M A G N E T I S M. 
as' reprefented by K L, on knocking the table, w ill gra¬ 
dually recede from the point G, (bowing as if the mag¬ 
net repelled it; which has given to this experiment the 
•name of magnetical paradox ; for, in fact, the magnet at¬ 
tracts the wire. 
This phenomenon refults from the directive property of 
the magnet acting at a greater diitance than the attrac¬ 
tive. In order to explain the immediate caufe of this 
phenomenon, it ■ np,uIt be confidered, that the wire K L, 
(fig. 30.) being rendered magnetic by the action of the 
magnet H, is inclined to it according to the laws of the 
dipping-needle ; but, on account of its weight, and be- 
caufe it is fupported not in its centre, but by one end, 
namely K, which (lands upon the table, it does not in¬ 
cline fo much as it ought to do, if it were freely fufpended 
by its centre, the end, K, now being a little higher than 
its proper fituntion. Let MN be the perpendicular, 
w hich pafl'es though the centre of the wire. Now, when, 
by the motion given to the table, the wire is tnade to 
jump; this, whilft remaining in the air, will take its pro¬ 
per inclination, as fhown by r Q, its centre remaining in 
the fame perpendicular M N ; for the directive power of the 
magnet, H, a£ts at a greater diitance than its attraflion. 
In this fituation it is evident, that a perpendicular P O, 
let fall from the lower extremity, r, of the wire, touches 
the table in a point farther from G than the point K ; 
and, as the wire after the jump comes down to the table 
again with the proper inclination, viz. parallel to r Q, it 
follows, that now its lower end mull touch the table at 
O; and thus every knock will force it to recede a little 
more from the point G, which lies juft over the magnet 
H. The fame explanation, applied to the firft part of the 
experiment, will (how that the wire mud in that cafe, viz. 
when the magnet is held above the table, approach con¬ 
tinually the point G. This experiment may be diverfi- 
fied by ufnvg iron-filings, inftead of the iron wire; for, 
in the firft cafe, the filings difperfed over the table will 
be gradually collected about the point G ; and in the lat¬ 
ter cafe, the filings placed about the point G, will be gra¬ 
dually forced to recede from that point. 
8. The Cylindrical Oracle.— Provide a hollow cylinder 
about fix inches high, and three wide, as A B, (fig. 31.) 
Its cover C D mult be made to fix on in any pofition. 
On one fide of this box or cylinder, let there be a groove, 
nearly of the fame length with that fide ; in which place 
a (mall fteel bar that is ftrongly impregnated, with the 
north pole next to the bottom of the cylinder, as S, N. 
On the upper fide of the cylinder deferibe a circle, and 
divide it into ten equal parts, in which are to be written 
the numbers from 1 to 10. Place a pivot at the centre of 
this circle, and have ready a magnetic needle. Then pro¬ 
vide a bag in which there are leveral divifions. In each 
of thefe divifions put a number of papers, on which the 
fame or fimilar queftions are to be written. In the cylin¬ 
der put leveral different anfwers to each queftion, and feal 
them up in the manner of (mall letters. On each of thefe 
letters or anfwers is to be written one of the numbers of 
the dial or circle at the top of the box. You are fuppofed 
to know the number of anfwers to each queftion. Then 
offer one of the divifions of the bag (dbferving which 
divifion it is) to any perfon, and defire him to draw one of 
the papers. Next put the top on the cylinder with that 
number which is written on the anfwer direitly over the 
bar. Then defire the perfon who drew the queftion to 
obferve the number at which the needle (lands, and to 
fearch in the box for a paper of the fame number, which 
he will find to contain the anfwer. The experiment may 
be repeated by offering another divifion of the bag to the 
fame or another perfon ; and, placing the number that 
correfponds to the anfwer over the magnetic bar, proceed, 
ing as before. 
9. The learned Fly, or the Syren. —This recreation is fome- 
what more complex than the preceding, and depends 
partly on philolophical principles, and partly on a little 
deception. You mult provide a table, with a box 1'unk 
m 
in its thicknefs, and the box muff be fu mi (lied with a- 
broad brim, inferibed with numbers, the hours of the day, 
or anfwers to certain queftions. You then delire a perfon 
to point out a number, or rp name any hour in the day, 
or to alk what o’clock it is, or to felect any one of cer¬ 
tain queftions written upon cards whidh you prefent to 
him. A fly, a fyren, or a fwan, floating in water, indi¬ 
cates, in their order, the figures of this number, or an¬ 
fwers the queftion propofed. 
All this is performed by means of a ftrongly-magne- 
tifed bar, fupported by brafs circles, concealed in the 
rim of the bafon which contains the water. It is evi¬ 
dent that, if the motion, nepelfary to point out the letters 
or numbers required for the anfwer, can be given to 
this bar, the fly or fyren, placed on a frnall boat contain¬ 
ing another magnetic bar, will proceed towards it, and ap¬ 
pear to anfwer the queftion. Such are the philolophical 
principles of this experiment; the deception is as follows: 
The table, which is forne inches in thicknefs, is hollow, 
and the cavity contains a certain mechanifm put in mo¬ 
tion by a firing, which, palling through the feet of the 
table, traverfes the floor, and is conveyed into a neigh¬ 
bouring apartment, feparated from that where the trick 
is exhibited only by a very .flight partition. This ltring 
terminates at a fort of table, on which are marked the di¬ 
vifions of the bafon ; and the whole is combined in fuels 
a manner, that, when the end of the firing is brought op- 
pofite to a certain figure, fuch as 4 for example, the mag¬ 
netic bar will be under the 4 inferibed on the edge of the 
veil'd. When the fyren then is delired to tell what o’clock 
it is; the perfon behind the partition, and who hears the 
queftion, has nothing to do but to pull the firing, and to 
bring the end of it oppofite to the required hour on the 
table, which is before him. The magnetic bar will ar¬ 
range itfelf below ; and the tradable (yren, beginning to 
move, will go and point out the hour. If a queftion has 
been feleded ; the perfon who exhibits the trick repeats 
it under a pretence of interrogating the lyren. The con¬ 
federate, who hears it, caufes the magnetic bar to move 
to the anfwer. It would not be difficult to eltablilh be¬ 
tween both a fecret communication of fuch a nature, that, 
without fpeaking, the fyren Ihould appear to guefs the 
queftion, and to give an anfwer to it. 
10. The Magnetic Curves. —Dr. Robifon ufed to relate a 
curious and inltruftive phenomenon, that he was long puz¬ 
zled to explain, rel'peding the mutual adtion ot large 
magnets. Amufing himlelf with fome experiments on 
magnetifm, with two large lirong magnets, as A, B, fig. 32, 
which were placed at about the diitance of three inches, 
with their oppofite poles fronting each other, he had 
placed a linall needle balanced on a point between them, 
as at D, which arranged itfelf in the fame line with the 
magnets; but, happening to fet it off to a confiderable dif¬ 
tance on the table, as at F, he was furprifed to fee it in- 
ltantly turn round on the point, and arrange itlelf in an 
oppofite diredion. When brought back to D, it re-af- 
fumed its former polition ; but, when he carried it out 
gradually along the line D F, perpendicular to N S, he 
found it grow lenfibly more feeble, vibrating more (lowly; 
and, when arrived at a certain point E, it Ihovved no po¬ 
larity towards either A or B, but retained any pofition 
given it; but, when carried farther out, it again acquired 
polarity to the magnets, though in a contrary diredion, 
arranging itfelf parallel to N S, with its north pole next 
to N, and fouth pole next to S. Being interrupted in the 
profecution of this experiment, but having marked the 
line D F on the table, he afterwards replaced the magnets 
and needle, placing the latter at E, where he expected it 
to be neutral; but it now turned its north pole to¬ 
wards B, and did not become neutral till carried further 
out. When (landing there, lomething happened to move 
the magnets A and B, which inftantly rullied together, 
and at the fame inftant the needle turned itfelf brilkly, 
and arranged itfelf as before at F. In Ihort, by gradually 
withdrawing the magnets frqin each other, he found that 
the 
