346 H O R O 
pafTmg between two little pulleys, which on either fide 
convert the direftion of the chain to the line of traftion 
of the fpring. 
Hence it is evident this fingle flat ring performs all the 
following operations, i. It receives the motion for ftrik- 
ing the hour from th?pendant. 2. The fame for ftriking 
the quarters. 3. It carries the pins, or teeth, which lift 
the hour-hammer. 4. The fame for the quarter-hammer. 
5. It contains the indentations by which the hour-fnail 
operates on it by its lever. 6.The fame by which the quarter- 
fnail operates on it. 7. It carries the part that recoils the 
movement which tells the hour to its firft pofition. S. 
It carries the part, for the fame purpofe, for the quarter- 
movement. 9. It contains a cavity, which moves over a 
fixed pin, that prevents the pendant from turning it too 
far. Thus in this ring, the fame parts, in three infiances, 
are made to perform double operations ; by which fim- 
plicity of conftrudtion is advanced, apparently to its 
greateft extent. 
For' this very ingenious and ufeful invention, Mr. El¬ 
liot was complimented with a prefent of thirty guineas 
by the Society of Arts: and, though we have pretty 
clearly explained the principle of it above, we hope to 
make its mode of adtion better underftood by a delinea¬ 
tion of the machine. The whole of the repeating work 
is fliown at fig. 88 and 89; the reference will be in 
general to fig. 88; but, as the fame letters denote the 
fame parts in both, whenever the part is not clearly feen 
in that figure, the deleft will always be fupplied by re¬ 
ferring to fig. 89. Let then AB, in fig. 88, repreient the 
pillar-plate, viz. its upper fide. C D is a circle of fteel, 
moving freely in the fteel pulleys EFGH. This fteel 
.wheel is put in motion by means of a fmall wheel a , faft 
on the axis, which goes through the pendant, which adts 
in the teeth on the under fide of the. circle reprefented 
by the arch b. I is the quarter-fnail. K the minute- 
wheel. The hour-wheel may eafily be fuppdfed, therefore 
requires no defcription. L is part of the hour-fnail, which 
is not drawn complete, as it would intercept the view of 
the rack the fteel circle at Q. O is the locking-lever. 
P the end which falls-upon the hour-fnail, to regulate the 
number of ftrokesit is to ftrike ; at the fame time, the 
end Q of the lever O locks into one or other of the 
notches in the fteel circle. The ftar-wheel, 1, is riveted 
or fcrewed to the hour-fnail, and moved by the fmall arm, 
2, on the canon pinion. M and N are the jumper and 
fpring. R is the quarter locking-lever. S ’.he end that 
falls upon the quarter-fnail. T the locking end, which 
locks into the notches at h. .At X is a pin in the fteel 
circle, to which is hooked a chain, and the other-end to 
the barrel V. This chain, when the pendant is put in 
motion, will unwind from the barrel, and, by acting be¬ 
tween the (feel rollers U W, will work both to the right 
and left, i, /;,-are fmall pallets, on which are the tails of 
the hammers afted upon by the pins as per figure, fo 
that, at the end of the ftrokes that have {truck, the fpring 
in V will draw back the fteel circle, and the pins 'will pals 
by fche hammer-tail, and a fmall fpring under l will bring 
it into its proper place. 
The method of ufing this repeater is as follows :—Hold 
the watch in the left hand, and apply the finger and 
thumb to the pendant; and, if it is required,to repeat 
the hour, turn it to the right till it gives a full itop. 
But, if it is to repeat the quarter, turn the pendant to the 
left till it gives a full ftop, as in winding up a common 
watch. In either cafe, the point P will fall on fome part 
of the hour-fnail; at the fame time the point Q will.ftop 
'the fteel circle. The hammer of courfe will give the 
hour, or the quarters, as the cafe may be. 
A Ruffian peafant, named Kulibi, has lately, made, a 
watch of Angularly curious workmanfhip. It is of the 
iize of an egg Within it are reprefented the fepulchre 
of our Saviour* the ftone clofing it, and the guards hand¬ 
ing nigh. The watch- is a repeater. When the hour 
itrikes, the ftone is fudden.ly rolled from the mouth of the 
LOGY. 
fepulchre, the guards fall on their faces, angels appear, 
the women arrive, and that melody is played which it is 
ufual to execute in Ruflia on Eafter-eve. 
III. Of TIME-KEEPERS. 
The method of determining the longitude by time¬ 
keepers pofiefles fome advantages beyond any other. In 
the firft place, it is attended with very little trouble; and, 
fecondiy, the longitude can be found oftener by a time¬ 
keeper than by any other means. There is not perhaps 
a more ftriking'proof of this pofition, than in the voyage 
of Peroufe round tho world in the fhip La Bouffole. Be¬ 
tween the ift of Auguft 1785, and the 8th of September 
1787, this celebrated navigator took only 72 lunar obfer- 
vations; but he took 393 obiervations on the longitude 
by his time-keeper. This is an unequivocal proof of th» 
benefits that navigation derives from this mode of finding 
the longitude. 
Could a clock or watch, therefore, be contrived to go 
uniformly at all 'feafons, and in all places and iituations, 
l’uch a machine, being regulated, for inftance, to London 
or Greenwich time, would always fhow the time of the 
day at London or Greenwich, wherever'it fliould be car¬ 
ried to; then, the time of the day at this place being 
found by obfervation, the difference between thefe two 
times would give the difference of longitude, according to 
the proportion of one degree to 4 minutes of time. See the 
article Astronomy, vol. ii. p. 454. 
It is to be regretted, however, that time-keepers are ex- 
penfive, and liable to ftop or go irregularly. Were it not 
for thefe inconveniences, no other method of finding the 
longitude need be fought after. But, in fpite of all the 
care which can be beltowed upon the execution of time¬ 
pieces, the vibrations, of the balance become lefs after a 
few months ufe; the main-fpring gives a little; friflion 
will increafe in the wheel-work, whether by the wear of 
the different parts or by the coagulation of the oil: and, 
iia proportion as thefe changes take place, the regulator 
receives weaker impulfions. We do indeed fucceed in re¬ 
ducing thefe variations to a fmall quantity by executing 
thevarious parts of the machine with the greateft nicety; 
but it cannot be done without great cxpence, fince firft- 
rate workmen are few in number; neither can any care 
or expence remedy the evil entirely. But let it be recol- 
lefted, that we (hould feek for perfection rather in the 
correclnefs of the principle of conffruction than in the 
nicety of execution, though the latter is not to be neglect¬ 
ed. How near we have approximated to the defired end, 
the following detail of fucceffive improvements will fhow. 
The firft perfon who propofed to afeertain the relative 
longitude of any place or Illip at fea, by means of an ho- 
rological machine for indicating the time of the firft me¬ 
ridian, was, as has been aflerted, Gemma Frifius, about the 
year 1530, in a trait, De Principiis Aftronomise et Cofmo- 
graphiae.. This method was deferibed and recommended 
in Carpenter’s Geography fo early as the year 1635 ; but 
the ftate in which liorological machines was, at that time, 
prevented his accomplifhing tlie defign. The idea, how¬ 
ever, once fuggefted, was valuable; and ftimulated inge¬ 
nious mechanics, in times fucceeding, to attempt the ac- 
complifhment of an objeCt of fuch national importance. 
The difeovery of the ifochronifm of the pendulum turned 
the minds of ingenious men to the improvement of clocks ; 
and we find that lord Kincardine tried a marine pendu¬ 
lum clock by Dr. Hooke in the year 1662; and that 
Chriftian Huygens, the celebrated Dutch mathematician 
and mechanician, contrived a time-keeper, aduated by a 
fpring, and regulated by a pendulum, which was tried at 
fea by major Holmes in. the year 1664, and fproken of by 
him in favourable terms. The fcapement was of the 
crown-wheel kind, which, from its nature, is almoft con- 
lcantly under the influence of the maintaining-power; but 
a fmall weight, connected with the crown-wheel, was 
raifed every half-fecond by the maintaining-power, and 
gave an impulfe to the pendulum, which, therefore, was 
not 
