406 
nexion of the air within the instrument with the external at- 
mosphere is cut off; but when one of the balls is heated, 
and the elasticity of the air within it thus augmented, and 
the intervening column of fluid driven towards the cool ball, 
the elasticity of the ‘air within the latter is also obviously in- 
creased by compression. Equal increments of temperature 
cannot, therefore, produce equal augmentations of volume, 
and, when the stem of such an instrument is divided into 
parts of equal capacity, the corresponding temperatures 
constitute, not an arithmetical series, but one which in- 
creases much more quickly. In fact, if ¢ be the tem- 
perature, » the volume of the air in the heated ball, and 
v” the volume of air in the cool ball, ¢ varies not as 2, 
but as <. 
This source of embarrassment in the air thermometer 
Mr. Grimshaw removes by a contrivance, a notion of which 
may be simply conveyed by describing his instrument 
as a differential thermometer, in the cool ball of which 
is placed a barometer, while to the side of the same ball a 
little syringe is attached, by means of which air may be 
pumped in or out, and the elasticity of the included air thus 
rendered invariably the same, before the temperature (exhi- 
bited upon the scale of equal parts attached to the stem in 
connexion with the hot ball) is registered. Two forms 
of this instrument were placed on the table of the Academy, 
which, however, Dr. Apjohn stated should be considered 
rather as rough models, than as finished instruments. Dr. 
Apjohn observed, that Mr. Grimshaw intended attaching 
to his thermometer a provision for keeping the barometer 
vertical ; and marking upon this latter instrument two addi- 
tional points of constant pressure,—one higher, the other 
lower, than the atmospheric standard,—by the use of which, 
when necessary, the scale of the instrument may be greatly 
extended, so as to comprehend with ease the entire of the 
atmospheric range of temperature. 
