The importance of larval viability in relation to the success of recovery techniques and to the interpretation of histological and migration studies was tested using mice infected with Toxocara pteropodis, T. canis and T. cati. Little difference was found in total numbers of viable larvae recovered from tissues incubated for 18 hours in 0·85% saline, 2% trypsin or 1% pepsin, but the rates of larval egress did vary. The recovery of dead larvae by digestion techniques was much lower. Larvae migrated in tissues undergoing fixation, which influenced histological findings. Larvae of T. canis did not undergo significant peritoneal migration in acutely-infected mice, but did penetrate their intestinal walls in large numbers within 4 hours of death. The implications of these findings in experimental studies are discussed.