ON THE EARTHQUAKE AND VOLCANIC PHENOMENA OF JAPAN. Ill 



or portions of the instrument by change of temperature, a variation in 

 the condition of the fibre by change in moisture, or as an effect due to 

 magnetic influences. Sometimes it appears that the southern side of the 

 column had risen or sunk within a period of fifteen minutes. At other 

 times the movement takes several hours. Usually it takes about as long- 

 to sink as it did to rise ; but this is by no means invariable, the upward 

 motion being sometimes accomplished in a few minutes, whilst the 

 lowering extends over several hours. The following are examples of 

 these movements, which for convenience are described as an actual lifting 

 or sinking of the south side of the column. 



February 9. — Daring the day the south side rose shghtly, and sank 

 again towards night. 



February 10. — About 4.30 A.M. the south sunk suddenly 2 mm., but in 

 one hour returned to its normal position. 



February 11. — Between the morning and the afternoon the south side 

 rose slowly. 



February 12. — No movement observable. 



February 13. — Steady rise of the south side, commencing in the morn- 

 ing, and continuing until 8 p.m. and on to midnight. 



February 14. — Fi-om midnight of the 13th the south sank, the greatest 

 fall taking place during the afternoon. This sinking continued steadily 

 to the 15th. On the 16th this sinking was irregular. 



February 17. — ^Between 3.30 and 4.30 a.m. the spot of light moved 

 2 mm. to the right and back to the left, indicating a rising and then a 

 sinking of the south side. A similar movement, but of from 4 to 5 mm., 

 took place in fifteen minutes, commencing at 9.30 p.m. 



February 18. — The south rose slightly during the day, and sank to- 

 wards night. About midday there was a deflection like the two which 

 took place on the 17th. 



February 19. — During three hours the south sank, and then rose in 

 three hours, the range of motion being about 6 mm. During the night 

 the south ag^in sank, and rose within a period of about three hours. 



February 21. — Early in the morning the south slightly sank, and this 

 continued until the afternoon, when within an interval of ten minutes it 

 quickly rose. There was a sudden but slight sinkage at midnight. 



February 24. — During the whole day, and also at night, the south was 

 slightly sinking. 



Febriiary 25. — During the day the south was rising. 



February 26. — From 4 to 7 a.m. the south sank, after which it rose to 

 ■9 A.M. In the afternoon it again sank ; &c. 



The deflections referred to as measured on the photographic record 

 are usually about 2 mm. ; it is not unfrequent that they reach 5 or 6 mm., 

 and once they were as much as 15 mm. 



When the mirror has a period of four or five seconds, and it is 29 inches 

 from the photographic plate, a deflection of 1 mm. might be due to a 

 tilting represented by '25 inch per mile. The sensitiveness of the 

 apparatus may be changed by altering the period of the mirror. The 

 approximation given may also be used in giving angular values for the 

 tremors which have been observed. 



The fact that the south side of the column appears to rise when the 

 sun is on the south side of the house would indicate that the tilting was 

 due to temperature, but as the same movements also take place during 

 the night, and at irregular intervals, it cannot be an immediate effect of 



