Blood collection is a common experimental procedure for which there are many different methods, each with its own advantages and disadvantages. Researchers should use methods that minimize pain, suffering, distress and lasting harm to animals while meeting study requirements. The authors evaluated stress, activity and tissue damage in BALB/cO1aHsd mice after collecting blood using one of six methods: retrobulbar bleeding with thin or thick capillaries, tail vein bleeding, saphenous vein bleeding, facial vein bleeding or jugular vein bleeding. The authors compared in-cage activity, corticosterone concentration and performance in open-field tests between treatment groups and collected histologic samples at 1 h, 3 d and 14 d after bleeding. Mice that underwent retrobulbar bleeding with a thick capillary had a smaller change in corticosterone concentration and higher in-cage activity immediately after blood collection, whereas mice that underwent jugular vein bleeding had a greater change in corticosterone concentration and lower in-cage activity and open-field activity. Mice that underwent saphenous vein bleeding had a high incidence of histological change at 1 h, 3 d and 14 d after blood collection, but few indicators of histological change were present in other groups at 14 d after blood collection. These results suggest that, when collecting a small volume of blood, retrobulbar bleeding with a thick capillary and without anesthesia causes the least stress in mice, whereas jugular vein bleeding and facial vein bleeding cause the most stress and saphenous vein bleeding causes the most lasting damage in mice.