This study tested four complementary hypotheses to characterize intrinsic and extrinsic influences on the order with which preschool children learn the names of individual alphabet letters. The hypotheses included: (a) "own-name advantage," which states that children learn those letters earlier which occur in their own names, (b) the "letter-order hypothesis," which states that letters occurring earlier in the alphabet string are learned before letters occurring later in the alphabet string, (c) the "letter-name pronunciation effect," which states that children learn earlier those alphabet letters for which the name of the letter is in the letter's pronunciation, and (d) the "consonant-order hypothesis," which states that children learn earlier those letters for which corresponding consonantal phonemes are learned early in phonological development. Participants were 339 four-year-old children attending public preschool classrooms serving primarily low-income children. Children's knowledge of each of the 26 alphabet letters was assessed, and these data were tested for the four hypotheses using a linear logistic test model (LLTM). Results from the LLTM confirmed all four hypotheses to show that the order of letter learning is not random, in that some letters hold an advantage over other letters to influence their order of learning. Implications for educational policy and practice are discussed.