A total of 5,000 one-day-old male broiler chickens were assigned to 8 different treatments in a 4 x 2 factorial design. Four wheat cultivars (Amiro, Guadalupe, Isengrain, and Horzal) and 2 levels (0 or 1 kg/t of feed) of an enzyme cocktail (Avizyme 1300, xylanase, 2,500 U/kg and protease, 800 U/kg) were used. Nutritionally complete mash diets contained 65 and 70% of the test wheat for the starter and grower period, respectively. Test wheats were used in diets for broilers, and growth performance and AME contents were measured. Broiler performance was measured in 4,800 broilers allocated to floor pens with 75 birds each and fed from 1 to 42 d of age. Digestibilities and AME contents of diets were measured in 200 broilers from 6 to 27 d of age individually allocated to battery cages. Chromic oxide (Cr2O3) at an inclusion rate of 0.5% in the diet was used as an indigestible marker. Apparent metabolizable energy was corrected by zero N balance to obtain AMEn. Wheat cultivar strongly influenced animal performance during the starter period (1 to 21 d of age). During the grower period (21 to 42 d of age), only BW and daily feed intake were influenced by wheat cultivar. Differences in daily feed intake were associated with differences in AMEn intake during the starter period, but not during the grower period. Nutrient digestibility was higher with the use of enzyme. Animal performance was not affected (i.e., wheat cultivar differences were not eliminated by using enzymes). During the grower period, significant interactions were detected with regard to nutrient digestibility and AMEn. Differences in AMEn content of wheat could not be explained by digestible starch.