From this we may conclude in a general way that it is essential that the special schools and classes should bring more than half of their defective pupils to the level of the intermediate course, and improve more than half of the ill-balanced, if they are to render public services superior to those of the ordinary schools. This must be the aim. How are we to know whether it is attained or not? By supervision exercised in the most serious manner, by well-kept individual records, in which are noted only facts which can be controlled. We remember, a dozen years ago, having turned over the records of young defectives in an asylum-school which had the reputation of perfect organisation, a reputation otherwise deserved, for everything that was shown to the public on visiting days was perfect. But a distinction must be drawn between what one sees and what one does not see. The records were kept with surprising negligence. They were dirty in appearance, torn, disordered, falling to bits. On reading them one only met with vague estimations, loosely expressed, about children who, as was repeated to satiety, "would make progress if they would work better." The less we say the better about contradictory diagnosis, such as one we noticed on a certificate of discharge: "Complete idiocy--very much improved"; or the too optimistic prognosis, really very naïve, if the writer has not had the bad taste to be ironical: "Vicious child--would make an excellent housemaid." If documents could be kept in this way, it is quite clear that those who so kept them felt pretty sure that nobody would ever read them. We demand that the notes which show the educational progress of the pupils should be written under the constant fear of control, in order that they may be guaranteed against negligence and interested optimism. The manner of control is very simple, and may be summed up in three paragraphs. 1. The estimation of the progress of the children should be made by the professors themselves, since they know each child well. The professor will always keep in mind that his notes will be checked by the inspector. With regard to instruction, notes will be kept with regard to reading, writing, arithmetic, spelling, according to the methods which we have indicated, and such remarks as "good," "very good," "passable," each signifying absolutely nothing, will be avoided. With regard to the manual work, it goes without saying that the record would have to be made in a somewhat different way. 2. The inspector will examine a certain number of cases chosen haphazard. He must carry out this control with an open mind and without prejudice. 3. He will make use of methods of control of a strictly impersonal nature.