/ . Embryo/, exp. Morph. Vol. 41, pp. 125-135, 1977
Printed in Great Britain © Company of Biologists Limited 1977
125
The effects of thalidomide and two analogues
on the regenerating forelimb of the newt
By ALLAN S. BAZZOLI, 1 JEANNE MANSON, 2
WILLIAM J.SCOTT 1 AND JAMES G.WILSON 1
From the Children's Hospital Research Foundation,
Department of Pediatrics and Department of Environmental Health,
College of Medicine, University of Cincinnati
SUMMARY
Oral administration (3 mg/day) of thalidomide during the dedifferentiation and early
limb-bud stages of newt forelimb regeneration produced a variety of specific limb deformities.
Proximal and preaxial skeletal elements were the most severely malformed, e.g., preaxial
hemimelia, severe proximal deformities, and preaxial polydactyly.
Likewise, oral, daily doses (3 mg) of the teratogenic analogue, EM12, on days 7 and 8
following bilateral amputation caused the same incidence and type of forelimb abnormalities
as did thalidomide. Conversely, the non-teratogenic analogue, EM87, when orally administered
(3 mg/day) on days 7 and 8 post-amputation resulted in a low rate of limb deformities,
similar in type to those seen in control regenerates.
The type of limb deformities observed in the regenerating newt forelimb following thalidomide treatment nearly mimic those seen in the human and monkey syndromes. Therefore, the
newt represents a possible model for investigating some of the problems associated with
thalidomide teratogenesis.
INTRODUCTION
After fifteen years of extensive investigation, the mode of teratogenic action
of thalidomide, the most crippling human teratogen, remains unclear. Even
the primary site of tissue insult is obscure.
One major obstacle hampering studies of the mechanism of thalidomide
teratogenesis has been the inability to reproduce the human syndrome, particularly the phocomelic-like limb defects, in common laboratory animals. Rats and
mice have proven virtually resistant to thalidomide-induced congenital malformations (see Cahen for review, 1964). One notable exception has been the
results of Kohler and Koch (1974) who reported a characteristic thalidomide
syndrome in one strain of rat and mouse. Attempts to duplicate those results
in both rats and mice in this laboratory have failed (unpublished observations).
Rabbits, although exhibiting fetal limb deformities in response to maternal
ingestion of thalidomide (Giroud, Tuchmann-Duplessis & Mercier-Parot
1
Authors' address: Division of Pathologic Embryology, Children's Hospital Research
Foundation, Elland and Bethesda Aves, Cincinnati, Ohio 45229, U.S.A.
2
Author's address: Department of Environmental Health, Kettering Laboratory, University of Cincinnati Medical School, Cincinnati, Ohio 45267, U.S.A.
9
EMB 41
126
A. S. BAZZOLI, J. MANSON, W. J. SCOTT AND J. G. WILSON
1962; Seller, 1962; Somers, 1962; Schumacher, Blake, Gurian & Gillette,
1968), lack both the consistency and severity of response attained in the human
syndrome.
Conversely, non-human primates such as macaque monkeys react almost
identically to thalidomide as does man (Delahunt & Lassen, 1964; Wilson &
Gavan, 1967; see Wilson, 1973 for review). However, the general shortage of
primates combined with the astronomical cost of monkey research effectively
limit their use for speculative mechanistic investigation such as those now associated with thalidomide.
For the aforementioned reasons, existing laboratory animals have proven
inadequate for studying the effects of thalidomide at the organ, tissue and
cellular level. Therefore, attempts were initiated to determine whether the
regenerating amphibian limb could be induced to respond to this potent primate
teratogen. Fundamental to this approach is the fact that limb regeneration and
ontogeny are governed by similar developmental processes: ectodermal-mesodermal interaction and proximo-distal development appear analogous in both
systems, as is the ability to legulate and compensate for injury (see Thornton,
1968; Faber, 1971; Stocum, 1975 for review.)
The present studies indicate that thalidomide produces specific limb deformities similar to the human syndrome in the regenerating forelimb of the newt,
Notophthalmus viridescens. Further, thalidomide teratogenesis occurs during a
specific time interval, paralleling the dedifferentiation and early limb-bud stages
of regeneration.
To examine the general sensitivity of regenerating newt limbs to thalidomidelike compounds, two analogues were tested during the dedifferentiation stage.
A teratogenic analogue of thalidomide, EM ]2 (Schumacher, Terapane, Jordan
& Wilson, 1972), produced the same incidence and type of malformations as
thalidomide. The non-teratogenic analogue, EM87 (Graudums, Miickter and
Frankus), an experimental drug of Chemie Griinenthal, caused a low rate of
limb defects similar to those observed in controls.
MATERIALS AND METHODS
Adult newts {Notophthalmus viridescens) used in these studies were collected
in southwestern Ohio and housed in large plastic containers (15 newts/cage).
A 12-h light/dark cycle and a constant temperature of 25-0 ± 0-2 °C were maintained for the duration of the experiment. Animals were fed a diet of diced rat
liver three times weekly; containers were cleaned and refilled with fresh
aerated distilled water after each feeding.
Immediately following transport to the laboratory, the newts were quarantined
for a ten-day period. An oral, daily dose of tetracycline (0-75 mg/animal) was
administered during the first seven days (modification of Gibbs, 1963), to
combat an endogenous bacterial infection (Red Leg). Prior to implementation of
Effects of thalidomide on regenerating newt forelimb
127
the above regimen, nearly 70 % of all experimental animals died within five
weeks post-amputation; once treatment began, the mortality rate dropped to
5 %. Also during the quarantine interval, Aqua Aid (methylene blue and acriflavin) was added to each container to control fungal infection. Only during the
quarantine period were newts subjected to tetracycline treatment and Aqua Aid
exposure.
Immediately after quarantine period, newt forelimbs were amputated bilaterally through the distal third of the humerus. Care was taken to trim back any
humeral protrusion and/or overlapping skin flap, insuring a clean, smooth
amputation surface. Following amputation, the newts were kept out of water
in moist containers for two hours to promote healing, then returned to an
aqueous environment and maintained as described above.
On specific days (see schedule, Table 1) a freshly prepared suspension of thalidomide in 0-3 % carboxymethylcellulose (CMC) was administered to each newt
(30 /d/newt) by gastric intubation. The thalidomide analogues, EM 12 and EM87
(supplied along with thalidomide by Dr F. Helm, Chemie Griinenthal), prepared
in similar fashion, were administered orally (3 mg/animal) to separate groups of
newts during one time interval (days 7, 8). Controls consisted of one untreated
and two CMC treated groups of animals.
Approximately 60 days after amputation, all newts were sacrificed, their
limbs collected and fixed in Bouin's fluid. Subsequent staining in methylene
blue and clearing in methyl salicylate allowed examination of the cartilaginous
skeleton (Bryant & Iten, 1974). The occurrence of absent, hypoplastic or supernumerary skeletal elements was recorded for each limb.
RESULTS
Variations in control regenerate limbs
Two types of controls were used: (1) Untreated controls (group I) whose
limbs were bilaterally amputated then allowed to regenerate undisturbed,
and (2) CMC-treated controls who received CMC orally (30/d/day) during
days 7-12 (group II) or days 15-20 (group III) post-amputation (Table 1).
Fig. 1A depicts what was considered to be a normally regenerated limb.
Deviations in the number of chondrified carpals (5-9) and phalangeal (4-10)
elements were considered within normal limits.
The occurrence of limb malformations in control regenerates is shown in
Table 1. Untreated control limbs exhibited a malformation rate of 10 % while
CMC treated controls (groups II and III) had rates of 15 % and 23 %, respectively. These slight increases in CMC controls were not statistically significant
by Chi-square analysis. Generally, the limb abnormalities found in both untreated and CMC-treated newts were much less severe than those observed in
thalidomide groups (Table 2). Except for the limbs of one animal in group II
which exhibited a characteristic thalidomide-like defect, i.e., preaxial hemimelia
9-2
128
A. S. BAZZOLI, J. MANSON, W. J. SCOTT AND J. G. WILSON
Table 1. The effects of CMC, thalidomide, and analogues on newt
forelimb regeneration
Morphology
Group
Treatment
Days treated
I
II
III
IV
V
VI
VII
VIII
IX
X
control
CMC
CMC
thai.
thai.
thai.
thai.
thai.
EMi,
EM 87
none
7-12
15-20
4-6
7,8
10-12
15,16
20,21
7,8
7,8
1
Normal limbs Malf. limbs
18
17
10
3
6
5
9
16
6
13
2
3
3
8
17
11
11
4
13
4
% Malf.
10
15
23
73 1
741
70 1
55
20
68 1
23
P < 001 compared with Group 11 vehicle treated controls.
control animals yielded nonspecific limb defects such as non-chondrified
phalanges (Fig. 1B) or slight radial and ulnar reduction (Fig. 1C).
Thalidomide syndrome in regenerating forelimbs
Severe, specific limb malformations resulted from oral, daily doses (3 mg) of
thalidomide to newts during the dedifferentiation and early limb-bud stages of
regeneration. The incidence of malformed limbs approached 75 % in treated
groups IV, V and VI (Table 1), all of which were statistically significant when
compared to control group II (x2's = 7-89, 13-50 and 8-66, P < 001). The
type of limb deformity produced by thalidomide changed with the posttreatment interval: early treatment (groups IV, V and VI) was associated with
proximal deformities and preaxial polydactyly almost exclusively, while later
treatment (group VII) resulted in an abundance of limbs with distal reduction
deformities. Although the malformation rate in group VII is not statistically
significant when compared to the corresponding control value (group III), many
of these malformations were considered to be thalidomide-induced because of
structural features similar to other treated groups. However, the data suggest
that the sensitivity to altered limb regeneration by thalidomide is waning by
day 15. Regenerating newt forelimbs appear insensitive to thalidomide-induced
defects by day 20 as shown by the low rate of malformations in group VIII.
Table 2 summarizes the types of skeletal forelimb abnormalities observed in
thalidomide-treated newts. Preaxial hemimelia, severe proximal deformities, and
preaxial polydactyly represent categories of limb malformations that were
considered highly specific for thalidomide treatment. Limbs with preaxial
hemimelia lacked the entire radius, and preaxial carpals, metacarpals and
digits 1 and 2 were frequently absent. The ulna was usually reduced in size and
Effects of thalidomide on regenerating newt forelimb
129
Fig. 1. Abbreviations: h, humerus; r, radius; u, ulna; c carpals; m, metacarpals;
p, phalanges. (Magnification x 10). (A) Day 55 untreated control regenerate.
Normal morphology. (B) Day 62 control regenerate. Received CMC (30 /*l/day)
during days 15-20 post-amputation. Distal reduction deformities: non-chondrified
phalanges. (C) Day 62 control regenerate. Received CMC (30 /*l/day) during days
15-20 post-amputation.Slight proximal deficiencies; both radius and ulna shortened.
(D) Day 64 treated regenerate. Received thalidomide on days 7, 8 (3 mg/day) postamputation. Preaxial hemimelia: radius absent; preaxial carpals and metacarpals
missing; ectrodactylous digits 1 and 2. (E) Day 64 treated regenerate. Received
thalidomide on days 4-6 (3 mg/day) post-amputation. Severe proximal deformities:
radius absent; rudimentary ulna fused with carpal; preaxial polydactyly. (F) Day
64 treated regenerate.Received thalidomide on days 4-6 (3 mg/day) post-amputation.
Preaxial polydactyly: arrow indicates bifurcation of digit 2; rudimentary radius;
ulna bowed and fused to humerus; supernumerary carpals.
fused to the humerus (Fig. 1D). Severe proximal deformities consisted of limbs
with an absent or small radial fragment, and an ulna that was usually malformed
and occasionally fused to the humerus. Pieaxial carpals, metacarpals, and digits
were often missing except in the cases where preaxial polydactyly was observed
(Fig. 1E). The final specific limb defect, preaxial polydactyly, affected either
130
A. S. BAZZOLI, J. MANSON, W. J. SCOTT AND J. G. WILSON
Table 2. Summary of specific forelimb malformations resulting from CMC,
thalidomide and analogue administration to adult newts during regeneration
Abnormal morphological categories
Treatment
(group)
Control (I)
Control (II)
Control (III)
Thai.
(IV)
Thai.
(V)
Thai.
(VI)
Thai. (VII)
Thai. (VIII)
EM12 (IX)
EM87
(X)
Total no.
limbs mal.
Preaxial
hemimelia1
Severe
proximal
deformities2
Preaxial
polydactyly3
Distal
reduction
deformities4
Slight
proximal
deficiencies5
2
3
3
8
17
11
11
4
13
4
0
2
0
4
6
2
3
2
4
0
0
0
0
4
7
7
4
1
5
1
0
0
0
3
3
0
0
1
2
0
1
0
1
0
0
1
4
2
1
2
1
1
2
0
3
1
2
0
3
1
1
Radius absent with or without preaxial ectrodactyly; ulna usually malformed.
Absent or rudimentary radial bone accompanied by ulnar reduction; occasional preaxial ectrodactyly or polydactyly.
3
Duplication of 1st or 2nd digit; supernumerary carpals usually accompanies polydactyly.
4
Missing or non-chondrified phalanges.
5
Shortened radius and/or ulna.
2
digits 1 or 2, and in most cases was associated with supernumerary carpals
(Fig. 1F). Less specific limb deformities involved distal reduction deformities,
i.e., missing or non-chondrified phalanges as in Fig. IB, and slight proximal
deficiencies, i.e., shortened radial or ulnar long bones as in Fig. 1C.
Malformations caused by thalidomide analogues
Oral administration (3 mg/day) of the teratogenic analogue, EM12, to newts
on days 7, 8 following bilateral forelimb amputation produced various severe
limb defects also mimicking in type and number those observed following thalidomide treatment (Table 1 and 2). EM 12 was equally specific in the production of
proximal and preaxial limb abnormalities as was thalidomide. The non-teratogenic analogue, EM87, similarly administered was associated with a low incidence
of limb deformities (Table 1) similar in type to those seen in controls (Table 2).
Only one limb exhibited a severe abnormality, a rudimentary radial bone.
Lack of regeneration in treated and untreated newts
The number of limbs in which regeneration was inhibited is shown in Table
3. Regeneration was classified as inhibited in two instances: (1) when no regeneration occurred as verified by absence of methylene blue stain in the limb stump,
Effects of thalidomide on regenerating newt forelimb
131
Table 3. The effects of CMC, thalidomide and analogues on the
incidence offorelimb regenerate inhibition in the newt
Morphology
r
Group
Treatment
I
II
HI
IV
V
VI
VII
VIII
IX
X
control
CMC
CMC
thai.
thai.
thai.
thai.
thai.
EM 12
EM 87
1
Days treated
none
7-12
15-20
4-6
7,8
10-12
15,16
20,21
7,8
7,8
Total
limbs
Regener.
inhibit.
23
26
20
24
34
30
28
25
30
24
3
6
7
13
11
14
8
5
11
7
%Lack
regener.
13
23
35
541
32
46
29
20
36
29
P < 005 compared with Group 11 vehicle treated controls.
or (2) when outgrowth ceased early in regeneration as indicated by the presence
of a humeral stump or spike.
Inhibited regeneration in newts subjected to six consecutive days of CMC
treatment was increased 2- to 3-fold compared to untreated animals. Although
Chi-square analysis suggested that these differences were not statistically
significant, it is possible that CMC itself, or more likely the stress from repeated
injections, could lead to an inhibition of regeneration. Only among treated
animals in group IV was the rate of inhibited regeneration significantly increased
compared to vehicle treated controls (x2 = 3-89, P < 0-05). However, the rate
in group VI was appreciably increased (doubled) over that in the corresponding
control (group II). Since groups IV and VI received the largest quantity of
thalidomide (3 vs 2 days of administration), it is tempting to speculate that some
regenerates were totally inhibited by thalidomide administration, thus simulating the amelia frequently seen in the primate thalidomide syndrome.
DISCUSSION
These results show that thalidomide administration to newts during early
forelimb regeneration produces a specific pattern of limb malformations,
proximal and preaxial skeletal elements being affected most severely. A similar
pattern of forelimb deformities has been observed in monkey and human fetuses
born to mothers ingesting thalidomide during the critical phase of pregnancy
(see Lenz, 1964; Wilson, 1973 for review).
Comparative morphological stages of forelimb development are shown in
Table 4. The dedifferentiation stage of newt regeneration is roughly analogous
132
A. S. BAZZOLI, J. MANSON, W. J. SCOTT AND J. G. WILSON
Table 4. Comparison offorelimb development and forelimb sensitivity to
thalidomide in human and monkey embryos and newt regenerates
Gestation age
Human1
Monkey2
pre-limb bud
early limb bud
late limb bud
paddle
27-29
30-32
33-35
36-37
24-25
ab. 27
28-30
thai, induced
forelb.deform.
25-30
22-27
Stages
1
2
3
31-
Regenerate age
Stages
dedifferentiation
early bud
late bud
palette
Sensitive period
Newt3
6-11
12-17
18-24
22-28
4-18
Wilson, 1973.
Heuser iind Streeter, 1941.
Iten and Bryant, 1973.
to the pre-limb-bud stage of monkey and human forelimb development.
Early and late limb-bud stages as well as the paddle stage (palette in newts)
are characteristic of all three species. In all, the susceptible period for thalidomide teratogenesis begins sometime prior to actual limb-bud formation and
ends at the beginning of the late limb-bud stage. The sensitive period for production of forelimb abnormalities in rhesus monkey embryos extends from days
22-27 post-conception (Wilson, 1970); in human fetuses the critical period for
severe proximal reduction deformities of the forelimb covers days 25-30 postconception (Lenz, 1964) and in newt regenerates the sensitive period lasts
approximately 14 days (4-18 post-amputation). Because cold blooded animals
such as the newt have slower metabolic and developmental rates, it is not
entirely unexpected that the sensitive period for eliciting forelimb abnormalities in the newt is proportionately longer than that of monkey and human
embryos.
One major difference between the production offorelimb deformities in newts
versus that in primates was the dosage required to elicit malformations. A
dosage of 10 mg/kg/day for 1-3 days in rhesus monkeys consistently produced
severe forelimb deformities when administered during the critical phase of embryonic development (Wilson, 1970). Each thalidomide-treated newt received 3
mg/animal/day for 2 or 3 days (ca. 1000 mg/kg). This large dose was administered to insure that each newt received a teratogenic amount, and even then
because of the recurring problem of regurgitation, some animals may never
have received an effective dose. In addition, much of the drug that was ingested
passed unabsorbed through the gastrointestinal tract as evidenced by white
bulbous fecal material excreted into the water one or two days after treatment.
Attempts to administer thalidomide dissolved in dimethyl sulfoxide (DMSO)
Effects of thalidomide on regenerating newt forelimb
133
intraperitoneally failed mainly because the solvent caused a high percentage of
distal reduction deformities.
Normal limbs were observed in each of the thalidomide-treated groups (Table
1). Two possible causes are (1) the regurgitation problem mentioned earlier,
resulting in some animals failure to absorb an effective dose, and (2) variation in
regeneration rate (Connelly, 1977) so that drug administration occurred before
or after the thalidomide sensitive period.
Gebhardt and Faber (1966a, b) have previously reported that oral thalidomide
administration via gelatin capsules (1 mg/animal) to Ambystoma larvae had no
influence on regeneration. The discrepancy between their report and the present
studies may be attributed to differences in dosage. Gebhardt and Faber gave
approximately 100 mg/kg/day, as opposed to 1000 mg/kg/day in the present
study. Since no mention of repeated thalidomide administrations was reported,
it is assumed that treatment was given only on one day in the previous study.
Quite possibly, longer as well as larger thalidomide dosage may be needed to
influence regeneration. Finally, inherent species differences in drug metabolism or
absorption rates between the Ambystoma larvae and the adult newt could
account for this disparity.
The analogues of thalidomide, EM 12 and EM87, were used to further delimit
the sensitivity of the regenerating newt forelimb to this type of compound.
EM 12 , which differs from thalidomide in the phthalimide moiety where a
methylene group is substituted for a carbonyl group (Fig. 2), produced identical
forelimb deformities as did thalidomide in the newt. Likewise in rabbits and
monkeys, maternal administration of EM12 resulted in the formation of fetal
limb anomalies identical to those produced by thalidomide administration.
Conversely, the experimental analogue, EM87, produced no increase in the
number of malformed limbs when compared to controls in newts. EM87 differs
structurally from thalidomide in the glutarimide ring where a methylene group
replaces a carbonyl group and in the phthalimide ring where a sulfonyl
group is substituted for a carbonyl group (Fig. 2). This analogue has proven
non-teratogenic in non-human primate, rabbit and rodent studies (personal
communication, Chemie Griinenthal).
Slight variations in the number of skeletal elements, specifically in the carpal
and phalangeal regions are intrinsic to normal forelimb regeneration in the newt.
The low incidence of observed forelimb abnormalities in untreated controls in
this study is consistent with those seen by Dearlove and Dresden (1976) and
Bryant and Iten (1974). Although CMC administration (30//I/newt) increased
the percentage of malformed limbs slightly (Table 2), it is unclear whether the
suspending agent itself or stress from handling and repeated oral treatments
(6 days) may be responsible for this increase. Experiments are underway, using
2- or 3-day CMC administration intervals in order to identify the causative agent.
The type of limb malformations observed in untreated and CMC-treated
control animals were non-specific in nature, e.g., they were located either
134
A. S. BAZZOLT, J. MANSON, W. J. SCOTT AND J. G. WILSON
TERATOGENIC
Thalidomide.
NON-TERATOGENIC
Fig. 2. Structures of thalidomide, EM-12 and EM-87.
proximally or distally and occurred without preference for the postaxial or
preaxial side of the limb. However, the forelimbs of one animal in control
group II exhibited severe preaxial hemimelia. One regenerate forelimb in group
X (EM87) had a severe proximal deformity. These limb defects were identical
to those observed following thalidomide administration, and are distressing
because of the specificity for proximal growth inhibition. A number of possibilities could explain these presumably spontaneous malformations but a more
definite resolution of this dilemma awaits further study of control regenerates.
The results presented herein suggest that the morphological effects of thalidomide on newt forelimb regeneration are similar to those seen in the limbs of
primate embryos. The obvious economical advantages strongly favor continued
investigation of the newt for further thalidomide studies. Questions which
must first be answered, however, concern the high rate of totally inhibited
regeneration and the magnitude of dosage necessary to induce abnormal limb
regeneration.
The authors wish to thank Kitty Juniper and Joan Randall for their expert technical
assistance.
This work was submitted by one of us (A.S.B.) as a thesis in partial fulfillment of requirements for the M.S. degree, Graduate Program in Developmental Biology, University of
Cincinnati.
Effects of thalidomide on regenerating newt forelimb
135
REFERENCES
S. V. & ITEN, L. E. (1974). The regulative ability of the limb regeneration blastema
of Notophthalmus viridescens: experiments in situ. Wilhelm Roux Arch. EntwMech. Org.
174,90-10.1.
CAHEN, R. L. (1964). Evaluation of the teratogenicity of drugs. Clin. Pharmacol. Ther. 5,480BRYANT,
514.
CONNELLY,
T. G. (1977). The relationship between growth, developmental stage and postamputation age of the regeneration blastema of the newt, Notophthalmus viridescens. J.
exp. Zool. 199, 33-40.
DEARLOVE, G. E. & DRESDEN, M. H. (1976). Regenerative abnormalities in Notophthalmus
viridescens induced by repeated amputations. /. exp. Zool. 196, 251-262.
DELAHUNT, C. S. & LASSEN, L. J. (1964). Thalidomide syndrome in monkeys. Science, N.Y.
146, 1300-1305.
FABER, J. (1971). Vertebrate limb ontogeny and limb regeneration: morphogenetic parallels.
Adv. Morphog. 9, 127-146.
GEBHARDT, D. O. E. & FABER, J. (1966a). The influence of aminopterin on limb regeneration
in Ambystoma mexicanum. J. Embryol. exp. Morph. 16, 143-158.
GEBHARDT, D. O. E. & FABER, J. (19666). The influence of aminopterin and other teratogenic
substances on the regenerating forelimb of Ambystoma mexicanum. Actaphysiol. pharmac.
need. 13, 480.
GIBBS, E. L. (1963). An effective treatment for red-leg disease in Rana pipiens. Lab. Anim.
Care 13, 781-783.
GIROUD, A., TUCHMANN-DUPLESSIS, H. & MERCIER-PAROT, L. (1962). Observations sur les
repercussions teratogenes de la thalidomide chez la souris et le lapin. C. r. Seanc Soc. Biol.
156, 765-768.
GRAUDUMS, L, MUCKTER, H. & FRANKUS, E. USP NO. 3, 951, 985, Lactams and process for
their manufacture, Assigned to Chemie Griinenthal, GmbH.
HEUSER, C. H. & STREETER, G. L. (1941). Development of the Macaque embryo. Contr.
Embryol. Carnegie Jnstn. 29, 17-55.
ITEN, L. E. & BRYANT, S. V. (1973). Forelimb regeneration from different levels of amputation in the newt, Notophthalmus viridescens: length, rate, and stages. Wilhelm Roux Arch.
EntwMech. org. 173, 263-282.
KOHLER, V. F. & KOCH, H. (1974). Teratologische Untersuchung der Thalidomid-ahnlichen
Verbindungen K-2004 and K-2604 an Maus und Ratte. Arzneimitell-Forsch. 24, 1616-1619.
LENZ, W. (1964). Chemicals and malformations in man. In Congenital Malformations (ed. Int.
Med. Congr., Ltd.), pp. 263-276. New York: Int. Med. Congr.
SCHUMACHER, H., BLAKE, D. A., GURIAN, J. M. & GILLETTE, J. R. (1968). A comparison of
the teratogenic activity of thalidomide in rabbits and rats. /. Pharmac. exp. Ther. 160,
189-200.
SCHUMACHER, H. J., TERAPANE, J., JORDAN, R. L. & WILSON, J. G. (1972). The teratogenic
activity of a thalidomide analogue, EM12, in rabbits, rats and monkeys. Teratology 5,
233-240.
SELLER, M. J. (1962). Thalidomide and congenital abnormalities. Lancet ii. 249.
SOMERS, G. F. (1962). Thalidomide and congenital abnormalities. Lancet i, 912-913.
STOCUM, D. L. (1975). Outgrowth and pattern formation during limb ontogeny and regeneration. Differentiation 3, 167-182.
THORNTON, C. S. (1968). Amphibian limb regeneration. Adv. Morphog. 7, 205-249.
WILSON, J. G. (1970). Teratogenicity of thalidomide and related compounds in nonhuman
primates. Paper given at 3rd International Congr. on Primatology, Zurich, Switzerland.
WILSON, J. G. (1973). Environment and Birth Defects. New York: Academic Press.
WILSON, J. G. & GAVAN, J. A. (1967). Congenital malformations in nonhuman primates:
spontaneous and experimentally induced. Anat. Rec. 158, 99-110.
{Received 9 February 1977, revised 1 April 1977)