December 26, 1895] 



NATURE 



181 



The remaining installations were on brick columns rising from a 

 -bed of concrete on the alluvium, the localities being chosen with 

 regard to the surface configuration and the proximity or absence 

 of a covering of forest or buildings which would influence the 

 effects of solar radiation. Underground and in the caves, which 

 represent seven out of the nineteen installations, the daily change 

 in temperature was not appreciable on the diagram from a self- 

 recording thermometer. At other stations this change was often 

 Aery great. 



The Instruments. 



The pendulums consisted of a horizontal boom about 5 feet in 

 length, held up by a fine brass wire. They had different degrees 

 of sensitiveness. Usually the adjustment was such that the 

 outer end of the boom was displaced i mm. by tilting the bed 

 plate of the apparatus from o"-i to o"'5. At the extremity of 

 the boom there is a light metal plate with a slit in it parallel to 

 the length of the boom. Underneath this floating slit, but at 

 right angles to it, there is a narrow slit in the top of a box. 

 Light, after passing through the two slits, goes into the box as a 

 point which is received on a drum carrying a photographic film, 

 which usually lasted one week. 



No accuracy of adjustment is required, and if the pendulum 

 is steady without the aid of a single lens, the photographic trace 

 is a remarkably clear line. 



If there are two slits — one broad and the other narrow — in 

 the floating plate, two lines may be produced, one of which is 

 thick, and the other extremely fine. For quick movements of 

 the pendulum, the best diagram is given by the former, whilst 

 for slow displacements the latter has the clearer definition. 



A modification of these arrangements is one which I used in 

 fapan, and which is now in operation at Shide, in the Isle of 

 Wight. In this instrument the boom is built up of straw and 

 reed about 2 "5 feet long, and weighs less than "25 oz. At its 

 outer end there is a small mica plate, which is blackened, and has 

 two slits in it parallel to the length of the lx)om. At a distance 

 of I "5 inches from its inner end, where there is a brass socket 

 and an agate cup, a pivoted weight is so arranged that it balances 

 the outer part of the boom. The apparatus is therefore equiva- 

 lent to an extremely light conical pendulum seismograph, which 

 multiplies tremor-like motion about sixteen times, but which is 

 at the same time very sensitive to changes in the vertical. When 

 the pendulum has a period of sixteen or eighteen seconds, one 

 millimetre deflection of the end of the boom indicates a tilting 

 of o"70. The record is received upon a roll of bromide paper 

 passing beneath a fixed slit in the lid of a box, above which is 

 the boom and mica plate. This moves at a rate of about forty- 

 two millimetres per hour. Should this paper move at a varying 

 speed, this is checked by an ordinary watch so placed upon the 

 lid of the box with the fixed slit, that its long hand every hour 

 crosses freely over a portion of this slit not covered by the float- 

 ing plate, and by eclipsing the light, causes hourly time-marks to 

 be made on the band of bromide. On many occasions I have 

 caused the pendulum to move at known times, and subsequently 

 determined the times at which these disturbances took place by 

 measurements made on the bromide film after de^^elopment. 

 The errors usually varied between three and ten seconds. 



These pendulums have worked well in damp caves and with 

 unskilled assistants. One assistant was a shopkeeper's daughter, 

 twelve years old, whose duty it was to refill and light a benzine 

 lamp every twenty-four hours. The records obtained were very 

 much freer from tremor eflfects than anything I have obtained 

 when working with a pendulum, the boom of which was a two- 

 inch piece of aluminium wire held in position by a quartz fibre, 

 and giving its record by mirrors and lenses. 



The movements recorded have been as follows. 



( I ) The Wandering and Long- period Movements oj Pendulums. 



All the horizontal pendulums wherever situated have slowly 

 wandered from their normal position. Those on the rock 

 have often gradually moved to one side and then returned, 

 the double excursion usually taking from two days to a week. 

 Underground on alluvium within twelve feet of water-level the 

 wandering was at times so great that the spot of light left the 

 film, which was two inches broad, and readjustments were often 

 necessary. P'rom the readings of the end of the pointers, it 

 would appear that in some cases at least, had the pendulums 

 been given sufficient time they might have returned to their 

 starting point. The pendulum in my house, one thousand feet dis- 

 tant from those underground, but thirty -six feet above water-level, 

 although usually more sensitive than those beneath the surface, 



NO. 1365, VOL. 53] 



wandered to a less d^ree. These wanderings might be due to 

 a local warping of the supporting column, but inasmuch as it has 

 generally happened that the periods of great movement and of 

 comparative rest of different instruments have coincided in time, 

 it would seem that the movements are in all probability due to a 

 more general cause. Because great movements have usually 

 been marked (but by no means always) at or after a rainfall, some 

 of them may be attributed to fluctuations in the volume and 

 flow of underground water, the pendulum nearest to this water 

 moving the most. 



The other movements, especially those of the pendulum, 

 situated in my house, and those of the instruments on the rock, 

 have been accompanied by local earthquakes, conveying the 

 impression that they represent actual rock bending, the earth- 

 quakes being interruptions in the process. As the number of 

 observations are limited, and as I am not in a position to deter- 

 mine which of the earthquakes recorded in Tokio are of local 

 origin, and which originated at a distance, although the materials 

 for making this determination have been accumulated at the 

 Central Observatory in Tokio, the matter is one requiring further 

 attention. 



Two pendulums on the alluvium, four hundred yards distant 

 from each other and very diflTerently installed, for periods varying 

 between four and forty days, have moved at the same time in 

 the same directions. 



These movements may therefore be due to some general cause. 

 A rough agreement with a barometric curve suggests the idea 

 that the cause is due to variations in atmospheric pressure, but 

 it is equally possible that they may be due to a greater evapora- 

 tion or precipitation of moisture on one side of these stations 

 than upon the other sides. 



(2) Diurnal Waves. 



By a diurnal wave I mean a deviation of the vertical during 

 one portion of a given period of twenty-four hours in one 

 direction, followed by a retrograde motion during the remaining 

 portion of this interval. On rock foundations my instruments, 

 which never had a sensibility exceeding i mm. deflection for a 

 tilt of o"' I, have failed to show such a movement. With in- 

 struments having a greater sensibility it is likely that its presence 

 would be detected. These movements, which may, for example, 

 be shown by a westerly displacement of a pendulum through a 

 distance of from 2 to 40 mm. from about 6 a.m. until about 

 3 p. m. , and an easterly motion until the following morning, may 

 be best studied from diagrams on films which have moved at a 

 rate of about 3 or 4 inches per day. 



In the underground chamber in the alluvium, within 12 feet 

 of the surface, where the daily change in temperature was less 

 than 1° C, the daily waves were often marked. On the surface 

 at my house, on a good foundation protected by a long building, 

 the wave was marked, but slight. 



At stations protected on all sides by high trees from the sun it 

 was feeble. At all other stations, where on one side at least 

 there was open ground exposed to the sun, with the exception of 

 rainy or cloudy days, when the photographic trace was a straight 

 line, it was very marked, the deflection sometimes amounting to 

 40 viillimetres. On one hill, from about 6 a.m. to 6 p.m. the 

 motion was westwards, while at the other side of a swampy 

 valley running N.N.W. the movements were at the same hour 

 almost in a contrary direction. It would appear that on every 

 fine day the trees on the hills on the two sides of this valley 

 bow first towards and then away from each other. 



For the present I am inclined to regard these phenomena as 

 being due to a tilting of the surface in consequence of the 

 evaporation of moisture. This effect extends to some depth. 



On open ground in the morning more moisture is taken 

 away from the eastern side of a station than from the westerrt 

 side, and therefore the eastern side, being relieved of a load, 

 rises, and the pendulum swings westwards. 



In the afternoon this action is reversed, and the pendulum 

 turns eastwards. 



The waves due to this effect difter in time of occurrence and 

 in amplitude according to the character of the locality in which 

 they are observed. 



The movement at night is slight. It may be a continuation 

 of the sun effect modified by the precipitation of dew from the 

 atmosphere, and by the condensation of aqueous vapour rising 

 in ground heated during the day but chilled on its surface after 

 sunset. The sub-surface condensation sometimes represents a 

 load one-tenth of that taken away by evaiwration. Like dew 



