Even after we have agreed upon a given series of pictures, the formula for giving the test, and upon the requisite number of passes, there remains still the question as to the proper degree of liberality in deciding what constitutes interpretation. There is no single point in mental development where the "ability to interpret pictures" sweeps in with a rush. Like the development of most other abilities, it comes by slow degrees, beginning even as early as 6 years. The question is, therefore, to decide whether a given response contains as much and as good interpretation as we have a right to expect at the age level where the test has been placed. It is imperative for any one who would use the scale correctly to acquaint himself thoroughly with the procedure and standards described above. XII, 8. GIVING SIMILARITIES, THREE THINGS PROCEDURE. The procedure is the same as in VIII, 4, but with the following words:-- (a) _Snake_, _cow_, _sparrow_. (b) _Book_, _teacher_, _newspaper_. (c) _Wool_, _cotton_, _leather_. (d) _Knife-blade_, _penny_, _piece of wire_. (e) _Rose_, _potato_, _tree_. As before, a little tactful urging is occasionally necessary in order to secure a response. SCORING. _Three satisfactory responses out of five_ are necessary for success. Any real similarity is acceptable, whether fundamental or superficial, although the giving of fundamental likenesses is especially symptomatic of good intelligence. Failures may be classified under four heads: (1) Leaving one of the words out of consideration; (2) giving a difference instead of a similarity; (3) giving a similarity that is not real or that is too bizarre or far-fetched; and (4) inability to respond. Types (1), (3), and (4) are almost equally numerous, while type (2) is not often encountered at this level of intelligence. This test provokes doubtful responses somewhat oftener than the earlier test of giving similarities. Those giving greatest difficulty are the indefinite statements like "All are useful," "All are made of the same material," etc. Fortunately, in most of these cases an additional question is sufficient to determine whether the subject has in mind a real similarity. Questions suitable for this purpose are: "Explain what you mean," "In what respect are they all useful?" "What material do you mean?" etc. Of course it is only permissible to make use of supplementary questions of this kind when they are necessary in order to clarify a response which has already been made. While the amateur examiner is likely to have more or less trouble in deciding upon scores, this difficulty rapidly disappears with experience. The following samples of satisfactory and unsatisfactory responses will serve as a fairly adequate guide in dealing with doubtful cases:-- (a) _Snake_, _cow_, _sparrow_