K. C. Chang has been instrumental in the West in propagating the view that North China in Neolithic times was lush, moist, and densely forested. A close scrutiny of data relating to palynology, soil science, and paleoclimate, however, indicates that the true loess areas of North China were, and have always been, semiarid steppe. In addition, the author considers Chang's belief that the earliest northern Chinese agriculture was a system of slash-and-burn or shifting cultivation to be invalid. Whereas the crucial problem in the slash-and-burn systems of the tropics and humid areas is fertility, the crucial problem in the agriculture of the loess areas is moisture, not fertility. The ability of loess soil to retain its fertility under semiarid conditions has accounted for the self-sustaining character of the agriculture of North China. These important issues, and others, are discussed in the volume under review.