We compared the locations of healed attack marks of the sea lamprey Petromyzon marinus on live lake trout Salvelinus namaycush with those of unhealed attack marks on dead lake trout to determine if the lethality of a sea lamprey attack was related to attack location. Lake trout were collected from Lake Ontario, live fish with gill nets in September 1985 and dead fish with trawls in October 1983−1986. Attack location was characterized by the percent distances from snout to tail and from the ventral to the dorsal midline. Kolmogorov−Smirnov two-sample tests did not detect significant differences in the distribution of attack location along either the anteroposterior axis or the dorsoventral axis. When attack locations were grouped into six anatomical regions historically used to record sea lamprey attacks, dead fish did not exhibit a significantly higher proportion of attacks in the more anterior regions. Even if the differences in attack location on live and dead fish were significant, they were too small to imply substantial spatial differences in attack lethality that should be accounted for when modeling the effects of sea lampreys feeding on lake trout. We suggest that the tendency for sea lamprey attacks to occur on the anterior half of the fish is related to the lower amplitude of lateral body movement there during swimming and thus the lower likelihood of being dislodged.