An observational study was conducted to determine the proportionate mortality of wild trumpeter (Cygnus buccinator) and tundra (Cygnus columbianus columbianus) swans that died during the winters of 2000-02 in northwestern Washington State, USA. Among 400 swans necropsied, 81% were lead poisoned (302/365 trumpeter swans; 20/35 tundra swans). Mortality started in mid-November and peaked from late December through mid-February; swan mortality that was not associated with lead poisoning was uniformly lower throughout the winter months. Lead poisoning was 24 times more likely to be the cause of death in swans found in Whatcom County compared to swans found in other locations in northwestern Washington State (95% CI: 12.7, 47.0). Mortality attributable to lead poisoning was twice as likely in adults as in juveniles (95% CI: 1.0, 4.2). Aspergillosis was documented in 62 trumpeter and two tundra swans, including 37 swans in which mortality was caused by lead poisoning. Males were twice as likely as females to have aspergillosis (95% CI: 1.1, 3.8). Traumatic injuries were documented in 37 trumpeter and seven tundra swans, including seven trumpeter swans with concurrent lead poisoning. Dead swans found outside Whatcom County were four times more likely to have traumatic injuries compared to those found in Whatcom County (95% CI: 1.6, 10.0). Overall, lead-poisoned swans were significantly less likely to have concurrent aspergillosis or traumatic injuries. There was no apparent association between grit ingestion (total mass or mass categorized by size) and lead poisoning or number of lead shot. Not surprisingly, lead-poisoned swans were more likely to have one or more lead shot compared to swans that died from other causes (OR 294; 95% CI: 92, 1,005); lead-poisoned swans were also more likely to have one or more nontoxic shot compared to swans that were not poisoned (OR 63; 95% CI: 19, 318). The source(s) of shot are unknown but likely are in or near Whatcom County, Washington.