June 9, 1887] 



NATURE 



141 



The following are the principal results for magnetic elements 

 for 1886 :— 



Approximate mean declination 



17° 55' West 



Mean horizontal force... ... j ,.0; -- . 



Mean dip 



57 (in Metric units) 



^ 67° 26' 38" (by 9-inch needles) 



< 67° 26' 45" (by 6-inch needles) 



f 67° 27' 40" (by 3-inch needles) 



The declination and horizontal force magnets were thrown 

 into vibration by the earthquake shock of February 23, the 

 extent of vibration being 20' in declination and 0*004 of the 

 whole horizontal force in that element. The motion commenced 

 at 5h. 37 ■6m. Greenwich civil time, and a second double dis- 

 turbance of much smaller amplitude (possibly accidental) was 

 registered from 7h. 39m. to about 7h. 57m. At the request of 

 M. Mascart, a copy of the photograph has been sent to him for 

 discussion with other records of the earthquake which he is 

 collecting. In view of the importance of the study of earth- 

 quakes, it appears desirable that a suitable seismograph should 

 be procured for the Observatory. 



V. Meteorological Observations. 



The mean temperature of the year 1886 was 48°7, being o°'6 

 below the average of the preceding forty-five years. The highest 

 air temperature in the shade was 89° '8 on July 6, and the lowest, 

 l6°*5, on January 7. The mean monthly temperature in 1886 

 was below the average in January, February (6°), March, June, 

 and December, and above the average in September, October, 

 and November. In the period of 156 days from 1886 December 

 16 to 1887 May 20 the mean temperature was 3°*i below the 

 average of twenty years, the daily temperature being below the 

 average on 1 1 5 days. 



The mean daily motion of the air in 1886 was 291 miles, being 

 7 miles above the average of the preceding nineteen years. The 

 greatest daily motion was 857 miles on December 8, and the 

 least, 56 miles, on October 8. The recorded pressures in 

 1886, exceeding 20 lbs. on the square foot, were 27*6 lbs. on 

 March 31, and 23^5 lbs. on December 9. 



During the year 1886, Osier's anemoiieter showed an excess of 

 about 17 revolutions of the vane in the positive direction 

 N,, E., S., W., N., excluding the turnings which are evidently 

 accidental. 



The number of hours of bright sunshine recorded by Camp- 

 bell's sunshine instrument during 1886 was 1228, which is about 

 twenty hours above the average of the preceding nine years. 

 The aggregate number of hours during which the sun was 

 above the horizon was 4454, so that the mean proportion of 

 sunshine for the year was 0*276, constant sunshine being repre- 

 sented by I. 



The rainfall in 1886 was 24*2 inches, being o'5 inch below 

 the average of the preceding forty-five years. 



VII. Chronometers, Time Signals, and Longitude Opera- 

 tions. 



The number of chronometers now being tested at the Obser- 

 vatory is 225. 



The first seven chronometers in the competitive trial of 1886 

 were exceptionally good, the first chronometer being superior 

 to any we have previously had on trial, except the first in 1882. 



For the annual trial of deck-watches, which commenced last 

 November, fifteen watches were entered, and of these nine were 

 purchased for the Navy, the first three being classed "A," or 

 equal, in performance, to an average box-chronometer. 



A supplementary trial took place in February and March, for 

 which nine deck-watches were entered, and of these seven were 

 purchased for the Navy, the first two being classed "A." 



The watches in each trial were rated for a period of nine 

 weeks, viz. two weeks (dial up) in the room at a temperature of 

 50° to 55°, four weeks in four different positions in the oven 

 (dial up, pendant up, pendant right, pendant left, arranged 

 symmetrically) at a temperature of about 80°, and three weeks 

 (dial up) in the room. When the period of rating in any 

 position was less than a week, weekly rates were inferred from 

 the rate for the period by simple proportion. 



In order to compare the performance of the several watches, 

 "trial numbers," representing deviations in weekly rates, have 

 been formed on the same general principles as for the chrono- 

 meter trials. The trials in different positions introduce, how- 

 ever, a new element, and an arbitrary weight must be assigned 

 to them in combining them with the trials "dial up." It has 

 been considered that when the watch is worn in the pocket the 



pendant will generally be "up," and that not more than one- 

 third of the deviation "pendant right " or "pendant left "is 

 likely to have practical effect. 



Putting a = Difference between greatest and least weekly 



rates "dial up," 

 b = Greatest difference between one week and the 



next " dial up," 

 c = Difference between weekly rates "pendant up" 



and "dial up," 

 </ = Difference between weekly rates "pendant 



right " and " dial up," 

 e = Difference between weekly rates "pendant 



left" and " dial up," 



the quantity c + - + '-may be taken as the measure of the 



3 3 

 deviation in weekly rate due to positions in ordmary wear. 

 Half weight has been given to this quantity in combining it 

 with the trial number "dial up" {a -f 2b), on the assumption 

 that the deck-watch would be usually lying "dial up" and 

 that it would not be carried in the pocket more than 

 eight hours a day on the average. Thus the quantity a -\r 2b 



+ ^ {c -f _ 4- 1 ), has been adopted as the trial number for 



3 3 

 deck-watches. It has been arranged that for the future all 

 pocket chronometers and deck -watches rated at the Observatory 

 after repair shall be tested in positions. 



The automatic drop of the Greenwich time-ball failed on one 

 day only during the past twelve months. On three days the 

 ball was not raised on account of the violence of the wind, and 

 on five days on account of accumulation of snow on the mast. 



As regards the Deal time-ball, there have been twelve cases 

 of failure owing to interruption of the telegraph connexions, 

 and on three days the violence of the wind prevented the raising 

 of the ball. For fourteen days after the snowstorm of December 

 26-27, 110 signals were sent to or received from the Deal time- 

 ball tower, telegrajDhic communication being interrupted. There 

 have been four cases of failure of the I p.m. signal to the Post 

 Office Telegraphs. 



The arrangements for hourly time-signals at Devonport to be 

 given by a local clock, corrected daily by the help of a time- 

 signal at Greenwich at 10 a.m., have been carried out under 

 Captain Wharton's directions, and a return signal from Devon- 

 port (serving as a test of the accuracy with which the local clock 

 was corrected) has been regularly received at Greenwich (at 

 I3h. om. 39s. G.C.T.) since November 22, with the exception 

 of 36 days following the snowstorm of December 26-27, when 

 there was interruption of the telegraphic communication with 

 the West of England, and of 23 days when no return signal 

 was received. The failures occurred for the most part on 

 Sundays. The plan appears to answer well, and it seems 

 desirous that apparatus should be provided by the Government 

 to enable the Committee of Lloyd's to e.-.tablish hourly signals 

 at the Lizard on the same system. 



The new contact apparatus of the Westminster clock was 

 brought into action on 1886 May 22, and the automatic signals 

 from the clock have been- received regularly from that date, 

 except on three days following the snowstorm of December 

 26-27. The error of the clock was insensible on 25 per cent, 

 of the days of observation, is. on 40 per cent., 2s. on 22 per 

 cent., 3s. on II per cent., and 4s. on 2 per cent. On one day 

 the signal was 15s. late, and on another day los. late. 



A suggestion has been made that in view of the importance 

 of the connexion of the British and Continental Surveys, the 

 telegraphic difference of longitude between Greenwich and 

 Paris, which was originally determined with great care in 1854, 

 should be confirmed in order to complete the network of tele- 

 graphic longitudes which have been determined of late years by 

 Continental astronomers. It seems desirable that Greenwich 

 Observatory, which, under Sir G. B. Airy's direction, took such 

 an active part in utilizing the telegraph for the determination of 

 longitude, should now assist in completing the cycle. The 

 necessary exchange of observers and signals could conveniently 

 be carried out in the summer of next year, when the French 

 geodetists will, I understand, be prepared for their share of 

 the work. 



The Report concludes with the following general remarks : — 

 "As the result of an International Congress on Astronomical 

 Photography held at Paris in April on the invitation of the 



