376 REPORT—1885. 
Their period varies from 15 to 60 minutes. In these records, which 
may possibly mean tips of the soil, the two instruments only occasionally 
agree. 
a, Earthquakes are clearly shown by the pointer swinging about and 
making a series of holes up to the edges of the bands of paper. The 
relation, if there be any, between these various movements I have not 
yet determined. As the instrument is simple and inexpensive and 
satisfactory in its working, I am anxious that it should be brought to the 
notice of all who are interested in these observations. In connection 
with these observations I may mention that in September of last year, in 
conjunction with Mr. W. Wilson, C.E., and Mr. Mano, of the Imperial 
College of Engineering, I carried one of these instruments to the summit 
of Fujiyama, which is about 12,365 feet in height. Whilst on the top 
we were unable to undress, wash, or eat anything but the plainest of 
food. These and other discomforts, amongst which were the difficulties 
of breathing, prevented our remaining on the mountain for more than 
five days. 
During this time we obtained observations during the day and night 
extending over a period of three days. These observations were made by 
observing the position of the end of the pointer as it moved across a scale 
of millimetres. The instrument was installed on the top of a very large 
block of lava, deeply buried in ashes, in the corner of a stone hut in which 
we slept. It was covered with a wooden case. Outside of this there was 
a tent made of oiled paper. The instrument was, therefore, well protected 
against currents of air. Before commencing readings the weight was 
suspended for about fifteen hours by the same wire on which it had hung 
for many weeks when in Tokio. To test the effects of moisture in the 
instrument, at the end of the observations I poured a large quantity of 
water all round the block of lava forming the support. This produced 
no visible change in the reading of the instrument. 
The scale from which the readings were made, and which was im- 
mediately below the end of the pointer, was a piece of metal on which 
there were a series of concentric circles at intervals of a millimetre. A 
series of straight lines crossing the centre of these circles gave the points 
of the compass. Had the scale been a series of lines at right angles to 
each other, the readings would have been more.definite. The results of 
interest connected with these observations are :— 
1. That the movements on the top of the mountain were much greater 
than those which I usually observe in Tokio. 
2. The tremors, or slight swing-like movements of the instrument, did 
not necessarily accompany the wind. 
3. That during the heavy south and south-east gales the direction of 
displacement of the pointer was towards the south-east, which is the same 
result as would be obtained if the bed plate of the instrument were raised 
on the south-east side, or if the mountain had tipped over to the north- 
west. 
My colleague, Mr, T. Alexander, treating Fuji as a conical solid made 
of brick, with a wind load of 50 Ibs. on the square foot, found the slope 
and deflection of a point. 100 feet below the apex of the cone. This 
calculated slope was two or three times greater than the greatest 
deflection which I measured. 
As it is difficult to imagine that a mountain could suffer deflection by 
a wind pressure, I will not insist upon the fact that deflection actually 
