for determining minute Differences of Temperature. 57 
differences of temperature which exist in animals 
of the same class. In pursuing this inquiry, I soon 
discovered that it was essential to devise other 
instruments than those in ordinary use. 
It was easy by enlarging the bulb and by select- 
ing a tube of extremely fine calibre, to render the 
common thermometer capable of more minute in- 
dications. But it was impossible to carry this 
change beyond a certain degree, the augmented 
length of the instrument becoming highly incon- 
venient. 
In order to obviate this difficulty, I devised the 
instrument which I am now about to describe. 
The form of this instrument is represented in 
the accompanying outline. The relative size of the 
bulb and calibre of the tube is such that the tenth 
part of a degree occupies a considerable space 
upon the scale. ‘The entire scale consists of ten 
degrees. At the upper part of the thermometric 
tube a small bulb is blown, which I shall desig- 
nate the reservoir; it is turned forwards so as to 
remain at a right angle with the tube. Al 
The bulb and the tube are filled with mercury, 
and a little of that fluid is included in the reser- 9, 
voir, when the whole is hermetically sealed. 
When an experiment is to be made, the mer- 
cury in the tube is to be brought into contact with 
the mercury in the reservoir, by placing the in- 
strument horizontally, with the reservoir upwards, in water of 
a sufficient temperature. 
I will now suppose that I wish to try the comparative 
temperature of the swallow which shuns, and the sparrow 
which abides, the rigours of our winter. The thermometer 
is removed from the water at the temperature of 110° Fahr., 
and placed upright. The contiguity of the mercury in the 
tube with the mercury in the reservoir being broken, the 
highest point in the scale will represent that degree, viz. 
110°. The lowest will consequently be the 100th degree. 
The entire scale is one of six degrees between these extremes, 
each degree being divided into tenths. 
The same plan is adopted for any other part of the scale. 
We have thus an instrument of the usual size, capable of 
measuring the tenths of a degree of temperature, at any 
part of the scale. It only requires the addition of a common 
thermometer to afford the extreme limit of the magnified 
scale. 
Third Series. Vo\. 8. No. 48. Jan. 1836. I 
