HOROLOGY. qqq 
dow ; for the number belonging to that divifion will be 
the hour. 
This dial however is attended with a very great incon¬ 
venience: as the boundary between the light and Iha- 
dow is always badly defined, it cannot be precilely known 
where it terminates; it will therefore be better to employ 
this dial in the following manner. Adapt to this globe a 
half-meridian, made of a piece of flat wire, feven or eight, 
lines in breadth, and half a line in thicknefs, and move¬ 
able at pleafure around its axis, which mu ft be the lame 
as that of the globe. Then, when you wilh to know the 
hour, move the half-meridian in luch a manner, that it 
fhall projeft the leaft lhadow poflible, and ^his lhadow will 
Ihow the hour on the equator. In this cafe however it is 
evident that the numbers naturally belonging to the points 
of divilion in the meridian flioul’d be infcribed on them; 
that is,- XII. at the meridian, I. at the following divifion, 
towards the weft, and fo on. 
Ring Dial, in an armillary fphere. —This dial is equally 
Ample as the preceding, and is attended with this advan¬ 
tage, that it may ferve by way of ornament in a garden. 
Conceive an annillary fphere, (fig. 16,) confifting only 
pf its two colures, its equator, and zodiac, and furnilhed 
with an axis palling through it. If we fuppofe this fphere 
to be placed on a pedeftal, in fuch a manner that one of- 
its colures (hall fupply the place of a meridian, and that 
its axis fhall be directed towards the pole of the place, it 
is evident that the lhadow of this axis, by its uniform 
motion, will fhow r the hours on the equator. If the equa¬ 
tor therefore be divided into 24 equal parts, and if the 
numbers belonging to the hours be infcribed at thefe di- 
vifions, the dial will be conftrufted. 
But, as the equator in general is not of fufficient thick- 
. nefs, the hours muft be marked on the inlide of the zone 
which reprefents the zodiac, and which on that account 
fhould be painted white. In this cafe, care muft be taken 
not to divide each quarter of the zodiac into equal parts; 
for the lhadow of the axis, which pafles over equal arcs 
on the equator, will rafs over unequal ones on the zodiac: 
thefe divisions will be narrower towards the points of the 
greateft declination of that circle ; fo that the divifion in 
the zodiac neareft to the folftitial colures, inftead of 15 0 , 
which are equal to the interval of an hour on the equator, 
ought to comprehend only i3°48'; the fecond 14 0 6'; the 
third 14 0 38'; the fourth 15 0 17'} the fifth 15 0 54'; and 
the fixth, or that neareft the equinoxes, 16° 17k It is in 
this manner that the zodiacal band, on which the hours 
are marked, muft be divided ; otberwife there will be fe- 
veral minutes of error ; but each interval may be divided 
into four equal parts for quarters, without any fenfible 
error. Tranlverfal lines may then be drawn through the 
breadth of the zodiac, taking care to make them concur 
in the pole. 
To conjlruf! a Solar Dial,’ by means of which a blind perfon 
may know the hours .—This may appear a paradox ; but we 
fhall fliovv that a fun-dial might be creeled near anliofpital 
for the blind, by which its inhabitants could tell the 
hours of the day. If a glafs globe, 18 inches in diameter, 
be filled with water, it will have its focus at the diftance 
of 9 inches from its furface; and the heat produced in 
this focus will be fo confiderable, as to be fenfible to the 
hand placed in it. This focus alio will follow the courfe 
of the fun, fince it will always be diametrically oppofite to 
it; and therefore, to conftrufttbe propofed dial, we may 
proceed as follows. Let the globe be furrounded by a 
portion of a concentric fphere, 9 inches diftant from its 
furface, and comprehending only the two tropics, with 
the equator, and the two meridians or colures; and let 
the whole be expofed to the fun in a proper pofition ; that 
is, with the axis of the globe parallel to that of the earth. 
Let each of the tropics and the equator be divided into 
,24. equal parts; and let the correlponding parts be con- 
nefted by a fmall bar, reprefenting a portion of the hour- 
circle comprehended between the two tropics. By thefe 
means all the horary circles will be reprefented in fuch a 
manner, that a blind perfon can count them, beginning 
at that which correfponds to noon, and which may be 
eafily diftinguilhed by lbme particular form. When a 
blind perfon then willies to know the hour by this dial, 
he will firft put his hand on the meridian, and count the 
hour circles on the bars which reprefent them ; when he 
comes to the bar on which the focus of the folar rays fall, 
he will readily perceive it by the heat, and confequently 
will know how many hours have elapfed fince noon; or 
how many muft elapfe before it be noon. Each interval 
between the principal bars, that indicate the hours, may 
be eafily divided by fmaller ones, in order to biave the 
half-hours and quarters. 
Method of arranging a Horizontal Dial, conJlruBcd for any 
particular latitude, in fuch a manner as to make it flow the hours 
in any place of the earth. —If the latitude of the place be lefs 
or greater than that of the place for which the dial has 
been conftrufted, after expofing it in a proper manner, 
that is, with its meridian in the meridian of the place, and 
its axis turned towards the north, nothing will be necef- 
fary but to incline it til! its axis forms with' the horizon 
an angle equal to the latitude of the place in which it is 
to be ufed. Thus, for example, if it has been conftrufted 
for the latitude of Paris, which is 49 0 50', and you wilh 
to employ it at London, in latitude 51 0 31'; as the dif¬ 
ference of thefe two places is i° 41', the plane of the 
dial muft make with the horizon an angle of i° 41', as 
feen in the figure, (fig. 17,) where SN is the meridian, 
ABCD the plane of the dial, and ABE, or abe r the 
angle of the inclination of that plane to the horizon. If 
the latitude of the primitive place of the dial be lefs than 
that of the placfe for which it is ufed, it muft be inclined 
in a contrary direction. 
Another Method .—When this fecond method of rendering 
a horizontal dial univerfal is employed, the hour-lines 
muft not be deferibed on it, but only the points of divi¬ 
fion in the equinoftial line. In regard to the ftyle, it 
muft be moveable in the following manner. Let ABC, 
(fig. 18,) reprefent the triangle in the plane of the meri¬ 
dian, where NBC is the axis or oblique ftyle, and AB 
the radius of the equator. The ftyle muft be moveable, 
though it always remain in the plane of the meridian, fo 
that the radius AB of the equator, having a joint in the 
point A, may form the angle B AC equal to a given an¬ 
gle ; that is, equal to the complement of the latitude. For 
this reafon a groove muft be formed in the meridian, fo 
as to admit this triangle to be raifed up or lowered, al¬ 
ways remaining in the plane of the meridian. When 
every thing has been thus arranged, to adapt the dial to 
any given latitude, fuch as that of 51° 31', for example, 
take the complement of 51 0 31', which is 38° 29', and 
make the angle B A 0=238° 29k The ftyle then will be 
in the proper pofition, and the dial being expofed to the 
fun, with its meridian correlponding to the meridian of 
the place, the lhadow of the ftyle, which ought to be 
pretty long, will Ihow the hour at the place where it in- 
terlefts the equinoftial. 
Refecting Dial .—A dial to Ihow the hours by reflection 
may be deferibed in the following manner, on a dark 
wall or ceiling. Defcribe a dial on a horizontal plane, 
that can be illuminated by the rays of the fun, fuch, for 
example, as the bottom of a window; but in fuch a man¬ 
ner, that the centre of the dial may be towards the north, 
and the equinoftial towards the fouth ; which will give 
to the hour-lines a pofition contrary to that which they 
ought to have in common horizontal dials. When the 
dial has been thus conftructed, and furnilhed with a fmall 
upright ftyle, apply a piece of thread to any point at plea¬ 
fure of one of the hour-lines, and extend it over-the end 
of the fl^de, till it reach any point of the wall or ceiling ; 
this point will be one of thole of the hour-line to which 
the end of the thread was applied. If four or five points 
be determined, in the fame manner, for each hour-line, 
by then drawing lines through thefe points, the required 
dial will be conftrufted. 
' 3 
To 
