Development of the amphibian pituitary with special reference to the neural lobe.

DEVELOPMENT O F T H E AMPHIBIAN PITUITARY
TT~ITH
SPECIAL REFERENCE TO THE
NEURAL LOBE
RICHARD M. EAKIN' AND FRANK E. BUSH*
Department of Zoology, University of California, B e r k e l t y
TWENTY-TWO FIGURES
The classic paper of P. E. Smith ( '20) on the effects of early
epithelial hypophysectomy upon anuran development records
evidence of morphogenic dependence of the pars nervosa of €he
pituitary body upon the pars buccalis. He found that the
neural lobe is " . . . profoundly modified by the loss of its
associate. It attains normal development neither as regards
size, form, nor position" (p. 72). The " . . . diminution varies
from 40 per cent to 80 per cent and transcends very distinctly
the considerable variability which this lobe [normally] ex(p. 71). We have re-examined this reported dehibits .
velopmental relationship between the neural and epithelial
components of the amphibian pituitary. A preliminary note
was published earlier in this journal (Eakin and Bush, '51).
. ."
MATERIALS AND METIIODS
Tailbud embryos of the Pacific tree frog, HyEa regilla, and
of the western spadefoot, Scaphioptts hnmmonndi, were adenohypophysectomized at stage 18 (see Eakin, '47). The resulting
albino tadpoles were raised, along with normal larvae, in
enamel pans containing pond water and a substratum of fine
mud. Poultry meal was sifted and added as food. The surface
of the water was occasionally blotted with a paper towel t o
' J o h n Simon Guggenheim Memorial Fellow at Stanford University during part
of the investigation.
Present address: Sacramento Junior College, Sacramento, California.
279
280
RICHARD M. E A K I N A N D FRANK E. RUSH
remove any film of oil which formed; otherwise the larvae died
of suffocation. When the albinos reached a large size (fig.2)
thyroid feeding was begun. At this time the controls were
nearing completion of metamorphosis. On alternate days a
half-grain tablet of beef thyroid substance was crushed and
added to the water. Albino tadpoles induced to metamorphose
by thyroid feeding became very feeble, especially in Hyla;
in final stages of metamorphosis they were susceptible to
drowning. Therefore, as soon as both forelimbs protruded
(stage &yI of Taylor and Kollros, '46) the larvae were removed from the water and kept on and under moist paper
towels. Thyroid substance was thereafter administered orally
by a small pipette. This was especially satisfactory with the
toads which were larger and more easily handled. When an
animal appeared to be approaching death (fig. 5) the head and
lower jaw were fixed separately in Bouin's solution for later
his€ological examination. A group of normal, completely metamorphosed frogs and toads was similarly preserved for comparisons. All specimens were sectioned at 8 ~1 and stained with
Galigher 's hematoxylin-ammonium sulfate and alcoholic eosin.
To compare the sizes of the neural lobes in albino and normal tadpoles the images of the microscopic preparations were
cast upon paper by a projector. The pars nervosa in each section was traced; the sketches were carefully cut out; blood
channels, which had been drawn, also were removed from the
paper reproductions. The cut-outs for each specimen were
then weighed. Procedures, materials, and personnel were kept
uniform throughout these quantitative determinations. We
gratefully acknowledge the assistance of Mr. Eugene Hessel,
Miss Lillian Pissott, Miss Patricia Tomlin, and Mr. Victor
G . Duran, University Photographer, and a critical reading of
the paper by Dr. Howard A. Bern.
RESULTS
Growth m d rnetnrnorph.osis
Data (table 1) on comparative growth and development of
albinos and controls are available for the Scnphiopus series
281
DEVELOPMENT OF AMPHIBIAN PITUITARY
only. It is clear that the hypophysectomized animals kept pace
with the controls in growth as represented by average total
length. By postoperative day 47, four albino tadpoles were
rapidly differentiating hind limbs and it was assumed that they
contained regenerated pars distalis. This was later confirmed
by microscopic examination. These four animals were, therefore, not included in the last two measurements, although they
would have changed the average length of the albino group
only slightly, reducing it by 0.2 mm. The albinos continued to
increase in length after the normal larvae completed metamorphosis. On postoperative day 70 one albino measured 108.5
TABLE 1
Total length and stage of development of albino and normal Scapphiopus hammondi
15
30
Albinos (13 animals) :
Average length (mm)
29.6
59.0
81.7 a
88.5
Iv
VI
VII'
VII
30.8
55.4
79.2
82.1 '
V
XI
Average stage
Controls (14 animals) :
Average length (mm)
Average stage
47
57
I ) A T H POSTOPERATIVE
xv
*
xx
Four albinos containing regenerated pars distalis not included.
Seven controls in last stages of metamorphosis not measured.
mm in total length and weighed 15.0 gm. Others were almost
as large. The longest control reached a maximal length of
90.5mm and the heaviest a maximal weight of 10.5gm. On
postoperative day 57, the nine albinos without pars distalis
averaged 9.6 gm,whereas the seven aquatic controls averaged
only 8.3 gm.
I n developmental age, however, the albinos of both Scaphiopus (table 1)and HgZa lagged behind the controls and did not
develop beyond stage V I I (early foot paddle), except those
with regenerated pars distalis. Some albino toads were induced by thyroid feeding to metamorphose completely ; others,
even after three months of thyroid administration, still re-
282
RICHARD M. EAKIN A N D FRANK E. BUS=
tained a small stub of a tail (stage XXIV, fig. 3). Although
over 100 frog albino larvae were started on a regimen of thyroid feeding, not one attained stage XXV. Most of them expired or would have died in stages XX to XXIII, if not previously sacrificed.
Shape and position of pars nervosa
I n general we confirm the observations of Smith ('20) on
the form and position of the pars nervosa in larvae developing
from hypophysectomized embryos of R a m boylei and Bufo
boreas. W e found that in H . regilla and 8. hammondi the
pars nervosa usually formed in the hypophysectomized tadpole as an irregularly oval body situated at the posterior tip
of the infundibulum (figs. 12, 13) whereas in the normal animal it was a dumbbell-shaped transverse bar lying dorsal to
the pars intermedia of the epithelial hypophysis (fig. 11).
I n the latter, the isthmus was relatively narrow; the lateral
parts, one of which is well shown, were broad and lobate.
Considerable variation in form and position of the pars
nervosa was observed in the hypophysectomized animal, more
indeed than that recorded by Smith. Commonly the pars
nervosa was medially situated, as figured by Smith, but
frequently it was a unilateral body connected to one side of
the infundibulum. Less often it appeared as a pair of symmetrically arranged lobes minus an isthmus (fig. 15). Except
for the absence of thc connecting piece the picture in the last
instance was essentially normal. In a few cases the pars nervosa was not a discrete structure but an area within the
median eminence (Rioch, Wislocki and 0 'Leary, '40), a midventral thickening of the infundibular floor. Figures 16 and
17 show the neural lobe in two different albino toads in which
it appeared as an unelevated region of the infundibular floor
in one and a large extension into the infundibular cavity in the
other.
Form o f i.lzfuuzdibzclum
W e found also a greater variation in the form of the infundibulum than that described by Smith. In his hypophysec-
DEVELOPMENT O F AMPHIBIAN PITUITARY
283
tomized animals the “pituitary wall,” as he termed the median
eminence, was invariably membranous whereas in the normal
tadpole or subadult it was “ . . thickened and camposed of a
high columnar ependyma with an ectally placed prominent
stratum of neuropilem” (p. 70). V e too noted in many of our
specimens the membranous character of the infundibular floor
and the complete absence of a median eminence. Yet in some
specimens without a trace of epithelial hypophysis the infundibulum appeared to be normal. Figure 21 shows a section
of an albino H y l a with a well-developed median eminence
which was continuous, a few sections caudally, with a large
pars nervosa.
The tractus preoptico-hypophyseus in H . regilla (Wilson,
Weinberg and Bern, ’57), although not studied with special
stains, could be followed in our hematoxylin-eosin preparations. I n the normal animal the fibers of this tract course
mesiad in the marginal layer of the diencephalic floor. They
then turn caudad into cables which run through the median
eminence of the infundibulum. At the end of the infundibulum the tracts swing laterad and dorsad and fan into
the paired lobes of the pars nervosa. The albino shows no
basic difference in pattern. I n those instances of a welldeveloped median eminence the picture is similar to that
of the normal animal, the eminence being only slightly narrower and less folded. In those cases of a membranous infundibulum the tracts can be identified, nevertheless, as several cables extending caudad within the thin floor and terminating in the pars nervosa.
.
Size of p a r s mervosa
I n our preliminary report (Eakin and Bush, ’51) we asserted that the hypophysectomized embryo of H. regilla developed a pars nervosa normal in size. No evidence was
presented, however, to support this conclusion. We have
now compared quantitatively, by the method described above,
the neural lobe in 35 albino tadpoles in various stages of
284
RICHARD M. EAKIN AND FRANK E. BUSE
metamorphosis with that in 20 normal subadults. The mean
weight of the paper cut-outs for the albino animals was
0.59 gm; the range was 0.1 to 1.5 gm. The mean weight of
the paper reproductions of the normal animals was 0.68 gm;
the range was 0.4 to 1.1gm. These data are presented graphically in figure 1. Applying a t-test for the significance of the
difference between the means it was found that in this
experiment the difference was not significant at the 5 per
cent level of confidence.
N
0.
II
10
9
8
7
6
5
4
3
Normal
2
I
0.0 0.2 040.6 0.8 1.0 1.2 1.4 1.6
GMS
0.40.6 0.8 1.0 1.2 1.4
A
MEAN 0.68
MEAN 0.59
Fig. 1 Distribution of sizes of pars nervosa (weight of reproductions) in hnophysectomized (left) and normal (right) Hyla regilla.
A
Mean weight and range of weights for the pars nervosa
cut-outs in 7 albinos of S. harnmortdi without regenerated
pars distalis were: 2.3 gm (1.1to 4.0 gm). I n 5 controls the
mean weight and range were: 3.5 gm (2.3 to 4.8 gm). The
difference between the means is significant at the 5 per cent
level of confidence but not at the 2.5 per cent level. Considering the small sample of toads we do not regard the difference
between albino and control as significant. I n one of the albinos the pars nervosa, consisting of bilateral lobes, was
larger than that in all of the controls except one.
DEVELOPMENT O F AMPHIBIAN PITUITARY
285
Regelzerat ed epithelial hypophysis
As in earlier studies involving hypophysectomy in embryos
of H . regilla (Eakin and Harris, ’51; Eakin, ’56) there were
many instances of regulation or regeneration of the epithelial
hypophysis. Apparently only a small fragment of the anlage,
inadvertently overlooked in the operation, develops into a
full complement of partes distalis, intermedia, and tuberalis.
These animals, recognized by normal pigmentation and progress toward metamorphosis, were not studied. The incidence
of regulation can be reduced by careful scraping of the ventral
surface of the forebrain and the roof of the foregut to which
hypophyseal cells frequently cling when the foregut is separated from the brain to expose the hypophyseal anlage.
I n several instances, in both the H y l a and Scaphiopus
series, a remnant of the hypophyseal anlage appeared to form
pars distalis alone. The absence of a pars intermedia, established by the albinism of the larvae, was confirmed by
microscopic examination. That the regenerate was pars distalis was indicated by the microanatomy of the body, the
histological evidence of thyroid activity, and the metamorphosis of the tadpole. Figure 6 shows an albino of R.regilla
which metamorphosed without thyroid feeding and ahead of
some of the controls. It left the water voluntarily, stayed
several days under moist paper towels until the tail was
completely resorbed, then crept up a side of the aquarium
and remained a few days under the cover. Nicroscopic examination revealed a distinct, albeit small, pars distalis (fig.
18). The pars nervosa although well developed was not so
large as that in several albino larvae completely lacking an
epithelial hypophysis.
Smith (’20) recorded one “partial albino” of R. boylei and
three of B. boreas among his hypophysectomized animals
which metamorphosed tardily and in which the regenerated
gland was a single structure containing both distalis and
intermedia components. We have observed similar instances
in H . regilla. Two larvae were particularly instructive. They
286
RICHARD M. EAKIN AND FRANK E. BUSH
exhibited marked albinism everywhere except on the dorsum
of the head anterior to the eyes. Careful search of the sections
of these animals disclosed a small hypophysis-like body situated beneath or beside the telencephalic hemispheres. Epithelial hypophysis was completely lacking at the orthotopic
site. Presumably an anterior remnant of the hypophyseal
anlage had been induced by telencephalon (Eakin, '56) to
differentiate into an ectopic hypophyseal body containing both
intcrmedin and thyrotropin-secreting cells.
Another specimen suggested the possibility that a pars
intermedia only had regulated. This very dark, almost hyperpigmented larva (fig.10) made only limited progress toward
metamorphosis. After three months in a pan containing other
larvae which completely metamorphosed, this specimen still
retained larval features such as rotund body, horny mouth
parts, elongated gut, etc. The hind limbs attained only stage
X in which the foot paddle shows weak interdigital notches
between the future toes. The larva had grown, however, to
a total length of 68.5 mm, which is very large for a H. r e g i l k
tadpole. Epidermal, dermal, and subcutaneous melanocytes
were unusually numerous and interlacing. Dorsal xanthocytes
were punctate. Histological examination revealed an elongated body at the normal site of the pituitary which was
pars intermedia-like in gross form and microscopic anatomy.
It lacked the cords and sinusoids characteristic of pars distalis. The sections were first stained with hematoxylin and
eosin. Later the coverslip was removed; the sections were destained and then restained by the periodic acid-Schiff (PAS)
technique (MeManus, '48). It was found that all of the cells
of the body, with one exception, were strongly PAS-positive
as is the pars intermedia of the normal pituitary of H. regilla.
This is further evidence that the hypophyseal body was an intermediate lobe. The exception was a small, oval PAS-negative
area on the ventral border of the body (fig. 19) which we
assume is pars distalis. Further support for this conclusion
is provided by the histological evidence of low-grade thyroid
activity (fig. 20), suggestive of some secretion of thyrotropin,
DEVELOPMENT O F AMPHIBIAN PITUITARP
287
and by the fact that the hind limbs of this animal had developed to stage X.
DISCUSSION
There is one point only which needs further comment,
namely, the difference between our results and conclusions and
those of P. E. Smith (’20) regarding the development of the
amphibian neurohypophysis. From Smith’s data on the size
of the pars nervosa (weights of models, table 6, p. 74) of eight
normal and eight albino frogs older than 60 days, we observe
that the average size of the pars nervosa of the normal is over
two and one-half times larger than that of the albino. Including six partial albino frogs would not appreciably change the
relationship. With the exception of one of Smith’s partial albinos, none of his hypophysectomized animals had a neural
lobe that compared favorably with the mean size of the normal
pars nervosa. I n the study here reported, however, we found
only a slight difference between the average size of the neural
lobe of the albino and that of the normal animal (fig. 1).Moreover we did not observe a “novel” neural lobe, as reported by
Smith, namely, an atypically situated pars nervosa formed by
the brain in response to an ectopically placed pars buccalis.
We believe, therefore, that in H . regilla and S. hammorcdi the
differentiation and growth of the neural lobe is not dependent
upon its anatomical associate, the epithelial hypophysis. Only
the form of the pars nervosa is affected by the absence of the
pars buccalis, as previously shown by Smith ( ’20). We suggest
that this effect on the morphology of the neural lobe is probably owing to purely physical factors.
Our findings are in accord with those of other studies. Holt
(’21) described an isolated, yet instructive, instance of aplasia
of the partes buccales in a 40mm pig embryo. Although the
epithelial hypophysis was “wholly absent . . the pars neuralis, on the other hand, was well developed, connected with the
brain, and normal in position, extent, and structure” (p. 207).
In a study on host antagonism to homoplastic and xenoplastic
grafts, Eakin and Harris ( ’51) transplanted the adult pitui-
.
288
EICHARD M. EAICIN A N D FRANK E. BUSH
tary from H . regiZZa and Taricka torosn into albino and normal
larvae of H. regilla. Some of their microscopic preparations
were examined by us. The pars nervosa of the albino hosts,
many of which were brought into late stages of metamorphosis
through thyrotropic stimulation by the graft, closely resembled
the neural lobes of the thyroid-fed albinos of this study in both
form and size. No quantitative determinations were made,
however. Figure 22 shows the pars nervosa of one of their
albinos. The work of Stein (’29) might be mentioned here as
a commendable attempt to test the developmental interrelationship between the epithelial and neural components of the
avian hypopliysis by grafting “hypophyseal ectoderm’ ’ of the
chick, with and without presumptive infundibulum, to the
chorio-allantoic membrane. She stated that there were general
indications that the development of the two anlagen were dependent upon each other, but her results were inclusive.
Whether thyroid feeding had any effect on the size of the
pars nervosa in our animals was not clear. I n some of the
thyroid-fed albinos of Hyla the pars nervosa was relatively
large (paper weights of 1.0 to 1.5 gm),but in many others it
was quite small (paper weights of 0.1 to 0.4 gm). The animals in the latter group were abnormally small, feeble and
wasted and, as pointed out earlier, in no instance did an hypophyscctomized Hyla larva completely metamorphose. It is
possible that either the dosage of thyroid substance was not
properly controlled or that other pituitary principles were
needed for complete metamorphosis. These very small neurohypophyses probably reflect, therefore, the generally poor
physiology of the animals rather than the absence of some
direct developmental stimulus from the epithelial hypophysis.
This explanation may apply to Smith’s findings in R a m boylei
and Bzcfo borens. On the other hand a few albino tadpoles of
HyZn raised to a very large size but not fed thyroid substance
developed fairly large neural lobes (paper weights of 1.0 gm).
I n the light of the studies on hypothalamic neurosecretion
(see review by Scharrer and Scharrer, ’54) it is not surprising
that early epithelial hypophysectomy may have no effect upon
DEVELOPMENT O F AMPHIBIAN PITUITARY
289
the differentiation and growth of the pars nervosa, since this
lobe of the pituitary is largely a depot of neurosecretion produced in diencephalic nuclei and transported along the tractus
preoptico-hypophyseus into the pars nervosa. The surgical
ablation of the primordial epithelial hypophysis, as performed
in these experiments, did not involve any perceptible trauma to
the diencephalon, in most instances, even though the anterior
part of its floor was scraped to ensure complete removal of
the pars buccalis. Usually it was not necessary to subject the
infundibulum to any manipulation because it lay posterior to
the area of operation. Moreover, the infundibulum is determined as early as the beginning neurula (Burch, '46) and has
already developed into a morphologically distinct region of the
brain By the early tailbud stage.
Although we do not regard the histodifferentiation and
growth of the neurohypophysis to be dependent upon the
epithelial hypophysis, the latter, particularly the pars intermedia, seems t o require a morphogenic stimulus from the
former (see Eakin, '56).
SUMMARY
1. Tailbud embroys of the Pacific tree frog, nyih regillu,
and of the western spadefoot, Scaphiopus harnmodi, were
adenohypophysectomized and raised to large larvae, whereupon those showing albinism were fed thyroid substance to
initiate metamorphosis. Normal larvae of both species were
maintained, without thyroid feeding, until metamorphosis was
complete.
2. The pars nervosa of the albino was compared with that
of the normal subadult as to morphology, position, histology,
and size.
3. The neurohypophysis in most albinos was irregularly
ovate instead of dumbbell-shaped as in the normal animal. Although usually situated at the tip of the infundibulum, considerable variation in position was noted, especially in albinos
of Scaphiopus. The histology of the pars nervosa of the albino
was normal in both species.
290
RICHARD 3%. EAKIN AND FRANK E. BUSH
4. The sizes of the neural lobes of 35 albino and 20 normal
Hyla were compared by weighing paper cut-outs of outline
drawings of the glands. Although the range of size variation
was greater in the albinos, there was no significant difference
between the means of the two groups. A study of a smaller
sample of Scaphiopus gave a similar result.
5. The morphology of the infundibulum was usually affected by early epithelial hypophysectomy. I n some instances,
however, the infundibulum of the albino was essentially
normal.
6. It is concluded that in the two species of amphibians
studied the differentiation and growth of the pars nervosa are
not dependent upon its anatomical associate, the epithelial
hypophysis.
LITERATURE CITED
BURCH,A. B. 1946 An experimental study of the histological and functional
differentiation of the epithelial hypophysis in Hyla regilla. Univ. Calif.
Publ. ZOO^., 51: 185-214.
EAKIN,
R. M. 1947 Determination and regulation of polarity in the retina of
Hyla regilla. Univ. Calif. Publ. Zool., 51: 245-288.
1956 Differentiation of the transplanted and explanted hypophysis
of the amphibian embryo. Jour. Exp. Zool., 131: 263-290.
EAKIN,R. M., AND F. E. BUSH 1951 The development of the neural lobe of the
pituitary in hypophysectomized embryos of the tree-frog, Hyla regilla.
Anat. Rec., 111: 5 4 4 4 4 5 .
EAKIN,R. M., AND M. HARRIS 1951 Function and fate of transplanted pituitary
in amphibians. Jour. Expo Zool., 116: 165-190.
HOLT,E. 1921 Absence of the pars buccalis of the hypophysis in a 40-mm-pig.
Anat. Rec., 32’: 207-216.
MCMANUS,J. F. A. 1948 Histological and histochemical uses of perioaic acid.
Stain Tech., 2’3: 99-108.
RIOCH,D. McK., G. B. WISLOCKI AND J. L. O’LEABP 1940 A precis of preoptic,
hypothalamic and hypophyseal terminology with atlas. Res. Publ. Ass.
Nerv. Ment. Dis., 8 0 : 3-30.
SCHARBER,
E., AND B. SCHABRER1954 Hormones produced by neurosecretory
cells. Ree. Prog. Hor. Res., 10: 183-240.
SMITH,P. E. 1920 The pigmentary, growth and endocrine disturbances induced
in the anuran tadpole by the early ablation of the pars buccalis of
the hypophysis. Amer. Anat. Mem., 11: 1-152.
STEIN,K. F. 1929 Early embryonic differentiation of the chick hypophysis as
shown in chorio-allantoic grafts. Anat. Rec., 4 3 : 221-238.
DEVELOPMENT O F AMPHIBIAN PITUITARY
291
TAYLOR,
A. C., AND J. J. KOLLROS
1946 Stages i n the normal development of
Rana pipiens larvae. Anat. Rec., 94: 7-24.
WILSON,L. D., J. A. WEINBERGAND H. A. BERN 1957 The hypothalamic neurosecretory system of the tree frog Hyla regilla. Jour. Comp. Neur.,
107: 253-272.
PLATE 1
EXPLANATION OF FIGURES
Photographs 6 and 7 by Victor G . Duran, others by Eakin.
2 Albino larva of H y l a regilla, stage V, 58 days after hypophysectomy but
prior t o thyroid feeding. Numerous and expanded xanthocytes mask punctate
melanoeytes. Controls in metamorphic stages XX to XXV (see fig. 7).
3 Albino subadult of Scaphiopus hammondi, stage XXIV, 81 days after hypophysectomy. One of four albinos which metamorphosed without thyroid feeding ; possessed regenerated pars distalis. Other albinos without pars distalis
and requiring thyroid feeding to induce metamorphosis had the same appearance as this specimen when they reached stage XXIV.
4
Normal (control) larva of S. hammondi, stage XXV, same age as albino
shown in fig. 3, same magnification.
5 Metamorphosing albino larva of H . regilla, stage XXII, 58 days after hypophysectomy plus 21 days of thyroid feeding. Numerous and expanded xanthocytes mask punctate melanocytes.
6 Albino subadult of H . regilla, stage XXV, about 67 days after hypophysectomy. Metamorphosed without thyroid feeding. Regenerated pars distalis
present (see fig. 18). No pars intermedia.
7 Normal (control) subadult of H. regilla, stage XXV, same age as albino
shown i n fig. 6, same magnification.
8
Albino larva of S. hammondi, stage IV, 38 days after hgpophysectomy but
prior t o thyroid feeding. Numerous and expanded xanthocytes mask punctate
melanocytes.
9 Normal (control) larva of S. hammondi, approximately stage XVII, same age
as albino shown in fig. 8, lower magnification.
10 Larva of H . regilla of normal coloration, stage X, 96 days after hypophysectomy. No thyroid feeding. Regenerated pars intermedia present ; small
pars distalis probably present (see fig. 19).
292
PLATE 1
DEYELOPNEKT OF A7vlPHIBIAIi PITUITARY
RICIldRL) X. X A K l N AND FRALIK E. BUSH
293
PLATE 2
EXPLANATION OF FIGURES
Photomicrographs by Victor G . Duran
A l l figures are cross-sectional views.
11 Pituitary of normal subadult IT. rrgilln, stage XXV (see fig. 7 ) . p . d., pars
distalis; p . i., pars intermedia; p . n., pars nervosa. X 98.
12
Pars nervosa of hypophysectomized, thyroid fed H . regilla, stage XXII. Consists of single nlrdiarl lobe; partes buccales completely absent. T t L , punctate
dcep melanocytes; p . n., pars nervosa; t., thyroid gland. X 48.
13 Pars nervosa of the hypophysectomized, thyroid-fed metamorphosing larva
of A. regzlla shown in figure 5. Consists of il single median lobe. Partes
buccales completely ahscwt. X 344.
14
Thyroid gland of the hypophyscctomixed, thyroid-fed animal shown in figure
5. X 450.
15 Pars nervosa of hypophysectomized S. hnmmondi, not fed thyroid, stage VII.
Consists of two unconnected lateral lobes. Partes buecales completely absent.
in., infundibular cavity; note absence of median eminence. X 69.
16
Pars nervosa ( p . n.) of hypophysectomized, thyroid-fed S. hnmmondi, stagr
XXIV, embedded in floor of infundibulum. X 273.
17
Pars nervosa of hypophysectomized, thyroid-fed 8. hamnondi, stage XXV,
inverted into irifundibular cavity (212.). Section rotated ninety degrees; ventral
to the right. Partes buccales completely lacking. X 288.
18 Pituitary of the albino sulmdult H . rt.grlZa shown in figure 6. p . d., regmerated
small pars distalis; p . n., pars nervosa, single lobe slightly to one side of midline. No pars intermedia present. X 204.
19 Regenerated epithelial hypophysis of the E.regdla tadpole shown in figure 10.
Sections first staiiied with hernatoxylin and eosin but later, after destaining,
restained by periodic acid-Schiff procedure. p . i., pars intermedia, strongly
PAS-positive; p . d., small ventral area, PAS-negative, assumed t o bc pars
distalis. X 142.
X 154.
20
Thyroid gland of tadpole shown in figure 10.
21
Median eminence of hypophyaectoinized, thyroid-fed Z. regiZla, stage XXII.
Partes buccales wholly absent. Well-developed pars nervoaa at a more posterior level. X 410.
22
Pars nervosa of hypopliysectoniized A. rfgzlla metamorphosing larva which
had held a homogrsft of adult pituitary in the orbit for 39 days, stage XXT.
One of the animals in a study on host antagonism to grafts by Eakin and
Harris, '51. Graft showii in their figure 6. X 288.
294
DEVELOPJIENT OF AMPHIBIAN P I T U I T A K P
PLATE 2
RICIIARU 31. EAKIN AND FRANK 1. R U S H
295