The principles of automation in animal development, as previously inferred from the concept of Cell Sociology do not fit in well with the current concept of sequential gene derepression. A more adequate explanation for those principles has been found in the literature dealing with the biochemical aspects of differentiation. Since oocytes and embryonic cells contain a greater variety of mRNAs than differentiated cells, as well as many tissue-specific (luxury) substances, it is concluded that the diversification of tissues consists of a (...) progressive selection of specific metabolic strategies, mediated by cell-to-cell contacts, from a broad range of pre-existing strategies. For each tissue, prior to its final determination, one luxury metabolic strategy is progressively intensified and becomes dominant. The others are either suppressed or maintained as latent metabolic strategies. The latter may on occasion become dominant again (transdifferentiation). These phenomena require a theory which considers gene regulation as the activation of otherwise repressed genes by specific activator RNAs. The high (apparently maximal) transcriptional activity on the lampbrush chromosomes may represent the synthesis of all the kinds of activator RNAs which are required for the reactivation of the genes during early development. A general conception is propounded of the automatism and programming of animal development, as inferred from the confrontation of these ideas with the concept of Cell Sociology. (shrink)