33) Suess, ' Das Antlitz der Erde ', Bd. I, p. 700. DiUo, Bd. I, p. 209. 



