A large area of boreal jack pine (Pinus banksiana Lamb.) forest in Canada is recovering from clear-cut harvesting, and the carbon (C) balance of these regenerating forests remains uncertain. Net ecosystem CO2 exchange was measured using the eddy-covariance technique at four jack pine sites representing different stages of stand development: three postharvest sites (HJP02, HJP94, and HJP75) and one preharvest site (OJP). The four sites, located in the southern Canadian boreal forest, Saskatchewan, Canada, are typical of low productivity jack pine stands and were 2, 10, 29, and 90 years old in 2004, respectively. Mean annual net ecosystem production (NEP) for 2004 and 2005 was −137±11, 19±16, 73±28, and 22±30 g C m−2 yr−1 at HJP02, HJP94, HJP75 and OJP, respectively, showing the postharvest jack pine stands to be moderate C sources immediately after harvesting, weak sinks at 10 years, moderate C sinks at 30 years, then weak C sinks at 90 years. Mean annual gross ecosystem photosynthesis (GEP) for the 2 years was 96±10, 347±20, 576±34, and 583±35 g C m−2 yr−1 at HJP02, HJP94, HJP75, and OJP, respectively. The ratio of annual ecosystem respiration (R) to annual GEP was 2.51±0.15, 0.95±0.04, 0.87±0.03, and 0.96±0.03. Seasonally, NEP peaked in May or June at all four sites but GEP and R were highest in July. R at a reference soil temperature of 10 °C, ecosystem quantum yield and photosynthetic capacity were lowest for the 2-year-old stand. R was most sensitive to soil temperature for the 90-year-old stand. The primary source of variability in NEP over the course of succession of the jack pine ecosystem following harvesting was stand age due to the changes in leaf area index. Intersite variability in GEP and R was an order of magnitude greater than interannual variability at OJP. For both young and old stands, GEP had greater interannual variability than R and played a more important role than R in interannual variation in NEP. Based on year-round flux measurements from 2000 to 2005, the 10-year stand had larger interannual variability in GEP and R than the 90-year stand. Interannual variability in NEP was driven primarily by early-growing-season temperature and growing-season length. Photosynthesis played a dominant role in the rapid rise in NEP early in stand development. Late in stand development, however, the subtle decrease in NEP resulted primarily from increasing respiration.