At that time there were practically as many different forms of statistical reports in the United States as there were hospitals. In the meanwhile almost every textbook published during the last fifty years has announced a new classification of mental diseases. They have been based on etiology, pathology, symptomatology and psychology. English, French, German, Italian and American classifications have appeared, each representing, as a rule, different schools of psychiatry. Kempf[143] would discard the term psychosis altogether and speak only of neuroses as "more consistent with the integrative functions of the nervous system." For diagnostic purposes he proposes to separate the benign from the pernicious processes and classify them according to their psychological mechanisms as suppression, repression, compensatory, regression and dissociation neuroses. The easiest way out of all these difficulties, as Southard[144] has said, would be "to deny the existence of entities in mental disease. There are two forms of this contention; first, that mental disease is nothing more or less than insanity, an entity itself, a genus with but one species, or secondly, that all victims of mental disease are individually to be provided with entities, that is, all examples of mental disease are sui generis. The development of psychiatry has killed the former contention stone dead, but the latter contention still flourishes to an extent among those who overstress the individual factor. And this latter contention is bolstered up by the existence of so many psychopathic patients of whom a diagnosis cannot be rendered for practical or theoretical reasons. However, there are no really consistent advocates of the sui generis plan of classification." It is interesting to note that he concedes ... "that the American Medico-Psychological Association's classification, adopted as it has been by a great number of American institutions and by the United States Government for war purposes, is a reasonably good classification and aware that its constituent elements fairly well correspond with what all American psychiatrists agree upon." Southard[145] raises the question as to how this classification can be used for diagnostic purposes. He answers this query by suggesting "A key to the practical grouping of mental diseases"[146] ... "to be followed, when necessary, like a botanical key in the search for the classification of a plant."... "It is a key to study and not an analytical classification with any pretence to finality."... "The plan is not so much an excursion into nosology as an essay in the technique of psychiatric diagnosis for the tyro."