Although we do not have an explanation for this overestimate, future studies should endeavour to use in situ pellets. If this is not possible and culture pellets are used, their characteristics should resemble those of in situ ones (e.g. similar algae for feeding, similar textures, C:N ratios, lipids contents, etc.). Interestingly, it was recently GSK J4 in vivo argued that degradation by bacteria and protozooplankton of culture (Rhodomonas sp.) faecal pellets of C. finmarchicus incubated in cold waters takes three days to be significant (32% after 3 days) ( Svensen et al. 2012).

In the present study, FP-CSD by bacteria and protozooplankton was measured over the first two days of incubation (about 10.1% d− 1 for culture pellets, Figure 2). Assuming a constant carbon to volume ratio, the degradation would thus be about 30% in 72 h, which is comparable to the results of Svensen et al. (2012). Those authors used a different method, where microscopic measurements Pirfenidone molecular weight were performed in order to estimate faecal pellet volume changes and

therefore degradation. It seems, however, that this microscopic method did not make it possible to determine statistical differences in volume during the first two days, in contrast to the respiration method used in the present study. The use of micro-respiration chambers may therefore be more sensitive. In addition, the method used in the present study is less subjective and less time-consuming than the microscopic method. Despite the limited data set, the novel Farnesyltransferase use of micro-respiration chambers for faecal pellet

FP-CSD in the present study highlights the importance of bacteria from the pellet matrix, free-living bacteria and protozooplankton for faecal pellet degradation. Bacteria and protozooplankton play an important role in faecal pellet degradation at the chl a max compared to deeper water, and it most likely an important factor in areas where primary production is high, as the abundance of bacteria and protozooplankton is correlated with primary production. Few studies have addressed the importance of protozooplankton in Arctic areas, but this knowledge will be crucial for our understanding of the role of protozooplankton for the vertical flux of faecal pellets, which have been underestimated in the past owing to the low temperatures. In addition, the comparison between in situ and culture pellets addresses the importance of using in situ pellets if we wish to extrapolate results to natural field conditions. The results obtained from the experiments with culture pellets should be treated with caution, as they may overestimate the degradation rates. Special thanks go to the chief scientist of the Conflux cruise, M. Reigstad. A ‘thank you’ also goes to S. Øygarde, C. Svensen and C. Wexels Riser for their help in the field, in the lab and for the carbon analysis. We are grateful to T. Tamelander for the bacteria data. This manuscript has benefited from the valuable comments and suggestions from C. Lalande, C.