AbstractOn the basis of experimental study and observations on the morphology of the legs of spiders, P. Friedrich described a cutting device within the trochanter which would sever that appendage from the body when the leg was stimulated by injury.A reinvestigation of the problem yielded entirely contrary results. Experiments calculated to induce spontaneous amputation through an automatic reflex in the leg were negative on scorpions, harvestmen, and more than a dozen species of spiders. A detailed morphological study of the skeleton and musculature of the arachnid leg showed that no autotomizing device exists, either in the skeleton or in the arrangement of muscles (as is usually supposed). In all species studied, except the scorpion, severance of the leg from the body occurs most readily at some given point. The point at which severance readily occurs is directly correlated with a definite structural weakness in the skeleton and musculature. The spider itself removes an injured leg by grasping the stump with its mouth‐parts. No such injured stump falls off spontaneously or without the application of tension. Scorpions do not autotomize appendages under any stimulation, nor do they show any anatomical weakness in the leg skeleton or musculature.Autotomy as an automatic reflex does not exist in the Arachnida.