I 36s

will make the experintem lefs craft and the/try,

it would be much name comenient, if the fea-water
could iotirely bc kept out ; which probably tray he
done by tying a bladder full of air to the neck of
the ball C., which will cc/Mud by 4he Treflitre of
the wake, without kithig any of it get ie.
If this can be done, the inflyument may be filled
with mercury and fpirits, jai like the at fig. t. lInt
it would bc more convernem to fill it with mercury
only in which cafe it may be made with as (mall
tuke and ball as common mercurial thermometers
or it may be filled with fpirits only. Thc infiru-
moot will thereby become rnuch left bulky which
will compenfate the want of the common kale of
heat.
It is better to put but little mercury into the hall
at top, for fear of its getting into the capillary nitre
by the motion of the infirumcnt.
The thermometer for finding the grracit sold, if
applied to this purpok, mull Ilk, bc left open at top.
Them is another inconvenience to be avoided which
is, that the mercury in the ball A, by the tolling of
the inftrument, might fornminics get into the kroner
leg of thc fyphon ; which would fpoil thc experi-
ment. To prevent fuch an accident, the moft con-
venient confiruftion, which occurs to me, is that of
fig. 4. which differs from fig. 3. in having the ball A
omitted ; fir that the mercury mining out of the
dexter leg will fall to thc bottom of the cylinder,
and will Let lac lo liable nr gct back into thc tube try
motion. Thc cylinder is made to dand not quite
parallel to the legs of the fyphon, that the tnersury
contained therein may more cafily be brought to
touch
3

To learn how to use PDF Compressio