On the Conduction of Water. 267 
very same reason as before, unless the caloric is applied at the bot- 
tom. ‘The experiments detailed in that paper were conclusive on the 
conduction of water when its temperature was above 40° Fahr., un- 
less the caloric were chiefly conveyed to the water along the sides 
and bottom of the vessel used. In some late experiments on this 
subject this difficulty has been removed, and the possibility prevent- 
ed by the following contrivance. A thermometer was immersed in 
water at 62° Fahr., so as to be three eighths of an inch deep on 
the bulb, in a large earthen dish. A hollow glass cylinder, four 
inches in diameter and two inches high, was then placed in the wa- 
ter so as to have the bulb of the thermometer in the middle of the 
cylinder. The cylinder was prevented from touching the bottom of 
the dish by three small pieces of wood placed under it. The ether, 
which was to be inflamed over the bulb, was thus confined within 
the hollow glass cylinder, so that the generated caloric could not 
come to the sides of the earthen dish. When heated oil was poured 
over the bulb, it was confined in the same way. ‘The influence of 
a heated iron was confined in the same manner. Yet when all these 
were repeatedly tried, the temperature rose about six degrees, except 
that the iron did not heat it so much. These experiments satis- 
factorily prove that caloric passed downwards. If it was not ra- 
diation, it must prove the conduction of water. The form of the ex- 
periment prevents the heating of the bulb by means of the dish. It 
was clear that the rise of the thermometer soon ceased, as it ought to 
do if it were conduction; for the heated particles, being made lighter, 
would be pressed upwards by the cooler and heavier particles around 
- them as soon as the conduction was much diminished by the cooling 
above. Hence after a few moments the thermometer would begin 
to fall, although the surface of the water was several degrees above 
that of the water in contact with the bulb. When an air thermom- 
eter, with its stem passing down through the neck of a funnel of 
glass, and made tight in the neck by a cork, is immersed to the 
depth of an inch in water over its bulb, and then the hollow cylin- 
der of glass is made to surround the bulb as in the other case, and 
kept from touching the funnel, and ether is inflamed within the cyl- 
inder, the experiment is clearly visible, beautiful and decisive. It 
is not obvious how any experiment can be more satisfactory than 
this ; any experiment of this character—for we must always except 
the common one of mixing heated and cold water, when the caloric 
must pass from particle to particle, as the temperature of the hotter 
is instantly diminished, and of the colder is instantly increased. 
