We used 4Ј,6-diamidino-2-phenylindole (DAPI) staining and fluorescent in situ hybridization to examine total bacterioplankton and archaeal distributions in surface waters and in deeper nepheloid layers and particle-poor waters across the Beaufort Shelf of the Canadian Arctic, including the Mackenzie River and Kugmallit Bay, as well as more distant Franklin Bay. Although the regional distribution of bacterioplankton was best explained by salinity (r s ϭ Ϫ0.89, n ϭ 28, p Ͻ 0.001) and indicators of primary production (chlorophyll a [Chl a], total organic carbon, and the ratio of Chl a to particulate organic carbon [Chl a : POC]), that of Archaea instead reflected measures of particulate matter, specifically microscopically determined particle concentration (r s ϭ 0.85, n ϭ 30, p Ͻ 0.001), suspended particulate matter, POC, particulate organic nitrogen (PON), and the beam attenuation coefficient. Moreover, when compared with similarly deep particle-poor waters, nepheloid layers were significantly enriched in Archaea (median concentration of 6.00 ϫ 10 4 mL Ϫ1 [15.5% of bacterioplankton] vs. 1.79 ϫ 10 4 mL Ϫ1 [3.6%]; p Ͻ 0.05), but not total bacterioplankton. The relationship between Archaea and particles, the dominance of the Mackenzie River as the regional particle source, the detection of highest archaeal concentrations (11.5-14.4 ϫ 10 4 mL Ϫ1) in the river, and the highly significant correlation (r s ϭ 0.97, n ϭ 12, p Ͻ 0.001) between Archaea in particle-rich waters and PON (the river providing the upper end member) suggest that many of these Archaea derive from the river.Archaea have now been recognized as ubiquitous and sometimes abundant members of the marine bacterioplank-1 Corresponding author (chimera1@ocean.washington.edu). 2 Present address: Department of Bioengineering, Rice University, Houston, Texas 77005.3 Present address: Department of Biology, University of Washington, Seattle, Washington 98195. AcknowledgmentsWe thank the officers and crew of the NGCC Pierre Radisson, Louis Fortier for leadership of the Canadian Arctic Shelf Exchange Study (CASES), and Martin Fortier for his work as chief scientist. We also thank J.-E. Tremblay (Laval University, Quebec) for generously providing us with nutrient data, S. Demers and K. Lacoste (University of Quebec at Rimouski) for the POC, PON, and Chl a data, and Y. Gratton (INRS-Eau, Terre et Environnement, University of Quebec) for the CTD and transmissivity profiles. We appreciated the hard work of the water column sampling team of G. Desmeules, E. Rail, F. Bazinet, and M. Robert and the TOC analyses of Nes Sutherland. T. Papakyriakou (University of Manitoba) kindly discussed wind data with us, and C. Lovejoy (Laval University) commented on an early version of this manuscript. We acknowledge with gratitude the availability of the Ribosomal Database Project II Versions 8.1 and 9.0, as maintained by Michigan State University. Finally, we thank S. Carpenter (University of Washington [UW]) for assistance preparing figures, E. Lessard (UW) for discussion, and...