THE THERMOMETER AS A COMPANION IN DAILY LIFE. 371 
that of the surface. This is easily proved by placing a thermo- 
meter under the tongue, where it will always record not more 
than a degree or so above or below 99°, and another in contact 
with the skin of any uncovered part of the body, where it will 
be certainly found to rise to a less extent. 
Before going further it will be necessary to describe the most 
suitable form of instrument to be employed in taking the tem- 
perature of the body. The thermometer which best answers the 
requirements of these experiments is that termed the clinical 
thermometer, which is employed by medical men. It can be 
obtained of any surgical instrument-maker, and often from 
chemists. The whole instrument is about five inches long. The 
bulb is small, but it should be of the same diameter as the 
stem, though it is frequently made smaller. The stem is uni- 
formly cylindrical ; graduated on its surface, and with a strip of 
porcelain running along its length, internally, to show up the 
mercury column, which is very fine in calibre, and flattened to 
render it more distinct. The instrument is self-registering on 
Phillips’ principle ; that is, a portion of the mercury column is 
detached from the rest, and prevented from rejoining it by the 
presence of a minute quantity of air between them. The 
graduated scale is a very open one, the fifteen degrees between 
95° and 110° occupying nearly four inches. Consequently, when 
the mercury cools down after an observation, it leaves the bulb 
partially empty, which makes it necessary that particular care 
should be taken, in carrying the instrument and shaking down 
the index, that the latter does not descend into the bulb, and in 
so doing mix with the body of the mercury, thus destroying the 
self-registering power. To prevent this inconvenience a simple 
and fairly effectual method has been recently adopted ; the tube 
between the scale and the bulb is constricted by being heated 
and slightly drawn out, so that the mercury has some impedi- 
ment made to its return, which is not sufficient to overcome 
the attraction of the particles of the continuous fluid, but is 
enough to prevent the index passing it, unless an extraordinary 
amount of force is employed. The figures and lines of the 
graduation being on the surface, are apt to become indistinct 
from the black wearing off. This can be replaced with facility 
by, after having first carefully freed it from all greasy matter, 
heating the stem — not the bulb — and rubbing a piece of heel- 
ball up and down it. The superfluous blacking which is left 
after the operation should be scraped off with a knife, which, 
held obliquely to the tube, is run up and down it several times. 
Next as to the method of employing the instrument. Before 
commencing an observation it is first necessary to see that the 
self-registering mercurial index has been sent down the tube 
since the last experiment. This is done by holding the thermo- 
