﻿21 



Ohsfi-nifion.-: of Xuiriiil^r 20 mnJ 27, 1901. 



These were taken with all three pendulums, Nos. ."ST, 38, 39, swung at atmosphoiie pressure, the 

 pressure being recorded l)y a mercurial barometer lent by Dr. Evans of Canterbury College, N.Z., 

 particulars of the errors of the instrument being supplied. 



The cistern of the barometer was placed on a level with the pendulums. 



A break-circuit sidereal chronometer. No. 6711, by KuLLBEUG, London, was used throughout, and its 

 rate was determined by telegraphic signals from the Astronomical Observator}' at Wellington, N.Z. 

 Eleven signals were sent at intervals of 30 seconds, commencing at 9.30 p.m. on the evenings of 

 November 25, 26, 27. The results arc as follows : — 



November 26, chronometer lost, in 24 hours, 0'87 second sidereal. 

 27, „ „ „ 0-66 



The arrangement of observing was as follows : — 



Twelve coincidences were observed, six to the right and si.x; to the left, then an interval of fifty 

 coincidences was allowed to pass, and then twelve more coincidences were observed. During this process 

 four readings of pressure, temperature, and arc were obtained. The penduhmi was then left swinging 

 for two hours, and then the same observations repeated. 



All three penduhmis were swimg in this manner. On the following day the case was taken off, the 

 pendulums reversed on the agate planes, and the whole process of observing repeated. 



The same thermometer. No. 753 (Centigrade), was used throughout the observations in noting the 

 temperature of the pendulums, of which the corrections were determined at the National Physical 

 Laboratory. 



Ohxcrvationf: of May 30, 190-1. 



These were conducted in the same house as those of 1901, and the method of observing was the same. 

 The pendulums, however, were not swung at atmospheric pressure, but at a considerably reduced pressure. 

 The manometer employed was an open U-tube kindly lent by Dr. Evans of Canterbury College. 



The barometer used in connection with the manometer was No. C 895, by Hicks, London, whose 

 corrections were known and applied. 



The rates of the chronometer. No. 6711, were determined by telegraph signals from Wellington 

 Observatory, and taken as - 1 ■ 20 seconds daily. 



The Antarctic Observations. 



Winter Harbour. Latitude 77" 50' 50" S. Longitude 166" -t-i' -45' E. 



Olmerralions of July 31 and August 1, 1902. 



The pendulum apparatus was set up in a wood hut on shore at Winter Harbour in a small room 

 partitioned off from the rest, where the temperature was kept as uniform as possible by means of lamps. 



A brick pillar firmly cemented was sunk below the surface of the frozen earth to a depth of 1 foot, and 

 rose 3 feet above the surface. The section of the pillar was 2 feet by 2 feet. 



Upon this the pendulum stand was placed, small holes being drilled into the surface of the Ijricks to 

 take the legs. 



The agate planes were then carefully le\'elled l)y means of the two small levelling pendulums, and the 

 pendulums were placed in the supporting V's, No. 39 at the back, No. 37 on the right, and No. 38 on the 

 left, the two latter being at right angles to the first. The light from the mirrors of the two pendulums 

 at the sides is thrown into the observing telescope by means of two prisms placed near the centre. 



