So long ago as 1859, Bidder and Kuyfer originated the conception that axis cylinders grow outward from the nerve cells, and that regeneration of nerve fibers is a result of proliferation of the distal end of the axis cylinders, each cylinder having at its end "an incremental cone of growth." This view, which long has held sway, is now being questioned. It has been found that undifferentiated cells from the early medullary tube or neural crest, after remaining undeveloped and dormant in the tissues, may take on activity and, it is asserted, develop into nerve fibers. As seen in neuromas, Weichselbaum interpreted such undifferentiated cells as cells of the sheath of Schwann. Kennedy,1 on examination of specimens from both ends of a nerve previous to its suture, found irregularly arranged groups of new nerve fibers both centrally and peripherally, and on the distal side were enormous numbers of spindle-shaped nuclei