Six best-selling introductory behavioral statistics textbooks that were published in 1982 and two well-known sampling theory textbooks were reviewed to determine the presence of rules-of-thumb--useful principles with wide application that are not intended to be strictly accurate. The relative frequency and type of rules are reported along with a discussion of their implications for teaching statistics at the introductory level. There were 14, 15, 20, 23, 29, and 30 such statements located in the 6 introductory textbooks, for a total of 131 rules-of-thumb. In all six textbooks, there appeared to be a large proportion of rules-of-thumb concerning sample size, either in distributional approximations or in inference-making. It is suggested that rules-of-thumb should be invoked only if absolutely necessary. Invoking a rule-of-thumb as though it were a condition or assumption should be avoided. Finally, textbooks should avoid inventing new rules-of-thumb. Two tables are included. (TJH)