Arguments are presented in answer to Surwillo's criticism of a statistical procedure used by Stennett (1957) to test the hypothesis of an inverted-U relationship between alpha amplitude and palmar conductance. Surwillo (1965), despite unwise methodology, produced data consistent with the hypothesis that alpha amplitude bears an inverted-U relationship to heart rate. In a recent paper, Surwillo claimed that Stennett (1957), in demonstrating an inverted-U relationship between alpha amplitude and palmar conductance, "misused x[SUP2] by violating the requirement that entries from which x[SUP2] values are computed must be independent," and suggested further that Stennett's "positive finding is really an artifact of the method" (Surwillo, 1965, p. 250). In attempting to support this assertion, Surwillo states that "The method is clearly invalid, because data.., were combined according to a rule which assumes that an inverted-U function relates the variables" (Surwillo, 1965, p. 251). He then refers to hypothetical data which are obviously arranged to "prove" his point (Surwillo, 1965, Fig. 2). The purposes of this paper are: (1) to clarify the rationale of the statistical technique involved; (2) to illustrate the fallacy of the conclusions to which Surwillo is drawn by his hypothetical data; and (3) to point out certain methodological problems and pitfalls involved in studies along "the arousal continuum." [ABSTRACT FROM AUTHOR]