Outbreaks of morbidity and mortality in double-crested cormorants (Phalacrocorax auritus) along Florida's Gulf Coast have occurred sporadically for at least 30 yr. During these outbreaks, the Clinic for the Rehabilitation of Wildlife, located on Sanibel Island in Florida, has admitted a substantial number of cormorants with consistent presentation of primarily neurologic clinical signs. In order to investigate the association of these outbreaks in cormorants with exposure to brevetoxin, we compared the timing of admittance of cormorants with outbreak-specific clinical signs to blooms of the brevetoxin-producing marine algae, Karenia brevis (formerly Gymnodinium breve), around Sanibel Island from 1995 through 1999. The clinic admitted 360 out of 613 cormorants with the common clinical sign of severe cerebellar ataxia in six outbreaks occurring during this period. The ataxia was characterized by a broad-based stance, truncal incoordination, hypermetric gait, and intention tremors of the head. The histopathologic findings in 10 cormorants euthanized in 1997 were mild and nonspecific. An immunohistochemical staining technique for the detection of brevetoxin in cormorants documented the uptake of brevetoxin in tissues from four cormorants admitted during an outbreak in 1997, but a modified technique used on samples from 11 cormorants admitted during a K. brevis bloom in 2000 produced indeterminate results. Admittance of cormorants with outbreak-specific clinical signs was positively correlated (P < 0.05) with concurrent concentrations of K. brevis in local water. The cross-correlation coefficient was also significant when increased K. brevis levels preceded cormorant admittances by 2, 4, 6, and 8 wk. This delay in time between K. brevis blooms and cormorant admittance and our clinical finding of neurologic abnormalities in cormorants without overt histopathologic features suggest an association between K. brevis blooms and local cormorant morbidity.