The diagram is simply a chart of the pressure of steam in a cylinder against the steam's volume. In 1796, Southern developed the simple, but critical, technique to generate the diagram by fixing a board so as to move with the piston, thereby tracing the "volume" axis, while a pencil, attached to a pressure gauge, moved at right angles to the piston, tracing "pressure".
The gauge enabled Watt to calculate the work done by the steam while ensuring that its pressure had dropped to zero by the end of the stroke, thereby ensuring that all useful energy had been extracted. The total work could be calculated from the area between the "volume" axis and the traced line. The latter fact had been realised by Davies Gilbert as early as 1792 and used by Jonathan Hornblower in litigation against Watt over patents on various designs. Daniel Bernoulli had also had the insight about how to calculate work.
Watt used the diagram to make radical improvements to steam-engine performance and long kept it a trade secret. Though it was made public in a letter to the Quarterly Journal of Science in 1822, it remained somewhat obscure, John Farey, Jr. only learning of it on seeing it used, probably by Watt's men, when he visited Russia in 1826.
Later instruments (illus.) used paper wrapped around a cylindrical barrel with a pressure piston inside it, the rotation of the barrel coupled to the piston crosshead by a weight- or spring-tensioned wire.