A developmental learn-to-swimming program is used to teach swimming skills based on a sequenced simple-to-complex order across different skills. This program is believed to be useful for children. The purpose of this investigation was to describe application of this program for children with learning and behavioral disorders. An action study design was employed without control group in this study since this action study was conducted in a naturalistic infield setting. Participants were 16 children in a semester and 15 in an adjacent semester. They had learning disabilities, mental retardation, attention deficit disorders, or autism. In a semester, they received a 60-min session per week for 10 weeks and were taught by student-instructors in a developmental program. The cornerstone of this program was a learn-to-swimming progression, including 10 basic swimming skills sequenced from water orientation to combined limb actions. Each skill was task-analyzed into 3-5 steps served as criteria for evaluating the success of this skill. The participant’s performance was measured at the beginning of each session. Its result was employed to determine a point on the progression to be focused on in teaching. The rule was that a following skill on the progression would be emphasized on only after all previous skills were performed successfully. If a participant could not correctly perform any steps of a skill immediately after the successful skill(s), then teaching emphasis would be placed on these steps. The number of correct steps by a participant in the first session and the last session during a semester was used in data analysis. This number was the sum of those correct steps of an unsuccessful skill (there was at least one step of this skill completed incorrectly) and those correct steps of the successful skill(s) listed immediately before this unsuccessful skill on the progression. The magnitude of increase, the dependent t test, and the effect size of d index were calculated. The results from the two semesters showed that the magnitude of increase from the pretest to the posttest was 43% and 39%, respectively; the difference between the pretest and posttest was statistically significant with t (15)=4.71 (p < .01) and t (14)=5.67 (p < .01), respectively; and the effect size of d index of this difference was large with d=2.44 and d=3.03, respectively. Thus, this developmental program was successfully applied for teaching these children these basic swimming skills.