2 86 Mr. Wollaston's description of the 
the bulb, and may change its place by a shake ; if too large s . 
a part of the mercury remaining in B may be detached by a 
shake, and occasion great inconvenience. To prevent that 
possible detachment, the long shape given in the figure is 
preferable to a spherical swell. A workman accustomed to 
blowing thermometers, though he will at first make it too 
large, will soon hit the size. When the metal is hot in this 
part, a slight pressure of the tube endwise will occasion a 
little thickening of the glass externally about the part C, 
which is of use for fixing the thermometer in its mounting. 
For the fine thread, the tube I> is chosen by comparison 
with other thermometers, such that if a bulb were blown on 
it of 0,4 inch in diameter, its scale would be about four 
inches between freezing and blood heat; that is, i6° to an 
inch ; when this tube is fitted to a bulb of an inch diameter,, 
its degrees will therefore be about an inch each. The tube 
is five inches long. Before the joining is made at E, the bulb 
is filled : and the upper end of D being broken off abruptly 
is joined by its edges to a small piece of tube F of the same 
external diameter, but of an open bore, so as to make a sort 
of bulb at the top by the cap F ; a blown bulb will not answer 
the purpose for which it is designed, of detaching a globule 
of mercury from the thread, and retaining it apart for future 
use. The joining at E must be neatly made without any 
crevices in which either mercury or air may lodge, and with 
as little swell in the thread as possible ; for if there be any 
thing that can be called bulb in that part which is protected 
by the mounting from the action of the heat, the thermo- 
meter will be long in feeling its whole expansion. 
Before F is sealed, the bulb and swell and F together con- 
