59 
Experiments upon Heal. 
Here, notwithstanding that the quantities of the silk were 
the same in the three experiments, and though in each of them 
it was made to occupy the same space, yet the warmth of the 
coverings which were formed were very different, owing to the 
different disposition of the material. 
The raw silk was very fine, and was very equally distri- 
buted through the space it occupied, and it formed a warm 
covering. 
The ravelings of taffety were also fine, but not so fine as the 
raw silk, and of course the interstices between its threads were 
greater, and it was less warm ; but the cuttings of sewing silk 
were very coarse, and consequently it was very unequally dis- 
tributed in the space in which it was confined ; and it made a 
very bad covering for confining heat. 
It is clear from the results of the five last experiments, that 
the air which occupies the interstices of bodies, made use of for 
covering, acts a very important part in the operation of con- 
fining heat ; yet I shall postpone the examination of that cir- 
cumstance till I shall have given an account of several other 
experiments, which, I think, will throw still more light upon 
that subject. 
But, before I go any farther, I will give an account of three 
experiments which I made, or rather the same experiment 
which I repeated three times the same day, in order to see how 
far experiments made according to the method here described, 
may be depended on, as being regular in their results. 
The glass globe of the passage-thermometer being filled with 
16 grains of cotton-wool, the instrument was heated and 
cooled three times successively, when the times of cooling were 
observed as follows : 
I 2 
