One major exception to this generalization is the work of Clement Juglar (1819-1905), who in 1862, published Des crises commerciales et de leur rétour périodique en France, en Angleterre et aux Etats-Unis. The second edition of this work, published in 1889, was considerably enlarged with historical and statistical material. Juglar is a spiritual predecessor of W C. Mitchell in that he did not build a deductive theory of the business cycle, but rather collected historical and statistical material that he approached inductively. His main contribution was his statement that the cycle was a result not of forces outside the economic system but of forces within it. He saw the cycle as containing three phases that repeated themselves in continuous order