We saw that some of the ancient Greek speculators imagined their infinite number of atoms as scattered originally, like dust, throughout space and gradually coming together, as dust does, to form worlds. The way in which they brought their atoms together was wrong, but the genius of Democritus had provided the germ of another sound theory to the students of a more enlightened age. Descartes (1596-1650) recalled the idea, and set out a theory of the evolution of stars and planets from a diffused chaos of particles. He even ventured to say that the earth was at one time a small white-hot sun, and that a solid crust had gradually formed round its molten core. Descartes had taken refuge in Sweden from his persecutors, and it is therefore not surprising that that strange genius Swedenborg shortly afterwards developed the same idea. In the middle of the eighteenth century the great French naturalist, Buffon, followed and improved upon Descartes and Swedenborg. From Buffon's work it was learned by the German philosopher Kant, who published (1755) a fresh theory of the concentration of scattered particles into fiery worlds. Then Laplace (1749-1827) took up the speculation, and gave it the form in which it practically ruled astronomy throughout the nineteenth century. That is the genealogy of the famous nebular hypothesis. It did not spring full-formed from the brain of either Kant or Laplace, like Athene from the brain of Zeus. Laplace had one great advantage over the early speculators. Not only was he an able astronomer and mathematician, but by his time it was known that nebulae, or vast clouds of dispersed matter, actually existed in the heavens. Here was a solid basis for the speculation. Sir William Herschel, the most assiduous explorer of the heavens, was a contemporary of Laplace. Laplace therefore took the nebula as his starting-point.