By cutting off the legs and tail of a salamander, Spallanzi got in the course of three months, six crops of these members. In one season one animal produced 087 perfect bones with their appropriate muscles and nerves. Kellogg tells us that "the long domesticated mulberry silkworm larva possesses the capacity of regenerating any of its legs, if the mutilation has not removed the whole appendage."

In man the power to reproduce a lost part is very limited. New pieces of skin may be grown; when a piece of diseased bone is removed, a new one sometimes gradually assumes the regular form, and all the attachments of muscles, ligaments, etc., become as complete as before. Children regularly grow new tonsils when they have been removed. Many fingers, the tips of which have been cut off in accidents, grow new tips including even the nails. A finger nail or a toe nail which has been seriously damaged is slowly thrown off as a new nail forms underneath, while in the embryonic state an entire limb, and the supernumerary digits in polydoctylism, are occasionally, though imperfectly reproduced after amputation. If a limb is torn from the body, provided the individual does not die from hemorrhage, a reparatory effort is established, and if the severity of the injury does not induce too much irritation in the system, the wound will gradually fill up, and the skin form over it. To a lesser extent we see this power exerted in the healing of ordinary wounds, and in cementing broken bones.

The regeneration of a nerve fiber is an interesting and instructive process. I present the following brief description after Howell (Textbook of Physiology), of course, stripping it of all theories and hypotheses of how it is accomplished.

When a nerve trunk is cut or killed at any point by crushing, heating or by any means, all the fibers from the point of injury to the periphery undergo degeneration. The definite changes included in this degeneration are observed only in a living nerve. A dead nerve or the nerve of a dead animal does not undergo these changes. The time required for the degenerative change to begin differs for the different kind of fibers found in the animal organism. In the dog they begin in four days, in the frog from thirty to a hundred and forty days depending on the season of the year. It has been found that if the frog is maintained at a high temperature (30 degrees C) degeneration will proceed as rapidly as in the mammal. In a cold temperature more time is required. In a dog it goes on so rapidly that the process appears to occur simultaneously throughout the whole of the peripheral stump. In the frog and rabbit it is observed to begin at the point of injury and progress peripherally.

The nerve fibers break up into ellipsodal segments of myelin, each of which contains a portion of the axis cylinder. These segments in turn break up into smaller pieces which finally are absorbed. The central stump, the fibers of which are still connected with the nerve cells, undergoes a similar degeneration for a short distance immediately contiguous to the wound.

In the peripheral portion of the nerve regeneration begins almost simultaneously with the beginning of degeneration. The nuclei of the neurilemmal sheath begin to multiply and form around them a layer of protoplasm so that, as the fragments of the old fibers disappear, their places are filled by numerous nuclei and their surrounding cytoplasm. This eventually forms a continuous strand of multi-nucleated protoplasm. This fiber bears no structural resemblance to 'the normal nerve fiber and is described as "embryonic fiber."

In the adult animal regeneration ceases at this point unless an anatomical connection is made with the central stump. However, such a connection is almost always accomplished unless special means are taken to prevent it. The two ends of the severed nerve find each other in a remarkable way and several ingenuous theories have been invented to account for this. After the two portions of the nerve have grown together the "embryonic fibers" of the peripheral end are gradually transformed into normal nerve fibers with myelin sheath and axis cylinder.

The earlier physiologists thought that if the severed ends of a nerve were brought together by suture they might unite by "first intention" without degeneration of the peripheral end. It is now known that this degeneration is inevitable once the living continuity of the fibers has been interrupted by any means. Any functional union that occurs is a slow process involving the degeneration and subsequent regeneration of the peripheral fibers. As if in disgust, nature tears down the old fibers and builds them anew.

Closely akin to the power to reproduce an accidentally lost part is the power to restore a long lost (suppressed) part. It goes beyond the individual. In close interbreeding of animals many characters are suppressed or "stopped down." In many cases of close in-breeding there is the loss of one or more vertebras in the spinal column. In cases of crossbreeding or in "spontaneous reversions" these suppressed characters are restored. Of this Darwin says: "No doubt the power of reparation, though not always quite perfect, is an admirable provision, ready for various emergencies, even for those which occur only at long intervals of time." Again: "That there is a tendency, in the young of each, successive generation, to produce the long-lost characters, and that this tendency, from some unknown cause, sometimes prevails." Lastly: "This subject has been here noticed, because we may infer, that when any part or organ is either greatly increased in size, or wholly suppressed, the coordinating power of the organization will continually tend to bring all the parts again into harmony with each other."

This shows that, with each species, there is a certain ratio of development of its several characters, which cannot normally be varied from, and which could not be varied from very greatly under any conditions, without resulting in such disproportionate development and loss of physiological coordination, and all their attendant evils, as are seen in the results of close-interbreeding. If Darwin had adhered to the principle he states above, in his treatment of the facts of breeding and variation, he could never have propounded a theory of the derivation of one species from another. He himself produced ample evidence to show that the more a form diverges from the primitive normal form, the less its chances of survival and the more need it has for "repair." On the other hand, crossbreeding and "reversion", by restoring one or more of the suppressed characters, increase the ability to survive.

Thus far, we have considered three phases of the living organism's power of self-repair. A few words here about the oneness of this power may not be amiss. Animals begin life as a single microscopic cell and by processes of cell proliferation, differentiation, and organization, grow into complex organisms with many dissimilar organs with their varied powers and functions. The organism is apparently self-evolving dependent only upon certain natural conditions. Each of the three phases of the power of self-repair we have considered are accomplished by the same processes and are subject to the same natural conditions as are involved in the development of the animal from ovumhood to adulthood.