Marine primary productivity studies pursued in the Canadian Arctic in 1961-1963, using standard techniques (oxygen and carbon-14) of the time, showed that maximum production developed rapidly in July at 5-m depth under very low light intensities and under intact but melting sea ice. The time of maximum production was correlated with depths of snow in spring. Low production levels were found in August in those years and estimated in 1956. Nitrate exhaustion occurred before maximum production and was followed by rapid sinking of chlorophyll and productivity and reduction of assimilation numbers. Gross production values greatly exceeded net production values in 1961-1963, and gross production in 2 years was about double that of a third year, which may be due to different initial conditions of nutrient concentrations in the euphotic zone. Interannual variability in primary production may be the result of changes in water masses in the surface layer, as driven by atmospheric pressure patterns.