^rAGNETIC SURVFA- OF JAPAN. 211 



Resfnrdinîr Tsliihnshi it seems difficult at first to siiocore.st a cause of such 

 a large (liscre])an('y. The Declination ohserved here was (neglecting 

 the îdtogether peculiar Hakone) the second smallest observed by the 

 Northern Party. Being the very first station of the whole survey, it 

 was feared f^r a time that some mistake had been made by the 

 oljserver. To make sure of this Mr. Nagaoka was despatched to- 

 wards the end of September to Ishibashi to make a redetermination 

 of the Declination. The observation was made at a slightly difi;erent 

 locality ; and the declination was measured at 0.45 a. m. by means of a 

 Sun's Azimuth. The result did not difter by one minute from the 

 value as found in June by means of Polaris. Here again, then, there 

 can be no question as to the very peculiar smallness of the Declination 

 at Tshiljashi. Geologically considered Ishibashi lies just within the 

 Eastern wing of the " Schaarung" (so called by Süss),* within which 

 lie the volcanic regions of Japan. The boundary between this vol- 

 canic region and the non-volcanic region to the East is the continuation 

 to the south of the similar boundary already described as coinciding in 

 position Avith the Kita-kami-gawa. Xow, although to all appearance 

 Ishil)ashi is situated on alluvial soil, yet, according to the geologists, 

 it just lies within this Schaarung. Hence it is quite probable that 

 volcanic rocks exist liidden from view but so distributed as to cause 

 a distinct magnetic disturbance. The values of the deolinatitui ob- 

 tained Ijy Sekino throughout this regi<m show consideraljle irregula- 

 rities, Ishibashi itself being characterised by a declination somewhat 

 smaller tlian the average. His value for this station (namelv, 4° 

 16' .6) is, however, much larger than ours. \\\' nuist therefore cii her 

 assume a change of 15' or so in five years, or, wluit is much more 

 probable, a very local source of dist in-bance. 



If we neglect these three stations (Nos. 1, 5, 12) the mean 



* See Dr. Naumann's Pamphlet already mentioned, pa^e 17 ; also Professor Koto's Paper 

 already pnlilishod in this Volume. 



