MAGNETIC SLKVEÏ Ol' JAl-AN. 213 



irregularities of (jiiite u local ilcseriptiijii, liciice, in dclaiilt of 

 evidence which could only be obtained by a series of observations 

 along the coast of the maiji island, it seems to me more prudent to 

 draw the isoi^onic line of 5° fairlv normal and re!)resent the disturb- 

 ance due to Sado by a small is;)Iated contour round that island. In 

 this way it is shown on our charts. In sh(jrt, knowing as we do that 

 an island is enough by itself to cause considerable distiirljances, we 

 are justified in explaining the peculiarities that are as being due to 

 the existence of the Ishmd of Sado rather than to the existence of the 

 Fossa Magna. Indeed tliere does not seem t<j l)e any very extrain'di- 

 narv disturbance after all in this rec^ion. There are as marked 

 disturbances in the north of Japan, Avhere no Fossa Magna exists ; 

 and these disturbances tind a sutHcient explanation in the presence 

 of volcanic rocks of all kinds. A mountainous volcanic reo"ion is 

 certain to present magnetic irregularities ; and in Japan there are two 

 regions specially to be noted as such. A glance at the four charts 

 will at once pick them out. The one is the great central mountainous 

 region, just where the Fossa Magna is. The other is the part between 

 the o8th and 40th parallels ; Init here there is nothing geologically 

 comparable to the Fossa Magna. If [)eculiarly abnormal irregularities 

 had been found to exist in the central retrion only, there miuht have 

 been reas(3n in connecting them with a very peculiar geological struc- 

 ture. l)Ut, since the northern region is quite as abnormal mairnetically 

 as the central one, the most logical course is to look upon these 

 abnormalities as due to something common to them b(jtli. And this 

 connnon element is sini[)ly a prodigitjus de\eIo[)ement of ^■olcanic rocks. 

 There are six stations which may be said t<j belong to the central 

 hilly region. These are Sekiyama, Ueda, Takanomachi, Kôfu, Ilara, 

 and Hakone — Nos. 36 to 41 inclusive. Excepting Hara, which is on 

 the coast, these stations all lie at considerable elevations. Thus Seki- 



