We have paid homage to the ability of the teachers and inspectors sufficiently often to be permitted to maintain here against those who would contradict us the necessity of our methods or of others of a similar kind. In order to determine the degree of intelligence or the state of instruction of a child one would require to have in mind the normal levels. Now, frankly, who knows what these are? Let any inspector, any teacher, glance over our test questions. He will be very much at a difficulty to say whether it is at nine years or at seven years that a child ought to be able to reply suitably to a particular question. We will go even farther. Let an inspector look at our scale, and say at what age reading is "fluent," at what age a child should write the third phrase with less than ten mistakes. Just let him try, and he will find the result. Let us add that people who are neither inspectors nor teachers will be still more embarrassed. We recollect that at the recent opening of a special class some eminent people appeared much astonished at the intelligence of the pupils. They were surprised at children of twelve years who made replies of which in reality normal children of eight should have been capable. It is impossible to form a correct judgment about matters so delicate unless one makes use of exact tests. We insist upon this because we foresee that all who visit the class for defectives will be subject to this illusion. All the more will they have an optimistic tendency to overestimate the intelligence and instruction of the children since they know in advance that they are going to see defectives, and consequently have a preconceived expectation of seeing degraded imbeciles with low foreheads and dirty habits. They will be quite surprised to find that the great majority of defectives do not answer to this description, and seeing that they have fallen into an error, they will correct themselves as usual by falling into the opposite mistake.