Smolt age and size, and individual growth during smoking were examined in a wild population of white‐spotted charr,Salvelinus leucomaenis, in a small stream in northern Japan. Smolt size (110–210 mm F.L.) differed neither between sexes nor among age cohorts of 2–7 years. Back‐calculation of riverine growth of smolts using otoliths showed a gradually increasing discrepancy in fork lengths of these age cohorts in both males and females, the faster‐growing individuals smolting at an earlier age. A mark—recapture study on individually‐identified fish revealed that the initial (spring‐summer) length‐frequency distribution of parr was unimodal, but became slightly skewed to the larger side in autumn, resulting in a bimodal distribution being evident by the following spring, with the upper modal group primarily composed of smolts. However, closer examination revealed that both male and female potential (subsequent) smolts were larger than the potential parr in the previous spring, the minimum lengths of the former being around 90–100 mm f.l. Specific growth rates of the potential smolts remained relatively high after summer through to the following spring, whereas those of the potential parr decreased after summer, despite having been similar between the two groups during the previous spring and summer. These observations revealed that the developmental triggering of smelting possibly occurs in or followin