The major objective of the experiment was to see which of various principles concerning classes, embodied in Grize’s axiom-system, the subjects understood. To this end 11 tasks exemplifying those principles as class-inclusion problems were prepared for the children. It was again necessary to control the children’s tendency to try to guess the correct answer to a problem that they could not solve inferentially. To limit guessing, 8 insolvable tasks were randomly mixed with the 11 solvable tasks administered to each child. The formulas of these insolvable tasks are given in (2.5b) in section 3 of Chapter 2. The subjects were warned repeatedly that some of the tasks had no answers, and that when they found such a task, the “best” thing to do was to say that not enough information was provided for the problem’s solution. Explanations for the children’s answers were not scored. Instead, the same question was asked in different forms in order to make chance success on a task improbable.