152 On the Conduction of Water. 
ing power of water, viz. that if water at 150° be mixed with an 
equal quantity at 50°, the temperature of the mixture will instantly 
be 100°. Now this can be done only by the passage of the ca- 
loric from particle to particle ; for, otherwise, the colder would sink 
to the bottom, and the change of temperature be effected very slow- 
ly. This is only one of the many cases in which philosophers have 
overlooked simple and familiar, and complete proof of the law to be 
ascertained. 
Although it is thus evident from the established facts of the ex- 
pansion of water both above and below 40°, that the experiment of 
Dr. Murray affords no proof of the conduction of caloric from parti- 
cle to particle, I thought it advisable to repeat the experiments, and 
notice the facts. Having procured a solid mass of ice, I hollowed 
out a place for the thermometer, to the depth of more than an inch, 
and covered the thermometer with water at a temperature below 40°. 
It had been previously ascertained that the mercury stood at 32° in 
melting snow. 
The water in the different setperitwaita was on the bulb of the 
thermometer, from one eighth to three eighths of an inch deep, and 
of course deeper over all the other parts of the thermometer. Care 
was taken too, that the bulb should not be in contact with the ice. 
The water on the thermometer soon fell to about 33°. Oil, heated 
to 160°, 173°, 184°, 260°, and 285°, was in successive experiments 
poured carefully on the water, and floated over the whole surface. 
The thermometer rose to 35°, and 36°, and even to 38°, as the oil 
was of higher temperature, and in greater quantity. Though the 
temperature rose to 38°, in several of the experiments, it never ex- 
ceeded that point. The ice was melted by the ol, so that the wa- 
ter was from one fourth to three fourths of an inch deep on the bulb. 
In two or three minutes the thermometer began to descend, although 
the temperature of the oil along the middle of it, and over the bulb 
was higher than that of the thermometer. These experiments show 
no proof of the conduction of caloric. The reason why the mercu- 
ry rose only to 36° or 38°, and so soon began to descend, is doubt- 
less to be found in the melting of the ice, and the temperature and 
weight of the water just formed from the ice, and in the fact that the 
water, heated to 36° or 38°, would on sinking to the bottom lose its 
temperature in melting the ice, and both it and the new melted water 
at 32°, would then ascend towards the surface. So soon as the tem- 
perature of the oil was reduced to a certain extent, this melting and 
