1. Experiments to test the proposition that systemic factors and not fixed properties of the cells within limbs determine loss of regeneration in frogs were performed upon the forelimbs of post-metamorphic R. clamitans and R. pipiens varying in body length from 4·4 to 5·7 cm.

  2. Control experiments consisting of simple amputations through the midradius ulnar or distal humeral regions showed that the frogs at this stage were incapable of detectable regeneration. The repeated removal of the cicatricial skin in another control series elicited regeneration, of an abortive type, in only one case among fourteen limbs tested.

  3. In twenty-five hypophysectomized R. clamitans adrenals from R. pipiens were transplanted heterotopically beneath the jaw concomitantly with amputation. Weak regenerative response was observed in eleven cases. It is concluded that the absence of the pituitary in the hosts lowers the activity of the transplanted adrenals.

  4. The transplantation of R. pipiens adrenals into normal R. clamitans hosts 7 and 14 days after amputation proved most effective, since among the twenty-one limbs examined only two did not show any regeneration.

  5. It is suggested that: (1) the transplanted adrenals were functional, although at a low level of activity; (2) the time of transplantation of the adrenals is critical, since in order to elicit regeneration it must coincide with a period in amputational wound repair that is most susceptible to corticoid activity; (3) the loss of regenerative potencies after metamorphosis in Anura is attributable to changes in the endocrine system of these organisms, rather than to changes in the properties of the cells of their limbs.

The frog provides among the vertebrates the best opportunity to investigate the factors which determine loss of regeneration within the ontogenetic boundaries of a single organism, for in tadpoles the capacity to regenerate tapers off along the proximo-distal axis of their limbs and in adult frogs limb regeneration is substantially absent.

It was generally accepted that metamorphosed frogs lose this capacity because of progressive complexity of structure in the growing limbs (Marcucci, 1916; Polejaiev, 1936; and Forsyth, 1946). However, the conclusion that it is the nature of the limb tissues which determines presence or absence of regeneration was shown to have only limited validity when regeneration was obtained in metamorphosed frogs. Thus Polezhajev (1945) demonstrated that repeated trauma to the amputational surface of limbs of adult frogs (which according to his own theory had irrevocably lost the powers of regeneration because of progressive differentiation) was sufficient for recuperation of this faculty in a substantial number of cases. By immersing metamorphosed frogs in hypertonic salt solutions, Rose (1944,1945) obtained limb regeneration; and finally, Singer (1954) (and indirectly also Van Stone, 1955), by increasing the nerve-supply of limbs, obtained regeneration from normally unresponsive amputational levels.

The papers of Walter (1911), Schotté (1926), Richardson (1940–5), Hall & Schotté (1951), and Schotté & Hall (1952) concurred in showing that both the pituitary and the thyroid glands influenced regeneration, but these findings could not explain the cessation of regeneration in Anura (Naville, 1927; Guyénot, 1927; and Schotté & Harland, 1943).

The suspicion that cellular properties did not play so dominant a role in regeneration as the above evidence seemed to suggest, became a near-certainty when the endocrinological factors in regeneration were re-examined in the light of discoveries made by mammalian endocrinologists in regard to the pituitaryadrenal synergism under stress conditions (Selye’s general adaptation syndrome, 1947; Selye & Stone, 1950). In 1952 Schotté & Hall proposed that, in a manner already known from mammalian endocrinology, the role of the pituitary in urodele regeneration was probably confined to stimulation of the adrenal cortex after amputational stress. The reality of a pituitary-adrenal synergism in respect to regeneration was demonstrated when, in newts deprived of their pituitary, replacement therapy with ACTH (Schotté & Chamberlain, 1955) and with cortisone (Schotté & Bierman, 1956) restored the regenerative capacities in these animals. The direct involvement of the adrenals in regeneration of urodeles became a certainty when Schotté & Lindberg (1954) induced regeneration in hypophysectomized newts by transplantations of frog adrenals.

Because of marked similarities between urodele and anuran regeneration, and because of the well-known involvement of endocrines in metamorphosis, a process which seemed to coincide with loss of regenerative processes in Anura, it became imperative to determine whether artificial changes in the endocrine system of frogs could modify regeneration of their limbs. The first attempts to investigate this problem were made a few years ago in this laboratory when it was shown that the transplantation of additional adrenals could lead to restoration of regenerative capacity in premetamorphic tadpoles at normally nonregenerating amputational levels (Lindem, unpublished thesis, Amherst College, 1954).

The realization that lost regenerative ability was regained in tadpoles by inducing a hyperadrenal state offered an irresistible invitation to conduct still further investigations on adult frogs. The purpose of this research, therefore, has been to determine whether or not the normally absent capacity for limb regeneration in adult frogs could be recovered by introduction of additional adrenal glands into these animals.

The effects of adrenal transplants upon the regeneration of forelimbs were studied in two American species of frogs: Rana clamitans, commercially procured from the south of the U.S.A.; and R. pipiens (donor of all the adrenal glands), secured from Wisconsin and northern Vermont. These postmetamorphic frogs were force-fed twice weekly on beef-liver with bone-meal supplement before and after the operations. For all the operations the frogs were narcotized by immersion in a solution of MS 222 (1: 1,000), a meta-amino-benzoic acid ethylester in the form of methanesulphonate (Sandoz, Basle).

Amputations were made at either the mid radius-ulnar level or at the distal humerus level of the frog forelimb. These amputations were performed under a dissecting microscope, and, immediately after the initial severance, the protruding bone elements were re-amputated to adjust them to the level of the retracted soft tissues.

For transplantation experiments both narcotized host and donor frogs were placed in petri dishes upon sterile surgical gauze moistened with sterile physiological salt solution. The adrenal glands in frogs appear as thin ribbons of a golden yellow colour extending along the median ventral surface of each kidney. The donors’ kidneys were dissected out and placed upon surgical gauze in another sterile dish and the adrenals were excised from the kidneys with iridectomy scissors and Swiss watchmaker’s forceps underneath a binocular microscope. Care was taken to minimize the amount of kidney tissue adhering to the adrenal bodies. All transplantations were made into the lower jaw because this region of the frog is highly vascularized. For the insertion of the prepared adrenals a small opening was made in the skin of the lower jaw with iridectomy scissors. A sterile probe was used to enlarge the subcutaneous pocket into which the pieces of adrenals were pushed as far anteriorly as possible to prevent postoperative extrusion; no transplants were lost.

The pituitary gland in the frog is easily identified through the semi-cartilaginous parasphenoid bone. The operative procedure was adopted from Levinsky & Sawyer (1952), with the difference that we extirpated the exposed pituitary with watchmaker’s forceps instead of with a pipette.

After any surgical procedure the animals were kept for one day at 13° C. (±1 ° C.) in finger-bowls, the bottoms of which were covered with surgical gauze and moistened with sterile physiologic solution to which a small quantity of sodium sulphadiazine was added. After 1 day the animals were transferred into demineralized water, about ten per aquarium, and were maintained at 20° C. (± 1° C.) for the duration of the experiment.

This research is based upon a study of sixty-five surviving animals from a much larger number of operated frogs. From these cases forty-seven forelimbs were studied histologically; and in-addition, the lower jaws of five frogs were sectioned serially to investigate the condition of the adrenal transplants especially in regard to vascularization; also, the heads of six hypophysectomized animals were examined in sections to verify the completeness of pituitary removal. The tissues were fixed in Bouin’s, decalcified in Jenkin’s solution, stained with Harris’s modification of Delafield’s hematoxylin and counterstained with orange G. For the Study Of slides of adrenal tissues the sections were stained by Mallory’s polychrome method.

1. Survival and functional state of adrenal transplants

At various times the conditions of forty-five heteroplastic or homoplastic adrenal transplants were studied among the fifty-three frogs having received similar transplants. For gross study the whole subcutaneous area of the lower jaws was exposed by dissecting away the skin and in all cases the transplants were found embedded within the richly vascularized sub-dermal tissues. Examination of these transplants in vivo under the dissecting microscope and study on slides both concur to show that the transplanted adrenals remained viable, normally pigmented, and fully vascularized for as long as 125 days, the longest period examined.

The colour photograph of Plate 1 illustrates the survival and the state of vascularization of two consecutive heteroplastic transplantations of adrenal glands from R. pipiens cut in both instances into four pieces to enlarge the surface area. (See explanation of plates for further details.)

The amount of tissue diminishes with time and gradual histolysis was unmistakable; but fragments fixed for histological study around 30 days after the last transplantation appeared normal and healthy: the frog erythrocytes were easily identified within the blood-vessels connecting the host’s tissues with the heterotopic adrenal bodies. The cortical elements appeared as oblong groups of cells and anastamosirig cords, separated by blood sinuses, and this aspect complies in all respects with the histological description of anuran cortices given by Jones (1957). Therefore, from both gross and histological evidence it is inferred that the adrenals ex situ were alive and received an adequate blood supply during the period considered. These indices of healthy survival of adrenocortical elements suggest, but do not prove, the functional activity of these tissues. Few of the many physiological effects of the hyperadrenal state indicated in the literature (Selye & Stone, 1950) have been detected in this study.

No visible atrophy of the adrenals was observed in normal frogs with the exception of pigmentation losses in cases with large dosages of adrenal transplants. Cameron (1953) described peripheral stasis of small blood-vessels and the presence of mucous exudates as symptomatic of a hyperadrenal state, but neither of these conditions was seen in these experiments. Our observations would suggest that the adrenal transplants were not acting cumulatively.

2. Morphological and histological effects of forelimb amputations in control frogs (24 postmetamorphic R. clamitans”, mean body length from nose tip to cloaca, 4-7 cm., range, 4-2 to 5-7 cm.; 16 cases studied histologically)

  • Simple amputations were performed in ten cases to study the histological aspects of non-regeneration in adult frogs, since no such investigations are known to us from the literature. Singer (1954) confines himself to descriptions of the morphological features of early cicatrization in amputated forelimbs, and we concur fully with his descriptions.

    Limbs were amputated either through the forearm or the upper arm and fixed at intervals between 15 to 50 days after amputation. No differential rate of cicatrization was noted between the two amputational levels and this is fully confirmed by slides. Two sections (one for each level) are presented: Plate 2, fig. A represents the usual features of non-regeneration in a longitudinal section of a limb amputated through the radius-ulna: the wound surface is completely healed with epidermis and dermis containing newly formed skin glands covering the amputation surface; directly beneath lies the conspicuous and uninterrupted basement membrane. Another feature that defines wound-healing in a non-regenerating limb is the presence of fibrous connective tissue and of muscle elements, oriented perpendicularly to the ends of the cut bone-shafts. These elements have infiltrated distally to the vacuolated regions of the perichondrium and the hemopoietic area of the bone-marrow; they constitute a fibroid wall and in spite of the presence of numerous ‘blastematous’ cells further progress of regeneration is always prevented in this and in similar cases. As time progresses the stratified layers covering the skeletal structures will become, in cases of more advanced amputation age, much thicker and serve as an impenetrable callus over the cut bones. Another invariable concomitant of inhibited regeneration is the transformation of the periosteal tissue into massive cartilaginous formations.

    Plate 2, fig. B is a section from the upper arm of a much younger frog. In contrast to the preceding case, this amputation surface completely lacks identifiable blastematous cells. Moreover, while showing the features of woundhealing enumerated above, this limb has a much thicker basement membrane and a more clearly defined cartilaginous cap immediately distal to the cut humerus at this later stage of cicatrization. It should be noted that this limb, whose humerus is made up exclusively of cartilage, and is therefore less structurally differentiated, demonstrates no greater capacity for dedifferentiation and regeneration than did the former limb.

    Among the ten controls examined we have found no regeneration on gross inspection, and on slides we found no free, unencapsulated blastematous masses. Singer (1954) records regenerative response in 15 per cent, of the controls in ‘recently metamorphosed frogs’ while our own experiments, performed on much older frogs, corroborate the findings of Thornton & Shields (1945).

  • Amputations combined with interference with cicatrization processes were performed in another series of fourteen R. clamitans, ten limbs of which were fixed 52 days after amputation and studied histologically.

Since early cicatrization is generally considered coincidental with nonregeneration it was thought advisable to test this proposition by means of surgical removal of the skin which, in frogs, precociously invades the amputation surfaces.

Plate 2, fig. C illustrates a case in which skin circumcision was performed twice, once on the seventh and once on the fourteenth day following amputation. The figure shows that normal wound-healing processes are operating, although they have been retarded. While the callus is well established over the skeletal shafts, and a connective tissue pad has formed distally, the basement membrane and skin glands have not yet been reconstituted over the whole amputation surface. The staining properties of some of the perichondrial elements visible in this section suggests some dedifferentiation and blastema-like cell formations may be discerned anteriorly to the cut bone-shafts. This attempt at regeneration, however, has been halted by the fibrous layer capping them. The generality of this aspect of early regeneration ‘interfered with’ has been verified in the remaining nine cases of this series. The skin-removal treatment employed twice on each limb was therefore only able to retard wound-healing, not to induce a recovery of lost ‘embryonic’ properties of cells requisite for regeneration.

3. Effects of adrenal transplants upon regeneration in hypophysectomized frogs (25 postmetamorphic frogs—16 R. clamitans and 9 R. pipiens’, mean body length, 4-4 cm., range, 41 to 4 9 cm.; 16 limbs investigated histologically)

The success in reawakening lost powers of regeneration in hypophysectomized newts by replacement therapy with cortical hormones (Schotté & Bierman, 1956) and with xenoplastic adrenal transplants (Schotté & Lindberg, 1954) suggested the use of a similar procedure in frogs. Concomitant with hypophysectomy one forelimb was amputated in all cases. The time and number of adrenal transplantations are summarized in Table 1 along with the data concerning the effects the transplants exerted on regeneration.

Table 1.

Results of heteroplastic transplantations of adrenals after forelimb amputation in hypophysectomized frogs

Results of heteroplastic transplantations of adrenals after forelimb amputation in hypophysectomized frogs
Results of heteroplastic transplantations of adrenals after forelimb amputation in hypophysectomized frogs

Macroscopic observation showed that during the 7 days preceding transplantation very rapid skin invasion occurred. Experiments to test the comparative rates of wound-healing in normal and hypophysectomized frogs confirmed that wound-healing was much more rapid in the latter: untreated hypophysectomized frogs at 15 days have a reformed dermis, epidermal skin glands, and a state of muscle-repair not yet present in limbs of normal frogs of the same amputational age.

In spite of rapid healing our data show that after adrenal transplantations, regeneration of some sort has occurred in eleveri cases among the twenty-five examined. Excellent regeneration may be observed in the case represented on Plate 3, fig. D. After amputation through the radius-ulna, coincident with hypophysectomy, this frog received a total of three adrenal transplants, administered separately on the seventh, fourteenth, and twenty-first day following amputation. This case afforded the most extensive regeneration of the hypophysectomized hosts, and yet the amount of differentiation and of proliferation is less than in the majority of the normal hosts to be reported below. The section reproduced on the photomicrograph illustrates an abortive attempt at regenerative activity, since the most distal region of the limb remains blastematous and is not engulfed in connective tissue at this stage.

It should be made clear that the kind of regeneration observed in this series was different from that observed in normal hosts to be described below, for only the case just mentioned underwent proliferation, while the remaining limbs did not progress beyond the point of ‘accumulation blastema’ characteristic of the ‘critical’ phase of regeneration described by Forsyth (1946) for anuran premetamorphic tadpoles at non-regenerating levels. This is true even for the group of nine cases that received as much as five consecutive adrenal transplants without any significant increase in the frequency of regeneration (Table 1), which indicates that in hypophysectomized frogs the required adrenal threshold to restore regenerative capacity is much higher than in normal frogs. This result is in keeping with the general observation that in the absence of the pituitary cicatrization takes place much more rapidly, and it also suggests that in the absence of adrenocorticotropic stimulation the transplanted adrenals were at a low functional level.

4. Effect of adrenal transplants upon regeneration in normal frogs (16 post-metamorphic R. clamitans; mean body length, 4·9 cm., body range, 4·7 to 5·3 cm.; 15 limbs studied histologically)

Because previous investigations had shown extreme delays in blastema formation, the amputational surfaces in all cases but four were reopened by skin circumcision simultaneously with the introduction of the first transplant.

All the results from this series are summarized in Table 2. Among the twenty-one limbs (from sixteen individuals) studied only two presented no regeneration. When one considers that among the twenty-four individuals of the previously reported control group only two limbs exhibited any regeneration at all the importance of the adrenal factor introduced in these experiments appears impressive.

Table 2.

Results of heteroplastic transplantations of adrenals after forelimb amputation in normal R. clamitans

Results of heteroplastic transplantations of adrenals after forelimb amputation in normal R. clamitans
Results of heteroplastic transplantations of adrenals after forelimb amputation in normal R. clamitans

Our data show that the first signs of regeneration, in the form of macroscopically detectable blastematous mounds, appeared 30 days after the last transplant, at an amputation age of about 45 days (at 20° C.) in all but one exceptional case (see explanation in Table 2). The second observation which results from our morphological and histological findings is that after an apparently normal start regeneration almost invariably becomes abnormal. A third general conclusion supporting the second centres around the disparity in stages of development at comparable amputation ages between anuran and urodele regeneration. This is well illustrated by the figures of Plate 4, in which five morphologically successive normal stages and one abnormal regenerate are assembled. A regenerate 50 days old (Plate 4, fig. G) corresponds to a newt regenerate of about 20 days of amputation age; the regenerate 60 days old (Plate 4, fig. H) corresponds to a newt regenerate of about 25 days; the regenerate 72 days old (Plate 4, fig. I) is comparable to the ‘palette’ stage of a newt about 35 days old. Finally, Plate 4, fig. J, of an amputation age of 110 days is comparable to the regenerate of a newt at 40 days. This particular regenerate, FJW26, was fixed, and a longitudinal section is represented on Plate 3, fig. E. Our records show that at the time of fixation this regenerate had not ceased to grow in length, an observation supported by the histological aspect of this section: while at its distal portion separate digital phalanges are discernible, the other bones of the forearm are still unindividualized.

The fifth normally regenerating case is represented by Plate 4, fig. K, and it shows clearly visible digital differentiations 130 days after amputation. A similar state of regeneration in an adult newt would have been achieved as early as 50 days after amputation. On slides this case showed clearly separated carpal, metacarpal, and phalangeal cartilaginous islets. However, a certain degree of syndactylism is evidenced by carpal fusion.

Frog regenerates of this type have the essential attributes of morphogenesis, growth, and differentiation that make these formations appear like true urodele regenerates. That this is not always the case has been observed in several instances where, instead of a regular terminal manus formation, a mushroom-like growth results. Such conditions may be discerned at the earliest beginnings of regenerative proliferation, and they are illustrated in statu nascendi, as it were, in Plate 3, fig. F. The lack of symmetry in cellular configuration is dramatized by the disorderly alignment of multiple cartilaginous and connective tissue whorls. It becomes understandable how after a long period (up to 8 months after amputation) continuous chaotic growth leads to a regenerate of abnormal appearance, such as has occurred in the left limb of case FJW19 (Plate 5, fig. M).

Not always, however, does abnormal regeneration of this type degenerate into a formation devoid of ‘order’. In other cases, such as represented by Plate 4, fig. L, axial growth is at first normal, but the distal portions of the regenerate, no doubt due to early morphogenetic disorientation, are abnormal. On slides this case exhibits a normal stem formation, but where the manus should differentiate there is within a large cartilaginous terminal mass only a vague suggestion of subdivision, even as late as 7 months after amputation.

Such cases, where gross observation and histological verification show early differentiation of multiple cartilaginous cores, always degenerate into abnormal appendages, and they have been observed several times in this investigation. Similar results have also been obtained from another research performed at this laboratory following administrations of ACTH and of cortisone to amputated adult frogs (Schotté, unpublished).

That true regeneration, still not perfect but of a type not yet reported, may result from adrenal transplants is exemplified by the corresponding right limb of the same frog, FJW19 (Plate 5, figs. M and N), the photograph being taken after 8 months of regeneration. This regenerate has always kept the appearance of a normal blastema, and at the time of writing there is a somewhat syndactylous manus with, however, one individualized digit and another finger (as indicated on the X-ray picture) with separated metacarpals and at least two phalanges fused with the rest of the manus.

To recapitulate the data from this series of experiments, it may be stated that the addition of the equivalent of the adrenal complement from one adult frog (administered within one to two weeks after amputation) has brought about regeneration in all but one post-metamorphic frog. The most surprising new finding was that in three cases the delayed transplantation of adrenals brought about regeneration in limbs which had visibly healed over their amputational surfaces.

This research has been undertaken to investigate whether or not the lost regenerative capacity in adult frogs could be reinstated by introducing additional adrenal glands. The results provide an affirmative answer to this question only so far as the positive action of the adrenals is concerned, for an investigation of the separate effects of other tissues, glands, or substances is not yet concluded. We are therefore unable at this time to attribute the restoration of regenerative capacity in frogs to the direct and exclusive action of the transplanted frog adrenals. In this respect it is imperative to test the possible role of kidney tissue, as it was impossible to entirely separate kidney elements from the adrenal islets by ordinary surgical procedures. Moreover, preliminary experiments performed with newts have revealed a complex situation in respect to the possible stressor effect of actively secreting and therefore poisonous and irritating transplanted kidney tissues. Since stress in Selye’s sense invariably involves the pituitaryadrenal axis, it would not be surprising to discover that transplanted kidneys also are capable of creating a hyperadrenal state conducive to regeneration (Schotté & Lindberg, 1954).

Before concluding that the adrenal transplants are the sole agents in bringing about regeneration in adult frogs, the importance of skin removal and of amputational stress must first be considered because of the possible stressor role of surgical trauma in newts (Pellman & Schotté, 1955; Lindberg & Schotté, 1955; Schotté & Bonneville, 1955). The amputation alone of another limb must be excluded as a stressor agent capable of promoting regeneration, since it was followed in all cases simply by the normal wound-healing mechanism. Secondly, concerning the effects of skin circumcision, only one of the fourteen forelimbs upon which repeated skin removal around the amputational surface was performed exhibited any regeneration, and such a regenerate did not continue growth beyond the accumulation phase. Thus, this treatment did not restore any regenerative activity comparable to that observed in the transplantation experiments.

These brief considerations will suffice to re-emphasize the belief that the results obtained are attributable to the introduction of additional adrenals. But then to what effect of the transplanted adrenals may one attribute the induced regeneration?

Since experimentation in this laboratory has implicated a pituitary-adrenal synergism in establishing conditions favourable to regeneration in newts, one wonders whether such a mechanism is operative in adult frogs also. Unpublished investigations regarding the effects of stress upon adult frogs and adrenal transplants upon regeneration in tadpoles would intimate that this is so. However, to support this hypothesis, an important piece of information is required. Were the adrenal transplants synthesizing cortico-steroids found to be essential for the initiation of regeneration in urodeles (Schotté & Bierman, 1956)? Besides the gross and microscopical condition of the adrenal tissues described in the experimental part there is additional evidence to indicate that the adrenal transplants were functioning, but at a basal level.

Firstly, in hypophysectomized frogs the bright yellow pigmentation of the adrenals faded in a manner identical to that in normal hosts receiving high adrenal dosages by transplantation. The paleness of the hypophysectomized host’s own adrenals was no doubt due to the fact that involution of the adrenal cortex, in the absence of ACTH, produces a lesser concentration of the yellowcoloured lipids in the cortical cells. The atrophy in the normal hosts receiving additional adrenals implies that the transplanted adrenals were functional, since in the endocrine system compensation for more than normal concentrations of a hormone is by atrophy of the organ synthesizing it. Secondly, bioassays of cortical secretions from isolated mammalian adrenal glands through which blood was perfused showed that deprivation of ACTH did not interrupt, but only lowered their secretory activity (Vogt, 1951). This result suggests that the frog cortices would not completely terminate their activity in the absence of the trophic hormone.

The next important problem concerns the duration of activity of the transplants. Both these results and those from a previous study in newts (Schotté & Lindberg, 1954) imply that there is a transitory period of corticoid activity, for in both experiments regeneration was only associated with transplantations made around two weeks after amputation, not when adrenals were introduced earlier. Apparently the period of cortical activity must be integrated with the time, late in wound repair, that is most susceptible to corticoid action. The failure to induce regeneration with smaller dosages in other experiments was probably not the result of inadequate quantitative dosage, but the result of premature administration, before the limb tissues were sensitive to proper interaction with cortico-steroids like cortisone. For this hormone is known to effect many changes in wound repair, such as diminishing cellular migration and infiltration, inhibiting fibroblast formation, and fibrin deposition (Cameron, 1953); and all these responses repress normal wound-healing processes which in turn interfere with the mechanisms of regeneration.

If these propositions regarding the delicate hormonal changes at the propitious moments when the frogs’ limbs are in a state of repair most receptive to their action are valid, why is it that the adrenal transplants are unable to facilitate regeneration in hypophysectomized frogs to the same degree? It seems that there may be two main reasons. Firstly, the total absence of ACTH in animals deprived of their pituitaries is no doubt responsible for a much lower level of activity of the transplanted adrenal glands in these frogs; in normal hosts, however, after vascular communication has been established, ACTH is present for cortical activation. Secondly, the much higher rate of cicatrization observed in hypophysectomized frogs possibly modifies the time at which the corticoid activity is most instrumental in preventing wound-healing.

The general results of this study, then, together with the still unpublished aforementioned experimental findings, support the proposition that a pituitaryadrenal synergism is operating in adult frogs as in newts; they also suggest that an artificially-induced hyperadrenal state is sufficient to determine recuperation of regenerative capacities in adult frogs. This, of course, does not preclude other factors (nerves, for example) from being just as effective in restoring regenerative potencies.

In conclusion it may be stated with a certain degree of confidence that the normal loss of regenerative potencies in Anura, more or less coincidental with metamorphosis, is attributable to endocrine changes in these organisms, rather than to irreversible modifications in properties of the cellular constituents of their limbs.

Thanks are due to the Division of Grants of the U.S. National Institutes of Health (Grant 2236) which has made this research possible. We also wish to thank Mrs. Jean Francis for excellent histological work and Mr. Carl Howard for consummate skill in photography.

Cameron
,
G. R.
(
1953
).
A survey of tissue responses to ACTH and cortisone
.
The Suprarenal Cortex
.
New York
:
Academic Press
.
Forsyth
,
J. W.
(
1946
).
The histology of anuran limb regeneration
.
J. Morph
.
79
,
287
322
.
Guyénot
,
E.
(
1927
).
La perte du pouvoir régénérateur des Anoures, étudiée par les hétérogreffes, et la notion de territoires
.
Rev. suisse Zool
.
34
,
1
51
.
Hall
,
A. B.
, &
Schotté
,
O. E.
(
1951
).
Effects of hypophysectomies upon the initiation of regenerative processes in the limb of Triturus viridescens
.
J. exp. Zool
.
118
,
363
88
.
Jones
,
I. C.
(
1957
).
The Adrenal Cortex
.
London
:
Cambridge University Press
.
Levinsky
,
G.
, &
Sawyer
,
W.
(
1952
).
Influence of the adenohypophysis on the frog water-balance response
.
Endocrinology
,
51
,
110
16
.
Lindberg
,
D. A.
, &
Schotté
,
O. E.
(
1955
).
Amputational trauma and regeneration of limbs in normal and in hypophysectomized newts
.
Anat. Rec
.
121
,
331
.
Marcucci
,
E.
(
1916
).
Capacità rigenerativa degli arti nelle larve di Anuri e condizioni che ne determinano la perdita
.
Arch. zool. (ital.), Napoli
,
8
,
89
117
.
Naville
,
A.
(
1927
).
La perte du pouvoir régénérateur des Anoures étudiée par les homogreffes
.
Rev. suisse Zool
.
34
,
269
84
.
Pellman
,
C.
, &
Schotté
,
O. E.
(
1955
).
Influence of repeated anesthesia upon regeneration of limbs in normal and in hypophysectomized newts (Triturus viridescens)
.
Anat. Rec
.
121
,
351
2
.
Polejaiev
,
L. W.
(
1936
).
Sur la restauration de la capacité régénérative chez les Anoures
.
Arch. Anat. micr
.
32
,
437
63
.
Polezhajev
,
L. V.
(
1945
).
Limb regeneration in adult frog
.
C.R. Acad. Sci. U.R.S.S
.
49
,,
609
12
.
Richardson
,
D.
(
1940
).
Thyroid and pituitary hormones in relation to regeneration. 1. The effects of anterior pituitary hormone on regeneration of the hind limb in normal and thyroid-ectomized newts
.
J. exp. Zool
.
83
,
407
25
.
Richardson
,
D.
(
1945
).
Thyroid and pituitary hormones in relation to regeneration of the hind limb of the newt, Triturus viridescens, with different combinations of thyroid and pituitary hormones
.
J. exp. Zool
.
100
,
417
27
.
Rose
,
S. M.
(
1944
).
Methods of initiating limb regeneration in adult Anura
.
J. exp. Zool
.
95
,
149
70
.
Rose
,
S. M.
(
1945
).
The effect of NaCl in stimulating regeneration in limbs of frogs
.
J. Morph
.
77
,
119
39
.
Schotté
.
O. E.
(
1926
).
Hypophysectomie et régénération chez les Batraciens modèles
.
C.R. Soc. Phys. Hist. nat. Genève
,
43
,
67
72
.
Schotté
.
O. E.
&
Bierman
,
R.
(
1956
).
Effects of cortisone and allied adrenal steroids upon regeneration in hypophysectomized Triturus viridescens
.
Rev. suisse Zool
.
63
,
353
75
.
Schotté
.
O. E.
&
Bonneville
,
M.
(
1955
).
The systemic effect of injury and of repair within dehumerized limbs of Triturus viridescens
.
Anat. Rec
.
121
,
364
.
Schotté
.
O. E.
&
Chamberlain
,
J.
(
1955
).
Effect of ACTH upon limb regeneration in normal and in hypophysectomized Triturus viridescens
.
Rev. suisse Zool
.
62
,
253
79
.
Schotté
.
O. E.
&
Hall
,
A. B.
(
1952
).
Effects of hypophysectomy upon phases of regeneration in progress. (Triturus viridescens
.)
J. exp. Zool
.
121
,
521
60
.
Schotté
.
O. E.
&
Harland
,
M.
(
1943
).
Amputation level and regeneration in limbs of late Rana clamitans tadpoles
.
J. Morph
.
73
,
165
80
.
Schotté
.
O. E.
&
Lindberg
,
D.
(
1954
).
Effect of xenoplastic adrenal transplants upon regeneration in normal and in hypophysectomized newts. (Triturus viridescens
.)
Proc. Soc. exp. Biol. Med. N.Y
.
87
,
26
29
.
Selye
,
H.
(
1947
).
Textbook of Endocrinology
.
Acta Endocrinologica. Montréal
:
Université de Montréal
.
Selye
,
H.
&
Stone
,
H.
(
1950
).
On the experimental morphology of the adrenal cortex
.
Springfield, Ill
.:
Charles E. Thomas
.
Singer
,
M.
(
1954
).
Induction of regeneration of the forelimb of the postmetamorphic frog by augmentation of the nerve supply
.
J. exp. Zool
.
126
,
419
72
.
Thornton
,
C. S.
, &
Shields
,
T. W.
(
1945
).
Five cases of typical regeneration in the adult frog
.
Copeia
,
1
,
40
42
.
Van Stone
,
J. M.
(
1955
).
The relationship between innervation and regenerative capacity in hind limbs of Rana sylvatica
.
J. Morph
.
97
,
345
92
.
Vogt
,
M.
(
1951
).
Cortical secretion of the isolated perfused adrenal
.
J. Physiol
.
113
,
129
56
.
Walter
,
F. K.
(
1911
).
Schilddriise und Regeneration
.
Arch. EntwMech. Org
.
31
,
91
130
.
plate 1

Ventral view of lower jaw of an adult R. clamitans (case FJW9) after removal of skin showing four prominent adrenal transplants (from R. pipiens donors) 125 or 118 days after transplantation. Several smaller islets below derive from other unsuccessful or regressing adrenal transplants. Note vascular connexions of the four main transplants.

plate 1

Ventral view of lower jaw of an adult R. clamitans (case FJW9) after removal of skin showing four prominent adrenal transplants (from R. pipiens donors) 125 or 118 days after transplantation. Several smaller islets below derive from other unsuccessful or regressing adrenal transplants. Note vascular connexions of the four main transplants.

plate 2

Fig. A. Photomicrograph of a longitudinal section of left forelimb of case FJW67 amputated through the radius-ulnar region and fixed after 23 days of amputation age. Note reformation of skin over amputational surface, complete with basal membrane. Also characteristic for arrested regeneration are the masses of procartilage formed on both sides of the periosteum at the expense of perios,teal and other connective tissue cells. ( × 25)

Fig. B. Photomicrograph of a longitudinal section of left forelimb of case AFC82, amputated through distal humerus and fixed 50 days after amputation. Note lack of blastematous cells and the cartilaginous nature of the humerus. (× 25)

Fig. C. Photomicrograph of a longitudinal section of right forelimb of case FJW93 amputated through the radius-ulnar level and fixed 50 days later. Note the incompletely reformed basement membrane, the crescent-like pad of fibrous tissue distal to the cut shafts of the bone collar, and the dense periosteal cartilage. (× 25)

plate 2

Fig. A. Photomicrograph of a longitudinal section of left forelimb of case FJW67 amputated through the radius-ulnar region and fixed after 23 days of amputation age. Note reformation of skin over amputational surface, complete with basal membrane. Also characteristic for arrested regeneration are the masses of procartilage formed on both sides of the periosteum at the expense of perios,teal and other connective tissue cells. ( × 25)

Fig. B. Photomicrograph of a longitudinal section of left forelimb of case AFC82, amputated through distal humerus and fixed 50 days after amputation. Note lack of blastematous cells and the cartilaginous nature of the humerus. (× 25)

Fig. C. Photomicrograph of a longitudinal section of right forelimb of case FJW93 amputated through the radius-ulnar level and fixed 50 days later. Note the incompletely reformed basement membrane, the crescent-like pad of fibrous tissue distal to the cut shafts of the bone collar, and the dense periosteal cartilage. (× 25)

plate 3

Fig. D. Photomicrograph of the forelimb of case FJW24 fixed 120 days after amputation. The amputation level is situated at the bottom of the figure where the old bony formations are still visible. Within the regenerated procartilaginous mass forming the distal portion of the radius and the general, still not separated mass of the manus, may be distinguished small hemopoietic islets. From the aspect of the distal skeletal formations it is doubtful whether digital formations would ever emerge. (× 25)

Fig. E. FJW26. Photomicrograph of a longitudinal section of left forelimb of case FJW26, at 110 days amputation age. Radius and ulna are well regenerated and elements of the basipodium of the metacarpals and even the phalanges are well indicated on this and on other sections. (× 25)

Fig. F. Photomicrograph of a longitudinal section of the right forelimb of case FJW9 fixed 60 days after amputation and 46 days after the second adrenal transplantation. The long cartilaginous shaft is the radius and ulna, probably fused, around which periosteum begins to form. From the distal area of these procartilaginous formations several whorls of cartilage are derived which have distinct and irregularly distributed centres. Since there is no regular connective tissue cap precluding further growth, these irregular blastematous centres may diverge into tridimensional, often digit-like formations which eventually form a mushroom-shaped regenerate. (× 25)

plate 3

Fig. D. Photomicrograph of the forelimb of case FJW24 fixed 120 days after amputation. The amputation level is situated at the bottom of the figure where the old bony formations are still visible. Within the regenerated procartilaginous mass forming the distal portion of the radius and the general, still not separated mass of the manus, may be distinguished small hemopoietic islets. From the aspect of the distal skeletal formations it is doubtful whether digital formations would ever emerge. (× 25)

Fig. E. FJW26. Photomicrograph of a longitudinal section of left forelimb of case FJW26, at 110 days amputation age. Radius and ulna are well regenerated and elements of the basipodium of the metacarpals and even the phalanges are well indicated on this and on other sections. (× 25)

Fig. F. Photomicrograph of a longitudinal section of the right forelimb of case FJW9 fixed 60 days after amputation and 46 days after the second adrenal transplantation. The long cartilaginous shaft is the radius and ulna, probably fused, around which periosteum begins to form. From the distal area of these procartilaginous formations several whorls of cartilage are derived which have distinct and irregularly distributed centres. Since there is no regular connective tissue cap precluding further growth, these irregular blastematous centres may diverge into tridimensional, often digit-like formations which eventually form a mushroom-shaped regenerate. (× 25)

plate 4

All figures here represented are regenerates from hosts which received two successive R. pipiens adrenal transplants 7 and 14 days after amputation of the forelimb.

Fig. G. R. clamitans forelimb regenerate at the stage of early blastema, 50 days after amputation.

Fig. H. R. clamitans regenerate at the blastema stage, 60 days after amputation.

Fig. I. R. clamitans regenerate at the stage of flattened palette, 72 days after amputation.

Fig. J. R. clamitans forelimb regenerate 110 days after amputation. (A photomicrograph of a section of that limb is reproduced on Plate 3, fig. E.)

Fig. K. Forelimb regenerate at the stage of prominent digital differentiation, 130 days after amputation, showing clearly defined three digital indentations.

Fig. L. Abnormal regenerate from forelimb amputated through the proximal radius-ulna showing a boxing-glove-like curvature at its tip, 7 months after amputation.

plate 4

All figures here represented are regenerates from hosts which received two successive R. pipiens adrenal transplants 7 and 14 days after amputation of the forelimb.

Fig. G. R. clamitans forelimb regenerate at the stage of early blastema, 50 days after amputation.

Fig. H. R. clamitans regenerate at the blastema stage, 60 days after amputation.

Fig. I. R. clamitans regenerate at the stage of flattened palette, 72 days after amputation.

Fig. J. R. clamitans forelimb regenerate 110 days after amputation. (A photomicrograph of a section of that limb is reproduced on Plate 3, fig. E.)

Fig. K. Forelimb regenerate at the stage of prominent digital differentiation, 130 days after amputation, showing clearly defined three digital indentations.

Fig. L. Abnormal regenerate from forelimb amputated through the proximal radius-ulna showing a boxing-glove-like curvature at its tip, 7 months after amputation.

plate 5

Fig. M. Ventral view of R. clamitans, 5 3 cm. body length, and amputated bilaterally, the left limb, however, amputated 45 days after the right (case FJW19). Two adrenal transplantations were performed 7 and 14 days respectively after amputation of the right limb. Note mushroom-like regenerate at left and excellent regeneration with digital differentiations at right.

Fig. N. X-ray photograph of the right limb of the above case taken 8 months after amputation and adrenal transplantations. Note the fused osseus rod of radius-ulna, some carpal formations and particularly phalangeal ossifications within the free digit.

plate 5

Fig. M. Ventral view of R. clamitans, 5 3 cm. body length, and amputated bilaterally, the left limb, however, amputated 45 days after the right (case FJW19). Two adrenal transplantations were performed 7 and 14 days respectively after amputation of the right limb. Note mushroom-like regenerate at left and excellent regeneration with digital differentiations at right.

Fig. N. X-ray photograph of the right limb of the above case taken 8 months after amputation and adrenal transplantations. Note the fused osseus rod of radius-ulna, some carpal formations and particularly phalangeal ossifications within the free digit.