The possibility of cultivating cells in vitro has been one of the most important technical breakthroughs of the last decades and is the basis of much of the progress in cell biology. The first aim of cell biologists was to find how to stimulate cell growth using more and more sophisticated culture conditions. The success of Rheinwald and Green for long term subcultivation of keratinocytes is a brilliant example of the importance of this approach. When achieving technical conditions which make possible long term subcultivation of normal cells, biologists have looked to chemically defined medium, making possible analytical study of the effect of peptides or drugs on cells growth and differentiation. When comparing cellular response in monolayer culture in vitro, with cells in vivo, it became clear that most of the cell responses observed on dedifferentiated cells in monolayer culture are not at all predictive for the response of the same cells in vivo. These differences seem to be due to cell-cell and cell-matrix interactions existing in vivo which are responsible for cell differentiation and pharmacological responses. In order to develop in vitro cell culture systems predictive for the in vivo cell response, and to make possible alternative model systems for animal drug testing, new sophisticated systems have been developed. These systems attempt to reproduce in vitro the cell to cell and cell-matrix interactions responsible for cell differentiation, with the hope of realizing, step by step, a real organogenesis in vitro. The first human organ for which this concept has been developed, is the skin. We would like to review the most recent development of this technique, making possible the realization in vitro of a living skin equivalent. The work of our Laboratory has been mainly concerned by the investigation of two questions: Are the cells, in the living skin equivalent, in a functional and differentiated state close to the in vivo one? Is it possible, using the skin equivalent system, to obtain, in vitro, cell responses predictive for the in vivo ones?