-fifty-two records of the appearance of x\urora Borealis (northern lights) 

 in Japan in our old documents such as " Nihon-shoki," " Azuma-kagami," 

 -&c. These records are shown in the following table : — 



List of Aurora Borealis in Japan 



Location. No 



3 



-t 



5 



6 



7 



S 



9 



10 



11 



12 



13 



14 



15 



16 



17 



IS 



19 



20 



21 



22 



23 



24 



25 



26 



Jan., 

 Sept., 

 Aug., 

 April, 

 June, 

 Mar., 

 July, 

 Aug., 

 Oct., 



July, 



Dec, 

 Dec, 

 Dec, 

 Dec, 

 Feb., 

 Jan., 

 Sept., 

 Mar., 

 Aug., 

 Aug., 

 Nov., 

 Dec, 

 Oct., 

 Sept., 

 Aug., 

 Oct., 



G21 

 083 

 839 

 843 

 939 

 1098 

 1119 

 1150 

 1150 

 1152 

 1170 

 1176 

 1177 

 1202 

 1204 

 1205 

 1227 

 1241 

 1247 

 1363 

 1370 

 1370 

 1371 

 1440 

 1455 

 1486 



— Tan., 1.500 

 IS Mar., 1582 



14 Mar., 1587 

 7 Sept., 1635 



Summer, autumn, wi 



— Jiily, 1656 

 27 Oct., 1673 



23 Dec, 1693 



11 Nov., 



15 Feb., 

 18 Dec, 



17 Sept., 



25 Sept., 



12 Oct., 

 (1 Jan., 



1 Mar., 

 22 Sept., 



24 Sept., 1822 



26 Nov., 1855 



2 Sept., 

 4, 6 Feb., 



25 Sept., 



18 Oct., 

 22, 26 Jan., 



31 Mar., 

 1 Mar., 



1726 

 1730 

 1737 

 1770 

 1770 

 1771 

 1781 

 1786 

 1810 



1859 

 1872 

 1908 

 1928 

 1938 

 1940 

 11U1 



Kyoto. 



Kyoto. 



Kyoto. 



Tokyo, Osaka, &cv 



Tokyo, Kyoto, &c. 



Tokyo. 



Kyoto. 



Ishikawa. 



Kyoto. 



Shiga. 



Tokyo, Kyoto, &c 



Every part. 



Kyoto. 



Kyoto. 



Chiba. 



Nagoya. 



Tohoku 



Osaka. 



Osaka. 



Wakayama. 



Shimane, Shizuoka. 



Ehime. 



Sapporo. 



Iwate. 



Rishiri. 



Anbetsu. 



The monthlj' frequency of the aurora in Japan bears a resemblance 

 to that of the aurorse which have been observed in other countries — 

 Norway, Sweden, Denmark, and so on (Fig. 2). It is already known that- 

 the aurora increases and decreases its number of appearances every 

 eleven years in accordance with the periodicity of the sun-spots. But 

 now I have found that the frequency of the aurora in Japan has some 

 relation to her 700-year periodic change of climate — that is to sa\% most 

 of the aurora have been sighted in the centur}-' of colder period, and that 

 in warmer period there are few or none. I have also proved that this 

 tendency is- even more conspicious in Korea (Fig. 3). Unfortunately, 

 we cannot get the records of the aurorse in Korea documents except 

 during the period from the eleventh to the seventeenth century. In 

 these records, however, are counted up two hundred and three aurorse. 

 At any rate, it is very interesting to know that there exists a definite 

 connection between the frequency of the aurora and the 700-year periodic 

 change of climate in Japan. Because the aurora appears as the result 

 of the influences of electrons from the sun upon the magnetic fields of 

 north and south of the earth in the upper region of the atmosphere, 

 the appearance of the aurora has a close connection with that of the 

 sun-spots as already known. And so the 700-year periodic change of 

 climate may also be due to some influences caused by the changes in the 

 solar activity. The fact that the frequency of the aurora in Japan has 

 similar periodicity with the climate will provide a serious subject to 

 jf'.tudy on this problem. 



120 



