A study at Fairbanks, AK, was started in 1984 to determine soil seed longevity of 17 weed species. Seeds were buried in mesh bags 2 and 15 cm deep and were recovered 0.7, 1.7, 2.7, 3.7, 4.7, 6.7, 9.7, and 19.7 yr later. Viability was determined by germination and tetrazolium tests. Seed viability data were fit to an exponential model, separately for each depth, and the likelihood-ratio test was used to determine whether seed-viability decline was affected by burial depth. Depth of burial had a significant effect on viability decline of prostrate knotweed, marsh yellowcress, bluejoint reedgrass, and wild oat. By 19.7 years after burial (YAB), all seeds of common hempnettle, quackgrass, wild oat, foxtail barley, and bluejoint reedgrass were dead. Seeds of 12 other species were still viable: corn spurry (0.1%), prostrate knotweed (0.3% at 2 cm, 0.8% at 15 cm), flixweed (0.5%), pineapple-weed (0.6%), shepherd's-purse (1.3%), wild buckwheat (1.5%), common chickweed (1.6%), rough cinquefoil (1.8%), common lambsquarters (3.0%), Pennsylvania smartweed (3.3%), marsh yellowcress (8.5% at 2 cm, 0.3% at 15 cm), and American dragonhead (62.2%). Seed dormancy at 19.7 YAB was very low for all species (< 4%) except for American dragonhead, common lambsquarters, Pennsylvania smartweed, and shepherd's-purse, which had seed dormancies of 100, 27, 25, and 38%, respectively. Seed longevity was not increased by cold, subarctic temperatures.