ON THE EARTHQUAKE PHENOMENA OF JAPAN. ^51 



highly probahle that an increase of vibratory motion in the coal strata 

 ■would tend to facilitate the outflow of gases contained therein in even a 

 greater ratio than would the diminution of atmospheric pressure. My 

 personal experience in mining tends entirely to support this deduction, 

 as I have invariably found that in places where tbe outflow of gas was 

 fairly constant as a general rule, it always increased to a greater or less 

 extent when what is technically called any ' working ' of the strata took 

 place. 



For the purpose of obtaining more reliable statistics on this head, I 

 have arranged with Mr. F. Ringer of Nagasaki to erect the microseis- 

 mometer at his observatory on the mainland, eight miles from this island, 

 and we shall be able, by making simultaneous observations, to connect his 

 notes with those made at the observatory at the mine. Of course it 

 would be preferable to have the mainland observatory situated at a point 

 much closer to the mines than this one is, but there is no place nearer 

 where we can be sure of having a reliable observer. 



Observations on Earth-currents, ijr. 



Owing to my inexperience in magnetic observation and the difficulty 

 in getting suitable apparatus constructed or erected down here, I have 

 only begun to make regular observations during: the current month. 

 Previous to this, however, I have been making isolated experiments with 

 such rough-and-ready apparatus as could be constructed on the spot. 



The results are briefly as follows: — 



With one wire connected with an iron bar fixed in the line of a 

 fault below-groixnd, and the other end fixed to a similar bar inserted m the 

 strata at a distance of about 100 yards, there was a considerable deflection 

 in a home-made galvanometer. 



With both ends connected with the coal strata below ground, but 

 away from the vicinity of any fault, the deflection was hardly perceptible. 



With one end fixed to the rod inserted in tbe fault below ground, and 

 the other connected with the surface strata, the deflection was again con- 

 siderable. I then erected the apparatus in my own house, connecting one 

 end with a rod inserted in the line of a fault, at a short depth from the 

 surface, and the other with the surface strata with the home-made gal- 

 vanometer I got slight though perceptible deflections, and with the 

 galvanometer which was received from Tokio I have since obtained 

 deflections quite capable of registration. 



Owing to my own inexperience and the meagreness of the statistics 

 which I have been able to collect, it would be altogether absurd to give 

 any opinion, as yet. as to the connection (if any) between the movements 

 of the galvanometer and the outflow of gas in the mine, and I think that 

 it is preferable to collect at least one year's statistics before saying any- 

 thing further on the subject, than to state that, so far as the experiments 

 have progressed, they tend to render me sanguine that such a connection 

 might be established, and also that I would be most happy to receive any 

 advice or suggestions on the subject from people more experienced in the 

 subject than myself. 



Tidal Observations. 



With regard to the observations to be made with a view to establish 

 the connection (if any) between the crushing together of the roof and 

 floor of the mine and the rise and fall of the tide, Mr. Stoddart writes: — 



