The author concludes: "(1) In a rigid statistical test to determine its effectiveness for discriminating between normal and schizophrenic subjects, both multiple-choice and individual clinical forms of the Proverbs Test yield mean scores whose differences are highly significant. (2) Cutting points for differentiating normals from schizophrenics are presented for three intelligence groupings which approximate the top quartile, middle half, and lowest quartile. ... (3) The effectiveness of the screening is consistent in the three intelligence levels for normal subjects. For schizophrenics the screening is least effective for the high intelligence grouping, probably due to the ineffectiveness of the technique when applied to chronic paranoid schizophrenics.".