The purpose of this study was to examine the division procedures of preschool children to determine whether such procedures involved one-to-one correspondence. Large and small numerosity trials were included so that the amount of effort and ease of using other procedures would vary. Odd and even number trials were included to determine whether children actually knew what the procedures they used implied about relative numerosity. Subjects were 20 four-year-olds and 20 five-year-olds, with equal numbers of males and females in each group. On each trial of the division task, children were shown two cookie monsters and a pile of round cardboard cookies and were asked to give both cookie monsters the same number of cookies to eat. When children finished distributing the cookies, they were asked whether the two cookie monsters had the same number of cookies. Children were videotaped while performing the task. Trials were scored as correct if the child had an equal number of cookies on each monster's plate after using all the cookies on the even number trials or all but one cookie on the odd. Findings indicated that although preschool children used one-to-one correspondence strategies to solve division problems, their use of such strategies was not consistent. Children used more strategies on large numerosity and odd number trials than on other trials. (RH)