In this article, four prediction methods, Staley-Hoag, Moyers, Gazilerli, Tanaka-Johnston, estimating the sum of the unerupted maxillary canine and premolar widths were studied on the measurement of periapical films and orthodontic casts of 16 males and 34 females. Paired t values between the sums of the actual and estimated widths of the canines and premolars were found statistically insignificant in Gazilerli's method and Staley-Hoag's eight formulas, and were found statistically significant in the others. It has been concluded that the use of the prediction methods in which t values are significant may cause important clinical problems.

Keywords: Canine width, Premolar width