The Pituitary Gland, Volume 1: Anterior Pituitary

This title is part of UC Press's Voices Revived program, which commemorates University of California Press’s mission to seek out and cultivate the brightest minds and give them voice, reach, and impact. Drawing on a backlist dating to 1893, Voices Revived makes high-quality, peer-reviewed scholarship accessible once again using print-on-demand technology. This title was originally published in 1966.

• Language: English
• Publisher: University of California Press
• Release date: Dec 22, 2023
• ISBN: 9780520309364

Book preview

THE

PITUITARY GLAND

THE

PITUITARY GLAND

Edited by

G. W. HARRIS, M.A., sc.D., M.D., D.M., F.R.S.

Department of Human Anatomy

University of Oxford

and

B. T. DONOVAN, B.SC., PH.D.

Department of Neuroendocrinology, Institute of Psychiatry

The Maudsley Hospital, London

Volume 1

Anterior Pituitary

UNIVERSITY OF CALIFORNIA PRESS

BERKELEY AND LOS ANGELES 1966

UNIVERSITY OF CALIFORNIA PRESS

BERKELEY AND LOS ANGELES, CALIFORNIA

Library of Congress Catalog Card Number: 66-19350

©

The several contributors named on pages vii and viii

1966

Printed in Great Britain by

William Clowes and Sons, Limited, London and Beccles

PREFACE

The subject matter of these three volumes has been subdivided in such a way that much anatomy and biochemistry relating to the pars distalis is contained in Volume 1, the control of the functional activity of this lobe, amongst other subjects, is in Volume 2, and information relating to the pars intermedia and neural lobe is in Volume 3. This arrangement was decided upon in the hope that, in these days of rising production costs and of scientific specialization, many readers may find their own field of interest represented in one of the volumes.

The original invitations to contribute to this book were issued in May, 1961, with the suggestion that the ‘aim is to produce an overall picture of pituitary anatomy and physiology in which recent ideas are set in perspective amid older well-established findings’. This aim, we feel, has been very happily achieved and we would like to express our very real gratitude to all the authors for their expert and distinguished co-operation. The alterations we have made to any manuscript have been slight and stylistic in intent. No constraint has been imposed upon our contributors concerning the scientific views expressed, and a uniform terminology has not been insisted upon. This work did not seem to be appropriate for decisions of the latter kind which, in any case, would have been arbitrary on our part. It was at first hoped that this book would be available by summer 1964. For a variety of reasons, quite unconnected with the publishers, this has not been found possible and some chapters have been much longer in press than was ever intended. As our contributors have told us, some of the opinions expressed are no longer valid in the light of more recent evidence, and other points could now be better documented. However, we have not aimed solely at enumerating ‘recent advances’ and the chapters have been allowed to stand as written. We, and not the contributors, are responsible for any shortcomings of the above kind. Three chapters have been regretfully omitted.

We are extremely grateful to Mrs. E. M. Collen for her indispensable aid in the editorial work and for checking many of the references. Our sincere thanks are due also to The Editorial Staff of Messrs. Butterworths for much help and tolerance.

September, 1965. B. T. DONOVAN

G. W. HARRIS

It is with profound regret that we write here of the sudden deaths of two contributors, Dr. Frank L. Engel and Dr. John D. Green.

Dr. Engel died at the height of an unusually productive career while Professor of Medicine and Director of the Division of Endocrinology in the Duke University School of Medicine. Widely known and honoured for his many fundamental investigations of the metabolic roles of the corticosteroids, adrenocorticotrophin and somatotrophin, Dr. Engel was no less an outstanding physician and stimulating teacher. A measure of the high quality of his postdoctoral programme in metabolic diseases is the impressive number of trainees who have been called to positions of responsibility in other medical centres. His untimely death leaves an unhappy hiatus in the ranks of medical science.

Dr. Green, a close friend of ours, started his research career with one of us in Cambridge, England. It was from Wayne University, Detroit, that he published his classical studies on the blood supply and innervation of the vertebrate pituitary and, later, whilst at the University of California at Los Angeles, he illuminated many aspects of neuroendocrinology with his work on the electron microscopy of the pituitary, on neurosecretion, on the relationships of limbic lobe structures to patterns of sexual behaviour, on the electrical activity of the hypothalamic supraoptic nuclei and many other fields. It is with a sense of great personal sorrow, as well as irreplaceable loss, that we record his death on December 10th, 1964.

CONTENTS 1

PREFACE

CONTENTS 1

CONTRIBUTORS TO THIS VOLUME

1 GROSS ANATOMY OF THE HYPOPHYSIS IN MAMMALS BERTIL HANSTRÖM

2 COMPARATIVE ANATOMY AND EVOLUTION OF THE HYPOPHYSIS KARL GEORG WINGSTRAND

3 THE COMPARATIVE ANATOMY OF THE PORTAL VASCULAR SYSTEM AND OF THE INNERVATION OF THE HYPOPHYSIS J. D. GREEN

4 CYTOLOGY OF THE ADENOHYPOPHYSIS H. D. PURVES

5 ELECTRON MICROSCOPY OF THE ANTERIOR PITUITARY J. D. GREEN

6

7 HORMONAL ACTIVITY OF THE PARS DISTALIS IN REPTILES AND BIRDS A. V. NALBANDOV

8 THE PHYSIOLOGY AND CHEMISTRY OF ADRENOCORTICOTROPHIN2 Herbert M. Evans, Lowell L. Sparks AND Jonathan S. Dixon

9 THE PHYSIOLOGY AND CHEMISTRY OF THYROID STIMULATING HORMONE Robert W. Bates and Peter G. Condliffe

10 THE PHYSIOLOGY AND CHEMISTRY OF THE EXOPHTHALMOS PRODUCING SUBSTANCE (EPS) OF THE PITUITARY BROWN M. DOBYNS

11 THE PHYSIOLOGY AND CHEMISTRY OF GROWTH HORMONE Herbert M. Evans, J. H. Briggs AND Jonathan S. Dixon

12 THE PHYSIOLOGY AND CHEMISTRY OF FOLLICLE-STIMULATING HORMONE CARL GEMZELL AND PAUL ROOS

13 THE PHYSIOLOGY OF LUTEINIZING HORMONE ALBERT SEGALOFF

14 THE PHYSIOLOGY AND CHEMISTRY OF THE MAMMOTROPHIC HORMONE WILLIAM R. LYONS and JONATHAN S. DIXON

INDEX

CONTRIBUTORS TO THIS VOLUME

BATES, ROBERT W.

Laboratory of Nutrition and Endocrinology, National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda 14, Maryland, U.S.A.

BRIGGS, J. H.

Metabolic Research Unit, University of California, and Charing Cross Hospital, London

CONDLIFFE, PETER G.

Laboratory of Nutrition and Endocrinology, National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda 14, Maryland, U.S.A.

DIXON, JONATHAN S.

Hormone Research Laboratory, University of California, Berkeley 4, U.S.A.

DOBYNS, BROWN M.

Dept. of Surgery, Western Reserve School of Medicine, Cleveland General Hospital, Cleveland, Ohio, U.S.A.

EVANS, H. M.

Emeritus Professor of Anatomy, University of California, Berkeley, U.S.A.

GEMZELL, CARL

Akademiska Sjukhuset, University of Uppsala, Uppsala, Sweden

GREEN, J. D.

Dept, of Anatomy, School of Medicine, and Brain Research Institute, University of California Medical Center, Los Angeles 24, California, U.S.A.

HANSTRÖM, BERTIL

Lund Universitets Zoologiska Institutionen, Lund, Sweden

HOAR, WILLIAM S.

Dept, of Zoology, University of British Columbia, Vancouver 8, B.C., Canada

LYONS, WILLIAM R.

Dept, of Anatomy, School of Medicine, University of California, San Francisco 22, U.S.A.

NALBANDOV, A. V.

Dept, of Animal Science, University of Illinois, Urbana, Illinois, U.S.A.

PURVES, H. D.

Endocrinology Research Unit, Medical Research Council of New Zealand, The Medical

School, Dunedin, New Zealand

Roos, PAUL

Institute of Biochemistryllniversity of Uppsala, Uppsala, Sweden

SEGALOFF, ALBERT

Alton Ochsner Medical Foundation, 1520 Jefferson Highway, New Orleans 21, Louisiana, U.S.A,

SPARKS, LOWELL L.

Metabolic Research Unit and Department of Anatomy, School of Medicine, University of

California Medical Center, San Francisco 22, U.S.A.

WINGSTRAND, KARL GEORG

Institut for Sammenlignende Anatomi, University of Copenhagen, Denmark

1

GROSS ANATOMY OF THE HYPOPHYSIS IN

MAMMALS

BERTIL HANSTRÖM

INTRODUCTION 2

PROTOTHERIA 4

Order: Egg-laying Mammals (Monotremata) 5

METATHERIA 7

Order: Marsupials (Marsupialia) 9

EUTHERIA 9

Order: Land-living Carnivores (Carnivora) 9

Family: Dog-like carnivores (Canidae) 9

Family: Bears (Ursidae) 12

Family: Raccoon-like carnivores (Procyonidae) 12

Family: Weasel-like carnivores (Mustelidae) 13

Family: Civet-like carnivores (Viverridae) 14

Family: Cat-like carnivores (Felidae) 16

Order: Seals and Walruses (Pinnipedia) 17

Order: Whales and Dolphins (Cetacea) 18

Family: Spermwhales (Physeteridae) 18

Family: Dolphins and Porpoises (Delphinidae) 19

Family: Rorquals (Balaenopteridae) 20

Order: Even-Toed Ungulates (Artiodactyla) 20

Family: Pigs (Suidae) 20

Family: Camels and Llamas (Camelidae) 21

Family: Deer (Cervidae) 22

Family: Giraffes (Giraffidae) 22

Family: Ox-like ungulates (Bovidae) 23

Order: Odd-Toed Ungulates (Perissodactyla) 25

Family: Horse-like ungulates (Equidae) 25

Family: Tapirs (Tapiridae) 26

Family: Rhinoceroses (Rhinoceridae) 26

Order: Aardvarks (Tubulidentata) 27

Order: Elephants (Proboscidea) 27

Order: Coneys (Procavidea) 28

Order: Manatees (Sirenia) 29

Order: Pangolins (Pholidota) 30

Order: Ant-eaters, Sloths, and Armadillos (Edentata) 31

Family: Ant-eaters (Myrmecophagidae) 31

Family: Sloths (Bradypodidae) 32

Family: Armadillos (Dasypodidae) 33

Order: Hare- and Rat-like Mammals (Glires) 34

Suborder: Hares and Rabbits (Lagomorpha) 34

Suborder: Rat-like mammals (Rodentia) 35

Order: Mole- and Shrew-like Mammals (Insectívora) 36

Order: Flying Lemurs (Dermoptera) 38

Order: Bats (Chiroptera) 38

INTRODUCTION

Studies of the structure of the mammalian pituitary gland (other than those concerned with common domestic species) have often been based on material that came to the hand of the investigator by chance, perhaps after the death of the animal in a zoological garden. Even if the possibility that the hypophysis of the animal in question may be pathologically influenced by the abnormal conditions of captivity (Ouboussier, 1948, 1951) is disregarded, the specimens upon which treatises have been written very often consist of a single gland. This is particularly unfortunate because specimens of the pituitary gland display marked variations. The heart or kidney of members of a certain mammalian species may vary in size and in minor details, but not from the viewpoint of comparative anatomy. These organs are not only typically constructed within a species but also within the genus or even the mammalian order to which they belong. But so far as the pituitary is concerned, although there are some features which may occur generally within an order, there remains a wide range of inter- and intra-species variation which may proceed as far, for example, as complete disappearance of two of the three components of the glandular lobe, namely the pars tuberalis and/or the pars intermedia. In order to fill the systematic gaps in our present knowledge of the mammalian pituitary it has been necessary to discuss the material available but, when studying the structure of a gland of a single specimen of a species, the reader should remember that the picture might perhaps have been rather different if the material had been richer.

The present survey of the gross anatomy of the pituitary generally follows the division of the class Mammalia used by Piveteau (1955). The mammals are there divided into: (1) the Prototheria with the single order of egg-laying mammals (Monotremata; p. 5); (2) the Metatheria with the single order of marsupials (Marsupialia-, p. 9); and (3) the Eutheria with no less than sixteen orders. According to Piveteau, nine of these may be derived from the fossil Proto-Creodonta and the other seven from fossil Proto-Insectivora. The first group includes the land-living carnivores (Carnivora-, p. 9), the seals and walruses (Pinnipedia-, p. 17), the whales and dolphins (Cetacea-, p. 18), the even-toed ungulates (Artiodactyla-, p. 20), the odd-toed ungulates (Perisso- dactyla-, p. 25), the aardvarks (Tubulidentata’, p. 27), the elephants (Probos- cidea-, p. 27), the coneys (Procavidea-, p. 28), and the manatees (Sirenia; p. 29). To the second group belong the pangolins (Pholidota-, p. 30), the ant-eaters, sloths, and armadillos (Edentata-, p. 31), the hare- and rat-like mammals (Glires\ p. 34), the mole- and shrew-like mammals (Insectívora; p. 36), the flying lemurs (Dermoptera; p. 38), the bats (Chiroptera; p. 38), and the tree-shrews, lemurs, tarsiers, monkeys, apes, and man (Primates; p. 39).

In all vertebrates the hypophysis is derived from two sources, an ectodermal dorsal évagination from the oral epithelium (Rathke’s pouch), which, after separation from its mother-layer, gives rise to the glandular lobe or adenohypophysis, and a similarly hollow ventral process of the floor of the diencephalon, the saccus infundibuli, which gives rise to the neural lobe or neurohypophysis (Figure 1). Initially Rathke’s pouch consists of a simple vesicle with thin walls, but soon a median unpaired larger portion becomes distinguishable from two smaller lateral lobes. Contact is quickly made with the ventrally extending saccus infundibuli and then the distal wall of the median ectodermal outgrowth, which has no contact with the developing neural lobe, and the proximal wall, which touches the saccus, come to show a distinct histological differentiation. The former becomes the pars distalis, distinguished by several differentiated cell types which are commonly divided into three main groups: acidophils, basophils and chromophobes. The latter becomes the pars intermedia which has an almost typical epithelial structure.

Simultaneously with the cell proliferation of the walls of Rathke’s pouch to form the pars intermedia and the pars distalis, cysts called Rathke-cysts may arise during the disappearance of the original lumen of the pouch by the fusion of its walls. In these cysts the distal wall consists of cells belonging to the pars distalis and the proximal wall of cells belonging to the intermediate lobe. Evagination cysts are parts of the epithelium of Rathke’s pouch which have been evaginated and isolated from the mother layer. More

considerable portions of the original lumen which persist between the pars distalis and the pars intermedia form the hypophysial cleft. Finally, ciliated cysts may be produced within the pars distalis and the pars intermedia because one or both surfaces of the ectodermal pouch are often covered by a ciliated epithelium. The paired lateral lobes of Rathke’s pouch (Figure 1) grow towards the rostro-ventral region of the saccus infundibuli and the adjacent floor of the hypothalamus and form the pars tuberalis. These paired outgrowths may also possess a lumen, and ciliated cysts are often found within the differentiated pars tuberalis which otherwise, like the intermedia, has an epithelial structure.

Remnants of the ventral portions of the lateral lobes which have coalesced with the pars distalis can be clearly demonstrated in some mammals, e.g. the cat, where they are found as delicate cell-strings along the rostro-ventral surface of the pars distalis. These have been called by the name pars tuberalis interna, while the tuberalis proper is distinguished as the pars tuberalis externa. In monotremes (Figure 2) and a few other mammals (Figure 4) this latter lobe retains its original independent position in relation to the pars distalis, but in the majority it coalesces with the rostral end of the pars distalis (Figure 6), grows around the infundibular stem in a dorso-caudal direction, and mostly forms a collar around it.

In reptiles and birds the embryonic neural lobe grows from the distal end of the saccus infundibuli by the formation of hollow buds which later become completely separated from the ordinary ependymal layer as closed vesicles. A similar phenomenon has been observed in some Tower’ mammals such as adult Monotremata, adult ant-eaters, and, during ontogeny, armadillos and man.

During development the neural lobe becomes histologically differentiated into a proximal and a distal region, the eminentia mediana and the processus infundibuli which are usually connected to each other by means of an infundibular stem. The eminentia, at least rostrally and dorsally, forms part of the diencephalic floor (the tuber cinereum) caudal to the optic chiasma, and in most cases establishes contact with the pars tuberalis, while the infundibular process is generally somewhat intimately connected with the pars intermedia.

The original cellular elements of the neural lobe are the pituicytes, which are transformed neuroglia cells. The rest of the tissue consists of connective tissue and non-medullated nerve-fibres, which are derived from nuclei within the hypothalamus. Thus, fibres from the nucleus infundibularis end within the eminentia mediana (sometimes also part of the fibres from the nucleus supraopticus and nucleus paraventricularis; see Bargmann and Scharrer, 1951) and large fibre bundles (tractus supraoptico-hypophyseus) from the nucleus supraopticus and nucleus paraventricularis terminate around the blood capillaries of the processus infundibuli.

PROTOTHERIA

Like reptiles and birds the Prototheria do not bear their young but lay eggs with thin shells and large yolks. Although they feed the young with milk they have no teats or nipples, the urogenital duct and the rectum open into a cloaca, and several features of the skeleton, not least the pectoral girdle, are of the same type as found in mammal-like reptiles. But the skin is covered with hairs, the red blood corpuscles have no nuclei, there are three auditory ossicles and seven cervical vertebrae, which, together with several other facts, proves that the Prototheria should be included with the mammals. The pituitary has been examined by Green (1951), Hanström and Wingstrand (1951), and Hanström (1954).

Order: Egg-laying mammals (Monotremata)

The mono tremes include the spiny ant-eaters (Tachyglossus) and the duck« billed platypus (Ornitorhynchus anatinus). Like so many other organs in these animals the pituitary has many features in common with reptiles (Hanström and Wingstrand, 1951; Hanström, 1954). Thus the pars tuberalis includes a chief part (pars tuberalis externa) which covers the ventral and lateral regions of the eminentia mediana and is separated from the pars distalis and the pars intermedia (Figure 2). The only connection between these lobes

Figure 2. (a) The Tasmanian ant-eater (Tachyglossus setosus); (b) the Australian ant-eater (T. aculeatus); (c) the Platypus (Ornitorhynchus anatinus). 0, Optic chiasma, M, corpus mammillare; median eminence dark criss-crossed, infundibular process checked, pars intermedia solid black, pars distalis coarse stipple, pars tuberalis fine stipple. In (a) there is no contact between the infundibular process and the glandular lobe and the pars intermedia and residual lumen are small. In (b) and (c) there is an intimate contact between the lobes, and the intermedia and the hypophysial cleft are large (After Hanström, 1953b)

are strings of tuberalis cells (the portal zone), which originate in the pars tuberalis proper (pars tuberalis externa) and accompany the portal vessels to the dorsal surface of the pars distalis. They do not, however, enter the main gland at this point but turn anteriorly and fuse with the rostral region of the pars distalis, where they join the chromophobic portion of its tissue and form the pars tuberalis interna (pp. 5 and 6). The significance of these facts in comparisons of the pars distalis in reptiles and birds and in mammals and for the correlation of the pars distalis in these vertebrates with the so-called oral and aboral lobes of Rathke’s pouch, will be found in the chapter on ‘Comparative anatomy and evolution’, by K. G. Wingstrand.

The pars distalis in those species available for histological study, viz. the Tasmanian spiny ant-eater (Tachyglossus setosus) and the platypus, consists of a rostral and a caudal subdivision. The rostral subdivision could be likened to the basophilic-chromophobic zona tuberalis of Dawson (1937, 1948) which is fairly commonly found in the Eutheria, but has been seen in only one specimen of Tachyglossus. With regard to the zona tuberalis it is emphasized (Harris, 1947; Hanström, 1952a, p. 191; see also p. 4) that the term pars tuberalis interna ought to be used for those portions of the pars distalis in adult mammals which prove to be remnants of the lateral lobes of the embryonic Rathke’s pouch, while the term zona tuberalis should be used for a chiefly chromophobic part of the median rostro-ventral area of the pars distalis without any reference to its ontogenetical or comparative-anatomical significance. It is possible that future embryological investigations will show that in several instances the region which is at present called a zona tuberalis is actually a pars tuberalis interna.

In the mammalian hypophysis the absence of contact between the proximal wall of Rathke’s pouch and the infundibular process during development is very often connected with a poorly developed, or absent, pars intermedia and residual lumen. In the monotremes these relations are demonstrated in the following manner (Hanström and Wingstrand, 1951; Hanström, 1954). In three specimens of Tachyglossus setosus the caudal region of the glandular lobe was separated from the infundibular process by a more or less thick lamina of connective tissue. Only one specimen (Figure 2a) had an intermedia, which was very small, and the residual lumen was poorly developed. In a fourth specimen there was slight contact between the neural and glandular lobes, a small intermedia and a small lumen. Finally, in the only specimen of Tachyglossus aculeatus investigated (Figure 2b) and the only platypus (Figure 2c) there was close contact between the two hypophysial lobes, a relatively spacious residual lumen, and a well-developed intermedia which, in Ornitorhynchus, almost completely enclosed the processus infundibuli.

The neural lobe of the Tasmanian ant-eater (Hanström, 1953b) (Figure 2) is sac-like with a spacious recessus hypophyseus and uniformly thick walls, and in these respects is simpler than that of many birds and snakes. The eminentia mediana (Figure 3) consists of an external (glandular) layer of delicate nerve fibres from the nuclei tuberis with few pituicyte nuclei, and an internal one, containing the ependymal cells, most of the pituicytes and their nuclei and bundles of longitudinal nerve fibres from the nucleus supraopticus and nucleus paraventricularis. The characteristic capillary loops of the external layer, belonging to the portal system of the pituitary in higher mammals (Green, 1951 and earlier), do not exist in the mono tremes, where these capillaries are restricted to the surface of the eminentia. In the Tasmanian ant-eater the thin eminentia reaches all the way from the caudal margin of the optic chiasma into the processus infundibuli, and there is between the thinner walls of the eminentia and the thicker ones of the process no region which may be called a true infundibular stem. In the Australian ant-eater the walls of the process have proliferated, and in the platypus, finally, the lumen of the recessus is almost entirely limited to the region of the eminentia. From the surface of the process thin connective tissue septa extend inwards, together with blood vessels, and cause a partial lobulation of the nervous tissue. In the central region of the process the tube-shaped tractus supra- optico-hypophyseus, which surrounds the lumen of the recessus, branches into bundles which join the lobules and terminate in layers of nerve endings adhering to the peripheral walls of the lobules. From a histological point of view these layers evidently correspond to the external glandular layer of the eminentia. The minor degree of differentiation of the eminentia and the process, the spacious recessus hypophyseus, and the absence of a true infundibular stem in the monotremes are primitive features compared with most other mammals.

METATHERIA

Like the succeeding Eutheria, the Metatheria are viviparous mammals but lack a well-developed placenta. Their eggs, which have no shells, absorb nourishment for a while from the walls of the uterus, but the young are born in an immature state and for a long time cling to the teats of the mother, which in the majority of species are situated in a marsupial pouch. Several other features in the anatomy of the genital organs, the skeleton, the dentition, and the brain are primitive or at least peculiar to the Metatheria. There is but a single order.

Order: Marsupials (Marsupialia)

The pituitary of marsupials has been studied by Parker (1917), Dawson (1938), Bodian (1939, 1940, 1951), Wheeler (1943), Oboussier (1948, 1954, 1955a), Hanström (1950, 1953a), and Green (1951). This organ is anatomically most interesting in a species of the kangaroo family (Macropodidae), viz. the quocca (Setonix brachyurd). Here the pars tuberalis shows morphological affinities with its counterpart in monotremes, while in other respects the gross anatomy of its pituitary could be a model for a simple type in the Eutheria (Figure 4).

As in monotremes the pars tuberalis consists of a chief part, the pars tuberalis externa, and a less prominent pars tuberalis interna (cp. pp. 4-6). The external pars tuberalis forms a collar around the infundibular stem (although weak posteriorly) and covers the eminentia mediana as well as part of the tuber cinereum rostral and lateral to the eminentia. The pars tuberalis interna is connected with the pars tuberalis externa by two narrow strings of cells only, which pass rostrally and laterally from the latter and

merge into the anterior region of the pars distalis (Figure 4). The strings then pass further downwards and unite in the middle, ventral surface of the pars distalis. They consist of chromophobes and basophils exclusively.

Figure 4. The pituitary of the Short-tailed wallaby or quocca (Setonix setosus). Indications as in Figure 2. The small caudal part of the region marked like the median eminence, which is situated between the caudal end of the pars tuberalis and the infundibular process, is differentiated as an infundibular stem. Although the thread of cells connecting the pars tuberalis proper with the pars tuberalis interna lies more laterally it is shown in the diagram (After Hanström, 1953a)

The residual lumen of Rathke’s pouch is fairly large and the pars intermedia covers the ventral and lateral, but not the dorsomedial and caudal, regions of the processus infundibuli.

The eminentia mediana ends rostrally at a point behind the anterior extension of the pars tuberalis externa, and both terminate before reaching the optic chiasma. The structure of the eminentia resembles that of the mono- tremes except that there are ‘capillary loops’ within the glandular layer, the external surface of which is corrugated as in most Eutheria, A short but

distinct infundibular stem exists behind the caudal end of the pars tuberalis, but the recessus hypophyseus only reaches a short way into the processus infundibuli, which itself shows a rudimentary peripheral lobulation of the same kind as occurs in the platypus (p. 7) and in other marsupials, such as the American opossums (p. 9).

Another member of the same family is the giant kangaroo of the genus Macropus (Oboussier, 1955a). In most respects the gland of this animal seems to correspond with that of the quocca, although no pars tuberalis interna has been described.

The axis of the neural lobe in the quocca and the giant kangaroo is horizontally directed, but in some other marsupials it is ventrally inclined. In the Australian brush-tail opossum (Trichosurus vulpécula) of the family Phalangeridae the axis slants backwards (Parker, 1917), whilst in the Tasmanian devil (Sarcophilus barrissi) of the family Dasyuridae (Oboussier, 1954) and the American opossums (Didelphis virginianus and D. aurita) of the family Didelphidae it slants forward (Dawson, 1938; Hanström, 1950, 1953a). In the three latter species (Figure 5) the processus infundibuli is deeply embedded in the pars distalis, which surrounds it except on the dorso-caudal surface, and is separated from it only by a very thin remnant of the hypophysial cleft and a unicellular layer of pars intermedia. The collar-shaped pars tuberalis is also Weakly developed. The anterior portion of the pars distalis, at least in Didelphis virginianus, is composed predominantly of basophils and chromophobes, forming the zona tuberalis of Dawson, which is continuous with the pars tuberalis proper.

In the neural lobe of Didelphis the eminentia resembles that of Setonix, but there is no infundibular stem since the eminentia ventrally borders directly on the processus infundibuli. As in the other species of marsupials thus far investigated, the recessus hypophyseus does not enter the process. According to Bodian (1951) the solid tractus supraopticus forms a hilum in the process, and each branch which enters and forms the hilum of a lobule repeats the structure of the tract itself. The glandular layer has the same position in the periphery of the lobules as in the Tasmanian ant-eater (p. 6), and the septa between the lobules contain the vessels of the hypophysial portal system. An equally beautifully lobulated process occurs in the brush-tail opossum (Parker, 1917).

EUTHERIA

These are the higher mammals, in which there is a true placenta and whose young are born as miniature adults after a long period of gestation. The brain is much more highly differentiated and possesses a corpus callosum which is absent from Proto- and Metatheria.

Order: Land-living carnivores (Carnivora)

Family: Dog-like carnivores (Canidae)

As shown by Oboussier (1948, 1955; cp. also Bargmann, 1958) the morphology of the pituitary of domesticated dogs is very variable, both individually and in different breeds. Common to full-grown specimens of most breeds (with few exceptions) is the fact that both the pars intermedia and the pars distalis have surrounded the processus infundibuli by growing around it from an original ventral position (Figure 6). The hypophysial cleft is generally fairly spacious ventrally but less so dorsally. The relations between the lobes and the cleft recall those in the American opossums (Figure 5), but the dogs have retained the original horizontal position of the axis of the gland. It should be mentioned that the pars distalis in domesticated dogs fairly often retains remnants (gland cells, cysts, colloid, etc.) of the embryonic connection with the palatal epithelium (Figure 1), and, in one specimen, an open canal (canalis craniopharyngeus) to the pharynx has been discovered (Oboussier, 1944, 1950, 1951). The pars tuberalis forms a collar at the junction of the median eminence and the infundibular process, whilst folds are evident in the pars intermedia, especially at its rostral end (= " vordere Umschlagszone’ of Romeis, 1940), which are often mixed with proliferations of the pars dis talis. In the fox there is a similar mixed zone, particularly at the caudal pole of the intermedia, where the inferior hypophysial artery enters the gland (Figure 7).

In the neural lobe the histology of the eminentia is similar to that of Setonix and the cat (Figure 34), and this type occurs in all carnivores thus far investigated (Hanström, 1953b). The median eminence occupies that region of the neural lobe which is in contact with the pars tuberalis, and the spacious

but very short infundibular stem, like the eminence itself, is penetrated by the appropriately short infundibular recess; this terminates at the rostral pole of the infundibular process (cp. also Figure 9). The structure of the neural lobe and its connections with the neurosecretory nuclei of the hypothalamus have been described by Bargmann (1949).

In most other members of the family Canidae that have been studied the pars intermedia and pars distalis have begun a similar dorsal folding and encircling of the infundibular process as in dogs. The lumen of the hypophysial cleft usually participates in this development, at least ventrally and

laterally, but the internal and external walls usually retain their original connection in the neighbourhood of the dorsomedial line. A gland of this type, although with smaller variations, is found in the black-backed jackal (Canis mesomelas; Oboussier, 1949; Hanström, 1952a), the maned dog (Chrysocyon jubatus; Oboussier, 1948), the fennek (Fennecus; Oboussier, 1948), and the raccoon-like dog (Nyctereûtes procyonides; Oboussier, 1948). The European red fox (Vulpes vulpes\ Hanström, 1947) seems to be a little different, in that the dorsomedial glandular layer of the specimen pictured in Figure 7 consists of intermedia tissue only, which has probably grown by proliferation and not by encircling.

According to Oboussier (1958) four pituitaries of the wolf (Canis lupus) that have been examined belong to the same type as that of the dog and the other species just mentioned (except the red fox) and show an infundibular process which is completely surrounded by a layer of intermedia and one of distalis tissue with remnants of the hypophysial cleft. In five specimens investigated by Hanström (1955) the process is enveloped by a dorsally continuous intermedia in the normal position and a co-extensive external layer of the same histological structure with a distinct lumen between them. Under those circumstances the external (dorsal) wall should normally have consisted of pars distalis tissue (cp. Figure 6), but as in the American mink (Figure 12) it has instead the structure of an intermedia (Figure 8). As pointed

out earlier (Hanström, 1947, p. 36), there is in the mink (as in the wolf) an external pars intermedia, separated from the internal one and from the infundibular process by the hypophysial cleft, despite the apparent necessity for contact with the neural lobe for differentiation of the intermedia structure (see p. 6). Two possible explanations are available to explain this situation in the mink and the wolf. Either the supposed formative influence originating from the embryonic neural lobe, and acting on the glandular one, must also affect the dorsal regions of the distal lobe on its way to encircle the processus infundibuli, or a very rapid growth of the most dorsal region of the internal intermedia layer has taken place, which may result in the development of an external intermedia layer.

The long-eared fox (Otocyon megalotis), which has a dentition very different from that of all other carnivores, also possesses a pituitary which is completely divergent from the complicated morphology in that of its relatives. Thus in the only investigated specimen the pars distalis does not show any tendency to encircle the processus infundibuli, the extension of the pars intermedia is restricted to the central ventral surface of the neural lobe, and in addition there is no residual lumen of Rathke’s pouch (Oboussier, 1948).

GROSS ANATOMY OF THE HYPOPHYSIS IN MAMMALS

Family: Bears (Ursidae)

The pituitary of the polar bear (Ursus maritimus) is very characteristic (Figure 9) and in its way just as complicated as that of the dog and the wolf (Hanström, 1947; Oboussier, 1955a). The pars tuberalis forms a collar around the thick and short infundibular stem and reaches the optic chiasma rostrally. The pars intermedia, which is separated from the pars distalis by a fairly spacious lumen, surrounds the whole processus infundibuli and

extends columns of intermedia tissue into the neural lobe, mostly without penetrating the connective tissue boundary. This is, however, a ‘pseudo- lobulation’ of the process (Hanström, 1953b) which has nothing to do with the distribution of the tractus supraoptico-hypophyseus, as is the case in the American opossums (p. 9).

In addition to the tuberalis and intermedia lobes the pars distalis has also proliferated around the neural lobe (and around the dorsal intermedia layer) and so formed an additional collar (= " Nackenhypophyse ’ of Romeis, 1940) around the infundibular process behind that of the pars tuberalis (Figure 9).

The recessus infundibularis has a very flat floor, from which the recessus hypophyseus penetrates into the neural lobe almost to its distal end in a way which recalls the pituitary of the monotremes (Figure 2b).

The pituitary of the European brown bear (Ursus arctos) largely resembles that of the polar bear (Hanström, 1947; Oboussier, 1955a).

Family: Racoon-like carnivores (Procyonidae)

The few pituitaries of this family which have been studied come from the lesser panda (Ailurus fulgens\ Oboussier, 1955b), the raccoon (Procyon lotor\

Oboussier, 1948), and the coati (Nasua narica; Hanström, 1950). In the glands of Ailurus and Nasua there is a collar-like pars tuberalis and a well-developed intermedia, which is separated from the pars distalis by a fairly spacious hypophysial cleft (Figure 10). The recessus hypophyseus terminates within the region of the eminentia mediana and thus the infundibular process is solid. The principal morphological difference in the raccoon is the presence of a thin layer of intermedia tissue dorsal to the neural lobe.

Family; Weasel-like carnivores (Mustelidae)

The pituitary gland of most weasel-like carnivores is almost as simple as that of the raccoon-like family. For instance the hypophysis of the ferret (Mustela furo) (Figure 11) is similar to that of the lesser panda (Figure 10),

despite the more flattened form. In one specimen investigated (Hanström, 1947) the intermedia surrounded the infundibular process except in the dorsomedial region; according to Holmes (1960; see also Thomson and

Figure 12. Sagittal section of the pituitary of the North American mink (Mustela vison). 1, recessus hypophyseus; 2, pars tuberalis; 3, pars distalis (also dorsal to the hypophysial stem); 4, hypophysial cleft; 5, pars intermedia (internal layer); 6, processus infundibuli; 7, entrance of hypophysial artery;

8, pars intermedia (external layer); 9, pars distalis; 10, mammillary body (After Hanström, 1947)

Zuckerman, 1954) the extent of the intermedia may vary to a considerable degree. Substantially the same architecture is found in the polecat (Mustela putorius), weasel (M. nivalis), and stoat (M. erminea), investigated by Hanström (1947), and the stoat of Oboussier (1948). However, in the weasel and stoat the intermedia may completely surround the infundibular process, and in the otter (Lutra lutra) there is a small bridge of pars distalis and pars intermedia tissue in the middle region of the dorsal surface of the process (Oboussier, 1955a). In the striped polecat (Ictyonix striatus) there is in addition a caudal double intermedia layer, which partially surrounds the infundibular process. Another form of this double intermedia is found dorsal to the process in the American mink (Mustela vison), where it and its lateral and ventral internal extensions together with a continuous intraglandular cleft (Figure 12) completely surround the processus infundibuli. A further complication is evident in the gland of the mink, for in addition to the collar-like pars tuberalis and the double dorsal intermedia the pars distalis too has developed a dorsal bridge which lies behind that of the pars tuberalis. This is similar to the condition in bears (see Figure 9). On the other hand the only wolverines (Guio guio*, Hanström, 1947) investigated had glands in which the pars intermedia seemed to be reduced in a pathological fashion.

The European badger (Meles meles) belongs to another group of weasellike carnivores and possesses a pituitary of a simpler type (Figure 13), The only morphological peculiarity is that, as in the weasel, the intermedia completely surrounds the infundibular process except, of course, at the entrance of the inferior hypophysial artery (Hanström, 1947).

In the neural lobe of all weasel-like carnivores the recessus hypophyseus terminates within the median eminence or the infundibular stem.

Family: Civet-like carnivores (Viverridae)

The few pituitaries of this family investigated by Hanström (1952a) show a considerable variation. The simplest is that of a mongoose of the genus

Myonax, in which the pars tuberalis does not completely surround the infundibular stem, the pars intermedia is confined to the rostral surface of the infundibular process, and the recessus hypophyseus reaches through the stem to the rostral border of the process.

EUTHERIA

Three other viverridae studied are the large-spotted genet (Genetta tigrina), the bushy-tailed miercat (Cynictis penicillata), and the ruddy mongoose (Galerella caesi). An interesting feature of Genetta is the very well-developed recessus hypophyseus, which, as in bears (Figure 14), extends into the heart of the infundibular process; in the other two forms it ends within the eminentia or the infundibular stem. In Genetta too, the pars intermedia consists of a single layer (with numerous cysts), whilst in Galerella and Cynictis there is a caudal double intermedia layer which may be seen in Galerella (Figure 15) to surround almost the entire infundibular process.

Figure 15. The pituitary of the ruddy mongoose (Galerella caesi). Indications as in Figure 6 (After Hanstrom, 1947)

The pituitary of Cynictis penicillata, which belongs to a full-grown animal, shows several primitive features (Figure 16). Thus the pars tuberalis externa contains remnants of the lumen of the lateral lobes of Rathke’s pouch, and within the pars distalis there are two lateral columns of cells which represent a typical pars tuberalis interna. In addition the large residual lumen possesses

Figure 16. The pituitary of the bushy-tailed miercat (Cynictis penicillata). Indications as in Figure 6. 1, chiasma; 2, remnant of the lumen of the lateral lobe of Rathke’s pouch; 3, epithelial stalk cyst; 4 corpus mammillare (After Hanström, 1947)

three distinct prolongations which correspond to the rostral, ventral, and caudal extensions of Rathke’s pouch at the stage in the embryonic cat, depicted in Figure 1. The middle ventral prolongation of the lumen, which is partially transformed into Rathke-cysts, is directed towards a large peripheral cyst which remains as a remnant of the original connection between the pharynx epithelium and Rathke’s pouch.

Family: Cat-like carnivores (Felidae)

The pituitary of the domestic cat (Felis silvestris silvestris) has been described in numerous papers; some of the most important ones are those by Brahms (1932), Dawson (1937), and Romeis (1940), and, for the grey wild cat (Felis silvestris ocreata), Hanström (1952a). The neural lobe and the innervation of the gland have been vigorously discussed in recent years, as, for example, by Nowakowski (1951), Spatz (1951), Engelhardt (1956),

Figure 17. The pituitary of the grey wild cat. Indications as in Figure 6. The short white strokes in the pars intermedia mark the border between two regions of different histological structure (After Hanstrom, 1947)

Metuzals (1959), and Martinez (1960). The morphology of the neurohypophysis (Figure 17) is, however, also interesting, for the recessus hypophyseus extends into the centre of the infundibular process as in bears and the genet, whilst the median eminence (Figure 34) is of a simple structure typical of the majority of the Eutheria (Hanstrom, 1953b).

The pars tuberalis forms a collar around the base of the infundibular stem and at least in the grey wild cat includes remnants of the pars tuberalis interna (Hanström, 1952a). This region probably is included in or perhaps is identical with the zona tuberalis of Dawson (1937). A double layer of the pars intermedia is always formed caudally, whilst the dorsal surface of the infundibular process is usually covered either by a single or a double intermedia layer. It is interesting that the most rostral region of the intermedia, which adheres to the ventral surface of the infundibular stem, has a different cytological structure from the main part, which adheres to the process, and much resembles the pars tuberalis. The development of this "paraneurale Umschlagszone’ of Romeis (1940; Abb. 180a) may perhaps be ascribed to a different formative influence of the different regions of the neural lobe on the histological structure of the embryonic glandular lobe (see Hanstrom, 1952a, p. 216).

EUTHERIA

The pituitary of the ocelot (Felis pardalis) resembles that of the cat except in that the pars intermedia does not cover the dorsal region of the infundibular process (Hanström, 1950). In the only pituitary of the tiger (Panther a tigris) that has been studied (Figure 18), the pars intermedia is much smaller

and covers the rostral surface of the process only; the double caudal layer, which exists in the two species of Felis, is absent (Hanström, 1946b). The axis is also different from that in the cat. The pars distalis is situated rostral to the infundibular process and does not extend beneath it as in most other mammals ‘below’ the anthropoids.

The lion (Panthera leo) pituitary is akin to that of the tiger (Hanström., 1952a; Kladetzky, 1954). In this species the dorso-ventral orientation is, however, still more emphasized because the gland is suspended from the base of the tuber cinereum, almost as in man and the anthropoids (Figure 19).

Order: Seals and Walruses (Pinnipedia)

The only descriptions of the hypophysis in seals and walruses are contained in a paper by Fuse (1939) on a fur seal (Callorhinus ursinus) and another by Oboussier (1955a) on the pituitaries of the South American sea-lion (Otaria by roñica) (Figure 20), the walrus (Odobenus rosmarus) and a systematically unplaced fur seal. In no instance is there any true recessus hypophyseus, whilst the hypophysial cleft is narrow but otherwise distinctly developed. In Otaria, the walrus, and Oboussier’s fur seal the pars tuberalis is collar-shaped, and in the last two species the infundibular process is completely surrounded

by the intermedia internally and the pars distalis externally with remnants of the residual cleft partially separating them as in some Canidae (Figure 6). In Otaria, however (Figure 20), a thin layer of the pars distalis alone seems to cover part (in a male) or the whole (in a female) of the dorsal surface of the process.

Order: Whales and Dolphins (Cetacea)

Family: Sperm whales (Physeteridae)

This family and the next following, the Delphinidae, belong to the toothed whales which have functional teeth but no baleens (plates of whalebone). Only two studies of the pituitary in a sperm whale (Physeter macrocephalus)

have been reported (Geiling, 1935; Wislocki and Geiling, 1936). The morphological type of this species (Figure 21) is, however, characteristic not only for another toothed whale, the bottle-nosed porpoise (p. 9), but largely for the baleen whales too.

EUTHERIA

In this group of mammals, the pars intermedia is completely lacking in all species investigated. Since the differentiation of an intermedia in mammals is dependent upon contact between the embryonic glandular obe and the infundibular process (p. 6), it is not surprising to find that in all whales the pars distalis and the process are completely separated by a wall of connective tissue which is sometimes very thick. There are no traces of an hypophysial cleft. The only communication between the glandular and the neural lobe is provided by the portal system in the region of the pars tuberalis where this enfolds the infundibular stem with a small clasp-like expansion. Apparently only the lateral paired prolongations of Rathke’s pouch have made contact with the embryonic neural lobe to produce the pars tuberalis, whereas the medial unpaired portion has produced either a primary solid adenohypophysis or one which has secondarily become solid. Because it has had no contact with the neural lobe an intermedia has not differentiated.

The position of the glandular and the neural lobes in the sperm whale, and still more in the blue whale (Figure 22b), is nearly the same as in the

lion. The external form of the gland also recalls that of the lion, because in the sperm whale its axis is almost perpendicular, and in the blue whale strictly perpendicular, so that the pars distalis is situated rostral to the infundibular process. However, the similarity between the neural lobes (Figures 19, 21) is only superficial; in the lion there is a recessus hypophyseus, which reaches deep down into the process, while in the whales there is just a rudiment of a recess within the median eminence. The relative size of the distal lobe compared with its counterpart is also much greater in the sperm whale than in the lion.

Family: Dolphins and Porpoises (Delphinidae)

In this family the pituitary of the bottle-nosed porpoise (Tursiops truncatus) has been investigated by Wislocki (1929) and Drager (1953), who showed that the sperm whale and the porpoise pituitaries are very similar. The pars tuberalis does not completely surround the proximal region of the neural lobe in either species.

Family: Rorquals (Balaenopteridae)

The pituitaries have been described in the following species: the finback whale (Balaenoptera physalus; Geiling, 1935; Wislocki and Geiling, 1936; Harris, 1950; Sverdrup and Arnesen, 1952), Rudolphi’s rorqual (Balaenoptera borealis; Hanström, 1944, Holzmann, 1960), the lesser rorqual (Balaenoptera acutorostrata; Harris, 1950), the blue whale (Sibbaldus musculus; Wislocki and Geiling, 1936; Valso, 1938; Harris, 1950), and the humpback whale (Megaptera novae-anglicae; Hanström, 1944). Several papers are mainly concerned with the neural lobe and its connection with the hypothalamus (Harris, 1950; Holzmann, 1960), another (Sverdrup and Arnesen, 1952) with only the distal lobe. However, present knowledge of the morphology of the rorqual pituitary establishes that this organ (Figure 22) has the same general structure as in toothed whales, which argues for the existence of a common ancestor for these two groups. A minute difference is that the pars tuberalis has completely enveloped the median eminence in at least two rorquals: Rudolphi’s species and the blue whale (Figure 22a,b). A more or less vertically orientated axis is characteristic of all glands in toothed and baleen whales. This is, however, as in Hominoidea (Figure 57), ontogenetic- ally a secondary phenomenon, since Harris (1950) has found that in a foetus of a Balaenoptera-species the axis is horizontal and the infundibular process lies dorsal to the pars distalis and not behind it, as in adults.

It ought to be mentioned that the glandular lobe in rorquals contains one of the most impressive and typical zona tuberalis (or pars tuberalis interna?) thus far detected in mammals.

With a reliably measured length of 33-27 metres and a weight of 147,000 kg the blue whale is the biggest of all living animals. The hypophysis of specimens weighing at least 100,000 kg had glandular lobes 30-45 mm long, 40-60 mm broad, and 20-25 mm high; the weight was maximally 53*5 g. The neural lobes, however, weighed at most 1*9 g (Valso, 1938). The proportions between these lobes agree with those in other giant mammals, such as elephants (p. 27) and giraffes (p. 22).

Order: Even-Toed Ungulates (Artiodactyla)

Family: Pigs (Suidae)

The histogenesis of the neural lobe of the domesticated pig (Sus scrofa) has been described by Shanklin (1944), the whole gland from the standpoint of domestication by Herre and Behrendt (1940) and Oboussier (1943) and, more generally, by Trautmann (1909a,b, 1911), Beato (1935), Green (1951), Herlant (1951), and Arvy and Buisson (1961).

In the newborn pig, according to Green (1951), the external shape and the anatomy is remarkably different from that in the adults, described by other authors, and resembles that of the cat, although the recessus hypo- physeus terminates within the median eminence (Figure 23a). It deviates from all other glands thus far examined in the dorsal extension of the pars intermedia, which in adult specimens pictured by Trautmann (1911), Oboussier (1943), Herlant (1951), and Arvy and Buisson (1961) is restricted to the region between the pars distalis and the infundibular process (Figure 23b), The hypophysis in Figure 23b is, however, shown as lacking the pars tuberalis dorsal to the median eminence, whilst such a collar-shaped tuberalis is present in the specimens described by Trautmann (1911), Green (1951), Herlant (1951), and Arvy and Buisson (1961). The hypophysial cleft

Figure 23. The pituitaries, (a) of a newborn pig (Sus scrofa), and (b) of a wild adult. Indications as in Figure 6 (a, after Green, 1951; b, after Oboussier, 1943)

may or may not have completely disappeared in adults. The American peccary of the genus Tayassu examined by Oboussier (1948) wholly resembles that of the wild European pig, as described by Oboussier (1943; Figure 23b).

Family: Camels and Llamas (Camelidae)

The few papers on the pituitaries of members of this family present rather scanty information concerning the morphology, although there is sufficient to permit a general picture. Thus in the Bactrian camel (Camelus bactrianus’, Stendell, 1913, 1914), the guanaco (Lama glama huanachus; Oboussier, 1948), and probably also the dromedary (Camelus dromedarius; Watermann, 1959) the pars tuberalis forms a collar around the short or perhaps histologically non-existent infundibular stem (Figure 24a). In the only Bactrian camel

Figure 24. The pituitaries, (a) of a Bactrian camel (Camelus bactrianus), and (b) of a llama (Lama glama glama). Indications as in Figure 6 (a, after Stendell, 1913; b, after Oboussier, 1955a)

studied and in the llama (Lama glama glama’, Oboussier, 1955a) the intermedia is partly separated from the pars distalis by a remnant of the hypophysial cleft (Figure 23a,b), while in two specimens of the guanaco the cleft was evidently obliterated. A profuse development of the pars intermedia is common for all glands. This completely surrounds the infundibular process with a very thick layer and in the guanaco is proportionally one of the biggest in mammals (Oboussier, 1948). In a male guanaco the pars distalis too has grown upwards and formed a second external adenohypophysial layer around the process; in a female specimen and in male and female llamas investigated by Oboussier (1948, 1955a) this was not the case. As in pigs the recessus hypophyseus terminates in the neighbourhood of the junction between the eminentia (or infundibular stem) and the infundibular process.

In camels Stendell (1913) and Watermann (1959) have described a lobulation of the infundibular process (not shown in Figure 24a) which seems to be of the same type as in the platypus in the monotremes (Figure 2).

Family: Deers (Cervidae)

In the deer pituitaries (Figure 25) the recessus hypophyseus terminates within the median eminence and in such species as the red deer (Cervus elaphus’, Oboussier, 1955a), fallow deer (Dama dama’, Waage, 1953), pudu

deer (Puda pudu\ Oboussier, 1955a), and the roe deer (Capreolus capreolus’, Waage, 1953), the pars intermedia, although very well developed, does not cover the dorsal surface of the infundibular process. However, the rostro- dorsal region of the process in the reindeer (Rangifer tarandus’, Waage, 1953) and almost its whole surface in the elk (Alces alces; von Mecklenburg, 1955) is surrounded by a thick layer of intermedia cells, as in the Camelidae.

Family: Giraffes (Giraffidae)

The hypophysis of the giraffe (Giraffa Camelopardalis) has been described by Hanström (1952a; see also 1952b), Kladetzky (1954), and Oboussier (1955a). The dimensions of the gland in a 15 years old bull (Hanström, 1952a) were: length 38 mm, breadth 20-5 mm, and height 14 mm. When compared with the largest dimensions (45 x 60 x 25 mm) of the blue whale pituitary, described on p. 20, the giraffe gland shows a much higher relative weight, since the giraffe bull may weigh at most 1,000 kg against the blue whale’s 147,000 kg.

As in the deer the recessus hypophyseus terminates in the region of the eminence (Figure 26), which itself fuses directly with the elongated infundibular process without any true infundibular stem. The eminentia is very thick and its external (glandular) layer (see p. 29) is divided into two subdivisions, of which the outermost is characterized by a great amount of connective tissue. The true glandular layer thus seems to be the middle region (the proximal subdivision of the external layer; see Hanström, 1953b, pp. 247-248). The pars tuberalis reaches the optic chiasma rostrally, and, as the rostro-ventral part of the median eminence almost touches the caudal region of the chiasma, consequently there is no pars oralis tuberis of the school of Spatz (1951; see Spuler, 1951). Between the dorso-caudal end of the collar-like pars tuberalis and the corpora mamillaria there is, however, a short pars caudalis tuberis.

As in most even-toed ungulates, the pars intermedia of the giraffe is large and completely surrounds the infundibular process. This probably takes place after proliferation and not as an embryonic folding of the walls of the growing Rathke’s pouch. Caudo-laterally the intermedia also covers part of the pars distalis. This lobe contains a large and richly vascularized zona

Figure 26. Mid-sagittal section of the pituitary of a giraffe (Giraffa Camelopardalis). Hypothalamus (with 1, optic chiasma; 2, mammillary body), median eminence, processus infundibuli, striated; pars tuberalis, fine stipple; pars intermedia, solid black. In the pars distalis, the zona tuberalis is left white, the rest, predominantly composed of acidophils, marked horizontally (After Hanstrom, 1952)

tuberalis, which is characterized by chromophobes and basophils (Figure 26), whilst the extensive pars distalis proper is completely predominated by acidophils with a small number of chromophobes and basophils. In the middle part of the hypophysis the pars distalis has proliferated dorsally into a thin lamina of distalis tissue above the equally thin dorsal intermedia. Although no exact calculation of the relative size of the neural and glandular lobes has been made, the transverse sections indicate that in the giraffe, as in whales, the glandular lobe is greater by far than the neural lobe.

In a giraffe calf (Hanström, 1952a), as in the specimen investigated by Oboussier (1955a), there was no pars intermedia and no pars distalis dorsal to the infundibular process. The latter animal was a bull, probably younger, because the length of the gland was only 24 mm.

Family: Ox-like ungulates (Bovidae)

Only a few of the many papers which deal with the pituitaries of such domesticated members of the group as the ox (Bos taurus), the goat (Capra sp.), and the sheep (Ovis sp.) can be mentioned here, thus: Trautmann (1909a,b, 1911, ox, goat, sheep), Stendell (1913, 1914, ox, goat, sheep), Wulzen (1914, ox), Atwell and Marinus (1918, ox), Lubberhuizen (1927, sheep), Gilmore, Petersen and Rasmussen (1941, ox), House (1943, ox, sheep), Garm (1949, ox), Basset and McMeekan (1951, ox), Waage (1953, sheep and the American Bison bison) Dellmann (1959, 1960, ox) and Sajonski (1959-60, sheep and goat). Nine wild South African species have been examined by Hanström; the Cape eland (Taurotragus oryx), the kudu (Strepsiceros strepsiceros), the impala (Aepyceros melampus), Grant’s gazelle (Gazella granii), the springbok (Antidorcas marsupialis), the grey duiker (Sylvicapra grimmia), the kongoni (Bubalis cokei), the Cape grysbuck (Nototragus melanotis), and the steenbok [Raphicerus campestris). Although there are great variations in the histology and in the development of cysts, follicles, and pseudofollicles, the following morphological features are common in both the domesticated (Figure 27) and the wild ox-like ungulates.

The recessus hypophyseus never extends into the processus infundibuli and terminates within the median eminence or the infundibular stem. The residual lumen of Rathke’s pouch is mostly obliterated either by fusion of the rostral surface of the pars distalis and the ventral surface of the pars intermedia, or by a transformation into cysts of Rathke. The pars tuberalis is always present and collar-like. A beautiful zona tuberalis is always present. The intermedia is always well developed in the median regions, and, in cattle, eland, kudu, gazelle, springbok, impala, sheep, and goat, surrounds the lower half of the processus infundibuli, but does not produce any double folds; in the duiker, kongoni, grysbuck, and steenbok it is reduced laterally and only covers the ventral region of the processus infundibuli. The neural lobe and its connections with nuclei of the hypothalamus have been described in detail recently by Dellman (1959a,b, 1960a,b).

A special feature of the ox and the sheep is the ‘cone of Wulzen’ (Wulzen, 1914; Atwell and Marinus, 1918; Lubberhuizen, 1927; House, 1943; Garm, 1949; and Waage, 1953). This attribute is a remarkable prolongation of the pars intermedia into the hypophysial cleft (Figure 27) which possesses the same histological appearance as the pars distalis with its acidophils and basophils. Figure 28 illustrates Wulzen’s cone in a 27-mm embryo of the sheep but this structure is recognizable in the same position as early as the 21-mm stage.

On account of its position far from the embryonic neural lobe, the peculiar histology of the Wulzen’s cone may be explained by the fact that part of the connective tissue in the vicinity of the embryonic hypophysis invades the space between the neural and glandular lobes in the region of the cone and probably prevents the action of the formative influence of the saccus

infundibuli on the region of the intermedia furthest from it. Consequently the cone does not assume the epithelial structure of the main part of the intermedia but develops histologically in the same direction as the pars dis talis. This