Culturing conditions:
Cultures of E. huxleyi StrainNCMA 2668 and 374 were inoculated at low cell density into media prepared from autoclaved filtered seawater with f/50 nutrient amendment. Cell populations were allowed to acclimate for approximately five generations, until cell density neared levels likely to significantly change the pH/pCO₂. Daily dilutions of cultures with pre-equilibrated media kept cell density low (<1x10⁵ cells/ml), ensured cells remained in exponential growth phase and prevented excessive drawdown of nutrients and CO₂. Cell density was determined by flow cytometry (model described below) and each flask was diluted with media that was continuously sparged with air containing 400, 750 or 1000 ppm CO₂. Air mixtures were created using CO₂ free air (Powerex air compressor, and Twin Towers CO2 scrubber) and pure CO₂ (Airgas) combined using a system of mass flow controllers (Sierra Instruments) and verified using a non-dispersive infrared CO₂ sensor (Licor 820). Cultures were maintained in 1-liter polycarbonate flasks at 15 degC under a 12/12 light dark cycle. Replicates (n=5) were placed in Plexiglas chambers which were supplied with a flow of the appropriate air mixture for each treatment. Preliminary experiments showed that gas exchange across the air/water surface significantly helped to maintain the target pCO2 in cultures without the mechanical disturbance of bubbling. Sedimentation was minimized by gentle mixing of the cultures by rotation of the bottles twice a day, during sampling and dilution. Cell densities ranged between about 30,000 cells/ml after dilutions to 80,000 cells/ml on the following day. The culture volume that was removed was used for analyses, and replaced with pre-equilibrated media. Cultures were maintained in this fashion for 14 days. This experiment was carried out twice, in 2011 and 2012.

Cellular carbon and nitrogen:
Samples for cellular particulate carbon and nitrogen were analyzed using a CE Elantech Flash EA 1112 elemental analyzer. In all, analysis blanks were run, and internal standards were inserted between samples, and remained within 1% of standard curve. For the calcifying strain (2668), samples were acid fumed for 24 h to drive off PIC. Values of organic carbon were subtracted from total carbon to yield cellular particulate inorganic carbon.