In the 18th century, a theory of thermodynamics (now obsolete) called the caloric theory of heat was developed; which claimed that changes in temperature are due to the transfer of an invisible, weightless fluid called "caloric".
Though this seems quite outlandish today, at the time it seemed viable based on two main assumptions:
Heat was 'self-repulsive' but attracted to ordinary matter
Temperature was the density of caloric
This theory did suffice to explain certain phenomena, such as why a hot teacup cools (the dense caloric would flow out of the cup and into the surrounding air). However, this theory was later discredited in favour of the kinetic theory of heat in modern thermodynamics.
Kinetic theory is based around the idea that matter is composed of small particles constantly in motion. The heat of a solid, liquid or gas, then, under kinetic theory, is the kinetic energy of the particles within it.
Kinetic theory is based upon the following assumption
A macroscopic volume contains a large number of these tiny particles.This means that for a volume that we would see every day, like the amount of water in a glass. there are incredibly large numbers of these tiny particles. The exact number of particles in a mole of a substance is known as Avogadro's number (6.0 x 10^23) and more info can be found here on our site.
There are no forces between the molecules expect when they collide.This basically means that the particles are not attracted to each other in any way that will affect their motion. The only time that they experience a force is when they are struck by other particles.
The seperation between the particles is large.The spacing between the particles is large when compared with the actual size of the particles.
The directions of the particles velocities are random. This means that it is likely to find a specific particle moving in any given direction. There is no preferred direction of their movement.
The collisions are elastic.This basically means that when any two particles collide both the momentum and the energy are conserved. For example, if both particles are moving at a speed of 20m/s and they collide head on, then they bounce backwards and return in the opposite direction moving at the same speed.
For example, in this image you can see the random movement of the particles of a gas:
The caloric theory of heat had an amount of evidence against it, such as in 1798 when Benjamin Thompson, better known as Count Rumford published a report on his observation of heat while manufacturing cannons. He had found that boring a cannon repeatedly does not result in a loss of its ability to produce heat, and therefore no loss of caloric. Rumford observed that the frictional heat generated by boring cannon in the arsenal in Munich was apparently limitless. To demonstrate this he immersed a cannon barrel in water and, using a specially blunted boring tool, found that the water boiled in under 3 hours. He then argued that this seemingly unlimited generation of heat was incompatible with the caloric theory and concluded that the only thing communicated to the barrel was motion. This suggested that heat could not be a conserved "substance."
Though this specific experiment did not completely disprove caloric theory at the time, it did inspire the work of James Prescott Joule and others towards the mid-19th-century.
Joule's work was instrumental to the development of the principle of conservation of energy, also known as the 1st law of thermodynamics.
Joule's paddle-wheel experiment is the most well-renowned of his experiments on the conservation of energy because it gave the most accurate results for the mechanical equivalent of heat (that we know now).
This excerpt from a Royal Society Publishing Article describes the apparatus well:
"A cylindrical vessel containing water or mercury was fitted with a paddle-wheel which rotated between fixed vanes on a vertical axis. The paddle-wheel was driven by two strings, wound round the shaft, which passed over two pulleys and were attached to falling weights. The work input was calculated as the product of the weight and the distance of fall to the floor of Joule's cellar. The heat generated was obtained from the temperature rise of the water or mercury (and the associated metalwork). In order to obtain a measurable increase in temperature of 0.5–2°F, the weights were wound back up and the experiment was repeated some 20 times as quickly as possible. Several paddle-wheel assemblies were constructed using different materials."
Joule carried out five series of experiments to determine the mechanical equivalent of heat, and decided that the results of series 1 were most reliable and made 2 conclusions:
that the quantity of heat produced by the friction of bodies, whether solid or liquid, is always proportional to the quantity of force expended, and
that the quantity of heat capable of increasing the temperature of a pound of water by 1°F requires the mechanical force represented by the fall of 772 lb from one foot high.
These results (especially the first) helped to corroborate the now axiomatic principle of conservation of energy.