HOROLOGY. 
351 
contrate-wheel driving another of the fame number round 
an outer cannon in the centre. 
The firft time-keeper which Harrifon made was in 1726, 
which Dr. Hutton afferts did not err a fecond in a month 
for ten years together ; but the firft time that one was 
publicly tried, was in a voyage to Lifbon in 1736 ; being 
placed in a box, hung in gimbols, it fully anfwered his 
expectation, and corre&ed the dead reckoning about a 
degree and a half; in confequence of which, according to 
Dr. Mackay, the board of longitude granted him a gra¬ 
tuity, and defined him to profecute his labours. 
In 1739 l ie finished a fecond piece more perfect than the 
firft ; and in 1749 according to Dr. Mackay, (but accord¬ 
ing to Dr. Hutton in 1758,) a third, which was pro¬ 
nounced more fimple in its conftruCIion than either of the 
former. But his labours did not flop here; in 1761, his 
fourth piece, of which we have given a deficription, was 
produced for trial ; and Mr. 'William Harrifon, the fion, 
offered to take charge of it, in a voyage to and from Ja¬ 
maica, which was accordingly performed in this and the 
following years. Mr. Robertlon, mailer of the academy 
at Portfmouth, was fixed upon to take tire rate of this 
piece, which he did, and reported that, on the 6th of 
November, 1761, at noon, it was 3 s flow after having loft 
24 s in nine days on mean Polar time. The Deptford, in 
which fhip the voyage was made, left Portfmouth on the 
18th of the fame month, and arrived at Madeira on the 
9th of December following, when it was found that the 
reckoning vras corrected by the time of the piece'about a 
degree and a half. In the run from Madeira to Jamaica 
the reckoning was corrected 3 0 ; and at the feveral iflands, 
where the fliip touched, the known longitudes agreed 
very nearly with thofe given by the time-keeper. Jan. 
19, 1762, the fliip arrived at Jamaica ; the time of mean 
noon was obferved, by equal altitudes, at Port Royal, on 
the 26th of the fame month, which, according to the 
piece, was q h 59™ 7 s- 5 ; but the original error on the 6th 
of November, 81 days 5 hours before, was 3 s flow', this 
quantity, therefore, was to be applied as a correction, 
together w ith the accumulation of the daily error of— 
in 8 i<i 5 h , viz. 3 m 36 s- 5 ; this fum of 3™ 39 s- 3 added to the 
time indicated, which, it has been faid, was 4 h 59 m 7 s *5, 
make 5 h 2 m 47 s for the difference of longitude between 
Portfmouth and Port Royal; which determination was 
only 4 s of time lefs than the determination at Kingfton 
of the fame, from the tranfit of Mercury over the fun’s 
difc. This fimall error in time cdrrefpor.ds to lefs than 
one nautical mile in the parallel of Jamaica. 
The Merlin, on-board of which the piece w'as now put, 
fet fail from Jamaica on the 28th of January, 1762, and 
experienced fuch a violent ftorra in the paflage, as obliged 
young Harrifon to remove his piece into an expofed fitu- 
ation; however, the fliip arrived at Portfmouth on the 
26th of March, and on the 2d of April the time of mean 
noon was found, from equal altitudes, to be n h 51“ 3i s- 5, 
to which its former error of 3% together with-the accu- 
24 s X i47 d 
initiation of the daily error, viz. - : —• = 6 m 32s, be¬ 
ing added, make the time of mean noon by the time¬ 
keeper u h 58 m 6 5- 5. From this report it appears, that 
from Nov. 6th, 1761, to April 2d, 1762, though the piece 
had experienced many violent agitations at lea, and had 
been expofed to great changes of temperature, the whole 
error amounted to only i m S3 s ‘S, or 28^', of longitude on 
the equator, which quantity (one fecond lefs than Dr. 
Hutton has ftated) is not quite 18 nautical miles in the 
parallel of Portfmouth. 
Though various objections were made to this trial, 
principally arifing from the obfervatio.os by which the 
longitude of Portfmouth and Jamaica had been afcer- 
tained, yet Harrifon, we are informed, obtained a reward 
upon it from parliament of 5000I. and was ordered to 
make a fecond trial to Barbadoes. But, previoufly to the 
iecond trial to the Well Indies, the board of longitude, 
on the 17th of Auguft, 1762, wilhed to place Harrifon’s 
piece in the hands of the aftronomer royal, at that time Mr. 
Blifs, for trial at the Obfervatory, which with was not 
complied with, by reafon of fome alteration to be made. 
The fame wifh was repeated by the board at their fitting 
on the 4th of Auguft of the year 1763, which was again 
not complied by the junior Harrilon. However, on being 
defired to fend the rate of going of his time-keeper, fealed 
up, to the fecretary of the admiralty, previoufly to bis 
failing, he confented to this requeft, and propoled to abide 
by the fealed rate on the trial to Barbadoes, which had been 
propofed. Accordingly, after having compared the time¬ 
piece with Mr. Short’s regulator, in Surrey-lfreet, London, 
which had its error newly afcertaihed by an excellent 
tranfit-inftrument, Mr. Harrifon junior went 011-board the 
Tartar on the 13th of February, 1764, and proceeded to 
Portfmouth, where he again compared it with an aftrono- 
mical clock in Mr. James Bradley’s temporary obfervatory. 
It may not be foreign to our purpofe to mention here, 
that the obfervatory juft fpoken of was fitted for the ex- 
prefs purpofe of obferving the eclipfes of Jupiter’s fiatel- 
lites, as well as of keeping the clock in juft time, in or¬ 
der that, to avoid future objections, the obfervations to 
be made by Meflrs. Maflcelyne and Green, at Barbadoes, 
on the fame fatellites, particularly the firft, when com¬ 
pared with Bradley’s, might ascertain the comparative 
longitudes of thefe two places, which was accordingly de¬ 
termined to be 3” 54 m 20 s . 
Before leaving Portfmouth, which took place on the 
28th of March, 1764, Mr. Harrifon took the rate of his 
time-keeper by equal altitudes, employed between the 
29th of February and the 26th of March ; and on April 
the 18lh found, from comparing his obfervations of the 
fun at 4 P. M. at the fliip with the corrected time given 
by it, that the fliip was at that inftant only 43 miles eaft- 
ward of Porto Santo, in confequence of wdiich‘determi¬ 
nation, fir John Lindlay, the matter, fleered accordingly, 
and faw the ifland in queftion before him at one o'clock 
the next morning, agreeably to expectation. Oil the 13th 
of May, the vefl'el arrived at Barbadoes, and on the four 
following days its error was afcerfained by Meflrs. Maf- 
kelyne and Green by equal altitudes of the lun, and alfo 
by a comparifon with the aftronornical clock at the ob¬ 
fervatory near Bridgetown ; and it was found that tiie 
amount of the daily deviations from mean-time was only 
43 s in excefs, or 10-J' of a degree of longitude. Mr. Har¬ 
rifon fet out on his return from Barbadoes in the New 
Elizabeth on the 4th of June; and, arriving at Surrey 
{fairs on the Thames on July the 18th, found, from a. 
comparifon with Mr. Short’s clock, the error of which had 
been afcertained on the very day, that the whole gain hi 
the 156 days was only 54 s , allowing the fealed rate , of one 
fecond gain per day, as a correction ; and it has been ob¬ 
ferved, that, if the allowances had been moreover made 
for the ftate of the thermometer, as ftated in the declara¬ 
tion, the piece in that cafe would have been found to 
have been 15 s only at variance with mean time, and this 
in the oppoiite extreme. 
Soon after this very fatisfr&ory trial, a committee of 
feven fcientific gentlemen and mechanicians were appointed 
by the board of longitude for examining the principles of 
Mr. Harrifon’s time-keeper, whole report was as follows: 
viz. “ That Mr. Harrifon has taken his time-keeper to 
pieces, in prefence of us, and explained the principles and 
confirmation thereof, and every thing relative thereto, to 
our entire fatisfaCtion; and that he alfo did, to our fatif- 
faCtion, anfwer every queftion propofed by us, or any of 
us, relative thereto; and that we have compared the 
drawings of the fame with the parts, and do find that they 
perfectly correfpond.” The committee were th? Rev. N, 
Maflcelyne, Rev. John Mitchell, Rev. William Ludlam, 
Mr. John Bird, Mr. Thomas Mudge, Mr. Larcum Ken¬ 
dal, and Mr. William Matthews.. Mr. Harrifon had then 
another 5000I. ordered him, with a promile that the refi- 
due of the whole parliamentary reward; which, by the aCt 
of queen Anne, was 2o,oool. Ihould be given liirn when a. 
1 proper 
