Forelimb regeneration from different levels of amputation in the newt,Notophthalmus viridescens: Length, rate, and stages

1. Some aspects of the influence of position on regeneration have been examined by comparing regeneration from two different levels along the newt forelimb. 2. We have defined a series of stages of forelimb regeneration in the newt,Notophthalmus viridescens, in order to facilitate this study. 3. Limbs amputated at either a proximal level (through the humerus) or a distal level (through the radius and ulna) pass through the same stages at the same times after amputation. 4. The histological sequence of events of digit regeneration was compared with that of limb regeneration from a proximal level of amputation and was found to be the same. 5. In limbs amputated at either proximal or distal levels, the rate of elongation of regenerates is the same during the phases of dedifferentiation, blastema accumulation, and blastema growth. 6. During the phase of differentiation and morphogenesis, the rate of elongation of regenerates from the proximal level is significantly greater than that of regenerates from the distal level. 7. The total length of regenerates from proximal and distal levels along the limb is significantly different. 8. The results indicate that positional information does not influence the developmental sequence of events of limb regeneration, but that it does influence the rate of growth of the regenerate and the specification of the structures to be replaced. Some aspects of the influence of position on regeneration have been examined by comparing regeneration from two different levels along the newt forelimb. We have defined a series of stages of forelimb regeneration in the newt,Notophthalmus viridescens, in order to facilitate this study. Limbs amputated at either a proximal level (through the humerus) or a distal level (through the radius and ulna) pass through the same stages at the same times after amputation. The histological sequence of events of digit regeneration was compared with that of limb regeneration from a proximal level of amputation and was found to be the same. In limbs amputated at either proximal or distal levels, the rate of elongation of regenerates is the same during the phases of dedifferentiation, blastema accumulation, and blastema growth. During the phase of differentiation and morphogenesis, the rate of elongation of regenerates from the proximal level is significantly greater than that of regenerates from the distal level. The total length of regenerates from proximal and distal levels along the limb is significantly different. The results indicate that positional information does not influence the developmental sequence of events of limb regeneration, but that it does influence the rate of growth of the regenerate and the specification of the structures to be replaced.

3. Limbs amputated at either a proximal level (through the humerus) or a distal level (through the radius and ulna) pass through the same stages at the same times after amputation.
4. The histological sequence of events of digit regeneration was compared with that of limb regeneration from a proximal level of amputation and was found to be the same.
5. In limbs amputated at either proximal or distal levels, the rate of elongation of regenerates is the same during the phases of dedifferentiation, blastema accumulation, and blastema growth.
6. During the phase of differentiation and morphogenesis, the rate of elongation of regenerates from the proximal level is significantly greater than that of regenerates from the distal level.
7. The total length of regenerates from proximal and distal levels along the limb is significantly different.
8. The results indicate that positional information does not influence the developmental sequence of events of limb regeneration, but that it does influence the rate of growth of the regenerate and the specification of the structures to be replaced.
One of the most interesting features of appendage regeneration in animals is that only the missing parts are replaced; each unique level along an appendage gives rise to an equally unique regenerate. This feature of regeneration has profound implications for our conception of the nature of developmental control mechanisms operating during regeneration, yet there is very little information about the factors which could be involved in ensuring that a regenerate appropriately completes an appendage.
We present in this paper the results of a study of three parameters of regeneration from different levels along the limbs of the newt, Notophthalmus (Triturus) viridescens: length, rate, and stages. The process of limb regeneration can be divided into three major phases: wound healing and dedifferentiation; blastema accumulation and blastema growth; differentiation and morphogenesis (Sehott6 and Butler, 1944;Sehott6 and Hall, 1952;Schmidt, 1966Schmidt, , 1968. To facilitate the accurate monitoring of these three phases of regeneration, we I established a series of normal stages based on morphological and histological characteristics. In general, these normal stages supplement previous descriptions of adult newt limb regeneration (l~ose, 1948;Singer and Craven, 1948;Singer, 1952;Schott6 and Hall, 1952;Manner, 1953;Cha]kley, 1954;Schmidt, 1958a, b;Liversage and Scadding, 1969;Pritchett and Dent, 1972).
In this paper it will be shown that amputation at different levels along the limb affects both the overall rate of elongation and the final length regenerated, indicating that there exists within the limb a gradient of developmental capacity. It will also be shown that at any given time during the first two phases of regeneration, the size of the regenerate and the rate of growth are the same regardless of the level of amputation, although in the third phase, both the length of the regenerate and its rate of growth are influenced by the level. However, throughout all three phases of regeneration, the general appearance and duration of normal stages are the same irrespective of the level of amputation. Therefore, the eventual amount to be regenerated and the rate of regeneration do not appear to influence the developmental sequence of morphological events seen in limb regeneration.

Material and Methods
Male and female adult newts, Notophtha~mus (Triturus) viridescent, were maintained in spring water at a constant temperature of 25 ~ C and a 12 hour light cycle. They were fed 2-3 times per week with Tubi]ex. Chloretone (Parke, Davis and Co.) was used as an anesthetic when amputations were performed and when camera lueida drawings were made.
For the establishment of normal stages, forelimbs were bilaterally amputated through the distal third of the humerus. The tissues at the site of amputation were trimmed to give a flat wound surface. The external morphological changes taking place during forelimb regeneration of between 28 and 50 regenerating limbs were observed every other day; camera lucida drawings of another 28 regenerating limbs were made at foul day intervals. All observations on the stages of regeneration were made during the fall and winter. For histological purposes, the limbs of 23 newts were amputated through the distal third of the humerus and between three and six limbs were placed in Bouin's fixative at each normal stage. Serial longitudinal sections of limbs were stained with either Maycr's hematoxylin and eosin or I-Ieidenhain's aniline blue.
For the observations on the influence of the level of amputation on limb regeneration, 14 newts ranging in length from 74-87 mm (mean 80 ram) and in wet weight from 0.9-i.7 g (mean 1.3 g) were used. The forelimbs of seven newts were amputated bilaterally through the distal third of the humerus and seven newts were amputated bilaterally through the middle of the radius and ulna. The progress of regeneration was observed in all limbs at two day intervals. To measure the length of the regenerates, camera lucida drawings of each limb were made once a week for 10 weeks. In the later stages of regeneration where digits are present, the regenerate was measured to the tip of the longest digit. When an elbow bend was present, the distances from the level of amputation to the elbow and from the elbow to the tip of the regenerate were added together to give the actual length of the regenerate. All observations on the rate of regeneration were made during the summer.
In order to determine whether the histological sequence of events seen in limb regeneration remains the same the farther apart two levels of amputation are from each other, the third digit of the forelimbs of seven newts was amputated through the joint between the metacarpal and first phalanx. The onset of chondrogenesis in a regenerate was chosen as the best histological change with which to compare regeneration from a proximal level (through the humerus) with that of a digit. Since a regenerating digit shows few changes in its external appearance, digits were placed in Bouin's fixative at various times after amputation: three digits at 14, 16, 18, and 21 days and two digits at 24 days. Serial longitudinal sections were stained with Mayer's hematoxylin and eosin.

Normal Stage8
The typical characteristics of 11 distinguishable stages of normal limb regeneration are presented below. A description of the external morphology of each stage is followed by a description of the internal organization of that stage. At the end of each stage description is a list of the comparable stages of adult  The abbreviations for each stage are: WH wound healing, EDD early dedifferentiation, LDD late dedifferentiation, MEB moderate early bud, EB early bud, MB medium bud, LB late bud, Pal palette, ED early digits, MD medium digits, LD late digits newt limb regeneration used by other workers. The comparison between our stages and those of other workers will be made only on the basis of their morphological and histological descriptions of regenerates since the level of amputation and the temperature at which regeneration took place have not been the same in all these previous studies. In general, the stages we describe here closely resemble those described by Singer (1952). The times given for each stage are taken from Fig. 1 which shows for any given day the range of stages which could be observed in a population of between 28 and 50 regenerating forelimbs.

Wound Healing Stage (0-5 Days after Amputation)
Immediately after amputation, the distal outline of the limb stump is completely flat (Fig. 2) and the stump epidermis terminates at the level of amputation. Within 24 hours after amputation, the distal portion of the stump becomes swollen (Fig. 2); the amputation surface is now covered by a thin transparent wound epidermis. Sub-epidermal vascular pools and engorged blood vessels can be observed; these show either vascular stasis or very sluggish blood flow. During the next two to four days the distal outline does not change noticeably, although 19" Fig. 2. Drawings of a dorsal view of a right limb amputated through the distal third of the humerus progressing through the eleven stages of regeneration. " A m p " is the abbreviation for a limb stump immediately after amputation; all other stage abbreviations are the same as those in Fig. 1. The stump has been tilted upward to show the wound area in the drawings of the regenerate limb at the stages of E D D and LDD. The stump has also been tilted upward in the first of two drawings of a regenerate at the stage of L B in order to show the distal dorso-ventral flattening and tilting of the distal end of the regenerate. • the stump becomes more swollen. Towards the end of this stage the vascular pools beneath the wound epidermis become smaller or disappear. From histological sections (Fig. 3) it can be seen that by three days after amputation the distal end of the stump is covered by a wound epidermis that is approximately twice the thickness (5-7 cells) of the stump epidermis. There are neither skin glands nor dermal pigment cells beneath the wound epidermis. Erythrocytes, tissue debris, and a few dead cells are seen both within and beneath the apical epidermis. There are no visible signs of dedifferentiation in either the transeeted muscles or bone. However, the nerves appear disorganized and degenerating at their distal ends.

Early Dedi//erentiation Stage (6-8 Days after Amputation)
The distal outline of the limb stump is a flat amputation surface with slightly rounded edges (Fig. 2), as in the previous stage. The wound surface itself now appears slightly swollen and translucent. When the wound epidermis is gently touched with a blunt probe, it appears to be blister-like, in contrast to the adjacent stump epidermis which appears to be firm.
The apical wound epidermis in this stage has thickened and has become approximately two to three times thicker (6-11 cells) than the stump epidermis (3-4 cells) (Fig. 4). There is no collagenous basement membrane beneath the apical wound epidermis, and tongues of epidermal cells project into the underlying wound area, as Singer and SMpeter (1961) have also described. Dead cells are seen in and beneath the wound epidermis. Tissue debris and loose erythrocytes can also be seen under the wound epidermis. Dedifferentiation of stump tissues is visible at this stage. Short segments of myotubes containing single nuclei can be seen in the vicinity of the wound and multinucleate osteoclasts can be seen around the eroding distal end of the humerus. A few blastema cells ~ are beginning to accumulate beneath the apical epidermis.

Late Dedi//erentiation Stage (8-11 Days after Amputation)
The distal outline of the limb stump at this stage is rounded. A small blisterlike swelling or protuberance usually appears in the ventro-posterior or central portion of the amputation surface (Fig. 2).
1 There seems to be some ambiguity in the stage descriptions of Pritchett and Dent (1972). They state in the text that limbs were amputated through the stylopodium (humerus or femur), but their figures show that amputations were through the zeugopodium (radius and ulna or tibia and fibula). 2 Blastema cells are defined here to be the apparently undifferentiated cells which accumulate between the mature tissues of the stump and the distal wound epidermis. Arrow points to tissue debris beneath wound epidermis. X 100 At this stage (Fig. 5), the epidermis of the small localized swelling seen externally is approximately 8-15 cells thick. A blister-like space usually separates this localized thickening from the stump tissue and a few dermal cells are often seen attached to the underside of the epidermis. The occurrence of this space has been previously reported by Rose (1948) and Singer aud Salpeter (1961). The irregular basal outline of the wound epidermis seen in the previous stage is less apparent. Less tissue debris and fewer dead cells are seen in or beneath the wound epidermis. Occasional melanocytes are now seen in the wound epidermis. Extensive dedifferentiation of bone and muscle as well as degenerating nerves are seen in the stump. At this stage, loosely arranged blastema cells fill the entire wound area beneath the apical epidermis.

Moderate Early Bud Stage (10-14 Days after Amputation)
The distal outline of the limb stump has become slightly rounded and the small localized swelling of the previous stage is no longer present (Fig. 2). The wound epidermis appears colorless and is no longer blisterlike.
There has been a visible reduction in the thickness (now 4-9 cells) of the apical regenerate epidermis (Fig. 6) since the previous stage. A small blister-like space, similar but smaller than that seen in the previous stage, is often present between the regenerate epidermis and the accumulating blastema cells. The blastema itself is larger than in the previous stage and occasional capillaries are seen in the blastema. Extensive dedifferentiation of stump tissues still appears to be taking place.

Early Bud Stage (12-17 Days after Amputation)
The distal outline of the limb stump is now dome-shaped (Fig. 2). The re-geneIate is colorless and is usually centered over the ventro-posterior oi' central portion of the amputation surface.
The epidermis is 5-10 cells thick at the apex of the regenerate dome and tapers off at the edges where it joins with the thinner stump epidermis (Fig. 7). Occasional blister-like spaces, similar to those seen in previous stages, are seen between the regenerate epidermis and the blastema; these are reduced in size and contain less tissue debris than before. Dedifferentiation of all stump tissues is continuing and it extends proximally into the stump tissues. The original site of amputation is only marked by the most distal skin glands in the stump. The blastema cells appear to be most closely packed in the central regions of the regenerate, or in those regions lying close to the end of a nerve. Although capillaries are not visible externally, occasional capillaries are seen in sections of the blastema. This stage appears to be equivalent to: "young blastema", Rose (1948); "early bud stage", Singer (1952); "blastema phase", Schott4 and Hall (1952); "bulb stage", Chalkley (1954); "bulb stage", Sehmidt (1958a and 1958b); "blastema stage", Liversage and Scadding (1969).

Medium Bud Stage (14-20 Days after Amputation)
The distal outline of the regenerate at this stage is cone-shaped (Fig. 2). The apex of the regenerate can be either rounded or pointed. With the aid of substage lighting, small vascular channels with flowing blood can be seen for the first time within the blastema.
There appears to have been a further reduction in the thickness of the apical epidermis (Fig. 8) since the previous stage (now 5-8 cells thick). Occasionally, the lateral regenerate epidermis is thicker than the apical epidermis, because the basal layer of cells have become columnar (also described by Carlson, 1967). As in the previous stage, the amputation site is marked by the skin glands at the stumpregenerate border. The blastema at this stage consists of an aggregation of cells which is considerably more dense than in previous stages; the cells appear most densely packed in the axial portion of the regenerate. Numerous capillaries are now seen within the blastema. Dedifferentiation is still continuing.

Late Bud Stage (18-24 Days after Amputation)
At this stage (Fig. 2), dorso-ventral flattening of the distal portion of the cone apex is observed. When viewed from the end, this flattened apex is not oriented along the anterior-posterior axis of the animal, but rather is slightly tilted in an antero-ventral direction. After this distal flattening has occurred, a bend begins to develop at the stump-regenerate border causing the regenerate to point anteriorly (Fig. 2). As the flattening of the regenerate apex becomes more pronounced the dorsal surface may even become slightly concave. Major vascular channels can be observed in the blastema. At this stage, the most apparent change in the regenerate epidermis is the thickening in the region where the anterior bend will form or has already formed (Fig. 9). Scattered dermal melanophores are now seen in the blastema in addition to melanoeytes and melanin granules previously observed in the regenerate epidermis. Chondrogenesis is observed for the first time in this stage; it is seen in the proximal region of the blastema and is taking place around the most distal portion of the humerus extending distally into the central dense accumulation of b]astema cells. An extensive network of capillaries throughout the regenerate has developed by this stage. Signs of active dedifferentiation are no longer apparent.

Palette Stage (22-28 Days a/ter Amputation)
At the distal tip, the regenerate has an irregular outline and appears flattened and paddle-like (Fig. 2). The pigmentation differences which previously marked the border between the epidermis of the regenerate and that of the stump are no longer clearly discernable.
At this stage (Fig. 10), chondrogenesis of the radius and ulna has begun. The regions of cartilage formation merge into the dense accumulation of blastema cells located distally. These distal blastema cells are arranged in a pattern foreshadowing the arrangement of the elements of the autopodium (hand or foot).
Occasionally, chondrogenesis appears to have begun in the autopodium. Myogenesis is seen in the proximal portion of the regenerate for the first time. Scattered blastema cells surround the regions of chondrogenesis and myogenesis. The thickened epidermis seen in the previous stage remains in the region where the elbow of the regenerate is forming ; there has been no change in the thickness of the regenerate epidermis elsewhere.

Early Digits Stage (24-33 Days a]ter Amputation)
At this stage, the onset of digit formation is apparent ; small digital projections can be seen at the distal edge of the regenerate (Fig. 2). The three developing digits are separated by two short interdigital grooves consisting of tissue which is pale and avascular. By four weeks after amputation, four digital projections and thIee interdigital grooves can be observed (Fig. 2). The bend in the regenerate is now situated distal to the stump-regenerate border. The stump is no longer swollen. There has been a slight reduction in the thickness of the regenerate epidermis since the previous stage (now 4-6 cells thick). The epidermal thickening at the elbow has now disappeared. The dista] epidermis appears to form thickened wedges which project deeply into the interdigital tissue (Fig. ]1). Skin glands are now seen forming beneath the epidermis in the proximal half of the regenerate. At the time that three small digits are seen externally, the zeugopodium (radius and ulna) has not separated from the autopodium, whose elements are essentially fused. Later, the zeugopodium becomes separated from the autopodium and almost all dements of the autopodium become recognizable, but not separated from each other. By the end of this stage, myogenesis has extended distally to the region around the developing radius and ulna. As in the previous stage, scattered blastema cells surround regions where chondrogenesis and myogenesis are taking place.

Medium Digits Stage (30-40 Days a/ter Amputation)
The distal outline of the regenerated hand consists of four short digits (Fig. 2). Digits II and III are longer than digits I and IV. The digits subsequently become more pointed and begin to separate from each other along the narrow interdigital grooves; complete separation usually occurs first between digits II and III.
There is no apparent change in the thickness of the regenerate epidermis, but skin glands are now seen beneath the entire regenerate epidermis (Fig. 12). In places where the digits have not completely separated, wedges of thickened epidermal cells are still seen projecting into the interdigital tissue. Few blastema cells are still seen in this stage since most of the regenerate appears to be cartilage, muscle, or connective tissue. Those blastema cells which can be seen are located in the distal regions of the autopodium. The skeletal elements of the autopodium have separated from each other, and they are more distinct than in the previous stage.

Late Digits Stage (34 Days after Amputation)
The digits are elongated and pointed (Fig. 2). Digit I is now separated from digit II along the interdigital groove. Digit IV subsequently becomes separated.
The epidermis of the regenerate is no longer thicker than the stump epidermis. The rest of the tissues of the regenerate have become fully differentiated. The cartilage of the radius and ulna is becoming hypertrophic, and some ossification is taking place in the peripheral regions.
Each of the stages we have described can be assigned to one of the three major phases of regeneration as follows: phase one, the phase of wound healing and dedifferentiation of the stump tissues, includes wound healing, early dedi//eren-tiation, and late dedi//erentiation stages; phase two, the phase of blastema accumulation and blastema growth, includes moderate early bud, early bud, and medium bud stages; and phase three, the phase of differentiation and morphogenesis of the regenerate, includes late bud, palette, early digits, medium digits, and late digits stages.

influence o/ the Level el Amputation
In this experiment forelimbs were amputated at either a proximal level (through the humerus) or at a distal level (through the radius and ulna). During the early stages of regeneration, the site of amputation can be easily discerned due to pigmentation differences between the stump and the regenerate, but this boundary becomes indistinguishable by the third or fourth week of regeneration. In order to make accurate observations, camera lucida drawings were made immediately after amputation and at weekly intervals during the subsequent 10 weeks of regeneration. Care was taken to draw limbs in the same position each time.
From Fig. 13, it can be seen that the increase in length of regenerates from two levels of amputation follows normal sigmoid growth curves with lag, rapid growth and stationary phases. Regeneration from the more proximal level results in the production of a significantly longer regenerate. By ten weeks after amputation the mean length of regenerates from proximal stumps was 6.41 mm (~ S. E. of 0.19), whereas, the mean length of regenerates from distal stumps was only 2.96 mm (~ S.E. of 0.09). However, regenerates from the proximal level do not become significantly longer than regenerates from the distal level until three weeks after amputation, i.e., at the beginning of the third phase of regeneration (Table 1). Fig. 14 shows the mean rates of regeneration at different times after amputation at the two levels. During the first two phases of regeneration, the rate of regeneration from both levels appears to be the same, but by three weeks after amputation, as regeneration goes into its third phase, the rate for limbs amputated at the more proximal level is significantly greater than the rate for limbs ampurated at the more distal level (a probability of less than 0.01 that this difference would occur by chance) ( Table 2). The rate of regeneration for limbs amputated at the proximal level continues to be greater than that for limbs amputated at the distal level until ten weeks after amputation. As can be seen from Fig. 14, the most rapid rate of increase in length of regenerates from both levels appears to take place three to four weeks after amputation at the onset of the third phase of regeneration. Neither the animal's length or wet weight can be correlated with the rate of elongation of its limb regenerate.
Despite the differences in the rates of elongation and eventual lengths of limb regenerates from p~oximal and distal levels of amputation seen in the third phase of regeneration, the data illustrated in Fig. 15 indicate that throughout all three phases of regeneration, limbs pass through the same stages at approximately the same times after amputation. The histological appearance of regenerates from the distal level of amputation was not examined since it is clear from the descriptions by Sehmidt (1958a and1958b) and Schott6 and Hall (1952) that regenerates from a distal limb level pass through comparable stages with comparable histo-      Fig. 1 20 Wilhelm l~oux Archiv, Bd. 173 LD logical appearances to those of regenerates from the proximal level examined here. From the histological examination of regenerating digits it was found that the onset of chondrogenesis in a digit occurs at the same time after amputation as it occurs in an upper arm regenerate. Fig. 16 shows that limb regenerates from either a proximal or a distal level pass through the same normal stages during the same phases of growth, and the same stages coincide with the maximum rate of limb regeneration. Therefore, the sequence of morphological and histological events seen in limb regeneration does not appear to be influenced by the level of amputation, the regenerate's eventual length, or its rate of enlogation.

Discussion
Spallanzani (1768) may have been the first to point out the correlation between the amount of appendage removed and the subsequent rate of regeneration when he observed that it took as long for a salamander to regenerate a toe as it took to regenerate an entire leg. Similar observations were subsequently made on tadpole tails (Ellis, 1909 ;Zeleny, 1916), newt tails (Morgan, 1906;Iten and Bryant, unpublished), lizard tails (Tassava and Goss, 1966;Bryant and Bellairs, 1967), and fish fins (Morgan, 1906;Tassava and Goss, 1966). Furthermore, a preliminary study of regeneration from different levels of amputation in adult newt forelimbs by Goss (1969 and personal communication) has shown that regenerates from proximal stumps elongate faster than regenerates from distal stumps.
A number of other factors, including temperature (Ellis, 1909;Maderson and Licht, 1968;Schmidt, 1968), light cycle (Maderson and Salthe, 1971), and season of the year (Schauble, 1972), have been shown to influence both the extent and the rate of regeneration in vertebrate appendages. It is unclear whether body size (length or weight) or ageing in adult urodeles has any influence on the rate of limb regeneration. According to Manner', Zapisek, and Vallee (1960), body size has no apparent influence on the rate of elongation of regenerate limbs; according to Goodwin (1946), ageing in adult urodeles has no apparent imfluence on the rate of regeneration. However, Pritchett and Dent (1972) assume that the age of the newt is proportional to its size and find that the rate of limb regeneration is inversely proportional to the size of the limb and therefore to the age of the newt.
In the experiments reported in this paper, the temperature and light cycle were kept constant, all animals regenerated their limbs at the same time of the year, and all animals were within the smallest category used by Pritchett and Dent (1972). Therefore, we are confident that these factors have not contributed significantly to the differences in rates of regeneration observed in this study.
The results reported here on the influence of the level of amputation on limb regeneration have shown that the first and second phases of regeneration, which last from the time of amputation until three weeks later, are not influenced by the level of amputation. During these two phases, regenerates grow at the same rate and are similar in size irrespective of the level of amputation. Phases one and two encompass wound healing through medium bud stages during which dedifferentiation, blastema accumulation and blastema growth are occurring (see normal stage descriptions). Preliminary results indicate that similar phases with similar characteristics are present during tail regeneration, and further that the sizes of both tail and limb regenerates are about the same when they enter the third phase of regeneration (Iten and Bryant, unpublished). Both the rate of growth and the length of the regenerate are dependent upon the level of amputation during the third phase of limb regeneration. During this phase which begins with the late bud stage (approximately three weeks after amputation), differentiation and morphogenesis of the new limb begins (see normal stage descriptions). Regenerates from either level regenerate seventy to eighty percent of their total length during the third phase and the transition from phase two to three is marked by the maximum rate of growth. However, despite the differences in the rate of lengthening for regenerates from proximal and distal levels, the appearance and duration of normal stages are the same throughout all three phases of regeneration.
The results of the histological study of digit regeneration have shown that the time of onset of differentiation remains the same even when the two levels of amputation are very far removed from each other. Therefore, it appears that the developmental sequence of events seen in limb regeneration is not influenced by the level of amputation. The possibility exists that the sequence of regenerative events could be influenced by the distribution of tissue types at the level of amputation (Stocum, personal communication). In this study, however, the sequence of regeneration following digital amputation at a joint appears similar to that following amputation through the upper ~rm.
The results of this study suggest that position along the limb influences either the rate of cell division, the duration of the maximum rate of cell division, or the percentage of cells actively dividing. The results also suggest that regeneration blastemas are limited to produce only those distal structures which are necessary to complete the appendage from any given level of amputation. The limitation imposed by position along the appendage could be an intrinsic one resulting from the position of origin of the blastema cells, or it could be an extrinsic limitation imposed during regeneration by stump tissues at particular levels (see Wolpert, 1971). Experiments ~re currently in progress to investigate these two possibilities, and to investigate the influence of position on the cellular kinetics within the regeneration blastema.