The author does not follow the customary method of attack on this problem, namely of studying the mechanisms of development in one species or a group; but he attacks the problem by first classifying all possible causes or combinations of causes of asymmetry in animals. He proposes the following classification: I. Entirely endogenous asymmetries due either (a) to asymmetries of genes or chromosomes, or (b) to asymmetries of the substratum (cytoplasm). II. Entirely exogenous (environmental) asymmetries. III. Asymmetries caused by the interaction of intrinsic and extrinsic factors. He assumes that all of the causes of asymmetry could not act asymmetrically but that at least one or more must act asymmetrically to produce asymmetry. A comparison of those possible types with the actual observations shows that most of them are actually met with in nature.