[ 659 ] 
weight nine grains 5 one of thefe was tempered and 
of a blew colour, and the other was quite hard. 
He alfo produced two iron weights, feverally weigh- 
ing 14 pennyweight 22 grains, and 15 pennyweight 
/grains, both nearly of a cylindrical form, but with 
one of the ends rounded off. 
The 2 large bars were then placed in a line, as 
in the former experiments, but with their ends fo 
near together, as only to admit of the cap of one 
the needles between them. 
The tempered needle was then placed flat upon 
the bars, fo that nearly one half of it refted upoit 
one bar, and the other half upon the other, the cap 
lying between the two. The needle was prefled 
clofe to the bars in this pofltion, after which the 
bars were drawn away, both at the fame time con- 
trary wife, till they were clear of the needles and 
this operation was repeated three or four times : after 
which that end of the needle which had refted upon 
the northern part of one of the bars, was found ftrongly 
to attrad the north end of the needle in the compafs- 
box 5 and the other end which had refted upon the 
fouthern part of the other bar, was found to attrad in 
like manner the fouth end of the fame needle in the 
box. The power of attradion alfo acquired by this 
needle appeared to be very conftderable, it lifting 
eafily with either of its ends, the two iron weights 
above mentioned, when cemented the one to the 
other with wax, and weighing together 1 ounce 
10 pennyweights 5 grains. 
The hard needle was then applied to the bars 
like the other and with the very fame fuccefs, it 
lifted alfo, as the other had done, both the weights 
together. 
The 
