House <y Garden 
Fi&.lII 
in the morning and eight o’clock in the even¬ 
ing; but for ordinary practical use from six 
o’clock in the morning to six o’clock in the 
evening covers all that is needed. 
In laying out a dial in this way, no allow¬ 
ance is made for the width or thickness of 
the stile or gnomon. If a thin gnomon is 
used, that is, of metal I/16 of an inch thick, 
it is scarcely necessary to make any allowance ; 
but if a heavy gnomon is to be employed, 
having, say, a thickness of 3/16 or 1/4 of an 
inch, then, instead of the single line AE 
(Figure II), there must be two 
parallel lines the same distance 
apart as the thickness of the gno¬ 
mon. In this case, instead of 
striking the circles from the cen¬ 
tral point E, two semicircles must 
be made, having their centers 
where the line Cl) intersects the 
two parallel lines; or what would 
probably be an easier method 
would be to cut into two equal 
parts the preliminary diagrams we 
have been describing and to place 
between them a strip of paper the 
exact thickness of the gnomon to 
be used. 
As the hours about the middle 
of the day are closer together than 
those early in the morning or late 
in the afternoon, it makes a much 
better looking dial to shift the 
center towards the twelve o’clock 
mark and to draw a new circle 
from this point. The lines radiat¬ 
ing from E should be extended to 
this new circle and the gnomon 
increased in proportion. That 
the dial should give the best 
results, a practical rule for the length of the 
gnomon is that the upper tip of its sun edge 
be directly over the outer line of the border 
containing the figures of the hours, (see 
Figure IV). The center of the new circle 
should not be moved, however, from side to 
side, but must always be on the line AE 
midway between the two six o’clock points, 
as shown in Figure III. In setting the dial, 
supposing it to be a horizontal one, great 
care should be taken that it should always 
be perfectly level, i. e. parallel with the plane 
of the earth, and that the gnomon should 
incline neither to the right nor to the left, 
but should point always to the true north, 
not the north of the magnetic meridian. 
The foregoing is the simplest of all prob¬ 
lems in dialling, as the gnomon pointing to 
the north and casting its shadow upon a 
level plane surface are the easiest given con¬ 
ditions. When these conditions are changed, 
mathematical principles may be carried into 
almost infinite complexity, as, for instance, 
in the case of the dial-face placed at an 
unusual angle with both the earth and 
5 1 7 
