Pre-eruptive fluoride exposure has been shown to be important for caries prevention. This paper aimed to determine the relative effects of water fluoride exposure during crown completion (CC) and maturation on caries experience in first permanent molars. Parental questionnaires covering residential history were linked to oral examinations of 19,885 6- to 15-year-old Australian children conducted in 1992 by the School Dental Services of South Australia and Queensland. The percentage of lifetime exposed to optimally fluoridated water at CC, maturation (MAT) and post-eruption (POST) was calculated. Combined exposure variables describing different levels of CC, maturation and post-eruption were created using a threshold exposure of 50%. Compared to the reference of CC <50%/MAT <50%/POST <50%, the categories CC ≧50%/MAT <50%/POST ≧50% [rate ratio (RR) 0.52], CC ≧50%/MAT ≧50%/POST <50% (RR 0.61) and CC ≧50%/MAT ≧50%/POST ≧50% (RR 0.67) had the strongest caries-preventive effect, followed by CC ≧50%/MAT <50%/POST <50% (RR = 0.79) and CC <50%/MAT ≧50%/POST ≧50% (RR = 0.81) in negative binomial regression models (p < 0.05). The categories CC <50%/MAT ≧50%/POST <50% (RR = 0.85) and CC <50%/MAT <50%/POST ≧50% (RR = 0.84) had weaker, non-significant caries-preventive effects. In conclusion a high exposure at CC was important for caries prevention irrespective of the effect of exposure at maturation and post-eruption. The strongest caries-preventive effect was produced by a high exposure at CC supplemented by a high exposure at maturation and/or post-eruption, but the latter two phases could not produce a significant caries-preventive effect on their own. Since most of the caries occurred on pit and fissure surfaces, the findings relate to this class of lesion.