pixiitiuD, mill niiotlier snmll 1n)|i wils hei-urvd in a Kiiiiilur mnnn<rr ii|xin tlic short 

 pnrt of the bar wliicli extended nl>ove the knife-edpe, for the juirpow «f inakini; 

 the pendulum liave a liuir-|>eri"»l "f nearly one second The tfitnl len^^th of the 

 |)enduhiin whs 135r. ni. : the bar was 3.8 c. n». wide and .42 e. m thick; the 

 cylindrical Ix^h was 10 o. in. in diameter and 1.1» r m tliick, and it« centre wo» 

 ai>i>roxiinately llt).r)c. m. from tlie knife-i-il^e 



Hotii of these pendulums were vil)nite<l at Tokio in thi- pliytical !al>omtory 

 of the Tniversity Itoth U'fore nn<l after the oliser\ationx at SHp|M)ro The 

 following description applies to the Tokio exjierinient. 



Position and stiKpension of the )K'nduluin. — The ixrululuin wai« Hwuni; in 

 one of the small rooms occtipyint^ the south wiiij; of the physical Ial>oratory. which 

 is well protected from sudden changes of teTn|ierature and from current« of air. 

 In this roon> there is a large stone pier about <)0 c. m. square in horizontal section, 

 built ui«in a solid fonndation and extending to a height nf two metres al>ove tlie 

 ground. A heavy bar of iron about 70 <• m. in length and having a cross 

 section of about 11 c. m. by 2.5 c. m. was place«! on the top of tlii« |>ier and wa« 

 secureil by means of heavy bhtcks of .stone placed alwve it. The end of tlie iron 

 Uir pr'rected just far eiiougli to allow a ri'Stini; plao' for a stout irofi plate which 

 was carefully levelled and upon which the knil'e-edge was made to r»-st. 'I'he 

 position of tlie pendulum was approximately as follows : — 



I^ititude 35° 42' N. 



T>innitude 139- 4(;' K. 



Height above sea level fj metn-s. 



Determination of the time of vibnition. — Time was obtained from a break 

 circuit sidereal chronometer (Negus 10291. The chrononi''ter n-mained in tlie 

 transit room of the astronomical observatory and its U-ats were received through 

 a telegraph line which connects the two jilaci-s. It is immaterial in the present 

 experiment what unit of time we use, and it was found convenient to use sid'-real 

 time throughout the ex|K'iiments. The chronograph iwd was by Alvan Clark 

 and Sons, the same instrument as was used in the Kujiiioyama fXiiedition. A« 

 to its uniformity of spec«! and other desirable charaiteristics we have only to 

 confirm the remarks maile by I'rof. Mendenliall in his report on tlie Fujinoyarna 

 expe<lition. A simple arrangement similar to that used by I'rof Mendenhall 

 was applied to the penibilum so that it could lie made to break the circuit 

 through the chronograph magnet at any desired vibration. Tlie fienAilum. 

 after lieing caused to break the circuit onci-, was swung fp-ely during 20 to 40 

 minutes generally, and was then ag.iin caus».'!! to break th«* circuit. The.«e two 

 breaks, together with seconds marks during the whole interval, were recordifl on 

 the chronograph. The whole numlier of second« reijuire«! for a given numl>er 

 of vibrations Ix-ini; known from pn-liminury ex|K'rimcnts (which had shown that 

 the tinn- of a single vibration of both pendulums was nearly one second», it only 

 remained to determine two fractional parts of a seconil at the Iwginning and end 



