Atlantic salmon (Salmo salar L.) parr (33 g start weight) were exposed to five different mean levels of carbon
dioxide partial pressure for about five weeks in an open flow system: 0.7 mmHg (1.9 mg L−1; control group),
3.9 mmHg (10 mg L−1; low group), 6.5 mmHg (17 mg L−1; medium group), 9.5 mmHg (25 mg L−1; high
group) and 12.4 mmHg (33 mg L−1; very high group). The water temperature was 15 °C. Plasma chloride concentration
decreased linearly with increasing carbon dioxide concentration on days 9, 20 and 37 (p b 0.05).
The mean reduction in plasma chloride was 0.65 mM/(mg L−1 CO2). The specific growth rate (SGR) decreased
with decreasing plasma chloride concentration (p b 0.05). The relationship between carbon dioxide concentration
and SGR can be described using a second-order polynomial model (p b 0.01). Based on this second-order
polynomial model, SGR is 100% until 15 mg L−1 is exceeded and minor reductions occur between 15 and
20 mg L−1. However, at higher concentrations the reductions in SGR increase very fast. That the relationship between
CO2 and growth was curvilinear rather than linear seems an important overall advance in the state of
knowledge. It should also be noted that some fish grew well even at high CO2 concentration (38 mg L−1).