The Structure-Function Relationship of the Lung of the Australian Sea Lion Neophoca Cinerea
The Structure-Function Relationship of the Lung of the Australian Sea Liont Neophoc e clnerea
by
Anthony Nicholson B.V.Sc.
A thesis submitted for the degree of Doctor of PhilosoPhY' Department of PathologY' UniversitY of Adelaide
February 1984 Frontispiece: Group of four adull female Australian sea lions basking in the sun at Seal Bay, Kangaroo Island. ËF:æ: oo',,, 'å¡ -*-d, l---
--a - .¡* É--- .-\tb.<¡- <}
b' \ .ltl '' 4
qÙ CONTENTS Page
List of Figures X List of Tables xi Abstract XIV Declaration XV Acknowledge m ents I I. Introduction Chapter \ I I.I Classification of Marine Mammals I I.2', Distribution of Australian Pinnipeds 2 I.3 Diving CaPabilitY 3 PhYsiologY 1.4 Diving 4 Cardiovascular SYstem ' l'.4.I B I.4.2 OxYgen Stores 1l L.4.3 BiochemicalAdaPtations L3 I.4.4 PulmonarYFunction I.4.5 Effects oi Incteased Hydrostatic Pressure T6 l-8 1.5 SummarY and Aims 20 Chapter 2. Materials and Methods 20 ?.I Specimen Collection 2I 2.2 Lung Fixation 2I 2.3 Lung Votume Determination 22 2.4 Parasite Collection and Incubation 22 2.5 M icroscoPY 22 2.5.I Light MicroscoPY Electron Microscopy 23 2..5.2 Trãnsmission 23 2.5.3 Scanning ElectronMicroscopy 25 Chapter 5. Norm al ResPiratorY Structure 25 t.r Introduction 25 Mam maI Respiratory System 3.2 Terrestrial 25 1.2.I MacroscoPtc 27 3.2.2 MicroscoPic 27 SYstem 3.3 Pinniped ResPiratorY 27 3.3.I MacroscoPic 28 3.3.2 MicroscoPic 3I 3.4 Results 3I 1.4.L MacroscoPic 32 3.4.2 MicroscoPic 7B 3.5 Discussion 7B 3.5.I MacroscoPtc 79 3.5.2 MicroscoPic 92 3.6 SummarY IV
Page
Chapter 4. Lung MorphometrY 94 4.1 Introduction 94 4.2 Methods 97 4.2.I Basic Principles of Point Counting 9B 4.2.2 Sampling Design 9B 4.2.3 Animal and Regional Differences 110 4.2.4 Interspecies ComParisoq tt0 4.3 Results 1r.1 4.1.1 Species Results t1l_ 4.3.2 Animal and Regional Differences lll 4.3.3 Interspecies ComParison 119 4.4 D iscussion I23 4.4.1 Discussion of Methods r23 4.4.2 Discussion of Results 135 4.5 Summary 153
Chapter 5- PulmonarY PathologY D5 5.1 Introduction r55 5.2 Results D5 5.2.I Macroscopic b5 5.2.2 M icroscopic 15B r69 5.3 D iscuss ion 5.3.I In vitro develoPment r69 5.3.2 PathologY r69 170 5.4 Su mm ary
Chapter 6. Discussion r7t 6.r Lung Structure r73 6.2 Lung MorphometrY L74 6.t Diving PhysiologY 175 6.J.I Aerobic Dive Limit 175 6.3.2 Bioche m istrY r77 6.3.3 Pulmonary Function r77 6.3.4 Increased Hydrostatic Pressure and Lung Collapse IB2 6.3..5 Nitrogen Absorption and Decompression Sickness tB5 6.4 Conclusion 187 6.5 Future Research r9l
Bibliography 193 LIST OF FIGURES
Page
Figure t. Dorsal view of excised lungs and trachea 3B 3B 2. Ventral view of excised lungs and trachea 39 3. Laleral view of left lung 4. Lateral view of right lung 39 5. Macroscopic view of lung samPle 40 40 6. Low power light micrograph of lung sample 7. Light micrograph of junction between tertiary and lobular bronchi 4I
B Higher power light micrograph of junction between tertiary and lobular bronchi 4I 9 scanning electron micrograph of junction between tertiary and lobular bronchi 41, 10. Transmission electron micrograph of transition region between tertiary and lobular bornchi 42 r1. Transmission electron micrograph of transition region 42 r2. Transmission electron micrograph of secretory cells in transition region 43 4J 13. High power transmission electron micrograph of secretory cell 44 14. Transmission electron micrograph of transition region 15. Transmission electron micrograph of non-secretory cells in transition region ¡ 44 16. Transmission electron micrograph of columnar ciliated cell in transition region 45 r7. Transmission electron micrograph of columnar microvillous cell 45 18. Transmission electron micrograph of migralory cell in transition region 46 19. Transmission electron micrograph of lobular bronchus 46 20. Transmission electron micrograph of cuboidal secretory microvillous cells 47 2I High power transmission electron micrograph of cuboidal secretory microvillous cells 47 22. High power transmission electron micrograph of cuboidal secretory micovillous cells 4B 2t. Scanning electron micrograph of lobular bronchus 4B 24. Transmission electron micrograph of lobular bronchus showing its relationshiP to an alveolus 49 49 25. Transmission electron micrograph of lobular bronchus 26. Transmission electron micrograph of columnar secretory microvillous cells in lobular bronchus 50 VI
Page
?7. High power transmission electron micrograph of columnar secretory microvillous cells 50 2B High power transmission electron microqraph of apex of columnar secretory microvillous cell 5L 29. Transmission electron micrograph of lobular bronchus 51 30. Transmission electron microqraph of non-secretory microvillous cell ,2 3r. High power transmission electron micrograph of non-secretory microvillous cell 52. 32. Trançmission electron micrograph of non-microvillous cells 5t 33. Transmission electron micrograph of non-microvillous cells 5t 34. Scanning electron micrograph of cuboidal ciliated cell 54 35. Transmission electron micrograph of distal region of lobular bronchus 54 16. scanning electron micrograph of junction between tertiary and lobular bronchi 55 37. Light'micrograph of two adjacent acini 55 38. Light micrograph of terminal lobular bronchus 56 39. Light micrograph of lobular bronchus 56 40. Low power transmission electron micrograph of distal region lobular bronchus 57 4r. Transmission electron micrograph of submucosal mast cell 57 42. Diagram of parenchymal subdivisions 62 43. scanning electron micrograph of tung parenchyma with lobular bronchus 63 44. Scanning electron micrograph of lung parenchyma 63 45. Transmission electron micrograph of thick septum in parenchyma 64 46. Transmission electron micrograph of thick connective tissue septum 64 47. Transmission electron micrograph of thick septum containing a pulmonary venule 65 48. Scanning electron micrograph of alveoli 65 49. scanning electron micrograph of termination of lobular bronchus 66 50. Scanning electrôn micrograph of alveolar duct 66 51. scanning electron micrograph of alveolar duct and sac walls 67 52. Transmission electron micrograph of alveolar duct wall 67 53. Transmission electron micrograph of alveolar duct wall 6B 54. Transmission electron micrograph of alveolar sac wall 6B 55. Transmission electron micrograph of part of alveolar sac wall 69 vil
Page
56. Transmission electron micrograph of alveoli showing alveolar 69 duct and sac walls 70 57. Transmission electron microqraph of alveolar septum 58. Higher power transmission electron micrograph of alveolar 70 septu m electron micrograph of alveolar type I epithelial 59. Transmission 7I cell 60. Transmission electron micrograph of alveolar type II epithelial 7L cell Scanningelectronmicrographofalveolarsurfaceshowingtypell 6r. 72 ce lls type 62. High power scanning electron micrograph of alveolar 72 II cells 13 63. Transmission electron micrograph of alveolar type II cells 64. Transmission electron micrograph of alveolar septum containing single capilìarY network 73 double 65. Transmission electron micrograph of alveolar septum rvith capillarY network 74 66. Transmissionelectronmicrographofalveolarseptumwithsingle 14 and double caPillarY networks 74 61. Transmission electron micrograph of alveolar septum 68. Transmissionelectronmicrographofalveolarseptumbetween adjacent alveolar sacs 15 69. Transmissionelectronmicrographofjunctionbetweenalveolar 15 septa from three alveoli 70. TransmissionelecLronmicrographofalveolarseptumwithtypel epithelial cell 75 electron micrograph of alveolar septum with 1L Transmission 15 "pseudo-single" capillary network 72. Transmissionelectronmicrographofalveolarseptumwithdouble capillary network 75
7t. Transmissionelectronmicrographofalveolarseptumwithdouble 75 capillarY network electron micrograph of alveolar septum with 74. Transmission 75 transeptal anastomosis 75. Transmissionelectronmicrographofalveolarseptumwithmixed capillarY networks 75 16. Transmission electron micrograph of double capillary alveolar 76 septu m
77. Highpoweltransmissionelectronmicrographofalveolarseptal 76 interstitiu m 17 78. Transmission electron microqraph of alveolar macrophage 71 19. Scanning electron micrograph of alveolar macrophage vlll
Page
100 80. Diagram outlining four-level morphometric sampling design BI. Logarithmic scale used to determine harmonic mean barrier 100 th ickness r20 82. Allometric Plot of lunq volume 83. Allometric plot of alveolar surface area l_20 I2L 84. Allometric plot of capillary surface area T2I 85. Allometric plot of capillary volumer 86. Allometric plot of tissue harmonic mean thickness and I22 cap.illary load ing r22 81. Allometric plot of oxyqen diffusing capacity showing progressive means of perce-ntage of interacinar BB. Graph r26 septal volume occupied by respiratory surf ace 89. Graph showing progressive means of perce-ntage of interlobular septal volume occupied by respiri tory surf ace r27 interlobular 90 Graph showing the contribution of the interacinar and r"pi" to the total respiratory surface volume r29 respiratory 91. Graph showing progressive means of total proportion of rutf""" volume within the true parenchyma I29 92. Graphshowingprogressivemeansforcombinedvolumeporportion of single and ãouble capillary alveolar septa within the true l_10 parenchym a percentage of airspace 93. Graph showing progressive means of the tll volume within true ParenchYma of pulmonary parasites in 94 Diagram showing the distribution I6I Neophoca cinerea 16r 95. Macroscopic view of rostral portion of maxillary turbinates 16r 96. Macroscopic view of ethmoid turbinates Scanning electron micrograph of trachea showing larval 97. 162 O. diminuata view of lung parenchyma with pathologic subpleural 9B Macroscopic r62 nodule oo Macroscopic view of lung parenchyma showing adult r62 P araf ilaro ides I62 100. Scanning electron micrograPh o f adult Parafilaroides r63 10I. Macroscopic view of larval mites on maxillary turbinate r63 102. Light micrograph of larval mite embedded in nasal turbinate r63 l0l. Light micrograph of embedded larva r63 104. High power light micrograph of embedded larval mite Scanning electron micrograph of larval mite on tracheal 105. 164 surface IX
Page r06. Scanning electron micrograph of larval mite on tracheal surface r64 107. Light micrograph of larval mite embedded in epithelium of nasopharynx r64 L0B. High power liqht micrograph of larval mite embedded in epithelium of nasoPharYnx r64 109. Liqht micrograph of larviparous female Parafilaroides Iying in an alveolar duct I r65 LIO. Light micrograph of larval Parafilaroides lying free within an alvêolar sac r65 I1l. Liqht micrograph of adult male Parafilaroides lying free within an alveolar duct r65 LIz. High power light micrograPh of adult Parafilaroides r65 rll. High power light micrograph of lobular bronchus showing lymphocytic interstitial infiltrate L66 114. Light micrograph of tertiary bronchus showing fibrinous exudate withi'n the lumen 166 ]]_5. High power light micrograph of tertiary bronchus showing cellular nature of bronchitic exudate 166 116. Light micrograph from bronchopneumonic region r66 lr7. High power light micrograph from bronchopneumonic region L67 ItB. Light micrograph of pneumonic reaction sumounding two larval r67 P araf ilaroides 119. Light micrograph of pneumonic lesion r67 r20. Low power light micrograph of subpleural paren"nyt" r67 I2I. Light micrograph of subpleural lesion 168 r22. Light micrograph of pneumonic region showing adult 168 P arafilaroides r73. High power light micrograph of pneumonic exudate t6B r24. High power light micrograph of larval Parafilaroides migrating through perivascular tissue r6B x
LIST OF TABLES
Page
Table
20 l_. Biological data of animals used in this study 2. List of morphometric sYmbols 97 3. Morphometric results - levels I and II IT2 4. Morphometric results - level III II2 5. Morphometric results - level IV 114 6. Morphometric results - absolute values 115 7. Morphomelric results - barrier thicknesses t15 l_16 B. Level I - analysis of variance 9. Level II - analysis of variance tl6 10. Level III - analysis of variance Ir_6 11. Level IV - analYsis of variance TL7 r2. t-test for differences between dorsal and ventral regions ltB 13. t-test for differences between anterior and posterior regions ltB 14. t-test for differences between predicted and measured absolute values It9 l-5. Results for grid systems trials - level I 124 16. Results for grid systems trials - level II r25 r7. Results of comparison between SQ and STWQ subsampling - level IV L32 18. Results of comparison between lI and 17 class lògarithmic scales for harmonic mean barrier thickness 133 19. Comparison between SQ and STWG subsampling - whole lung densities 134 20 species comparison of parenchyma, airspace and tissue volumes 139 expressed as Percentages 2r. species comparison of absolute and specific morphometric parameters 145 22. Species comparison of alveolar diameter index, capillary diameter and haematocrit r47 23. Effective capillary surface area and corrected oxygen diffusing capacity for the Australian sea lion r49 24. oxyqen conductance values for individual components of the morpho- *"iii" model for estimating the oxygen diffusion capacity 152 25. Recorded incidence o pulmonary parasites in otariid species 156 XI
ABSTRACT
to The lungs of the Australian sea lion, Neophoca cinerea, wele examined determine the arrangement of ultrastructural features of the terminal airways and whether gas-exchange region. Morphometric studies were conducted to assess the observed structural modifications affected the oxygen diffusing capacity of lung.
The nìajor structural differences between the Australian sea lion lung and those of terrestrial mammals are the generalized reinforcement of both the and inter- airways, by cartilage, and the parenchyma, by pronounced interlobular acinar septa and the general thickening of the alveolar septa' of the thickened alveolar septa a significant proportion carried a double capillary network whilst the majority possessed a single capillary network similar to that of terrestrial
mammals.
Morphometrically the parenchyma was found to occupy little more than 7070 of the lung volume, of which only 14% was respiratory tissue. Approximately 27o/o of the alveolar septa contained a double capillary network' The interstitial com- grealer ponent of the alveolar septa showed significant regional differences being ventrally and posteriorly, whilst the alveolar surface area was significantly greater ventrally. The mean values of adult of this species for various morpho- metric parameters wele: lung volum e, 6.95L; alveolar surface area, 183m2; total capillary surface area, 2O9m2; capitlary volumer 3g3 ml; tissue harmonic mean thickness, 0.59 ymi and maximal oxygen diffusing capacity, 4'582 ml of O2.sec-l.mbar-Ì. When compared to predicted values for terrestrial mammals similar body weight, lhe total capillary surface area was 2370 greater, and the capillary loading was B0% greater; both these results were statistically signi- ficant. of the other parameters, the lung volume was 50% greater than predicted, gteatet, the capiliary volum e 6o0/o greater, the oxygen dif fusing capacity 260Á whilst the alveolar surface area was ll% less than predicted; however, none of xil
these results were statistically significant. Of the double capillary septa, and the interlobular and interacinar septa, approximately only one-half of the capillary surface areas of these structures are effectively involved in gas exchange. When
these are taken into consideration, the effective capillary surface area was f47o
Iess than predicted. Inclusion of this effetive value into the calculation of the maximal oxygen diffusing capacity resulted in a value of 4-287 ml O2.sec-l.mbar-l which was about l-B% g.""t", than that predicted for a ter- restrial mammal of similar body weight. From the regions of pathology two halarachnid mites, Orthohalarachne
diminuala and O. attenuata. and a nematode lu ngworm of the g enus Parafilaroides were identified. The mites were found attached to the mucosa of the nasal tur- binates, nasopharynx and trachea where they caused ulceration of the epithelium with a localized infiltration of mononuclear cells and eosinophils. Within the lung
parenchyma the most prominent pathology was associated with free worms, both
adult and larval, which caused an acute bronchitis and bronchopneumonia which in some areas developed into a purulent pneumonia with destruction of alveolar
walls. The reinforced airways are important for enabling the high expiratory flow rate achieved during normal respiration, and together with the increased con- nective tissue of the septa for assisting the near complete alveolar collapse at the
end of expiration and when diving to depth. The parenchymal reinforcement will also assist the reinflation of the alveoli during inspiration, as will the acinar geometry by increasing the bulk flow of inspired air into the gas exchange region.
The large tidal volume will result in greater peripheral bulk flow of the inspired
air, and so reduce the distance for passive diffusion of oxygen in the aìveolar air'
The high oxygen utilization which results is due, in part, to this shorter diffusion path and the inspiratory pause of normal respiration. These will enable more efficient use of the increased maximal oxygen diffusing capacity as determined morphometrically. The presence of mites in the nasal turbinates may limiL the gas xilt flow rate during respiration, whilst the parenchymal pathology may reduce the alveolar surface area available for gas-exchange. The role of these structural and morphometric features are discussed in relation to the diving ability of these animals and their ability to avoid decom- pression sickness when diving to depth. xrv
DECLARATION
This thesis contains no material which h"s'be"n accepted for the award of any other aeqreê or diploma in any university. To the best of my knowledge and belief, this thesis contains no material previously published or written by another person' except when due reference is made in the text of the thesis. I consent to this thesis being made available for photoeopying and loan.
Anthony Nicholson XV
ACKNOWLEDGEMENTS
The research described in this thesis was performed in the Department of pathology at the University of Adelaide. It was supported by a University
Research Grant Postgraduate Scholarship awarded by the University of Adelaide.
Support was also provided by a University of Adelaide Equipment and Maintenance Committee Special Research Grant for 1979, and the Australian Marine Sciences
and Techrlology Advisory Committee for l980-l-981.
I wish to first thank my supervisor, Dr Joseph C. Fanning, for his assistance
in alt aspects of this study and with other problems that inevitably appear during
phD studies. .Professor Barrie Vernon-Roberts is also gratefully acknowledged for
his timely support and advice. Of the many people who offered advice and assistance, there are those for
whom special mention needs be made:
Miss Marjorie Quin, Department of Pathology, for her help and expertise in
preparing the histological material; Dr Oliver Mayo, Biochemistry Section, Waite Agricultural Research Insti-
tute, and Dr Lindsay C. Richards, Department of Restorative Dentistry, for their
advice and help with statistical analysis of the data;
Dr Brendon Griffin, Electron Optical Centre, for his advice and support in
general, but in particular for that in regard to the morphometry; Mr Manohar Hullan, Medical Artist, Institute of Medical and Veterinary
Science, for his artistic skill and understanding; Mr Dale Caville, Department of Pathology, for his professional help throughout this project, and in particular for his enthusiasm and skills in the production of the prinLs for this thesis; Mrs Dorothy Wagstaff , Department of Pathology, f or her support and
secretarial assistance; XVI
Mrs Carmen Carter, for her typing skills during the embryonic stages of this thesis, and in particular, Ms Christine Gradolf, for her excellent typing and advice in the final production of this thesis; Finally, my family and many friends, for their continued support and
encouraqement throughout this project, to all of whom I am.deeply grateful. Chapter I
INTRODUCTION
The efficiency with which marine mammals have adapted to their aquatic environment has intrigued and fascinated humans for many years' Perhaps that feature of their existence which has aroused most interest is their ability to
sustain long periods of submersion. This marked breathhold diving ability enables for some aquatip species to reach great ocean depths and remain under water
extended periods of time.
I.I CLASSIFICATION OF MARINE MAMMALS There âre three exlant groups of marine mammals; the cetacea which
includes the totally aquatic whales and dolphins, the Pinnipedia which comprises
the seals, sea Iions and walrus which are amphibious, and the lesser known totally (Scheffer,1976)' The aquatic Sirenia, represented by the manatee and the dugong
pinnipeds are classed into two superfamilies which are most readily differentiated by external characteristics, in particular their hindlimbs and the presence or hind- absence of external pinnae. The Phocidae, or true seals, cannot flex their
limbs for terrestrial locomotion and have no external pinnae, whilst the Otarioidea have can use their hindlimbs for terrestrial locomotion and, except for the walrus,
small external pinnae. The Otarioidea are further subdivided into two families, the Otariidae which includes the sea lions and fur seals, and the Odobenidae
represented by the walrus (Repenning et al', 1979)'
I.2 DISTRIBUTION OF AUSTRALIAN PINNIPEDS All of the pinniped species resident in and around Australian waters belonq to the family otariidae. These are the Australian fur seal, Arctocephalus Pusillus Australian dor if erus the New Zealand fur seal, Arctoc eohalus forsteri. and the the sea Iion, Neophoca cinerea (Marlow and King, I974). The distribution of 2.
Australian seal extends south from the mid-north coasl of New South Wales to southern Tasmania and west along the Victorian coast, whilst in Australian waters the New Zealand fur seal is present along the coasts of Western Australia and south Australia (Bonner, l981). The present range of the Australian sea.lion is the Abrolhos islands and coasts of southern and southwestern Australia, from Houtman (2BoS, tI4oE) in Western Australia to a rocky outcrop, The Pages (35o47's,
llBoIT'E) in South Australia (Walker and Lin|, lgBI)' Althoùùh the Australian sea lion has a wide distribution, the estimated total
population is only 4r000-5r000, of which 60-70% occur in South Australian waters
(Ling and walker, I977). Due to this small population Neophoca is considered to be a threatened but not endanqered species whose numbers have been relatively stable since,the cessation of mass slaughtering by commercial sealers late last population century (Stirling, I97Ð. As a consequence of the relatively small total of this species, only a limited number of animals were made available for this study from which other maLerial has been utilized and incorporated into a multi- disciplinary study of this species. Other aspects of this overall project have involved investigations into the population status and reproductive biology' and
the levels and effects of industrial and agricultural pottutants.
L' DIVING CAPABILITY Among marine mammals the larger cetaceans are regarded as the most from capable divers. However, much of the data relating to these species is indirect evidence. The maximum depth attributed to a cetacean is 1,L00 m, the depth at which a sperm whale, Physeter catodon, was reported to have become reported duration entangled in submarine cables (Heezen, Ig57), whilst the longest is I20 min., attributed to a harpooned bottle-nose whale, Hvoero odon amDullatus (lrving, Ig3Ð. Many of the earliest dive duration records for pinnipeds were from forced dives by restrained animals (scholandet, 1940). However, more recently telemetric devices have been attached to freely-diving animals in the wild with 3
the greatest performances achieved by the Weddell seal, Leptonychotes weddelli. The greatest recorded depth was 600 m (Kooyman, 1966) whilst the greatest duration was 7J min. (Kooyman et al', 1980)' The diving capability of otariids is generally considered to be less than that of phocids, but similar to that for most smaller cetaceans (Ridgway, I976). From limited studies of voluntary diving otariids the maximum depths recorded are 2O4 m for Lhe Afro-Australian fur seal,
Arctoce alus usillus (su bspecies not specified) (Kooyman, 1982) and 207 m for the Norther'n fur seal, Callorhinus ursinus (Gentry et al., lg8l). The maximum duration. recorded for these two fur seals is 6.4min. for A. pusillus and 7.6 min. forc.ursinus(KooymanrJ912;Gentryetal"rgïÐ'Theonlyotherdirectly
measured records for a freely-diving obariid were those obtained from a trained
California sea lion, Zalop hus californianus which reached a depth of 25O m and
remained submerged for B min. (Ridgway, I972i 1976). Little is known of the diving ability of the Australian sea lion in regard to either depth or duration, although indirect evidence indicates that pups are capable of dives to f6 m
(Walker and Ling, 19Bl) and there is no evidence to suggest that they differ in
their divinq ability from other otariids.
I.4 DIVING PHYSIOLOGY The earliest attempts to elucidate those aspects which contribute to the
enhanced capabilities of diving animals were those reported by Paul Bert in l870
and Charles Richet in the late lB90's (Andersen,1966).5ince that time there have
been several periods of intensive investigation of diving species. The first of these was during the mid i.9f0's and 1940's from which came several hallmark
monographs (Irving, 1939; Scholander, J-940). The work of this period was followed
some twenty years later by a number of investigators, much of whose work is still
in progress today utilizing more sophisticated technology and methods to gain new
insights into diving physioìogy and behaviour. As with the earlier studies, much of the current work being undertaken is conducted on pinnipeds, and even then is 4
generally timited to a small number of species. The reason for this is largely due to pinnipeds being easier to maintain and manage in captivity. Some pinniped
species have even proven to be tractable and suitable experimental animaìs in the wild. Limited studies have been conducted on captive cetaceans. However, their very large size and pelaqic behaviour makes study of them in the wild a mammoth operation or mole usually impossible. For this reason much of the following
discussion on diving physioloqy is restricted to pinnipeds. The experimental results
are often inconclusive, because of the array of experiments performed on a wide variety of species which can make interspecies extrapolation uncertain'
The general conclusion amongst physiologists, anatomists and biochemists is that the diving ability of marine mammals is not due to the development of any which certain bio- new 'rapparatusrr but rather is the result of the degree to chemical and physiological features common to all mammals are utilized. These features include a respiratory and cardiovascular diving response, the availability
and utilization of the body's oxygen stores, and the ability to minimise the ill- effects of the metabolites of anaerobic pathways utilized during maximum duration diving. I.4.I Cardiovascular SYstem In l870 Bert demonstrated a profound bradycardia in ducks in response to
forced submersion. Some thirty years later Richet demonstrated that this response
was neurally mediated by administering atropine to ducks prior to submergence
and noting that the bradycardia was blocked. Working with restrained seals, Irving level of et al. Qg35) and Scholander (1940) produced a similar bradycardia to a onty l0-I5yo of the pre-dive heart rate. They also observed that in these animals the resting respiratory cycle often includes a prolonged inspiratory pause during
which the heart rate declines, but to a lesser extent than that during submersion'
since then a number of groups have shown that the reduction in the heart rate of
pinnipeds during forced submersion is gneater than that during voluntary diving
(Elsner, 1965; Harrison et al., I972; Kooyman and campbell, J-972). Kooyman and 5
the decline in Campbe¡ e97Z) demonstrated in free-swimming Weddell seals thal the dive heart rate is correlated with the duration of the dive so that the longer 20 min' the more profound the bradycardia up till about 20 min. duration. Beyond that for the heart rate remained at 16 beats per minute which was about 50% of
5 min. dives.
In the California sea lion the stroke volume was maintained during the diving in bradycardia so that the cardiac output varied in accordance to the changes heartrate(Elsneretat.,1964).Morerecently,studiesofsimulateddivesbythe by about harbor seal, Phoca vitulina indicated that the stroke volume decreases )0% together with an B0% decrease in cardiac output (sinnett et al', l-978)' A similar decrease in stroke volume was calculated by Blix et al', 1983' The the earlier observed.differences in the stroke volurne between the harbor seal and from study on the california sea lion have been suggested to possibly result brady- differences in experimental technique, differences in the degree of diving between the cardia which was more intense in the harbor seal, or the differences regard to the species studied (sinnett et al., 1978). Despite these differences in of stroke volume, the decrease in cardiac output has been observed in a number I979; BIix eb other studies (Elsner et al., 1964; Murdaugh et al., I96Bi Zapol et al.,
al., 19B3). Atthough the heart rate and cardiac output decrease durinq diving' maintained or measurements of the mean arterial blood pressure show that it is (Irving et al'' even slightly increased due to extenstve peripheral vasoconstriction Sinnett et al', 1942; Van Citters et al., 1965; Elsner et al., 1966; Stevens, I97I; visualized in I97B; Zapol et al., Ig7Ð. The extent of this vasoconstriction has been the harbor seal by the use of arteriograms (Bron et al.' 1966)' In the non-diving was seen seal injected contrast medium was rapidly cleared from the aorta and very little entering the peripheral organs, whilst during simulated diving there was the aorta' evidence of peripheral organ perfusion with delayed clearance of 6
Angiograms of the cerebral circuìalion during diving and non-diving showed no differences, indicating that perfusion to the brain is maintained. More recently the effects of peripheral vasoconstriction on blood flow distribution during simulated diving have been studied in the weddell seal (Zapol et al'' 1979)' the qrypus (Blix spotted seal' Phoca vitulina h and the grey seal, Halichoerus et al. tgBl), by the injection of radioactive microspheres into the left ventricle. The
results of these two studies are in agreement for the most part, with the notable exception of the pulmonary blood flow. In both studies the cardiac output
decreased by B5-9Oo/o as did blood flow to the left and right ventricles' The the heart decrease in coronary flow is associated with the decreased work load of
during bradycardia. Cerebral blood flow was constant in lhe Weddell seal whilst in the spotted and grey seals there was an initial marked decline in cerebral blood flow which then gradually increased throughout the l0 min. dive, although the lower parts of the brain were less perfused than the higher structures. Dormer et al. (1977), using a Doppler blood flow transducer, observed a similar initial
decrease in cerebral blood flow followed by a marked increase associated with
voluntary diving in the california sea lion. The blood flow to the adrenal glands, in both microsphere studies, decreased by onty 25-4To/o from pre-dive values' In the in the weddell seal nearly 3OVo of the injected microsphere dose was detected the lung, whilst in the spotted and grey seals the pulmonary f low was only 60/o of
cardiac output. These differences in the amount of pulmonary flow may have been of a resuìt of differences in the degree of flow through the two pulmonary sources blood, these being the bronchial artery and the recirculation of blood through
peripheral arteriovenous shunts (Zapol et al., I979; BIix et al., I9B3). In both these was studies the blood f low to all other organs and the skeletal m usculature
reduced by at least 90% from that at rest. This selective distribution of blood due to the peripheral vasoconstriction therefore maintains an adequate blood supply'
and hence oxygen supply, to those tissues most susceptible to hypoxia as originally
postulated by Irving (1939). 1
have also The mechanisms which control this cardiovascular diving resPonse into the possible been subject to intensive investigation with the first insight (Andersen, 1966)' He mechanisms arising from the work of Richet in the IB90's blocked the found that administration of atropine to ducks prior to submersion system in the bradycardia and so illustrated the involvement of the nervous since been identi- control of the diving lesponse. other control mechanisms have the light of species fied, although the role each plays is uncertain, particularly in response differences. from increased The bradycardia of diving is considered to result primarily (Butler, I9BZ)' The parasympathetic activity mediated via the vagus nerve is considered by initiation of increased vagal tone during the apnoea of diving mediated through branches some to result from the stimulation of facial receptors is evidence to of the trigeminal nerve (Dykes, L974; Daly, l981). Howevet, there part, under hiqher support the suggestion that the bradycardia is, at Ieast in seals frequently central nervous conlrol. During bouts of voluntary diving harbor an increase in heart rate displayed a pre-dive anticipatory bradycardia, and oflen the dive (Jones et al" as the animals approached the surface toward the end of Paqophilus I97Ð. Similar observations have been made in freely-diving harp seals, (now groenlandica, Ronald qroenlandicus (Casson and Ronald, Ig75) called Phoca toward the end of diving has and Healey, 1981). A similar anticipatory tachycardia (stevens, which have also been also been observed in california sea lions I97l) prolonged apnoea trained to decrease their heart rate on land in the absence of increase in (Ridgway et al., Ig75). The Weddell seal shows a similar anticipatory its heart rate in heart rate toward the end of voluntary dives and also decreases longer dives there is a accordance with the duration of the ensuing dive so that for
more pronounced bradycardia (Kooyman and Campbell' L972)' efferents actinq The peripheral vasoconstriction is mediated by sympathetic been suggested as via a-adrenoceptors (Butler, I1BZ), the stimulation of which has (Irving 194I, 1942; Murdaugh being independent of changes in the heart rate et al', B
(Elsner et al', 1966; Blix' et al., 1968) and that it is under separate neural control by circulating cate- Ig75). This vasoconstriction is also likely to be influenced due to the minimal cholamines (Butler, Lg82) which would appear possible diving (Zapol et al'' reduction in perfusion of the adrenal glands during simulated
I979; BIix et al., I9B3). Theonsetofapnoeahasalsobeendemonstratedasimportantforthe initiation of diving, bradycardia, both in conjunction with other control marked respiratory mechanisms (Butler, l-982) and alone as evidenced by the mammals (lrving et al.' associated cardiac arrhythmia displayed by many marind et al', 1975; Casson 1963; Kooyman and campbell, 1977; Lin et al., I977i Tanjii l9B0)' Peripheral chemo- and Ronald, 1975; Dormer et al., I977i Pasche and Krog, considered to be receptors responding to changes in arterial blood gas tensions ale (Tanjii eL al., I975; Daly' important in the maintenance'of the diving bradycardia in lgBI), rather than in its onset due to the immediacy of the initial decrease
heart rate.
L.4.? Oxygen Stores
(a) Blood capacity was The potential of diving vertebrates for an increased oxygen he concluded first reported by Bert in lB70 when, after bleeding ducks and hens, per kilogram as did that ducks possessed approximately twice'the volume of blood has also been hens (Andersen, 196::6). The blood volume of diving mammals the weddell seal measured and found to be higher than terrestrial mammals' In been estimated to be and Dall porPoiset Phocoenoides dalli the blood volume has L969) whilst in the equivalent to about I4o/o of total body weight (Lenfant et al', to be at Northern elePhant seal, Mirou a ust irostr is it has been estimated volume tends to least 20% of total body weight (simpson et al., 1970). The blood (Lenfant et al'' be greater in phocids than otariids on a total body weight basis active lean 1970) and represents an even greater proportion of the metabolically total body weight body weight since the blubber fat may account f or up to 3\o/o of Jones, I9B2)' in phocids (Lenfant et al., 1969; Ferren and Elsner, I979; Butler and 9
Apart from the blood volume there are a number of haematological para- meters which are different in marine mammals. In particular. the haematocrit value and haemoglobin concentration are generally much higher than in terrestrial mammals. The haematocrit in marine mammals ranges from about 4oo/o to a little more than 6070 (Lenf ant, 1969; Ridgway et al., I97O; Simpson et al., I97O; Geraci and smith, 1975) as compared to values of 32o/o and 4lo/o in terrestrial mammals
(Lapennas and Reeves, 1982). The mean corpuscular volume of marine mammals is
generally greater than 100 ymJ whereas for most terreslrial mammals it is less volume than 100 ¡rml (Riagway, I972). However, this increased mean corpuscular only results in a slightly larger erythrocyte diameter (Lenfantr 1969). The total haemoglobin concentration in marine mammals ranges from L4 to 26 qm values Hb/l_00 ml blood (Lenfant, 1969; Clausen and Ersland,1969) with the higher generally occurring in phocids, whilst the lowest values are similar to those of con- humans, l-4 to J-5 gm Hb/100 ml blood (Clausen and Ersland, 1968)' As a which sequence of the higher haemoglobin concentrations the volume of oxygen
combines with the blood is increased resulting in an increased oxygen capacity
ranging from lB to 36 mI 02/100 ml blood (Lenfant, 1969; Clausen and Ersland, follows 196Ð. The occurrence of these oxygen capacities within different species a similar distribution to that of the haemoglobin concentrations so that the
highest capacities occur in phocids with the Iower limit of the range again being similar to that found in humans (Clausen and Ersland, 1968). These differences in
the oxygen capacities of the blood of these animals show a trend whereby those
species which normally endure long dives have the higher blood oxygen capacities (Lenfant et al., 1970). In studying the haematology of three genera of porpoises,
Ridgway and Johnston (1966) found that the blood volume, haematocrit, haemo- globin concentration and hence total blood oxygen capacity were all greater in the faster swimming, deeper diving pelagic species than the coastal dwelling species'
Whether or not such a direct correlation is present amongst all marine mammals' the increased blood oxyqen store, due to an increased blood volume and an 10.
diving increased blood oxygen capacity, is an important source of oxygen during for many specles. other properties of marine m ammal blood are a sigmoidal oxygen- an dissociation curve similar to that in terrestrial mammals but with a higher P5gr (Lenfant, 1969; increased Bohr effect, and an increase in the buffering capacity clausen and Ersland, I969i Lenfant et al., I97O; Dhindsa et al., L974; Lapennas the and Reeves, 1982). The lower affinity of the blood for oxygen, together with
increased Bohr effect, will tend to retard the onset of severe hypoxia during (Lenfant, diving by enhancing the unloading of blood-bound oxygen to the tissues a of 1969; Lutz, IgBÐ. The increased buffering capacity in these animals is result to an increase in the Haldane effect, which reflects the ability of reduced blood carry carbon dioxide. The advantage of an increased buffering capacity is that it with minimises the effects of shifts in the blood pH (Lutz, r9BÐ which, together the other respiratory properties of the blood, led Lenfant et al' (1970) to conclude thal the blood of marine mammals is apparently betler adapted for ensuring a
rapid recovery after diving rather than for sustaining prolonged dives.
(b) Myoglobin The myoglobin present in the skeletal musculature is not only an oxygen depot but when in sufficient concentration is also important in facilitatinq the diffusion of oxygen from the capillaries to the mitochondria (weber et al', 1981)' in a The concentration of myoglobin present in skeletal muscle has been measured variety of marine mammal species and was generally found to be increased above the levels present in terrestrial species (Man'kovskaya, 1975; Blix, 1976; Glazova' myoglobin L977; Castellini, lgBt; castellini and somero, 198I). This increase in
represents another important source of oxygen to marine mammals when divinq'
(c) Lung greater Many marine mammals, except perhaps for the large whales, have a lung volume that terrestrial mammals of a similar size (Goudappel and Slijper' l95B; Kooyman, L97t). The tidal volume of many marine species is B0% or more of the total lung capaeity (lrving et al., I94Ii Olsen et al., 1969)' In general phocids 11.
exhale either before a dive or during'' the early descent phase so that the actual lung volume only represents ZO-6t% of the total Iung capacity (Scholander, I940;
Kooyman et al., I97Ii I972). Otariids and cetaceans are generally regarded as diving on inspiration so that the lungs represent a relatively large oxygen store during diving (Kooyman and Andersen, 1969; Ridgway et al., 1969; Kooyman'
I973). Atthough there is some evidence to suggest that, for the california sea lion at least, dives are undertaken on only partial inspiration (Ridgway, I977) or that theyexhalesomegasonsubmersion(Keremetal.,I975)whichresultsinabreath- hold volume of 50-90% of their vital capacity (Matthews, 1977)' Thus there appear to be differences between the "slow" phocid breathers and the "rapid" breathers,
the otariids and cetaceans, in regard to their reliance on lung oxygen stores during diving which tends to be reflected in the differences in their haematological properties (section 1.4.2(a)). Another feature of diving which tends to minimise the contribution of lung oxygen stores is the alveolar collapse which occuts when
diving to depth, as first predicted by Scholander (1940) and since demonstrated in several species (Ridgway et al., 1969; Kooyman et al., I97O; I97Ð. The significance of this movement of air from the alveoli into the non-respiratory airways will be further considered in regard to the effects of pressure whilst
diving to depth in ÏSection I.4.5 .
1.4.f Biochemical AdaPtations Recent studies indicate that the majority of voluntary dives by free-diving
animals are well within the maximum limits observed in restrained and voluntarily
diving animals (Kooyman et al., I976; Kooyman et al., I9B0; Gentry et al', 19Bl)' were Kooyman 4_4. (1980) found that 9770 of voluntary dives by weddell seals within a limit of about 26 min., and that this was about the length of time beyond which the lactic acid concentration in blood increased during the recovery; the in Lhe increase in lactic acid is an indicator of anaerobic metabolism, particularly skeletal musculature. They therefore concluded that for the weddell seal the
aerobic dive limit is in the order of 2o-25 min. Based on estimated body oxygen 12.
stores during a breath-hold dive and the rate of whole body aerobic oxygen utiliza- tion, Hochachka (l98I) calculated a maximum aerobic dive limit of lB min., very similar to that observed by Kooyman et al. (1980). These results suggest that the vasoconstriction observed in forced dives is not nearly as important in the majority of voluntary dives as originally predicted. These results also stress the
importance of large oxygen stores in the body, particularly those observed in the blood and to a lesser extent those of the muscle. However, during maximum duration diving the cardiovascular responses are of great importance to maintain the an adequate oxygen supply to the obligate aerobic tissues of which the brain is
most important and sensitive. That there is an accumulation of anaerobic metabolic end-products during
sustained diving, which aPpear in the blood during the recovery period, was first reported by Scholander (1940). To account for this marked anaerobic capacity a to number of investigators have examined the tissues of various marine mammals
determine if there are any biochemical differences when compared to terrestrial
species. The two glycolytic enzymes most frequently studied have been lactate
dehydrogenase and pyruvate kinase; pyruvate kinase because it is a useful index of glycolytic activity (5imon gL-e!., I97Ð and its pivotal role in gtycolytic control
(storey and Hochachka, I974), and lactate dehydrogenase because it can function in the mofe usual pyruvate to lactate direction, and under certain conditions in the oxidation of lactate to pyruvate (Castellini et al', l9B1)' In a comprehensive
study of various tissues from a wide range of marine mammals and control ter- restrial mammals, the activibies of these two enzymes, assayed at physiologic temperatures, displayed no differences in the potential for anaerobic glycolysis
(castellini, 1981; Castellini et al., 1981). These authors concluded that the meta- bolic stress of an extended dive is within the range of stresses encountered by
exercising terrestrial mammals. The only consistently increased enzymic activity observed was that of lactate dehydrogenase, in both directions, in the marine diving mammals but this was not correlated with the known differences in the r3.
abilities of the species studied. Rather it rvas proposed that this increased activity reflected an adaptation for the removal of blood lactate during or after diving (Castellini et al., tg8L). One observed apparent adaPtation to the need for in the occasional prolonged anaerobic activ ity was the signif icant increase buffering capacity of skeletal muscle of marine mammals (Castellini and Somero, I981). This buffering capacity was found to be strongly eorrelated to the
myoglobin concentrations and to the muscle lactate dehydrogenase activities' of muscle consequently, when anaerobic glycolysis is activated, with the depletion pH oxygen stores, this buffering capacity is able to m inim ise the associated
changes which would otherwise occur due to the accumulation of lactate'
I.4.4 Putmonary Function
The extent to which the lung contributes to diving ability appears to differ in part' between two groups of divers. The division into these two groups is based, on their respiratory patterns. The phocids tend to have periods at the surface after a dive during which they ventilate a number of times, and have been called ilslow,,breathers, whilst otariids and cetaceans have been called "rapid" breathers respiratory because on surfacing from -a dive they will often only perfotm one in cycle before resubmerging (Kooyman, I973). tnis difference is also reflected dive on the volume of air retained in the lungs when diving. Phocids generally store expiration and otariids and cetaceans on inspiration, so that the lung oxygen (see 1.4.2(c))' All would appear to be more important in "rapid" breathers Section par- marine mammals have large tidal volumes which, in the "rapid" breathers ticularly, are associated with high flow rates' Studies with excised lungs have (VC) per second demonstrated a peak expiratory flow rate of eight vital capacities in the California sea lion (Kerem et al', 1975) and up to l0 VC/sec' for the lungs of the harbor porpoise, Phoceana phoceana (Kooyman and sinnett, L979)' In live than those from animals the peak expiratory flow rates recorded have been lower lion (Kerem excised lungs but still very high; up to 5 VC/sec. in the California sea porpoise, et al., L975; Matthews, Lg77) and 6.1- VC/sec. by the bottle-nosed 14.
TursioÞs truncatus (Kooyman and Cornell, l-9Bl). In the few larger cetacean species studied, the peak expiratory flows measured were only about 2'5 VC/sec' (spencer et al., 1967 Kooyman et al., 1975) which is still higher than the values attainable by humans, about 1.5 VC/sec. (Pride et al', 1967)' Another striking feature of expiration in marine mammals is that these high flow rates are maintained even at low lung volumes (Kerem et al., 1975; Matthews, 1977; the Kooyman and sinnett, 1979; Kooyman and cornell, l9B1) whereas in humans
peak expiratory flow is attained at 80% of the vital capacity, below which volume the flow rate decreases rapidly (Pride et al., 1967)' The conclusion reached by
these investigators was that the cartilage reinforcement of the terminal airways of otariids and cetaceans maintains the patency of these smallest airways during the periods of these high flows and at low volumes (Kerem et al., I975; Matthews' in contrast 1977i Kooyman and sinnett, 1979; Kooyman and cornell, I9BI). This is the to humans where the expiratory flow rates are limited by the collapse of relatively thin-walled bronchioles (Hyatt et al., I95B; Macklem and Mead, 1968)' result in The large tidal volumes and high flow rates exhibited by marine mammals able reduced residual volumes as evidenced by the degree to which the lungs are to empty (Denison et al., I97I; Matthews, I977i Kooyman and SinneLL, 1979)' the high These factors contribute to a high turnover of alveolar 9as which aids
oxygen utilization achieved by these animals (scholander, 1940). for The oxygen utilization by marine mammals tends to be greater than that te¡estrial species. In many marine species the oxygen utilization is generally B- ZLo/o is reduced 10%, so that the inspired oxygen concentration of approximately to ll-ll% in the expired air (Scholander, 1964; Andersen, 1966; Ridgway et al', 5-7% 1969; Shapunov, 1971). Although in phocids the oxygen utilization is only
(Andersen, 1966). The degree of oxygen utilization has been shown to depend on the dive' several factors (Kooyman et al., I97Ð; the lung volume at the start of the pre-dive ventilatory behaviour, and the depth of the dive which will be dis-
cussed in Section l-.4.5. r.5.
During breath-hold dives marine mammals are able to withstand lower arterial oxygen tensions and higher carbon dioxide tension than terrestrial
mammals (Scholander, 1940; Robin et al., I963i Ridgway et al', 1969; Lenfant et eL., I97O; Pasche, I976i Gallivan and Best, 1980; Gallivan, l9B0)- Despite an
increased tolerance to high carbon dioxide tensions marine mammals are not
insensitive to it (Pascher I976), in fact it has beerr suggested that carbon dioxide is the major factor in controlling ventilatory behaviour and diving duration in the
manatee (Gallivan, l9B0). In phocids alterations to the diving pattern appear to be an important adaptation to variations in oxygen and carbon dioxide con- centrations. In experimental dives with an increased inspiratory carbon dioxide concentration, the duration of each dive was reduced (Pasche, I976) and more time was spent at the surface (Craig and Pasche, 1980). Similar alterations were
observed in the diving patterns of harp seals in response to increased metabolic
oxygen requirements (Gallivan, l98la), whilst in the manatee metabolic adjust- (Gallivan ments after- a long dive are made during a series of short dives and Best,
l_980). The need to alter diving patterns rather than ventilatory patterns stems from the limitations of ventilatory behaviour. The tidal volume of many marine (scholander, mammals is very large, often reaching 80-90% of total lung capacity although 194O; Scholander and Irving r Ig4I Irving et al., J-941; Olsen et al., 1969), it tends to be somewhat less in the large whales (Laurie, 1933; Andersen, 1966;
Wahrenbrock et al., Ig74). As a consequence the tidal volume approaches the vital capacity and the end-expiratory volume is very small (lrving et al., l94I; Kerem et al., 1975; Matthews, 1977; Kooyman and SinnetL, 1979; Kooyman and cornell, lggl). Due to this large tidal volume, the minute ventilation is relatively high
even though the resting respiratory frequency is very low in most species and only slightly higher in phocids (Andersen,1966i Ridgway, I972)' In the Weddell seal the minute ventilation increases after diving with increasing dive time up until a dive of l5 min. when the minute ventilation remains constant (Kooyman et al., I97l)' r6.
The maximum increase in minute ventilation is achieved by doubling the tidal volume and increasing the respiratory rate to three times that of resting so that overall the minute ventilation is increased six times above that at rest- This increase in minute ventilation is quite small when compared Lo the twenty times increase attainable by exercising humans (Kooyman et al., l97l-). In the harp seal the minute ventilation and oxygen extraction were observed to be at or near their maximum so that any increase in one of these variables leads to a decrease in the other (Gallivan, 1981a). In cetaceans, with very large tidal volumes, increases in pulmonary ventilation are limited because any major increase in respiratory frequency results in a decrease in the tidat volume (Scholander and Irving, )'94Ii
Andersen, 1966). 1.4.5 Effects of Increased Hydrostatic Pressure
As a consequence of the large tidal volumes and high oxygen utilization, the alveolar oxygen and carbon dioxide tensions can vary quite markedly during the various phases of the respiratory cycle. These gas concentrations are subject to
the particular activ ity of the ani m als. Working with a trained bottle-nosed porpoise, Ridgway gl__d. Q96Ð observed that the oxygen concentration in the
expired air immediately after diving to depth, about 100 m' was greater than that after surface breath-holds of similar duration. Similar results were obtained by freely-diving Weddel seals in which the alveolar oxygen tension was greater after
a dive of 20 min., which are usually to a depth of 100-400 m, than during resting
apnoeic periods at the surface which averaged 4.5 min. (Kooyman, 1967; Kooyman et al., I97r. The potential for this respiratory gas behaviour was outlined by
Scholander (1940) who predicted that when diving to depth, due to the increase
hydrostatic pressure, the alveoli would collapse, before the airways so that the air
would be sequestered away from the gas exchange surfaces' This was suggested as
a mechanism to limit the degree of gas exchange at depth and so prevent the absorption of large quanities of nitrogen and hence minimise the potential for decompression sickness during ascent. Scholander (I940) predicted that this r7.
alveolar collapse would be complete at a depth of 100-200 m for cetaceans and
10-40 m for the grey seal. Based on alveolar oxygen concentrations after diving,
Ridgway gl_4. Q96Ð predicted that in the bottle-nosed porpoise alveolar collapse would be complete at about J-00 m, whilst in the weddell seal, Kooyman et al. (1971) predicted alveolar collapse to occur at a mean depth of 70 m. More recently, Ridgway and Howard (I979), by measuring the nitrogen tension in the dorsal epaxial muscles of bottle-nosed porpoises, estimated that the collapse of the alveoli is nearly complete by 70 m but emphasised that this exclusion of gas from the respiratory surface occurs gradually. They also remarked that despite peripheral vasoconstriction, the skeletal muscles still receive an effective blood (I980) supply during diving which agrees with the conclusion of Kooyman et al. and
Hochachka (L98l) that the majority of dives are aerobic so that the peripheral
blood supply wilt be maintained. However, in regard to shallow dives, Ridgway and
Howard (Ig79) concluded that the degree of alveolar collapse is insufficient to limit nitrogen absorption during dives to less than 70 m' scholander The ability f or alveolar collapse, as originally suggested by (1940), is dependent on the rigidity of the term inal airways. In otariids and
cetaceans these airways are reinforced by cartilage down to the level of the alveolar sacs, whilst in phocids the term inal portions of these airways a¡e supported by a thickened muscular layer (Denison and Kooyman, I97Ð' These
authors predicted that the airways need only be about five times more rigid than the alveolar walls to allow alveolar collapse to occur with increasing hydrostatic
pressure. They also noted that an important feature for alveolar collapse is that the ratio of the alveolar volume to the dead space volume needs to be small so that the gas from the alveoli wilt fitl the dead space. This movement of gas due to
increased pressure has been demonstrated radiographically in Weddell and Nor- thern elephant seals (Kooyman et al., 1970). At the maiimum pressure to which
these animals were subjected, lI ATA equivalent to a depth of 100 m, the alveoli were completely collapsed, the dorsoventral tracheal diameter was reduced more lB.
in size' There- than 5070, whilst the third to fifth order bronchi showed no change fore the lung volume estimated to be about 100 ml at ll ATA was contained within the conducting zone of the lung. These authors also observed that the major
reduction in alveolar volume occurred by 3 ATA so that, when diving, the amount of alveolar surface available must be greatly reduced before a depth of l0 m is
reached. Alveolar collapse is dependent not only on airway reinforcement but also a flexible thorax, Iarge distensible blood vessels within the thorax, large amounts of elastic tissue in the lungs, and a resilient trachea (Ridgway, 1972)' A flexible thorax, in the form of an increased number of floating ribs, is necessary to reduce
the resistance to the increased hydrostatic pressure. Distensible blood vessels such as the reLe mirabilia of cetaceans and the pericardial venous plexus of many to pinnipeds and some cetaceans (Harrison and Tomlinson, 1963) are thought as the air in engorge with blood at depth and so partially fitl the thoracic cavity the lungs the lungs is compressed (Ridgway,I972i Hui, I975). The elastic tissue of the allows them to become atelectic without separaLing from the chest wall, and resilient trachea allows the respiratory airways to collapse beyond the limits of
the sea-level dead-space volume (Ridgway, I972)'
I.5 SUMMARY An important aspect in regard to the diving ability of marine mammals is that the vast majority of dives undertaken are apparently within the aerobic dive Iimit for each particular species. To enable the aerobic processes to function an an enlarged adequate oxygen supply is essential, which is achieved principally by blood oxygen store as compared to that of terrestrial mammals' This oxygen muscula- source is supplemented by smaller amounts contained within the skeletal ture and to a varying deqree that within the lun9s. other physiological and utilization biochemical mechanisms involved in diving, function to enable efficient rapid and of this oxygen supply. The role of the lung in diving is to ensure a 19.
sufficient supply of oxygen for the blood and muscle stores, and to minimize or prevent the ill-effects associated with increased hydrostatic Plessure when diving to deþth.
The aims of this study were therefore: i) to examine and describe in detail the normal structure of the terminal airways and gas exchange regions of the lung of an otariid species, the Australian sea lion Neophoca cinerea, since many of the previous studies of
examples from this family have been cursory in reqard to these structures; ii) to quantify, using morphometric techniques, those structures involved in gas
exchange and to determine the oxygen diffusing capacity of the lung, as
many detailed morphometric studies have been conducted on the lungs of a
wide variety of terrestrial mammals, birds, reptiles and amphibians, but no report for a marine mammal species has yet appeared; iii) to examine the lungs for evidence of patholoqy and where present to
describe the associated changes, and to isolate and identify the aetiological
agents; and iv) to consider Lhe role of the normal respiratory structure and the morpho- the oxygen metric dif f using capacity in the availabiliLy and supply of required to perform aerobic dives and in the prevention of the problems pul- associated with increased hydrostatic pressure' and the influence of
monary pathology on normal gas exchange' 20.
ChaPter 2
MATERIALS AND METHODS
2.I SPECIMEN COLLECTION of these animals' an The lungs of five animals have been examined' Three (A52l were collected immature female (AS4) and two adult females and AS24)' on 26 March 1979 and from Dangerous Reef (l4o 4g'5, 136012'E)' South Australia two, both adult females (4525 28 and 29 October 1980 respectively. The remaining (55050'5' ll7oI5'E)' Kanqaroo Island' and A526), were collected from Seal Bay animals were obtained under South Australia, on IB March l9B1 (Tabte 1). All permit from the National Parks and witdlife service of south Australia'
Table I Biological data of animals used in this study'
Weight Curv ilinear Reproductive Animal Location Date (kq) lenqth (cm) Status
119.0 ApparentlY immature AS4 Dangerous Reef 26.3.79 þ5.5 Non-lactating AS23 Dangerous Reef 28.10.80 60.56 L57.O Pregnant: I mth foetus AS24 Dangerous Reef 29.10.80 65.32 158.5 Lactating with B week PuP AS25 Seal Bay 1B.l.Bl 87.O9 161.0 Lactating with 6 week PuP 4526 Seal BaY 18.1.81 95.O3
Eachanimalwascapturedandrestrainedinalargenet.Anaesthesiawas into the right brachio- induced bY intravenous injection of sodium thiopentone 20 ml (5 mg/ml) of cephalic vein using a l0 cm l5G needle. Approximately the animals this was supple- anaesthetic was required for each animal' In four of halothane and oxygen applied mented by gaseous anaesthesia using a mixture of A tracheostomy was through a face mask and a stephenrs anaesthetic apparatus' the gaseous anaesthesia' performed and a cuffed endobracheal tube inserted for of a lethal dose of anaes- The animals were then killed by intravenous injection
[hetic agent. 2T
2.2 LUNG FIXATION Fixation of the lungs in situ was achieved by instillation through the endo- tracheal tube, of Karnovsky's paraformaldehyde-glutaraldehyde fixative (Karnovsky, 1965) in either 0.05 M sodium cacodylate buffer at pH 7'3 and (by 400 m0sm (animal AS4), or phosphate buffer at pH 7'4 and lB0 m 0sm freezing point depression). A conslant head of pressure of l0 cm of water above the
sternum was maintained for one hour before the tube was clamped. The apex of the left ventricle was also cannulated, through an upper' midline abdominal incision and the same fixative perfused at a pressure of 120 mm of mercury' and main- Pressure in the cardiac cannula was measured by a sphygmomanometer tained by a foot pump connected to a 20 litre reservoir of fixative. Perfusion was
mainlained for one hour before the cannula was clamped' The nasal cavities of
animals AS25 and A526 were fixed by packing the nasopharynx with cotton wool' fillinq the cavities with l0% buffered formalin, and packing the external nares with cotton wool. The carcases were then placed in large polythene bags partially filled with fixative and transported to the laboratory for detailed examination.
2.] LUNG VOLUME DETERMINATION
The trachea, lungs and heart were lemoved in toto four days to three weeks afler arrival at the laboratory where the carcases had been stored in a cold room from the at 4oC. The heart and major arteries and veins wele carefully dissected Iungs. The lungs were reinflated to l0 cm pressure of water above mid-lung height, with fresh fixative administered through a plastic tube inserted into the trachea. Fixative from the lungs and tracheas of animals AS25 and A526 were were dnained off for collection of any free parasites within the airways. The Iungs
then reinflated with fresh fixative to l0 cm pressure of water.
The left and right major bronchi were ligated as close as possible to the lung
parenchyma and the lungs divided. The volume of the right and left lung from each animal was measured by water displacement. For this, each lung was carefully 22.
sub- Iowered into a large plastic tub filled to the brim with water, completely merged and held in this position until no more water overflowed' The lung was quantities of carefully removed from the tub and the tub refilled with measured to water until the level at the brim was reached, the volume of water needed refill the tub being the volume of the lung. Samples for morphological examination whilst were taken from peripheral and central areas of the right lung in each case, the left lungs wete reserved for morphometric study'
2.4 PARASITE COLLECTION AND INCUBATION Any fluid which escaped from the lungs during sampling for morphology and morphometry was collected. All of this fluid collected from the lungs and and mites tracheae was filtered through fine silk and any specimens of worms
recovered were transferred to 70% ethanol' midline and The skulls of animals 4523, 24' 25 and ?6 were bisected on the the nasal cavities examined in detail for the presence of mites' Despite fixation of with l0% formalin and subsequent storage in a cold room at 4oC for several a number of live weeks, upon examination of the nasal cavities of animal 4525 larval mites were recovered. These were incubated by placing them into a shallow layer of physiological saline in a petri dish at room temperature of approximately develop- ZOIC, as described by Furman and Smith (I97r, to observe any in vitro
ment.
2.5 MICROSCOPY 2-5.L Light MicroscoPY with setected samples of lung tissue, up to 1.0 x 2.0 x 0.5 cm in size, alonq
parasitized regions of trachea, nasopharynx, and nasal turbinate were dehydrated stained with in alcohol and embedded in paraffin wax. Five micron sections were (Miller, Parasitized haematoxylin and eosin (H & E) or Miller's elastic stain 1971)' in regions of lung, nasopharynx and nasal turbinate were also embedded 23
glycolmethacrylate according to the method of sims (I974), except that only were o.?5 g of benzoyl peroxide was used rather than 0.5 g. Two micron sections glass strips cu.t on a SorvaII JB-A microtome with Ralph knives made from either fB mm wide and 6.5 mm thick (Bennett et al., I97O or ordinary glass slides (Semba, IgiÐ. The sections were stained with Massonts Trichrome (Foote
Modification) as used routinely in this laboratory'
2.5-2 Transmission Electron Microscopy samples from selected regions of the lungs were diced into small pieces, (pH7'3' l-l mmJ, and rinsed in several changes of 0.I M sodium cacodylate buffer were post- 120 m0sm) and Ieft in buffer overnight. Following this the specimens at fixed for one hour in l% osmium tetroxide (oso4) in cacodylate buffer, rinsed least five times with distilled water to prevent precipitate formation, and then block stained with 0.5% aqueous uranyl acetate for one hour' Specimens were rinsed again with distilled water, dehydrated in increasing concentrations of alcohol, and embedded in spurr's low viscosity resin held in Beem capsules
(Glauert, Ig75). These specimens were cured in a 70oC oven overnight' The blocks
were trimmed and I pm sections cut using glass knives. The sections were stained with methylene blue - azure II and examined to enable specimen orientation' ultra-thin sections, pale gold to silver, were cut with a diamond knife on a Reichert 0m U2 ultramicrotome, mounted on either 200 mesh Cu/Rd grids or 75 x
J00 mesh Cu grids, and stained with 50% ethanolic uranyl acetate and Reynoldrs lead citrate (Lewis and KnighL, L977). These were examined in a Philips EM 100
electron microscope with high resolution stage at 60 kV'
2.5.3 Scanning Electron Microscopy Samplesoflung,upto5.0xJ.0x0.5cminsize,wereselectedto by demonstrate the terminal airways, and prepared by the method described Malick and wilson (Murphy, Ig78). Specimens were rinsed in sodium cacodylate 1-2 hours buffer, pH 7.3 and 32O mOsm, Ieft overnight, post-fixed in J-% oso4 for
then processed as follows: 24.
l. Rinsed in six changes of distilled water over 15 minutes. Z. Immersed in freshly prepared l% thiocarbohydrazide (TCH) (Eastman Kodak)
zO-tO minutes.
3. Rinsed in distilled water as in step J'' 4. Immersed in 1% aqueous OsO4 'for 2-3 hours' 5. Rinsed in distilled water as in step I'
6. Step 2 repeated. 7. Rinsed in distilled water as in step 'I'
B. Step 4 repeated. 9. Rinsed in distilled water as in step I' Throughout the procedure specimens were placed on a rotator aL 6O levo- lutions per minute. The TCH solution was prepared by rinsing 0.5 g TCH in distilled water several times until the majority of crystals turned from a dark grey
to an off white colour. Twenty-five mitlilitres of distilled water was added to the beaker containing the TCH and then heated to 60oC for several minutes. The solution was allowed to cool to room temperature in a cupboard to avoid direct
sunlight. prior to being added to the specimens the solution was filtered through a
0.47 pm Millipore filter.
The specimens were dehydrated by passing them through 71o/o acelone for
10-15 minutes and then kept in 100% acetone overnight. This was followed by two
changes of dry 100%o acetone over the next 24 hours. specimens were critically- point dried in a Denton critical-point drier, mounted on stubs with colloidal silver (Pelco) and examined in a siemens Etec Autoscan at 20 kv. some specimens,
which were dissected following initiat examination to expose suitable areas' were coated with 10 nm of carbon and 20 nm of gold-palladium in a Denton vacuum
evaporator and re-exam ined. 25.
ChaPter 5
NORMAL RESPIRATORY sTR UCTURE
}.I INTRODUCTION pinnipeds in Research into the diving capabilities of marine mammals, the depth particular, has been principally by physiological methods to determine the utilization of and duration of dives (Scholander, 1940; Kooyman' 196Ò and et al', 1969; oxygen during such dives (lrving, 1939; Scholander, I94O; Ridgway of the respiratory Kooyman, et al., I97I1 I97Ð. Detailed microscopic examination (wislocki, 1929; 1942; system has been, for the most part, Iimited to the cetaceans (wislocki, L935) whilst most Bélanger, J-940; Fanning, I97Ð and the sirenians the upper airways studies of the pinniped respiratory system have been limited to 1973; Boshier and Hill' and light microscopic examination (Denison and Kooyman,
L974; Boyd, 1975).
f.2 TERRESTRIAL MAMMAL RESPIRATORY SYSTEM f.Z.I MacroscoPic into three classically the mammalian respiratory system has been divided distinct functional zones: the larynx; i) the conducting zone: which contains the nose and nasopharynx; the the trachea; the two major, or primary, bronchi; the secondary bronchi; tertiary bronchi and their subdivisions including the terminal bronchi; and
the bronchioles. ii) the intermediate zone: containing the terminal bronchioles; the respiratory
bronchioles; the alveolar ducts; and the alveolar sacs' iii) the respiratory zone: which contains the alveoli and their associated
caPillarY network. well developed The components of each zone ate not always present nor as
in all species. 26.
The lungs are also divisible by structural features into zones which possess some functional characteristics. The first division is into two lungs, left and right, one lying on either side of the thoracic midline, each being supplied by a major bronchus. The next zone) present in most terrestrial mammals, is into lobes; generally there are more lobes on the right side. Each lobe is supplied by a branch of the primary bronchus, the secondary ot lobar bronchus. The division into lobes is by fissures in the surface of the lung which penetrate into the parenchyma for a variable distance. These fissures are lined by a continuation of the visceral pleura which covers each lung. Each lobe is subdivided into bronchopulmonary segments which are supplied by tertiary, or segmental, bronchi. The segments are struc- turally separated by connective tissue septa which are most apparent subpleurally but become tenuous and indistinct deeper within the parenchyma (Krahl, 1964;
Reid, 1967). The next zone is the secondary pulmonary lobule which, despite certain
disagreements, mainly of a semantic nature, is described as the smallest region of the lung outlined by connective tissue septa (von Hayek, 196O; Thurlbeck and Wang, I974; Murray, L916). The secondary pulmonary lobules are the smallest
macroscopic subdivisiong of the lung (Pump, 1964) and contain 3-5 terminal bron- chioles together with the parenchyma they supply (Reid, 1967; Murray, I976;
Inselman and Mellins, IgBl). The final structural zone also has a functional basis and for this reason a
number of definitions have been presented, with two in particular gaining favour with pulmonary anatomists and physiologists. The first of these definitions and the term pulmonary acinus were introduced by Rindfleisch in IB72 (Nagaishi, 1972) Lo
describe the terminal morphological and functional unit of the lung. This defini-
tion considers those structures distal to, but excluding, the terminal bronchiole as
the acinus, a view still held by many (Reid, 1967; Gamsu et al., I97Ii Thurìbeck and wang, L974; Murray, I976; Inselman and Mellins, 19Bl). Then in I92I Loeschcke (Nagaishi, Ig7Ð considered the pulmonary acinus to comprise the 27.
terminal bronchiole together with the structures distal to it, a definition still adopted by others (Pump, 1969; Boyden, I97Ii Nagaishi, 1972; Hansen et al', L975;
Young et al., 1980; Schreider and Raabe, lg8l).
3.2.2 Microscopic
The normal microscopic structure of the respiratory system has been studied in a wide variety of terrestrial mammals. Details of the normal structure, with mention of species differences, can be found in the reviews of Weibel (1973),
Rhodin (I97Ð and Breeze and Wheeldon (1977).
].] PINNIPED RESPIRATORY SYSTEM
The pinniped respiratory system is typically mammalian in its overall design, having the three functional zones. f.f.I Macroscopic The division of phocid lungs, into lobes, vaties greatly between different species. The ribbon seal, Histriophoca fasciata (now called Phoca fasciata Burns,
l-981), at one extreme, has no external fissures to divide either lung into lobes and
has no accessory lobe (Sokolov et aI., I97L), On the other hand, in the bearded
seal, Eriqnathus barbatus nauticus and the ringed seal, Pusa hispida kraschennikovi
(now called Phoca hi ida his ida Frost and Lowry, lgBl) each lunq is divided into three distinct, independent lobes, as well as an accessory lobe on the right (Sokolov et al., l97l). Between these extremes are the Weddell and the grey seals
which have only two lobes on the left, and three (grey seal) or four lobes (Weddell
seal) on the right side (Hepburn, J-896; I9I2; Boyd, I975)' Less variation appears to exist in the lung lobation of otariids. In the
Stellerrs sea lion, Eu m atop ius iubatus. the left lu ng is incompletelY divided into
two lobes whitst the right lung is deeply fissured to form four lobes (Sokolov et al., l97l). The lungs of the California sea lion have been described as Possessing three
distinct lobes on each side (Green, I97Ð' 28.
Inphocidlungstheintrapulmonaryairwaysaresupportedbyhyalinecar- tilageplatesdowntotheleveloftheterminalbronchioles,whilstinotariidsthe carlilage plates are present to the airway termination'
5.J.2 MicroscoPic
(a) IntrapulmonarY airwaYs membrane; a The walls of the intrapulmonary airways consist of a mucous an adventiLia. muscular layer; a submucosa; a f ibrocartilaginous layer; and membrane and The mucous membrane comprises an epithelium, a basement
a lamina proPrta. ciliated (i) Epithelium: The conducting airways are Iined by a pseudostratified columnar epithelium columnar epithelium with goblet cells. This becomes a simple of phocids without ciliated cells and fewer goblet cells in the smallest bronchi (BoshierandHilì,IgT4;Boyd,1975;WelschandDrescher'1982)' basement membrane' (ii) Basement membrane: The epithelium rests on a thick propia contains (iii) Lamina propria: The thick, predominantly circular lamina
collagen and elastic fibres' connective tissue elements (iv) Submucosa: The submucosa of smooth muscle and medium-sized contains many large mucous glands, Iymphocytic accumulations, bloodvesselsand,intheWeddellseal,gland-Iikediverticulae.Thesmoothmuscle is continued down to the bundles of the submucosa form a helical network which bronchial glands level of the bronchioles. The simple, branched tubulo-alveolar wiLh small numbers of contain a predominance of cuboidal mucous secreting cells which are closely pure serous and seromucous alveoli. The large diverticulae, epithelium of low associated with the glands in the trachea, have a non-secretory 1974; Boyd, I975; Welsch and cuboidal and simple squamous cells (Boshier and Hill,
Drescher, I9BZ). and more (v) cartitage: In phocids the airway cartilage plates become smaller at which point the irregular distally, down to the level of the terminal bronchioles walt (King distinct smooth muscle layer alone forms the support for the airway 29
and Kooyman, 1971; and Harrison, 196I; Simpson and Gardner, I972i Denison BoshierandH.LllrI974 Boyd,IgT5iWelschandDrescher'1982)' by irfegular plates of carti- The terminal airways of otariids are supported which point the cartilage plates lage down to the level of the alveolar sacs' at and Gardnet' I972; Denison form a terminal ring (Denison et al., I97I; Simpson and Kooym an, L973).
(b) Bronchioles Bydefinitionabronchioleisanintrapulmonaryairwaywhichdoesnothave in the literature the cartilaginous support in its wall (Reid, 1967). However, generally been referred to as terminal airways of both phocids and otariids have their walls' For this reason bronchioles despite the presence of cartilage within I972; Denison and Kooyman' the ,,bronchioles,, of otariids (simpson and Gardner, Ig73)havebeenincludedinthisdescriptionofthebronchioles. bronchioles is formed of (i) Epithelium: In phocids the epithelium of the terminal cells' This continues into the simple cuboidal cells with no ciliated or goblet outpouchings in their walls respiratory bronchioles which carry occasional alveolar (BoshierandHill,LgT4;Boyd,IgT5iWelshandDrescher'1982)' are present in the submucosa' (ii) submucosa: In phocids no glands or diverticulae airways eontinues down to this level' and the outer elastic fibre layer of the upPer otariids, the arrangement of some surrounding the smooth muscle (Boyd, L975)' In airways' In a small bronchus of a of these structures is quite variable in the lower abundant but no sub- steller's sea lion epithelial goblet cells were extremely airway in a california sea lion mucosal glands were evidence' whilst in a similar 1972)' the reverse situation was observed (Simpson and Gardner'
(c) ParenchYmal subdivisions terminal respiratory unit in The secondary lobule is considered to be the size and is supplied by an phocid lungs. This lobule is about 1.5 x 2.0 mm in which branches into several occasionally alveolated muscular terminal bronchiole This structure is considered respiratory bronchioles (welsch and Drescher, IgB2)' 30.
tobeasecondarylobulebyBoydQgT5)whosta|edthattheprimarylobuleisthat a cartilage-supported "bronchiole" connective tissue-bound region supplied by bronchioles' which then branches to form B-L0 terminal into a pair of alveolar sacs' In otariids the terminal airway Ieads directly within a fibrous connective tissue called a secondary lobule, which is contained Kooym an, 1973). These fibrous sheaths, sheath (Denison et al., I97L; Denison and lobules, are characteristic of pinniped which divide the parenchyma into discrete (simpson lungs of otariid seals than phocids Iungs and are more pronounced in the an' 1973)' and Gardn et, I972i Denison and Kooym
(d) Alveolar ducts and alveolar sacs are continued distally as the The respiratory bronchioles in phocid lungs simple squamous epithelium' The ducts alveolar-lined alveolar ducts which have a into alveoli by the alveolar septa terminate as alveolar sacs which are divided Drescher, I9B2). In otariid lungs (Boshier and Hill, I974;Boyd, I975; Welsch and therearenoalveolarducts;thealveolarsacsarisedirectlyfromtheterminal
airways (Denison and Kooym an, 1973)' the formation of myoelastic Several descriPtions of phocid lungs report alveolar ducts (Belanger' 1940; sphincters in the resPiratorY bronchioles and but and Hill, I974; Welsch and Drescher, l9B2) Simpson and Gardn et, LgThiBoshier (Denison and Kooym an' 1973; Boyd' 1975)' these have not been reported by others (e) Alveoli respiratory bronchioles in phocid The alveoli first aPpear in the walls of the forming part of the walls of the alveolar lungs and become more abundant distally, ductsandsacs.Inotariidsthealveoliarepresentonlyinthealveolarsacs. phocids is very similar to that of (i) Alveotar septum: The alveolar septum of I epithelial cells and their cyto- terrestrial mammals, with typical squamous type the septal walls (Boshier and Hill' plasmic extensions which cover the majority of 1982)' The secretory type II epithelial Ig74; Boyd, I975i Welsch and Drescher, to be a source of alveolar surfactant' cells contain the lamellar bodies considered 3r.
Adjacent epithelial cells are joined by desmosomes and tight junctions; the epi- thelium lies on a distinct basement membrane which in some regions is formed in part by fusion with the basal lamina of the underlying, typically mammalian non- fenestrated capillary endothelium. The fibrous framework of the alveolar wall consists of fibroblasts, some smooth muscle cells and collagen and elastic fibres. (ii) Capitlary arrangement: The alveolar walls of pinniped lungs are generally
considered to carry a single capillary network as has been well demonstrated in
phocids (Simpson and Gardner, I972; Boshier and Hill, I974; Boyd, I975; Welsch
and Drescher, l9B2). For otariids, however, there are two preliminary reports
which describe the presence of a double capiltary network in the alveolar wall of the Steller's sea lion (Simpson and Gardner, 1972) and the Australian sea lion
(Fanning, 1977). This double capillary arrangement is generally regarded as being
unique to the Cetacea and Sirenia (Simpson and Gardnerr IgT2; Fanning, 1977)' (iii) Alveolar macrophages: Occasional macrophages containing lysosomes were
found free in the alveoli (Boshier and H ill, I974; Boyd, I975)'
,.4 RESULTS f.4.1 Macroscopic
(a) Trachea
The trachea and main stem bronchi consisted of dorsally incomplete rings of hyaline cartilage. The ends of the cartilages overlapped and were joined by a strong fibrous connective tissue sheath. The trachea divided into the right and left major bronchi one-quarter of the distance from the larynx to the hilum (Fi9s. I
and 2).
(b) Lungs
Each lung was triangular in shape with the apices lying cranially, caudally
and ventrally (Figs. I and 4). The lungs were divided by complete pleura-lined fissures into lobes. The left Iung had only one fissure which divided the lung into
cranial and caudal lobes (Fiq. l). The right lung was divided into three major lobes; 32.
cranial, middle and caudal (Fiq.4). On the medial aspect of the right caudal lobe
[here was a small accessory lobe (Fig. 2). Apart from the complete fissures which divided the lungs into lobes, all surfaces of the lungs carried a variable number of similar but incomplete fissures (Figs. J and 4). The division of the parenchyma into (Figs. lobules by connective tissue septa was readily apparent through the pleura 5
and 4).
(c) Lobular bronchi The tobular bronchi were up to 0.25 mm in diameter (Figs. 5 and 6). The
Iength of these bronchi was highly variable and depended on the region of the lunq.
This has not been studied in detail.
3.4.2 MicroscoPic
(a) Lobular bronchi
The Iobular bronchi arose from the tertiary bronchi; the junction between the two was well demarcated by an abrupt change from the pseudostratified ciliated columnar epithelium with goblet cells of the tertiary bronchus to the pseudostratified to simple columnar non-ciliated cells of the lobular bronchus
which became simple cuboidal distally (Figs' 7, B and 9)' (i) Epithelium: The epithelium of the lobular bronchus has been divided into two
regions; an initial transition region of columnar cells, and the main part of the
lobular bronchus with cuboidal cells.
The epithelium of the short transition region was columnar in type and the cells were either secretory or non-secretory. occasional ciliated and microvillous
cells were also Present. For most of the tength of the lobular bronchi the epithelium was cuboidal in
type and could be classified into microvillous or non-microvillous cells. Occasional
areas of the epithelium contained squamous cells and cuboidal ciliated cells' The lateral surfaces of adjacent epithelial cells were joined by tight junctions near the lumenal surface with pronounced intercellular interdigitating
processes. 3t.
The transition region contained four types of epithelial cells,and migratory cells. i) Secretory cells: The secretory cells which measured from -l-0 to lB pm in height contained a round to ovoid nucleus, up to 6 ¡rm in diameter, which
was basally situated and occasionally appeared bi-lobed in section (Fiqs. L1
and I2). The apical cytoplasm contained a variable number of membrane-
bound granules which ranged in diamter from 0.6 to l-.0 um. Many of [hese granules wele homogeneously electron-denser whilst others were of
moderate electron density and contained an obvious electron-dense core of variable size (Figs. l2 and Il). Other organelles present included a supra- nuclear Golgi complex, scaltered mitochondria, many small clear vesicles up
to 0.1 um in diameter, profiles of rough endoplasmic reticulum, free ribo-
somes, and occasional bundles of fine fibrillar material. The apical surface
camied a small number of microvillous-like projections up to 0.5 pm high
and 0.ll um wide (Fiqs. 12 and 1l)' ii) Non-secretory cells: Much of the cytoplasm of these columnar cells, up to
15 um in height, was occupied by the large ovoid nucleus which was B pm
x 4 um (Figs. 14 and l5). The remainder of the cytoplasm contained endo-
plasmic reticulum, most of which was rough, mitochondria, a Golgi complex,
an occasional secretory granule 0.25 to 0.5 um in diameter, an occasional
multivesicular body, and ofLen a large lysosome. On the apical surface was a small number of blunt cytoplasmic projections 0.7-0.9 x 0.4-0.6 um which
sometimes contained a few organelles such as ribosomes or profiles of
endoplasmic reticulum. The lateral surfaces had long thin cytoplasmic
processes which interdigitated with those of adjacent cells. iii) Citiated cells: The transition region contained small numbers of columnar ciliated cells, ll to 2l Um high (Figs. L0 and ll). These cells had many typically structured cilia, up to 6 um long, projecting from the apical surface between which were many slender microvilti 1.5 x 0.08 ¡rm in size 34.
(Fiq. 16). The cilia could be seen to originate from the basal bodies of the
apical cytoplasm. The oval nucleus was basally situated, above which was a
zone of many elongate mitochondria I.6 x 0.1 um. Other organelles present included clear vesicles about 0.2 ym in diameter' an occasional membrane- bound granule up to 0.5 pm diameter, some endoplasmic reticulum, free
ribosomes and occasional bundles of cytoplasmic filaments. iv) Microvillous cells: A small number of columnar microvillous cells, 9 to tl ym in height, were present in the transition region (Figs. l-l and 14). They carried a variable number of regularly-shaped microvilli, 0.9 x 0.15 Fm, on their apical surface. The microvilli contained fine axial filaments which continuedintotheapicalcytoplasm(Fiq.u).Theovoidnucleuswasbasally situated, and the majority of the organelles occurred in the apical cyto-
plasm. These included mitochondria, a small number of free ribosomes, an
occasional vacuole and multivesicular body, and small bundles of fine fila-
ments (Figs. 14 and l7). v) Migratory cells: Occasional migratory cells were observed in the epithelium of the transition reqion and the lobular bronchi (Fig' lB)' The shape of these cells was hiqhly variable, depending on their location. The cytoplasm was pale staining and contained many round, homogeneously electron-dense granules 0.6 to 1.0 um in diameter. Other organelles included a small
number of mitochondria, a few profiles of endoplasmic reticulum and many
scattered ribosomes. The nucleus often appeared bi-lobed. There were six types of epithelial cells present in the main part of the
lobular bronchi as well as a small number of migratory cells.
vi) Cuboidal secretory microvillous cells: These cuboidal cells, from 6 to B ,ym high, contained a variable number of large vesiclesr 0.6 to 2 vm, which often
contained osmiophilic lamellae or amorphous material (Figs. 19, 20 and 24). Also present were smaller vesicles,0.2 to 0.5 um, with granular contents and an occasional multivesicular body (Figs. 20 and 21). The large, often 35.
indented, nucleus tended to be centrally situated, and numerous mito- chondria which varied in shape from round, 0.1 to 0.4 ¡'rm, to elongate, I.j x O.j um, were scattered throughout the cytoplasm. A Golgi complex was generally present adjacent to the nucleus, together with many profiles of rough endoplasmic reticulum. Also present within the cytoplasm wete many free ribosomes and scattered bundles of fine filaments (Fiqs. 2Or 2I
and ZZ). The short microvilli, 0.5 x 0.9 Um, occurred predominantly on the
apical surf ace and contained slender axial f ilaments which entered a
terminal web within the apical cytoplasm (Figs. 2or 22 and 23). vii) Columnar secretory microvillous cells: These secretory microvillous cells
appeared columnar in shape, B to l2 ¡rm high, due to their apical cytoplasm which bulged into the airway lumen (figs. 23r 25 and 26). The apical cyto- plasm contained many profìles of rough and smooth endoplasmic reticulum which often appeared dilated (Figs. 27 and 2B). Scattered throughout the
cytoplasm were several large vesicles,0.5 to 1.0 ¡im diameter, which con- tained osmiophilic lamellae or amorphous material, and a small number of moderately electron-dense secretory granules 0.5 ym in diameter' The
mitochondria were most often elongatedr 0.l-0.6 x I.0-l'0 pm, and generally
possessed distinct cristae. Other organelles present included free ribosomes,
a prominent Golgi complex and an occasional multivesicular body. The large ovoid nucleus was centrally located. A small number of microvilli,
0.6 x 0.I um, which contained slender axial fitaments, projected mainly from
the periphery of the apical surface of the cell' viii) Non-secretory microvillous cells: The non-secretory microvillous cells were present in small numbers in the epithetium of the lobular bronchi (Figs.25
and 29). These squat cells tended to be wider, 7 Lo 9'¡rm, than they were high,4 to 6 ym (Figs. 29 and 30). They had many regularly-shaped, closeìy- arrayed microvilli, 0.7 x 0.12 pm, which wete generally located on the central region of the apical surface. The prominent axial filaments within 36.
the microvilli extended into the apical cytoplasm (Fiqs. f0 and ll). The celts
contained a basally situated ovofd nucleus, a Golgi complex, elongate mito- chondria 0.1-0.2 x 1.0-2.0 pm, free ribosomes, smooth endoplasmic reticulum, an occasional vacuole or membrane-bound vesicle, 0.2 y m
diameter, multivesicular bodies and bundles of cytoplasmic f ilaments. ix) Non-microvillous cells: These squat, cuboidal cells, 7.6-8.8 x 6.6-7.6 ¡tm, had round to oval, centrally-located nuclei, some elongated mitochondria
O.3 x 2.5 ¡rm, profiles of rough endoplasmic reticulum, and variable numbers
of vesicles up to 0.5 Um and membrane-bound granules up to 0.3 pm in diameter. Some of these cells possessed small, blunt cytoplasmic plo-
jections, 0.09-0.16 xO.2O-O.28 Ìrm, on their lateral and apical surfaces (Figs.
29,32 and ll). x) Ciliated cells: The ciliated cells were found in the lower regions of the lobular bronchi in very small numbers (Fiqs. 23 and l4). These cells were similar, in most aspects, to those of the transition region. However, they
tended to be more cuboidal in shape, being only about 7 ¡rm high.
xi) Squamous cells: The squamous cells were similar to the type I cells of the alveolar septal walls. They occurred infrequently in the more distal seg- ments of the lobular bronchi. The thin cytoplasmic extensions of these cells,
which most commonly contained only small pinocytotic vesicles, covered the
occasional capillaries which protruded between the cuboidal cells (Figs. )5
and 36).
(ii) Basement membrane: The basement membrane underlyinq the epithelium
consisted of a thin distinct basal lamina with variable numbers of associated
collagen and elastic fibres (Figs. 12, 14,26 and 32).
(iii) Lamina propria: This was a prominent fibrous connective tissue layer that
supported the basement membrane. It consisted of collagen and elastic bundles in varying proportions (Figs. IIr 25 and 26). The extent of the lamina propria was evident in LM sections stained with Millerrs elastic stain (Figs. 17 and lB). 31.
(iv) Muscularis: Distinct bundles of smooth muscle fibres wete arranged into spirals which continued down the bronchi to the terminal cartilaginous rings (Fiqs.
39 and 40). The arrangement of the muscularis was not studied in detail. (v) Submucosa: Underlying the muscle layer was a submucosa of variable thick- ness. The connective tissue of the submucosa, which was continuous with that of the muscle layer, was composed of a loose arrangement of elastic and collagen fibres and fibroblasts and contained a rich plexus of thin-walled capillaries, both vascular and tymphatic. Occasional larger, thicker-walled vessels were present, as were eosinophils, neutrophils and mast cells (Figs.35r 40 and 4l). (vi) Cartilage: The irregularly-shaped plates of hyaline cartilaqe continued throughout the lobular bronchi down to the origins of the alveolar ducts, at which point they formed a terminal ring (Fiqs. t7r i9r 39 and 49). The cartilage plates were surrounded by a prominent perichondrium containing many elastic fibres which fused with those of the submucosa. The structure of the cartilage plates .was that of typical hyaline cartilage (Fiq. 40). (vii) Adventitia: The loose connective Lissue surrounding the cartilage layer formed an adventitia. Where the neighbouring structutes were other airways or larger blood vessels, the adventitia was relatively thick and contained blood vessels and lymphatics (Figs. 7, B, 37 and 39). However, if the bronchus was
surrounded by parenchyma then the adventitia tended to be quite thin (Fi9s. 24r 37
and l9). In either case the adventitia was continuous with the interstitial con- nective tissue framework of the lung parenchyma (Figs. 7 and J7). Figure l. Dorsal view of excised lunqs and lrachea. Note major pleural-lined fissures which divide lungs into lobes ( t ). One-fif th life-size.
Figure 2. Ventral view of excised lungs and trachea. Note cranial division of the trachea into the two major bronchi ( t ), and the small accessory lobe of the right lunq (AC). One-fifth Iife-size. 38
1 Figure i- Lateral view of left lung showing lhe complete fissure which divides the lung into cranial (CR) and caudal (CR) toUes. Note Lhe incomplete superficial fissures ( I ) and the parenchymal connective tissue septa (sep) apparent through the pleura. One-quarter life-size.
Figure 4- Lateral view of right lung showing the division into cranial (CR), middle (MI) and caudal (CR) tones. Note the incomplete fissures ( I ) and parenchymal connecLive lissue septa (sep). One-quarter life-size. 39
p I
3
4 scannlng Figure 5. Macroscopic view of lung specimen prior to preparation for (cart); electron microscopy. Note large bronchus with obvious cartilage plales smaller lobular bronchi (LB); and septa which divide the parenchyma into lobules and acini ( t ). 5x.
I
septum Figure 6. Light micrograph of lung section showing pleura, intersegmental (i the division of (i s s), interlobular septa (i I s)'and interacinar septa a s)' Note (path)' H and parenchyma into lobules G--) and acini (...); and subpleural pathotogy E. lOx 40 Figure 7. Light micrograph of junction between tertiary and lobular bronchi showing change from pseudosbratified columnar epithelium (ps c) to simple cuboidal epithelium (s c). Note the inLerlobular septum (i I s) and the interacinar septum (i a s) which are conlinuous wilh the bronchial adventia; the alveolar ducts (AD) and alveolar sacs (AS). H and E 50x Scale - 250 ¡rm'
Figure B. Light micrograph of junction belween tertiary and lobular bronchi. In the upper bronchus there is an abrupt change in the epithelium from pseudo- stratified columnar to simple cuboidal, whilst in the Iower bronchus a short transition region of simple columnar cells is present ( t ). note cartilage plates (cart) and smooth muscle bundles (s m b) surrounding entrance to alveolar duct. H and E. 140x. Scale - I0 um.
Figure 9- Scanning electron micrograph of junction between terLiary and lobular bronchi showing ciliated cells (CC) of the tertiary bronchus and columnar sec- refory cells (sc) of the transition negion. 5,600x. scale = 2.5 vm. 41
n
1
AD
I I
1
AS
. ils
r. ; tt? I +r
'¿ /? .\ _2. z¿ t \ , -çi ¡
¿ a t , ¡ ) ¡r-l-l \ I t .;a /- /l ¿ .2.* à l> f , .¡ r-( A -..É 'tL f 4 I 1 Figure 10. Transmission electron micrograph of transiIion region between bertiary and lobular bronchi. Note columnar ciliated cells (CC), secrelory cells (5C), and a goblet cell (GC). 2,9OOx. Scale = 5 pm.
Figure ll. Transm ission electron m icrograph of transition region showing columnar secrebory cells (SC), non-secrelory cells (NSC), a ciliated cell (CC) and a non-secretory microvillous cell (MVC). Note also the connective tissue fibres of lhe lamina propia (l p), and the smooth muscle cells (s m c) of the muscularis. 3,500x.Scale-5 um. 42
a
a taz Figure 12. Transmission electron micrograph of secretory iells in the transition region. Note the bi-lobed nucleus (N), the membrane-bound secretory granules (s g), some with electron-dense cores, a Golgì complex (G), the microvillous-like projections (MV) on the apical surface, and the basal lamina (b D. 7'000x. Scale = 2 um.
Figure ì-5. High power transmission electron micrograph of one secretory cell of the transition negion. Note lhe membrane sumounding the secretory gnanules ( 1), and the Golgi complex (G). 26'I00x. Scale - 500 nm' 43 Figure14.Transmissionelectronmicrographoftransitionregionshowingseveral (sc) an¿ a microvillous cell (Mvc)' non-secretory cells (NSC), a secretory cell cells' and the blunt Note the occasional granule (g) within the non-secretory 2'5 cytoplasmic projections (c p)' 5,100x' Scale - um'
Figure15.Transmissionelectronmicroqraphoftransitionregionnon-secretory projections (c p)' and long' cells. Note, lysosomes (lys), blunt apical cytoplasmic 2 thin laberal cytoplasmic processes ( t ). 7,800x' Scale = sm' 44
(-l ciliated cell in Figure 16. Transmlsslon eleclron m icrograph of columnar microvilli (MV). transition region with typical cilia (C) between which are slender Note the elongate mitochondria (m)' 6,700x' Scale - 2 um'
microvillous cell Figure 17. Transmission electron micrograph of columnar (vac)' 25,5OOx' Scale showing microvilli (MV) with axial filaments, and vacuoles = 500 nm. 45
ét of Lransilion region' Nole migratory Figure IB. Transmission electron micrograph and eleclron-dense granules cell with bi-tobed nucleus (N), pale-staining cytoplasm (q). B,gOOx. Scale = 2 Pm'
of lobular bronchus showing cuboidal Figure 19. Transmission electron micrograph (b v) and muscularis (musc)' epithelium, lamina propria (l d with blood vessel cells (SMVC) and non- Epithelial cells present include secretory microvillous microvillous cells (NMVC). 1,600x. Scale - 5 um' 46 microvillous Figure 20. Transmission electron micrograph of cuboidal secretory (MV)' 9'400x' cells. Note numerous larqe cytosomes (cyt) and apical microvilli Scale = 2 gm.
i
microvillous 1 Figure 21. Transmission electron -micrograph of cuboidal secretory (N), (cyt)' vesicles cells showing indenled, centrally-situated nucleus cytosomes (m)' I6,600x' containing amorphous material (veÐ, and scattered mitochondria Scale = J- irm. 47
7 :
-c
t Ð ':!..; 17 L2I Ð €) æ (. ,È'?e d iÛiJ
rt
'4-: Figure ZZ. High powen transmission electron micrograph of secretory microvillous (cyt), cells showing lerminal web (t w), elongale milochondria (m), cytosomes and lateral interdigitating processes ( t<). 20,400x' Scale = I ¡rm'
Figure 2J. Scanning electron micrograph of lobular bronchus showing surfaces of cuboidal secretory microvillous cells (cub) and bulging columnar secretory micro- villous cells (col). 8,100x. Scale = 2 vm. 48
L 3 n
/1
t1¡. I
; (tB) its Figure 24. Transmission electron micrograph of lobular bronchus showing relationship to the alveolus (alv). Note the similarity between the cuboidal secretory microvillous cells (SMVC) of the lobular bronchus and lhe alveolar Lype (adv) II epithelial cell (ep ID; the hyaline cartilage (cart) and the thin adventitia separaEing the bronchus and the alveolus. 2,300x. scale = 5 um.
Figure 25. Transmission electron micrograph of lobular bronchus showinq the epithelium which contains a non-secretory microvillous cell (MVC) and columnar secretory microvillous cells (SMVC) wiLh butging apical cytoplasm' Note the thin (cart); lamina propia (l p) which separates the epithelium from the cartilage and the intralumenal macrophages (mac). 3,400x' Scale = 5 ¡rm' 49
I
,dr...:: t: Ð
"tO,Ê'' Figure 26. Transmission electron micrograph of columnar secretory microvillous cells. Note large nucleus (N), and large vesicles (ves) containing amorphous matenial; the thin lamina propia (l p) and Lhick muscularis (musc). 5r?OOx. Scale = 2.5 ü,m'
Figure 27- High power transmission electron micrograph of columnar mierovillous secretory cells. Note dilated endoplasmic reticulum (e r), osmiophilic lamellae within large vesicles (ves), and peripherally siLuated microvilli (MV). 17,500. Scale =1gm. 50
ç" êr., \ micrograph of bulging apex of Figure 28. Hiqh power transmission eleclron I both rough columnan secretory microvillous cell which contains many profiles of (e ribosomes' and secretory granules and smooth endoplasmic reticulum r), free I (s g). tl,l0Ox. Scale = l- Fm.
Figure 29. Transmission electron micrograph of lobular bronchus showinq epi- thelium, Iamina propria and muscularis. The epiLhelium contains two non- (NMvc)' secretory microvillous cells (MVC) and a non-microvillous cetl Note basally situaled ovoid nuclei of microvillous cells, elongale mitochondria, and clustered regularly-shaped microvilli 8,000x' Scale = 2 gm' 51 I
I - non-secrelory microvillous cell Figure J0. Transmission electro n micrograph of the large ovoid nucleus and showing the centrallY-situated, reqular microvilli, L elonqate mitochondria' 13,700m' Scale = vm'
micrograph of apical portion of non- Figure 1I. High power Lransmission electron secretorY microvillouscell.No|etheaxialfilamentsofthemicrovilliwhich (vac)' 39'800x' Scale - 500 nm' extend in to lhe cytoplasm (f ). and the vacuoles 52
{'ui FigureJ2.Transmissionelectronmicrographofnon-microvillouscellsshowinq (N), mitochondria (m), and blunt cyto- large, centrally-Iocated nucleus elongate basal Iamina (b l). I0,400x. Scale = 2 Fm. plasmic projec|ions (c p). Note the
Figure]J.Transmissionelectronmicrographofnon-microvillouscells.Notelarqe(c p)' (g), and blunt cytoplasmic projections nuclei, membrane-bound granules 9,B00x.Scale=2 Ym' 53
,
f o a 'ÐC c a C t a o t.: q
rl
+ cell showing long Figure f4. Scanninq electron micrograph of cuboidal ciliated 2 cilia (C) and centrally-situaLed microvilli (MV)' 8,000x' Scale = ym'
bronchus' Figure J5. Transmission electron micrograph of distal region of lobular extensions of Note superf icial capillaries (cap) covered by thin cytoplasmic blood vessels squamous cells; muscularis (musc); and submucosa with thin-walled (b v). 1,900x. Scale - I0 um. 54
.ti junction be[ween terliary Figure J6. Low power scanning electron microqraph of by and lobular bronchi. Note cilialed cells (CC) of tertiary bronchus covered (cub) cells of the mucus and red blood cells, the columnar (col) and cuboidal (cap) by lobular bronchus, and an occasional superficial capillary covered squamous cells. 700x. Scale = 20 um'.
septa Figure 37. Liqht micrograph of two adjacent acini outlined by interacinar the bronchi, (i a s). Note distribution of elastic tissue in the lamina propria of (adv) alvolar duct walls and alveolar septa ( t ); the bronchial adventitia which varies in thickness and is continuous wibh the interstitium of lhe inleracinar and ducts alveolar septa ( t 1); the terminal ring of cartilage plates; and lhe alveolar (AD), alveolar sacs (45), and alveoti (alv). Miller's elastic stain' 70x' scale = 200 um. 55
r-Ul
I t ! .zt¡'fa, /+
I I alv AD
I ( AS
\t s
- --- 37 t. lobular bronchus showing dis[ribution of Figure 18. Light micrograph of [erminal and submucosa' Note also the the elaslic lissue (el) within the lamina propria l00x' Scale 50 um' cartilage plates (cart)' Miller's elastic stain' =
showing the helical arrangement Figure 19. Light micrograph of lobular bronchus cartilage plates (cart) which of lhe muscularis (musc) and Ihe irregularly-shaped of the advenIitia (adv)' form a terminal rinq. No[e the variations in the Lhickness H and E. 100x. Scale - 50 ¡rm' 56
; -Ê'-i- i 38 J a a a ¿D a a I t t'.a t ,i, I ( \ å I a \ t-, 'a a 'a \D \ I I ) I I I t t a \ a I \ 5 - ta .lr a Ç 4- I ¡ I ç tll'. ,)1, ìt a¡ ft o tr.l. , \ a I ,Q. t/¿lr \ t o , !ô ,'raD o-' t ¡ I t a , 39 I i Figure 40. Low power transmission electron micrograph of lobular bronchus showing epilhelium, muscularis (musc), submucosa (s m), and hyaline cartilage (cart). Nole bhe submucosal blood vessel (b v) and probable lymphatic channel (lym). 1,200x. Scale = 20 ym. Figure 4l- Transmission electron micrograph of submucosal mast cell with dis- tinctive granules. 20,500x. Scale = I um. 57 58. (b) ParenchYmal subdivisions Theparenchymaofthelungwasdividedintoanumberoflargesegments (Figs' 5 and 6)' prominent connecLive tissue sepla which were separated by thick, wete septa which divided the lung parenchyma The thinner connective tissue These septa magnification (Figs' 5' 6' 7 and l7)' a prominent feature even at Iow a lobular acini' Each lobule was supplied by divided the lung into lobules and bronchusandincludedalltheparenchymalstructurestowhichthebronchusqave rise(Figs.6,42and4]).Thelobulewasfurtherdividedintoanumberofacini. Eachacinus,onterminalrespiratoryunit,containedthoseparenchymalstructures distaltoaterminalcartilagerinqinthewallofthelobularbronchus.Those structureswithinanacinusgenerallyincludedtwoormorealveolarductsandthe and which they gave rise (Figs. 6, J7,47,43 various alveolar sacs and alveoli to 44). Theconnectivetissueoftheseparenchymalsubdivisionscontainedtypical .collagenandelasticfibres,fibroblasts'Somesmoothmusclecellsandoccasional mastcells;itwascontinuouswiththeconnectivetissuecoreofthealveolarsepta (rigs.45and46).Thesethickerseptacontainedthepulmonaryarterioles,venules andlymphatics.Occasionallyanalveolarcapillarycouldbeseentodraindirectly into a venule (Fiqs' 47 and /+B)' on each surface by a single capillary AII of these thicker sePta were lined epithelium, of type I and type lI cells' network and covered by typical alveolar 46)' Thus these the alveolar septa (Fiqs' 45 and which was continuous with that of thickerseptaformedthebaseofsomealveoli(Figs.6,7,'7and4B). (c) Alvolar ducts and alveolar sacs Eachlobularbronchusterminatedbygivingrisetoaseriesofalveolarducts andalveolarsacs.Theentrancetoeachgroupofductswasdelineatedbyaringof cartilageinthewalloftheairway(Fiqs.37,'gand49).Atthispointtherewasan abrupLchangeintheepithelium;fromthecuboidalbronchialcellstotheextenstve squamoustypeofthealveolarsepta(Fig.49).Thealveolarductswerelarge 59. of passages lined with alveoli which either branched to form further generations ducts or gave rise to the alveolar sacs. Each sac was a collection of alveoli which opened into a common area for 9as mixing (Figs.7r37r 43r 44 and 50). those of The walls of the alveolar ducts tended to be more pronounced than prominent the sacs (Fiq.5l). Bundles of collagen and elastic fibres were the most feature of these walls, which also contained occasional smooLh muscle cells and fibroblasts (Figs. 52,53,54 and 55). The walls of the ducts and sacs were generally lined on both surfaces by alveolar capillaries which were covered by the These cell attenuated cytoplasmic processes of the type I alveolar epithelial cells' sac processes extended beyond the capillaries to cover those parts of the duct and walls that were exposed to the duct or sac lumen. The connective tissue of the walls' duct and sac walls was continuous with the interstitium of the alveolar (d) Alveoli The alveoli formed part of the walls of the alveolar ducts and sacs. They (Fiqs' 37' were open-ended polygons which faced on to the alveolar ducls and sacs network 44, 48r 50 and 56). The walls of the alveoli consisted of a f ine capillary covered by suspended in a connective tissue framework, the whole of which was alveolar epithelium. The alveolar septal walls carried either a single capillary network, or a double capillary arrangement whereby the connective tissue core supported a capillary network on each surface' (i) Epithelium: There were only two cell types which formed the alveolar lining' in niches of i) Type I cells: The nuclei of the type I cells tended to be located the capillary network (Fig. 57); however, they occasionally protruded into the alveolar lumen (Fig. 5B)' The nuclei were surrounded by a thin cyto- plasmic layer which contained the majority of the sparse organelles' which included a few mitochondria, a Golgi complex, free ribosomes and some profiles of endoplasmic reticulum (Fi9s.57r 58 and 59)' The thin attenuated sheets of cytoplasm which radiated from the perinuclear region generally (Figs' contained numerous pinocytotic vesicles and occasional mitochondria 60. 59 and 60). Occasionally a cytoplasmic process was seen to cross the septum on to the surface of an adjacent alveolus; this was only observed in areas of single capillary network. ii) Type tI cells: The type II epithelial cells tended to occur in groups of two to three, although they did occur singly (Figs. 57,60r 611 62 and 6l). They most often lay in gaps between the capillaries of the septal wall. The cells were ovoid to cuboid in shape and contained a wealth of organelles; many mito- chondria, a prominent Golgi complex, smooth and rough endoplasmic reticulum, free ribosomes, multivesicular bodies and, most characteristi- cally, osmiophilic lamellar bodies or cytosomes. These cytosomes, which were 0.6 to 1.7 pm in diameter, contâined an irregular array of lamellae of variable thickness which generally formed concentric layers. Infrequently a cytosome could be seen discharging its contents on to the alveolar surface. Many short microvilli, 0.7 x 0.1¡rm,projected from the apical surface of the type II cells, whilst the lateral and basal surfaces carried slender inter- digitating processes which communicated with adjacent cells (Figs. 60r'62 and 63). The alveolar epithelium lay on a thin distinct basal lamina which at times fused with that of the capillary endothelium (Figs.60 and 76). Adjacent cells of the epithelium were attached by tight junctions (Fiqs.60 and 6l). iii) Type III cells: No type III cells were observed in the alveolar septum. (ii) Capillaries: The capillaries of the septal walls formed a dense meshwork which was most easily appreciated in scanning electron micrographs of an en face view of an alveolar wall (Figs. 48, 51 and 56). These vessels were lined by a non- fenestrated endothelium (Figs.59, 64 and 76). The perinuclear region contained the majority of cytoplasmic organelles; mitochondria, a Golgi complex and some endoplasmic reticulum (Figs. 59 and 6l). The cytoplasmic processes contained occasional mitochondria and many pinocytotic vesicles. The endothelial cells were surrounded by a thin distinct basal lamina (Fiqs. 76 and 77). 6I. (iii) Interstitium: The interstitium, which was composed of fibrous connective tissue, formed a supporting framework for the capillary network' The major (Fiq.77). components of the interstitium were typicat collagen and elastic fibres Associated with this fibrous component were fibroblasts and smooth muscle cells that often were apparent only as cytoplasmic processes' Thin cytoplasmic processes and an occasional cell body could be found in close association wiLh Lhe interstitial aspect of the endothelium. These contained few orqanelles, principally mitochondria, and probably represented pericytes (Fig.76)' Also present was an occasional mast cell with its distinctive crystalline granules. The arrangement of the connective tissue elements varied according to the capillary arrangement. Where the capillaries were most closely applied to the alveolar epithelium the interstitium consisted of the fused basal laminae of the epithelium and the endothelium. Generally in those parts of the alveolar wall with a single capillary arrangement, one side of the capillary was in close association with the epithelium, whilst the other side abutted the connective tissue core of the septum (Figs. 55, 64 and 7l). This connective tissue core occasionally con- tained only small numbers of collagen and elastic fibres but more often a marked thickening together with some cellular elements. Where there was a double capillary alrangement the two capillary layers were usually separated by a well developed connective tissue core which contained fibrous and cellular components (Figs. 6i,661 67 and 72). There were occasional areas in which the central con- nective core was intact but carried a sparse capillary network on both sides, so that gaps were present between adjacent capillaries on the same side of the septum (Figs.7l and 7f). Occasionally the capillary network from either side of (Fig. the septum joined together to form a typical single capillary arrangement 66), whilst at other times there were transeptal anastomoses between the two networks (Figs. 65 and74)- (iv) Macrophages: Typical macrophages were occasionally found on the surface of the alveoli and lobular bronchi (Figs. 25r 78 and 79). Figure 42. Diagram of parenchymal subdivisions' 62 Cartilage Tertiary bronchus ACINUS ñr!ìliítÌltrirl Alveolus LOBULE Alveolar duct Lobular bronchus Figure 4f. Scanning electron micrograph of parenchyma with a lobular bronchus (LB) supplying a pulmonary lobule which contains several acini (En). Nole the alveolar ducls (AD) and alveolar sacs (AS). lOx. Scale = 500 pm. Figure 44. Scanning electron micrograph of parenchyma showing division of Iobular bronchus (LB) to supply individual acini. Note interacinar septa (i a s). alveolar duct (AD), alveolar sacs (45) and individual alveoli (alv). JOx. Scale = 500 pm. 63 Figure 45. Transmission elecLron micrograph of thicker septum. Note the con- nective Iissue elements, particularly the collagen (coll), the fibroblasts (fib), and the mast cell (MC); the conlinuily between the connective tissue of the thick septum and that of the alveolar septum ( 1); and the respiralory epithelium which covers the thick septum ( tt) Z,¡OOx. Scale = 5 Fm. Figure 46. Transmission eleclron micrograph of thick connecbive tissue septum whose elements are continuous with those of the alveolar septum. Note also the respiratory epithelium lining the thick septum. 2rZOOx, Scale = 5 ¡rm. 64 \. 4:I ,.rf (ç, fib --t-f L t !! /. - - a pul- Figure 47. Transmission electron micrograph of thick septum containinq ( L'900x' Scale monary venule. Note the capillaries draining into the venule d' ) = I0 ym. pul- Figure 48. Scanning electron micrograph of several alveoli' Note larqer (t the thicker monary blood vessel (b v) with communicating alveolar capillaries ); and the conneclive tissue septum (sep) which forms the base of the alveoli' absence of alveolar pores of Kohn' 26Ox' Scale - 50 vm' 65 '( \q 47 \ of lobular bronchus' Nole Figure 49. Scanning electron mierograph of termination ring, and abrupt chanqe in cartilage plates (cart) in wall of bisected Lerminal to predominantly squamous cells epithelium from cuboidal cells (cub) of bronchus (sq) of alveolus. l00x' Scale = 50 Pm' duct' Note thick connective Figure 50- scanning electron micrograph of alveolar tissuebands(t)formingwallsofduct;individualalveoli(alv);andtheabsenceof alveolar pores of Kohn. BOx' Scale = 200 um' 66 -.¡l duct and alveolar Fiqure 51. scanning elecLron micrograph of walls of alveolar ( 640x' Scale sac. Note thickness of ducb wall (t ) compared to sac wall tt )' = 20 um. wall' Note elastic Figure 52. Transmission electron micrograph of alveolar duct septum ( and the (el) and collagen (coll) fibres which continue into the alveolar 1); duct wall ( )' alveolar type I cell cyloplasmic extensions which cover the tt 3,500x.Scale-5 um. 67 a alv alv 52 duct wall covered by Figure 5J. Transmission electron mtcrograph of alveolar (ep cells' Note the capillaries (cap) and alveolar type I (ep D and type II II) duct wall inlersti[ium' 1,800x' collagen ( x ) and elastic (el) fibres and cells of the Scale - l0 um. sac wall' Note small Figure 54. Transmission electron micrograph of alveolar is continuous with lhat of amoL¡nt of connective tissue (CD of sac wall which covet the sac wall ( l-)' alveolar septum; type I cell cytoplasmic extensions which and alveoli (alv). 2,000x. Scale - 10 um' 68 ,.1 alv :i' alv d alv ç t Figure55.Transmissionelectronmicrographofpartofalveola[Sacwall.Note cell ( ¡¡ç ) of sac wall; alveolar elastic and collagen fibres and smoolh muscle alternaling thin ( t ) and thick ( t t ) septum with single capillary network, and (alv)' 1,900x' scale = I0 ym' interstitium; and alveolar sac (AS) and alveoli several alveoli showing alveolar duct Figure 56. Scanning electron micrograph of capillary network (cap) covered by ( l ) and sac ( t l ) walls, and dense alveolar (i s) which contains a large pulmonary epithelium. Nole the interacinar septum a blood vessel (b v). 400x' Scale = 50 um' 69 o \ ) t ói; AS ¿ rt alv alv .i showinq type I Figure 57. Transmission electron micrograph of alveolar septum Note the thin peri- (ep D and type u (ep II) cells lying in niches of septal wall' organelles' 4'200x' nuclear cytoplasm of the type I cell which conlains very few Scale = 5 ym. septum Figure 58. Higher power transmission eleetron micrograph of alveolar Note the with protruding type I cell body containing small numbers of organelles' pinocytolic vesicles type I cell thin cytoplasmic extensions which contain many ( t ). f0,500x. Scale = 2 um 70 :ri sparse perl- Figure 5g. Transmission electron micrograph of type I cell. Note non- nuclear organelles and pinocytotic vesicles of cytoplasm ic extensions; (N) organelles and fenestrated endothelium of capillaries with sparse perinuclear cytoplasmic pinocytotic vesicles ( t )' 7,700x' Scale = 2 um' Br600x' scale 2 um' Figure 60. Transmission electron micrograph of type II cell' = 71 l' ¿¡' ¡ '' l Figure6l.ScanningelecLronmicrographofalveolarsurfaceshowinggroupsof 2'2OOx' Scale J'0 pm' type II cells (ep u) lyinq wilhin niches between capillaries' = type II cells' Figure 62. High power scanning electron micrograph of alveolar junction line between cytoplasmic Note numerous superficial microvilli (MV); and sheets of type I cells ( t ). e,fOOx' Scale = 2 uin' 72 c Ð t .Ì q ,:1 i a f Ê t a ** t t "l a ¡ -A a¡. t q Ò I to' a ì ì a ,le, ç t'èa .t a l' I õ /." ì I' I . C at' .,Ò ('frt' ri :Ìrl+ ¡ù'" ] .7 l'" I \- Ì ç4. f _.. @ 1' Note the dense array Figure 6J. Transmission electron microqraph of type Il cells' (cyt) mitochondria (m); of organelles, in particular the numerous cy[osomes and and the tight cell junctions ( t )' B,l00x' Scale = 2 ¡rrTì' septum containing single Figure 64. Transmission electron micrograph of alveolar of the type I (ep I) capillary network. Note the attenuated cytoplasmic processes ( regions of the blood-gas and endotheli,al (en) cells; and the thick ( t ) and thin tt ) along the barrier which generally lie on opposing sides of the septum and alternate length of the septum. 4r000x' Scale = 5 um' 73 ì epl septum with double Figure 65. Transmission electron micrograph of alveolar lissue core (cT) and capillary network. Note well developed central connective transeptal anastomosis ( 1)' l,B00x' Scale - l0 um' septum' Note junction of Figure 66. Transmission electron micrograph of alveolan ( and interstitial connective double capillary network lo form a single network t ), tissue with a fibroblast (fib)' 2,000x' Scale - 7'5 ¡rm' septum' Note arrange- Figure 67. Transmission electron micrograph of alveolar the sobin ment of capillaries and interslitium which form thersheet'andtposLs'of l0 pm 'sheet-flow' model of the lung' 2,300x' Scale - 74 / I 65 -/ I 66 67 -1 Figures 68-75. Morphometric series of transmission eleclnon micrographs from anlerio-ventral region of animal A54. 1,600x. Scale = L0 Fm. Figure 68. Alveolar septu m between adjacent alveolar sacs. Note thick connective lissue core (CT) and lymphalic channel (lym). Figure 69. Junclion between alveolar sepla from three atveoli (alv) Figure 70. Alveolan septum with type I cell (rp D lying in a niche belween two capillaries. Figure 71. Alveolar septum with central connective tissue core and alternat- ing capillaries to produce a "pseudo-single'r capillary network. Figure 72. Alveolar septum with double capillary network and type I cell (ep D. Figure 7f. Alveolar septum with double capillary network. Note example of the occasional gaps which occur between adjacent capillaries on the same side of the cenlral conneclive tissue core ( I ). Figure 74. Alveolar septum with transeptal anastomosis ( t ). Note cellular components which form alveolar sac wal (+ç ); type I (ep D and type II (ep tr) c e Ils. Figure 75. Alveolar septum with mixed capillary networks. Note fibroblast (x ) ot interstilium. 75 alv alv alv -- 68 E TI E L * I 73 / ---- u \ ep 75 ^ås. / Figure 76. Transmission electron micrograph of double capillary septum. Nole endothelial cell nucleus (N), probable pericyctes (peri), and fused basal laminae (b t) of epithelium and endothelium.7,200x. Scale - 2 pm. Figure 77. High power transmission eleclron micrograph of alveolar septal inter- stitium. Note collagen and elastic fibres, and fibroblast (fib). 13,900x. Scale = I um. 76 Figure 78- Transm ission electron m icrograph of alveolar m acrophaqe. Note adhesion of macrophage to seplal surface, and pseudopodia ( t ). ff.fOOx. Scale = I um. Figure 79. Scanning electron micrograph of alveolar surface with macrophage attached by pseudopodia ( t ). Z,9OOx. Scale - 5 um. 77 - 18. DTSCUSSION '.5 f.5.I MacroscoPic (a) Trachea ThestructureinN.cinereaissimilartothatfoundinterrestrialmammals (Rhodin,I974).However,thedorsalwallofthetracheaisaffordedgreaterpro- the cartilage rings which overlap each oLher' tection in Neophoca by the ends of seals studied (Sokolov et al'' I97Ii Green' In Neo ohoc a. as in other otariid I912),thetracheadividesintothetwomajorbronchiwellforwardint'hethoracic cavity,sothatthebronchiareaslongas,andsometimeslongerthan,thetrachea. Thisisincontrasttot,hesituationinterrestrialmammals(BloomandFawcett' 1972; Tarasoff seals (Sokolov et aI. , I97I Green, 1968; Rhodin, Ig7Ð and phocid andKooyman,IgT3b)wherethetracheabifurcatesat,ornear'Lhehilumofthe lung. (b) Lungs Thedivisionofthelungsintolobesbycompletepleura-Iinedfissures,is morelikethaLofmanyterrestrialmammals|hanthatofsomemarinemammals. into two the riqht lung into four lobes and the left The division, in Neo hoca of (Miller al'' 1964; in cats, dogs, sheep and pigs et lobes is the same as that found fissured to the left cranial lobe may be deeply Robinson, rgfjÐ.As in these species sea lion The lobar division of the Australian indicate apical and cardiac segments' of the cetaceans (Green' 1972; Fanning and lung is in marked contrast to that (sokolov et aI'' IgTI' Tarasoff and Kooyman' Harrison, I974) and some phocids I973a)wherethereisno,olVerylittle,evidenceoflobation.Thelunglobationin NeophocaisthesameasthatdescribedfortheSteller'ssealion(Sokolovetal., ta ]97I)andWeddellseal(Byd,|975),andisverysimilartothatoftheCaliforn sea lion (Green, L972)' (c) Terminal airwaYs Thedivisionofthecartilage-containingterminalairways,fromthetertiary bronchustothelobuìarbronchuswhichdirectlygivesrisetothealveolarduct,is 19. arrangement present in terrestrial in marked contrast to the bronchial-bronchiolar (Simpson and I974; Banks' l9B1) and phocid seals mammals (Krahl, Lg64iRhodin' Gardner,I972;DenisonandKooyman,J.gT}).Inphocidsthebronchiolesarere- inforcedbythickenedbundlesofsmoothmuscle(BoshierandHill,IgT4;Boyd, presence of cartilage-reinforced terminal Igl5;Welsch and Drescher, l9B2). The (wislock\' I929i I9J5; I942i Wislocki and airways has been described in cetaceans Bêlanger,I94O;Bélanger,1940;SimpsonandGardner'I972iKooyman'1973; Bälanger, 1977), sirenians (Wistocki, L935; Fanning and Halrison, L974; Fanning, 1940)andotherotariids(DenisonetaI.,IgTIiSimpsonandGardner,IgT2;Denison andKooyman,IgT3;Kooyman,LgT,.TheterminalairwaysoftheAustraliansea lioncontaincartilaginousreinforcementtogetherwithprominentbundlesof smoothmuscle.Thereinforcementofterminalairwaysinallmarinemammalsis and in (Ridgway et al.' 1969i Denison et al.' r97I)' important for alveolar collapse (Kerem al'' I975i high expiratory flow rates et cetaceans and otariids to enable Matthews,I97l;KooymanandSinnetL'!919;KooymanandCornell'1981)' 5.5-Z M.icroscoprc (a) Lobutar bronchi (i)Epithelium:Thefourcelltypesobservedinthetransitionregionareall columnar in shaPe' Thesecretorycellswhichcontainelectron-densemembrane-boundqranules are,inmanyrespects,similartotheepithelialserouscellsdescribedintherat (Jeffrey and Reid, 1977), The (Jeffrey and Reid, 1975) and in human foetuses secretorygranulesinNeophocameasureduptol:ilmindiameterwhichiswithin thesizerangeof0.6to].2pmforthesegranulesintherat(JeffreyandReid, I975iSpiceretaI.,r9B0).However,otherfeaturesofepithelialserouscells rough endoplasmic reticulum and lumenal described in rats, such as abundant microvilliwereabsentfromthesecellsintheAustraliansealion.Therefore cytochemicaltechniquesneedtobeappliedtodeterminethetruenatureofthese cells. BO ThemosLnumerousnon-secretorycellinthetransitionregionwasacell devoidofanystrikingcharacteristicfeatures.Thesecellscorrespondtothe intermediatecellsdescribedinmanyspecies(JeffreyandReid,]'975;L971i BreezeandWheeldon,ISTT).Similarcellsinhumansandhamstershavebeen (McDowe' their undifferentiated appearance termed indifferent cells because of etaI.,I97BiBeccietal.,I97B).Inseveralspeciesthesecellshavedisplayedsigns cells or mucous cells (Breeze and wheeldon' of differentiating into either ciliated L97\iMcDowelletaI.,I97B),andhavebeenmorecommoninregeneratinqres- piratoryepithelium(McDoweIIetaI.,IgTB).Itappearslikelythatthenon- function' sea lion perform a similar replacement secretory cells of the Australian asoccasionallythesecellsindicaiedsomedifferentiation.Thesecellscontaineda cells; to lhose found in the secretory serous-like small number of granules similar displayed signs of ciliogenesis' however, none of the cells observed Theciliatedcellspresentinthetransitionregion,andthoseinthelobular bronchi,showednoessentialdifferencestoLhosefoundinterrestrialspectes 1977)' (Rhodin, I966i I974; Breeze and Wheeldon' Thenon-secretorymicrovillouscells,withtheirdistinctivemicrovilliand prominentaxialfilaments,werepresentinsmallnumbersinboththetransition dis- contained many organelles, the most region and the Iobular bronchi. They tinguishingofwhichwerethebundlesofcytoplasmicfilaments.Theappearanceof terrestrial the brush cetl in the airways of many these cells was similar to that of mammals(Baskerville,lgT0;Rhodin'I914;JeffreyandReid'1975;Breezeand wheeldon,rgTtiAllan,rgTB;Kennedyetal.,rgl}iHydeetal',1979)andinthe 1968;Weibel' I973i Hijiya et al., 1977)' alveoli of the rat (Meyrick and Reid, Thegranularmigratorycellsobservedintheepitheliumofthelobular bronchiappealidenticaltosomeofthemigratorycellsofthesubmucosaandto cells alveolar capillaries' of the two miqratory granulocytes observed within the globule species' the lymphocyte and the described in the epithelium of other 1977), those found 1975; 1977; Breeze and Wheeldon, Ieukocyte (Jeffrey and Reid, 81. granular globule leukocyte' Globule in the lobular bronchi most resemble the large, refractile qranules and may possess leukocytes are mononuclear cells with are considered to arise from the subepithelial microvillous projections. These cells (Breeze and wheeldon' I97Ð' The migratory cells mast cells which they resemble electron-dense a bi-lobed nucleus, unif orm Iy in Neo ohoc a. however, Possess granulesandnocytoplasmicprocesses.Thedistributionofthesecellsin Neooho Câ. togetherwiththeirstructure'indicatethattheyaremostprobably generally regarded as containing a crystalloid eosinophils. Eosinophil granules are inclusions has been observed in the eosino- inclusion, atthough the absence of such (Schalm al', I975)' Eosinophils have been found phils of the horse and the cow et tooccurparticularlyinlooseconnectivetissueassociatedwithvarioustypesof epithelium,includingthatoftherespiratorytract(Schalmetal.,1975),aS observedhere.Parasitism,particularlyinassociationwithmigratinglarvae' causesaneosinophilia(Schalmetal.,I97Ðwhichisacommonfindinginthe Iion (Needham et al'' 19B0)' These findinqs' peripheral blood of the Australian sea togetherwiththeobservedpulmonaryparasiticinfestationoftheseanimals (Chapter5),wouldsuggestthatthesecellsareeosinophilsratherthanglobule Ieukoc Ytes. Thereweresixtypesofepithelialcellsobservedinthelobularbronchus. Themostprominentfeatureofthecuboidalsecretorycellsofthelobular bronchiwasthelargenumberofsecretoryvesicles.Thesevesiclesqenerally were similar in appearance to the lamellar contained osmiophilic Iamellae which cytosomesfoundinthealveolartypellcells.Thecolumnarsecretorycellsonthe but any, vesicles with osmiophilic lamellae other hand, contained very few, if rathertheycontainednUmerousprofilesofendoplasmicreticulumandan occasionalsecretorygranule.Thesecellspossessedmanyofthefeaturesdescribed of rals (Smith et al'' I974; Jeffrey for the non-ciliated bronchiolar (Clara) cells andReid,I975;SmithetaI.,I979;Plopperetal.,IgBOa),mice(Nidenand Yamada,1966;SmithetaI.,I979;Plopperetal.,l9B0a;PacketaI.,tgBt),rabbits 82. (cutz Plopper et al., l980a) and.humans (cutz and conen , rg7r; smith et al., 1979; al., 1975; Smith et al., 1979; Plopper eL al., and Conen , I97Il André-Bougaran et ].980c).Unlikeinmostterrestrialspecies,thesecolumnarsecretorycellsinthe secretory granules' However' similar small Australian sea lion contained very few in the non-ciliated bronchiolar cells of numbers of granules have been observed l980b), cattle (Plopper et al'' L980b) and dogs (Hyde et al., I97B; Plopper et al., granules are (Plopper et al" ferrets (Hyde et al', Ig79), whilst in cats no Present 1980b).Clara-Iikecells,inwhichsecretorygranulesarepresent,haveonlybeen describedinthreeothermarinemammalspecies,theSouthAustraliandophin (FanningandHarrison,IgT4;Fanning,Lg77)andtheWeddeltandcrabeater' (Welsch and Drescher, I9BÐ' The occurrence of Lobodon carclno ha US seals non-ciliated broncholar cells has been species differences in the appearance of et al', 1979; Plopper et al'' I9B0c; well recognized (Breeze et al., I976; Smith with differences within the one specles widdicombe and Pack, l9B2) together (Smithetal.,I974;Packetal',1981;WiddicombeandPack'I9BÐ' Thepresenceofcellsintheterminalairwayssimilartothealveolartypell regions of the respiratory bronchioles of cells has only been described in the distal the third order bronchioles of humans macaques (Castleman et al., Ig75), in bronchioles of the weddell (Andre-Bougaran et al., I975) and in the respiratory J-982)' Thus it would appear that the and crabeater seals (Welsch and Drescher, contains both clara-like cells and epithelium of the lobular bronchi of Neophoca alveolar tYPe ll-like cells' Thepresenceofthesetwotypesofsecretorycellinthelobularbronchiis and the extracellular lining layer in interesling in terms of lower airway function boththeairwaysandthealveoli.Thepresenceofanextracellularlayerinthe behaved similarly to alveolar sur- bronchioles of terrestrial mammals which (1970)' This layer has since been factant, was demonstrated by Macklem et al' observedelectronmicroscopically(GilandWeibel,IgTI;KuhnetaI.,1974)and hypophase covered by a thin, strongly was found to consist of an amorphous 83. osmiophiliclayer.ThisagreeswiththeEMappeatanceofalveolarsurfactant(GiI andWeibeI,1969),andthetwolininglayerswerefoundtobecontinuousatthe Kuhn and alveoli (Gil and weibel' L97I; transition between respiratory bronchioles et al., I97Ð- Theexactsourceofsurfactant,bothalveolarandbronchiolar.isinsome Clara cells it was produced by the bronchiolar doubt. Niden Qgi|)suggested that type ll cells' This suggestion has qained and then phagocytosed by the alveolar has and autoradiographic examination little support, and subsequent cytochemical with the non-ciriated bronchiorar cets, together red Lo the concrusion that the alveolartypellcells,contributetothehypophasewhilstthealveolartypellcells alonesynthesisethetensio-activesurfacefilm(CutzandConen,IgTI;Kuhnet a|.,I974;PetrikandColleL,Igl4;BreezeetaI.,I916;WiddicombeandPack. I9B!;GailandLenfant,Lgas).Theremayhoweverbeabronchiolarsourceofthe osmiophilicsurfacefilm,thatis,theoccasionallyobservedtypell-likecells presentintherespiratorybronchiolesofmacaques'humansandSomeseals (CasLlemanetal.,LgT5;André-Bougaranetal',1975;WelschandDrescher,lg82). Anotherpossiblebronchiolarsourceisthatofthenon-ciliatedbronchiolarcellsin (Nagaishi, I972; f igures have been observed which occasional osmiophilic lamellae BreezeandWireeldon,IgTÐalthoughnofunctionhasbeenascribedtothese. also found to the lobular bronchi of Neophoca were occasionat clara-like cells in contain osmioPhilic lamellae' Theperipheralairwaysofmarinemammalshavebeendemonstratedtohave greaterstabilityLhanthoseofIerres|rialspecieswhichhasgenerallybeen attributedtothereinforcementoftheseairwayseitherbycartilaqe,smooth I975i Leith' Denison et al., IgTIiKerem et aI., muscle, or bo|h (Scholander, 1940; I976;Viatthews,1977;KooymanandSinnett,IglglKooymanandCornell,IgBl). ThisreinforcementisconsideredtoenablenearCompletealveolaremptyingin drvinq to I97I) as was first proposed for animals excised lungs (Denison et al. , (Ridgway et al'' demonstrated in the porpoise depth (scholander, 1940) and since 84. Ig6Ð,Theairwayreinforcementisalsoconsideredtoassistinthemaintenanceof highexpiratoryflowrateswhichhavebeenrecordedfromtheseanimals(Kerem etaI.,L915;Matthews,ISTTiKooymanandSinnetL,Ig7giKooymanandCorneII, airways may' if they do release sur- IgBl). The type ll-like cells in the terminal factant,contributetotheabilityforalveolalre-eXpansionafterdivingtodepth' aswellasthegreaterflowproperties.However,preliminarysurfactantextraction studiesperformedonthelungsofWeddetlsealsfoundthattheamountof in terrestrial species and associated with surfactant was much less than that found of alveolar type II cells present this was an apparent reduction in the number (Boyd,Ig75).ThisisincontrasttotheobservationsofBoshierandHill(]974)and impressed by the relative abundance of Wetsch and Drescher (1982) who were alveolartypellcellspresentintheWeddellseal.IntheAustraliansealionthe alveolartypellcellswerereadilyapparentandfrompreliminaryobservations werefoundtooccupyapproximately5%ofthevolumeofthetissuecomponentof proportion of the type II cells was the lung parechyma. In the rat the volume nearly]0%oftheparenchymaltissuewheretheycomprised]4.57oofthetotal lgBI)' This finding' together with the parenchymal cell number (Haies et al., fraction of the lung and the reduced measured reduction in the parenchymal parenchyma (section would appear proportion of septal tissue within the 'l'I(b))' in the number of alveolar type II to indicate [hat there is an overall reduction lung' confirmation of this will require cells present in the Australian sea lion me of the cell types present in the determination of the number and volu there is a reduced number of type II cells parenchYma. Howevert if as is indicated, be compensated for by the type ll-like present in the alveoli, then this may well cells in the lobular bronchi' Thenon-microvillouscellsofthelobularbronchiweresimilartothenon- transition region but no indications of secretory intermediate-like cells of the or ciliogenesis were observed' These differentiation such as granule formation (Jeffrey and cells described in many species cells were similar to the intermediate 85 Reid,I975;BreezeandWheeldon'LSTTiJeffreyandReid'L971)andthe I97B) and hamsters (Becci et al'' indifferent cells of humans (McDowell et al'' l97B). of the muscularis are predominantly (ii) Muscularis: The smooth muscle bundles arrangedinaspiralfashion,andcontinuedistallytotheterminalcartilagerings ofthelobularbronchi.Thepersistenceofprominentmusclebundlesdowntothe terminationoftheairwaysisafeaturecommontoallmarinemammalsstudied I972; Denison and (Wislocki, 1935; Bélanger, 1940; Simpson and Gardner' 1973; Boshier and Hill, 1974; Fanning' Kooyman, I973; Kooyman, I971 Boyd' in Neophoca, as in other otariids' the Ig77;Welsch and Drescher, I9B2). However sphincters or valves as they do in smaller muscle bundles do not form myoelaslic (Wislocki' I973; Fanning' 1977)' the manatee cetaceans (Bélanger, I94O;Kooyman' (Bélanger, I94Oi Boshier and Hill' I974i Ig35i Bélanger, L940) and some phocids Australian Sea Iion, aS in other matlne Welsch and Drescher, IgB2). ln the beyond the terminal airways to form the mammals, this muscle layer continues ducts and sacs' pillars of muscle within the walls of the alveolar to the airway termination is a (ii¡) cartilage: The presence of cartilage down Sirenia and otariidae (Wislocki, L9Z9; feature unique to the lungs of Cetacea, 1935;Bélanger,l-940;SimpsonandGardnet'I97?iDenisonandKooyman'I973; Fanning, 1977). InNeophocathecartilageisfoundasirregularlyshapedplates lobular bronchus gives rise to alveolar which form distinct rings wherever the ducts. (b) ParenchYmal subdivisions lobules and acini in Neophoca is readily The division of the parenchyma into apparent.SinceallintrapulmonaryalrwaysintheAustraliansealionpossess by definition (Reid, 1967). There are cartilaginous support, they are all bronchi the walls of any airways' so no structures also no alveolar structures arising from equivalenttorespiratorybronchiolesarepresent,althoughsmallnumbersof squamoUscoveredcapillariesdooccurlnthedistalregionsofthelobularbronchi. 86. divisions are: the tertiary bronchus' In the Australian sea lion the terminal airway non-ciliated cells; and the lobular with pseudostratified columnar ciliated and predominantly non-ciliated' The lobular bronchus, with cuboidal cells which are from the terminal cartilage rings' This bronchus gives rise to the alveolar ducts arrangementdiffersmarkedlyfromthatfoundinterrestrialmammalsandsome the terminal cartilage-containing other marine species. In terrestrial species bronchioles which divide for several bronchi give rise to the non-cartilaginous are reached' Each of these gives rise to generations until the terminal bronchioles alveolar outpouchings in their walls' a number of respiratory bronchioles, with and sacs (Krahl' 1964; Rhodin' I974; Inselman beyond which are the alveolar ducts andMellins,lg8l).Respiratorybronchiolesarenotpresenl'nor'aswelldeveloped insomespecies,inwhichcasetheterminalbronchiolesleaddirectlytothe possess the general terrestrial arrange- alveolar ducts (Banks, IgBl)' Phocid lungs ment;however,thecartilagereinforcementextendsfurtherperipherally'downto (King and Harrison' I96l; the Ievel generally Lermed the "terminal bronchioles" l9B2) although Boshier and Hill f(ooyman, 1913; Boyd, 1975; Welsch and Drescher' before the smallest "bronchi" were (1974) described the cartilage support stopping general agreement that the respiratory bron- reached. Beyond this point there is the cetacea and sirenia all terminal chioles lead into the alveolar ducts' ln airwayscontaincartilagesupportdowntotheleveloftherespiratorySacor Wislocki and Belanger', J'940; Simpson alveolar duct (wislocki, 1929; 1935; L942i andGardner,IgT2;Kooyman,I973;FanningandHarrison,IgT4;Fanning,L977). Theterminalairwaysincetaceanshavemorerecentlybeentermedrespiratory segments (Fanning, L977) bronchi (Fanning and Harrison, L974) or sphincteric an extensive capillary plexus' In otariids because of the presence of cartilage and have also generally been referred to the terminal airways which contain cartilage r972i Kooym an, r973i Denison and Kooyman' as bronchioles (Simpson and Gardner, Ig7Ðbutapartfromthisdifferencetheterminalairwayarrangementinthose otherspeciesexaminedissimilartot'hatobservedinNeophoca. 87. connective tissue septa is in The partitioning of lobules and acini by distinct the presence and extent of interlobular contrast to terrestrial mammals in which septaishightyvariable,andinwhichnointeracinarseptaarefound.Distinct in cattle, sheep and pigs; in horses and complete interlobular septa are present absent in the dog, cat and monkey humans they are incomplete; and are et al'', 1975)' The lunqs of (Mclaughlin et al., L96I; Tyler et al., I97I; Mariassy which divide the lungs into lobules or acini cetaceans contain no connective septa (Fanning,L977),whilstinthelunqsofallpinnipedsexaminedtheseseptaare the lobule varies greatly (see particularly distinct although the definition of J-940; Pizey, r954; Denison et al'' r97L; section 3.2.r) (Hepburn, l,gr2; Bélanger, SimpsonandGardnet,LgT2;DenisonandKooyman,IgT3lBoshierandHill,I9T4; Boyd, 1975 Welsch and Drescher' l9B2)' Thepresenceofsuchextensiveparenchymalseptaislikelytoaffectatleast twoaspectsofrespiratoryfunction.Firstly,theseseptaformacontinuous surface and persists throughout the paren- network which originates at the pleural chymatoconnectwiththeadventitiaoftheairwaysandbloodvessels,andwith This is Iikely to assist the re- the interstitial component of the. alveolar septa' phase of inspiration and after diving to inflation of the alveoli during the initiat have a high expiratory flow rate and depth. It has been demonstrated that otariids the lung (Kerem et al'' L975; Matthews' an associated near complete emptying of (Denison et al'' 197I)' This would Ig77)with a great potential for alveolar collapse reopen the alveoli' The inflationary then require a stronq inflationary force to intrathoracic pressure which is force is exerted bY an increasingly negative diaphragm and the inspiratory intercostal generated by contraction of the oblique through the parenchyma by these muscles. This force could then be transmitted septawiththealveolarreinflationbeingassistedbytheforceoftheinspiredair Another factor which may assist in the and the bronchial and alveolar surfactant' of the connective tissue fibres, in eãrly reinflation of alveoli is the elasticity particular,thecollagen.Thismayresultfromtherecoilofthesefibresafter BB. complession'associa|edwithalveolarcollapse,toreturnthemtotheirrelaxed state.Ifcollagendoesinfactpossesssuchamechanicalproperty,thenthiscould beexpectedtobeofparticularsiqnificancewhenreturningtothesurfaceafter diving to dePth. prevenL collateral ventilation' In SecondlY, these thickened septa would identif ied which enable collateral terrestrial mammals, three pathways have been ventilation;however,notallpathwayshavebeenobservedinallspecies.These pathwaYs are: i)theinteralveolarpolesofKohnwhichallowcommunicationbetweenindi- vidual alveoli, described by Lambert (1955) which ii) the accessory bronchiole-alveolar canals and surrounding alveolar allow communication between distal bronchioles sYstems, and respiratory bronchioles in iii) the collateral ventilation which occurs between respiratory bronchioles dogs (vtàrtin, 1966) and through the intersegmental l9B0)' in humans (Andersen and Jespersen' marine mammals studied, no alveolar In the Australian sea Iion, as in other poresofKohnhavebeenobserved.DemonstrationofoLhercollateralpathways (Lambert, 1955; Martin' 196Ö or resin requires either extensive serial sectioning J-980)' In the present study no casting of the airways (Andersen and Jespersen, in the Iimited number of serial sections resin casts were made of the airways, and which would indicate possible collateral examined no channels were apparent pulmonary pathology (Chapter 5) it appears ventilabion. From examination of the between adjacent lobules' although unlikely that there is any collateral ventilation the same lobule' some may occur between acini within (c) Alveolar ducts and alveolar sacs Thewallsofthealveolarductsandalveolarsacshaveasimilarstructure. alveoli together with the rings of They consist of the openings of the individual and appear as pillars in the duct interstitial tissue which surround lhese openings 89. walls are generally thicker than those of the and sac walls. The pillars of the duct interstitium is continuous with both thaL sac and contain more interstitium. This the alveolar septa, as occufs in terrestrial of Lhe airway adventitia and thal of mammals(Rhodin,I974;Murray'I976;WeibelandGil'L977)' (d) Alveolar sePtum of Lhe alveolar septal surface and (i) Epithetium: The type I cells which line most osmiophilic lamellar bodies' have a similar the type II cells with their distinctive mammals (Rhod\n' I974i Weibel' I91)i appearance to those described in terrestrial Murray,I976;WeibelandGiI,IITT)andinothermarinemammals(Boshierand and Drescher' l9B2)' Hill, 1974; Boyd, I915;Fanning' I977; Welsch of the pulmonary capillaries in (ii) capillaries: The structure of the endothelium theAustraliansealionisverysimilartothatinterrestrialmammals(Weibel' I97:);Rhodin,]978).Thecytoplasmextendsoutfromtheperinuclearregionin and nU merous pinocytotic vesicles thin, non-f enestrated sheets which contain occasional m itochondria' which form the septal inter- (iii) Interstitium: The connective tissue components terrestrial species (Rhodin' L914; Hance stitium are similar to those described in most abundant components are the collagen and Crysta I, I9l5;Forrest, IglÐ.The andelasticfibres,togetherwithfibroblastswhicharemostlyapparentasthin cytoplasmicprocesses.Othercellspresentinsmallnumbersaresmoothmuscle cells containing contractile fibres cells, pericytes and mast cells. No interstitial in the septa of rats, lambs, monkeys and as described by Kapanct et al. Q974) examination of many septa' humans were observed despite extensive In Neophocathearrangementofthealveolarseptumpossessesfeatures common toboththeCetaceaandSirenia,thatisadoublecapillarynetwork,and single capillary network' in temestrial to the terrestrial mammals, that is a mammalsthesinglecapillaryplexusissupportedbytheseptalinterstitiuminsuch awaythatbothsurfacesofeachcapillaryareexposedtoalveolarairatsome (Nagaishi, I972; Weibel, I973; Rhodin' I97Ð' The poinr along the .septum 90. connecLiVetissueofthesep|aisgenerallyarrangedsothatonesurfaceofthe through contact with the alveolar epithelium, capillary endothelium is in direct the ept- the opposite surface is separated from Lheir fused basal laminae, whilsL of interstitium (Weibel, Ig7ll. In the lungs thelium by a variable amount of cetaceansandsireniansthereisaLhickconnectivetissuecorewhichliesbetween thecapillarynetworkoneithersurfaceoftheseptum,sothatonlyonesurfaceof eachcapillaryisexposedtoalveolarair(Fiebeger,1916lWisIocki,I9z9;1935; Simpson and Bélanger, 1940; Bélanger, L94o; Haynes and Laurie , 1937i Wislocki et aI., Harrison, I974i Fanning, 1917; Haldiman and Gardn er, I972; Fanning and 19B0).Thecapillalyallangementinphocidseal.sissimilartothatofterrestrial I912; Boshier and plexus only (SimRson and Gardner, mammals containing a single Hill,l974;Boyd,I975;WelschandDrescher,IgB2),whilsIinotariidsthereis generallyconsideredtobeasinglenetwork,althoughadoublecapillarysupplyhas N eophoca beenobservedintheStellelrssealion(SimpsonandGardner'I912)and eoohoca may contain a single capillary (Fanninq, IglÐ. The alveolar sePla inN alrangement,adoublecapiliaryartangement,ortheoneseptummaycontainboth arranqements.Nosignificantregionaldifferenceswerefoundinregardtothe or double capillary arranqements' occurrence of either the single ThemajordifferencebetweenthealveolarseptaoftheAustraliansealion within the organization of the two networks and terrestrial mammals is in the walls;thatisthecapillarynetworkandtheinterstitialnetwork.Severaltheories havebeenproposedtodescribethealrangementofthesenetworksinterrestrial mammals.Inoneofthese,thecapillariesaresaidtoformadensehexaqonal fibres, so the flat meshwork of the interstitial network which is intertwined with thatthecapillariesappeatfirstononesideoftheinterstitiumandthenonthe other(Weibel,L963;1971;WeibeIandGiI,LgTl).Thisisincontrasttothesheet- (Fung and sobin' 1969; Sobin et al'' I91O; flow model of sobin and his colleagues are model the walls of adjacent capillaries Rosenquist et al.. Ig73). ln this separatedbypillarsofconnecLiveLissuetocreateanextensive,essentiallyflat 9r. with blood flows' Ryan Q961; I973)' from work space through which a 'rsheet" of thehamsterlung,hassuggestedthattherearetwoseparatecapillarynetworks core of connective tissue with occasional isolated from each other by a central anastomosingbranchestoconnectthetwonetworks.Thisisinagreementwith generally mammals (Burr\' L974)' but is not observations of [he lungs of neonatal terrestrial mammals' Although' similar observa- considered to be the case in adult tionshavebeenmadeinthelungsofSomereptileswhichhavebeentermed recently Rosenquist (I981) ,,pseudo-single,, capillary network (Perry , I9B3). More hasshownihecollagencontentofthehumanalveolarwalltobearrangedpartly low model and parLly as described by the sheet-f as described by Ryan Q969; I97J) Rosenquist et al'' f973)' (Funq and Sobin, L969; Sobin et al'' I97O InNeophocathealveolarseptahavesimilaritiestoeachofthesedes- criptions.WhereLhereisasinglecapillarynetworktheseptalwallappearssimilar of the which the capillary crosses the midline to Weibel's description (1971), in septumtoexposethe,,thin,,portionoftheair-bloodbarriertothealveolarlumen ononesideoftheseptumandthentothealveolarlumenontheotherside.often portion of network the capillary occupies a major where there is a single capillary of ,,thick,,connective tissue Space scattered the septal waII with only smaII areas oneithersideofthecapillary.AlsopresentarealeaSofconnectivetissue betweencontiguouscapillarieswhichapproximatetherrpostS||ofthesheet-flow model. arrangement each network of capillaries In lhose areas of a double capillary central core of interstitium' capillaries are lies on either side of a well developed occasionallyfoundwhichclossthiscentralcoreandSoarecapableofqas exchangeonbothsurfaces.ThisarrangementissimilartothedescriptionofRyan (1969;IgT,andcorrespondstothereptilian',pseudo-single''arlangementdes- in the Australian sea Iion the capillary network cribed by Perry (I98l). However, whichlineseachsurfaceoftheseptumis,forthemostpart,complete'sothat therearefewareasof,,thick'|interstitiumexposedtothealveolarlumen. 92. lnpreliminaryobservationsofotherotariidspecies,amixtureofcapilìary found in the Steller's sea lion and the network tYPes, as in Ne oohoc a. was in the Australian fur seal and the California sea Iion, and to a lesser extent unpublished observations)' This apparent Northern fur seal (Nicholson and Fanning' parallels the evolutionary appearance of increase in specialization of the lung groups, Arctocephalus and the diverqent these species. The two earliest otariid identifiable from about six million years ago and U allorhinus, are first separately capillary plexus. The other extant together show the least amount of double speciesexamined,evolvedfromabranchoftheArctocePhaluslineabouttwoto 1977; Repenning et al'' 1979)' three million years ago (Repenning and Tedford, Thesespeciesshowagreaterproportionofdoublecapillarynetworkintheir alveolarsepta.Thismaywellrepresentaformofconvergentevolutiontoward thatofthecetaceansofwhichtheearliestextantformsevolvedsometwenty millionyearsago(Gaskin,19B2).However,somelowervertebratessuchasthe lungfish,amphibiansandSomereptilesalsopossessadoublecapillaryplexus I976 Hughes, I91B; Perry, l9B]) within |heir alveolar septa (Hughes and Weibe|, mammals (Krahl' 1964; Burri' 1974)' It as do late foetal and neonatal terrestrial hasbeensuggested(Fanning,Ig77)thatthePresenceofthedoublecapillary plexusincetaceansrepresentsapersistenceofthemammalianfoetalpattern.To the case in otariids will require exami- determine whether Lhis is to some degree material which is presently not available' nation of both foetal and neonatal lungs, ].6 SUMMARY ThelungsoftheAustraliansealionpossessseveralstructuralfeatureswhich include a generalized reinforce- differ from those of terrestrial mammals' These the parenchyma. The airways are reinforced ment of both the lower airways and bycartilageplatesandsmoothmusclebundlesdowntotheirtermination.These and cuboidal epithelium whose cells lobular bronchi are lined by a simple columnar aregenerallysimilartothosefoundinthelowerairwaysofterrestrialspecies. 93. tissue component The parenchyma is reinforced by an increased connective Each pulmonary lobule which forms pronounced interlobular and interacinar septa' number of distinct acini' This is supplied by a lobular bronchus and contains a alveolar septa so that these connective tissue thickening also occurs within the capillary network' septa carry a typical single capitlary network, a "pseudo-si¡gle" or a double caPillarY network' flow rates These structural modifications permit the high expiratory of the nearly completely exhibited by these animals, and assist in the reinflation divinq to depth' collapsed alveoli as occurs at end-expiration and after i I I 94. ChaPter 4 LUNG MORPHOMETRY 4.I INTRODUCTION Stereologyisthatbodyofmathematicalmethodswhichenablesthethree- from examination of two-dimensional dimensional interpretation of structures stereological principles to quanti- images of these structures. The application of (Weibel, I979a; EIias and Hyde' tate structures is the realm of morphometry morphometric estimation of lung I9B0). The develoPment of methods for the in the twenty years since the first structural parameters has progressed rapidly G962) and weibet (1961)' In this extensive examinations by weibel and Gomez periodtheprinciplesandapplicationsoflungmorphometryhaveexpandedtothe pointwheretherespiratoryorgansofmanyanimalspecies.bothmammalianand and various experimental con- non-mammalian, have been studied under normal ditions. have most often been morphometrically The structural features of lungs that the calculation of the structural pul- estimated are those which contribute to oxyqen' O'O, (Weibel' 1970a)' The monary diffusing capacity of the lung for pulmonarydiffusingcapacityisameasureofthelung'sconductanceforoxygen fromalveolarairtocapillaryblood.Thiscanbeexpressedbytheequation' (Bohr' r9o9 weibel' 1970a) oro, = i grle gor-Fóo, of O2 from air to blood by molecular diffusion where V..v2 is the flow tOO, is the 02 tension of alveolar air' and Þ"O, is the mean alveolar capillary blood 02 tension' Thestructuralpulmonarymodelfromwhichmorphometricparametersaredeter- been described by weibel (1963'' I97Oa' mined and the procedures undertaken have I979a)andWeibelandKnightf]964).Thestructuralfeatureswhichneedtobe capacity are the areas of the quantified for calculation of lhe pulmonary diffusing surface exposed Lo the air and diffusing surfaces, that is, the alveolar epithelial 95. of the diffusing barriers, particularly the alveolar capillary surface, the thickness barrier and the plasma barlier that of the tissue which forms the air-blood betweentheendotheliumandtheerythrocytes;andthevolumeofthealveolar capi Ilaries. AllorsomeofthesefeatureshavebeenstudiedinLhenormallunqsofa varietyofmammalianspecies.Theseincludeavarietyofspeciesofshrews' shrew (Gehr et al'' l9B0; weibel et including the smallest mammal, the Etruscan I97I; Hugonnaud et al., ]-977); the aI., l9B0); the mouse (Geelhaar and Weibe|, (Kennedy et I9B2); the syrian golden hamster Epauleted fruit bat (Maina et al., Burri et aÌ.,I974; Crapo et al., a!., ]978); the rat (Burri and Weibel, I97la,b; L97ka,b;Gehretal.'IgTBb,BurriandSehovic,IgTg;PinkertonetaI.,]-982);the dog(Siegwartetal.,IgTIiGehrandWeibel,IgT4;Gehretal.,l9Bla;Berendsen (Boatman et al'' L979); the horse and De Fouwt tgBZ); the pigtail macaque J-980; Gallivan' l9B1b); a ranqe of (GillesPie and Tyler, L9671 Gehr and Erni, fg8lb); and the human (Weibel and African bovids and viverrids (Gehr et al', Gomez,I962;Weibel,)'963;Dunnill,1964;Thurlbeck,1967iHasleton,I9l2;Weibel et al., I975i Gehr et al', 1978a)' a wide range in the values of the mor- The results of these studies present phometricparameterswhichisareflectionofthewiderangeinthebodysizeof Etruscan shrew with a body weight of only bhose species studied. Thus for the weight of Some 500 kg, the lung volume 2.5 g through to the horse with a body rangesfromabout0.]mlupto]TL.Similarlythestructuraldiffusingcapacity for the shrew up to 34'080 ml for oxygen ranges from 0.00018 ml O2.sec-l.mbar-l (Gehr et al., IgBlb)' Not all morphometric para- O2.sec-f.mbar-l for the horse increase in body size' The harmonic meters increase so dramaticatly with this up only increases from 0.27 um in the shrew mean thickness of the tissue barrier to0.60uminthehorse(GehretaI.,IgBIb).Thisreflectsthebasicneedofall animalstomaintainaneffectivebarrierofminimalthicknesstoallowo2 d if f usion. 96 Guantitationoftheapparentproportionalityofmorphometricparametersto bodysizehasbeenachievedbythedeterminationofallometricreqression equations.Thiswasfirstattemptedforawiderangeofmammalianspeciesby TenneyandRemmers(196]).Theyfoundthatlungvolumewasnearlylinearly proportionaltobodyweightandthatthealveolarsurfacearearelaLedtothe wholebodyoxygenConsumptioninasimilarfashion.Sincethattimeanumberof para- relationships for various morphometric studies have established allometric (Stahl, 1967i Weibel' Igl2'' 1913; Weibel melers, both between different species andGiI,IgTTiLechner,IgTB;WeibeI,I9796;GehretaI.,l9Blb;Heusner,l9B]) (siegwart et species with different body weights and between animals of the one aI.rI97I;Gehretal',198la;Weibeletal''IgBÐ'Theconclusionfromthese variousstudiesrvasthaLmostofthemorphometricparametersofthelungscale about linearlY with bodY weight' Thisinterestinmorphometrystemsfromitsabilitytodescribestructural featuresofthelungintermsoflungfunction.Hencethestructuraldiffusing capacityofthelungforoxygen'DLo',canbecomparedtothephysiologically determinedoxygenconsumption'UOr'thesameanimals'oranimalsofthe Samespecies(Weibel,L972,1973;WeibelandGiI,ISTT;Lechner,I91B;Gehret al.,l-980;WeibeletaI.,19B]).Comparisonoftheresultsfromthesetwomethods capaqity exceeds the physiologic oxygen indicates that the morphometric diff using consumption.Thisdiscrepancyarisesbecausethemorphometricmethodmeasures themaximumpossiblediffusingcapacity,asthemodelassumesthatallregionsof at the one point in time (weibel' the lung are venLilated and perfused maximally most morphometry IS presentlY the I97Oa, I91Ð. Despite this discrepancy' suitable methodavailabletogainquantitativefunctionalinformationfrompul- monarY structure' 97. 4.2 METHODS Tocalculatethestructuralpulmonarydiffusingcapacity,thedimensionsof estimated' These features include lhe area various structural features needs Lo be diffusing barriers' and the volume of of the diffusing surfaces, the thickness of the by measuring the following the alveolar capillaries. These features are estimated tissue in t'he lung' the alveolar and parameters: the volume of the respiratory capillarysurfaceareas'thethicknessofthetissueandplasmabarriers,andthe their definitions as applied in this capillary volume. Table 2lists the symbols and chapter. Table 2: List of morphometric symbols Vx: volume of feature x Sx: surface area of feature x th: harmonic mean barrier thickness thickness T: arithmetic mean barrier Vv(x): volume densitY of feature x x P(x): number of test Points on feature Prt total number of test points applied Sy(x): surface densitY of feature x with feature x I(x): number of test line intersections Lrt total length of test lines aPPlied feature x I: intercept length of test line on lengths lh: harmonic mean of intercept L: lung p: parenchYma within lung parenchyma alveolar surface or alveolar airspace within capillaries within ParenchYm a Hct: capillarY haematocrit otort oxygen diffusion caPacitY of lung Pto2: conductance of tissue barrier for oxyqen for oxygen OpO2 conductance of plasma barrier o"o, rrconductancett of erythrocytes for oxygen 9B 4.2.I Basic Principtes of Point Counting (a) Volume densitY were estimated by point The relative volume densities of lung components examined was covered by a test grid of counting of thin sections. Each section linesandpointsandthenumberofpointslandingonthatfeaturewerecounted. was determined by the equation: The volume density of that feature, Vy(x), (Weibel' L979a) Vy(x) = P(x)/PT feature x where p(x) is the number of poinls tanding on PT is the total number of test points counted' (b) Surface densitY density, of a feature was estimated by The relative surface area, or surface the surface of the feature with a series of counting the number of intersections of Sy(x), was then calculated using test lines of known length. The surface density, the equation: (weibel', r979a) sy(x) = 2.I(x)/L1 wherel(x)isthenumberofintersectionsoftestlineswiththesurfaceof the feature x LT is the total length of the test lines' (c) Harmonic mean bamier thickness Alsonecessarytothedeterminationofthelunglsconductanceforoxygenis (¡h), is proportional to the diffusion the harmonic mean barrier thickness which 1964)' This was estimated by the resistance of the membrane (weibel and Knight, equation: (Weibel and Knight 1964) rh = 213 . lt-t ' wherelhistheharmonicmeanofatlinterceptlengthsofthetestlines. 4.2.2 SamPling Design Amulti-levelsamplingdesign,withfourlevelsofincreasingmaqnification (19BIa)' Stratified sampling (Gehr et al'' was used, as described by weibel 9l-4. 1978a)waSalsoapptiedtodetermineifanyregionaldifferencesexistedas 99 and weibel, I97Ð and the described for the dog lung (Glazier et al., 1967; Gehr the sampling design is presented in horse lung (Gehr and Erni, l9B0). An outline of Fiq.B0.Throughoutthisdescriptionofthesamplingtechniquestheterm aveoli and alveolar sacs' and parenchyma refers to: the alveolar septa, air in the and interlobular septa' The the respiratory surface lining the abundant interacinar include: the airways' remaining features, which constitute the non-parenchyma thicker connective tissue the blood vessels larger than capillaries, the core of septa, and the Pleura. and ttcoarset' non- At level I the volume density of "coarse" parenchyma' parenchymalstructureslargerthan].mminthelungwereestimated.Atlevelll less than I mm' in the the volume density of "fine" non-parenchymal structures, estimation of the ,,coarse" parenchyma wele estimated' Levet III was for the parenchyma' At level IV volume density of respiratory surface within the "fine" were estimated (weibel et the parameters of features within the alveolar septa al., l98la). (a) Level I TheleftlungofeachanimalwasdividedintothreestrataasusedbyGehret strata consisted of an approxt- at. (1978a) for examination of the human lung' The middle and basal regions of each mately I cm thick slice taken from the apical, lung (Fiq. B0). knife. The first slice was The lungs were sliced with a J5 cm microtome of the lung' The bulk of taken by cutting off, approximately, the first one-third thelungwasthenplacedcutsurfacedownonapelspexcuttingboardwithlcm to cut the slice' A similar technique high borders along which the knife was moved was used for each subsequent slice' enlarged to twice life- Both surfaces of each lung slice were photographed, size,andprinted.Theprintsfromoneanimalwereinitiallyexaminedonalight transparent overlay with lines 22 mm box with a weibel multi-purpose test system that very few points fell on in length (weibel et al., Lg66).This led to the finding Figure 80. Diaqram outlining four-level morphometric sampling design 100 Level I 2x Lung volume il ll il il il il il \ o GD oo rP o Level It 2Ox Level Itr 2OOx Level trt 9,2OOx po the test area, so a f-cm squate labtice grid overlay was also used. The dete intersections of the horizontal and vertical lines, were counted to larger than I mm whether they fell on parenchyma, or non-parenchymal structures life -s i ze. To record the number of points on each feature a ten channel electronic used (designed and counter in two rows of five, with summation for each row, was of points built by Adelaide university Medical school workshop). The total number the slice, so that the Ianding on any one lung slice was determined by the size of total number of points varied between slices' the range The results for the two grid systems were compared by estimating for the true volume density, using the equation' (Exner, Vyt aVy=Pp* PP . (l-PP) PT I976) whereVVisthevolumedensityofthefeatureofinterest' volume density' A VV is the estimated range of the PPisthevolumefractionofthefeatureofinterest,and PT is the total number of test points counted' was subsequently used The I cm-square lattice grid proved to be more precise and for all animals. (b) Level II 2.5 x From each lung slice used in level I two blocks of tissue, approximately one from the 1.5 x 1.0 cm, were selected; one from the dorsal end of the slice and in paraffin ventral end (Fig. B0) (Gehr et al., 1978a). Each block was embedded at 20x wax from which a 2 um section was cut, stained with H & E, and examined into a Durst magnification. The procedure undertaken was to mount each section 55 mm lens' The enlarged L1000 Photographic enlarger with point light source and points falling on image was projected on to a test grid of points. The number of as were non-parenchymal structures less than l- mm diameter, were counted, were: points falling on parenchymal features. The parenchymal features counted the alveolar septa, the alveolar airspace, the duct airspace, the intersaccular' interacinar and interlobular connective tissue septa. 102. animal and the Three grid systems wefe tested on one set of slides from one test system ranges for the true volume densities estimated. The most efficient as employed, con- was subsequently used for all five animals. The term efficient but also takes into siders not only which grid resulted in the greatest precision, (Mathieu et al'' consideration the time taken to perform each series of counts 19Bl). The three grid systems used were: i) a f cm-square lattice' ii) a Weibel multi-purpose test system with lines of 1l mm lenqth, and iii)aWeibelmulti.purposetestsystemwithlinesof22mmlength. test system The totat number of points per section was determined by the points counted used and the area of each section, so that the total number of varied between each section. to be the The weibel multi-purpose test system with lines of ll mm proved four animals' most efficient, in the trial procedure, and was used for the other (c) Level III microscopy with a The z pm sections used in level II were examined by light of the Wild M50l Sampting Microscope (Wild Heerbrugg, Switzerland; courtesy to the Department of Anatomy and Histology, Adetaide university)' Attached which was placed microscope was a ground-glass viewing screen on the surface of in lenqth' Points a 42 point weibel multi-purpose test sysLem, with lines lB mm the falling on single alveolar septa, double alveolar septa, intersaccular septa, epitheliun and non-parenchyma were respiratory ¡i-'*"änd the core of larger septa, counted. tissue septa In morphometric studies of terrestrial mammals all connective The reason for have been included in the non-parenchyma (weibel et al', lg8la)' this is that in most terrestrial species there are few connective tissue septa tissuerhey when present ,nyg"s-exchange may carry would be insignificant "nd volume of epithelium sea lion lunq there are compared to the ;"t-"h;teitator!'- A-.ln the Australian integral part a large number of interacinar and interlobular septa which form an t0l. of these septa is Iined by res- of the alveolar structure. In section each surface to the total respiratory tissue piratory epithelium and capillaries which contribute many of these thicker septa had been within Lhe parenchyma. At levels I and II includedwiththeparenchyma.Thiswasbecausetheresolutionattheselevelswasepithel iun the point landed on the respirator/ or insufficient to determine whether a A to For this reason it was considered necessary connective tissue of the septa' include these structures at level III' Todeterminethemostsuitablemagnificationandthemostefficientnumber counts was undertaken using one section. of fields to be eXamined, a trial series of screen magnification' 60x' l20x' and 240x' This was examined at three levels of ThewholeIissueareaofthissectionwassampledbyconsecutivefieldsofviewat for each successive ten fields summated' each magnification level and the results Thevolumedensitiesforthefeaturescountedweredeterminedforeach groupoftenfields,andthentheprogressivemeanforthetotalnumberoffields plotted on the graph were the t 570 limits was calculated and plotted on a graph. for the volume density of the paren- of the final value for each feature, except were used' chymal airspace where + 1% limits ThegraphsforallfeaturesweÎeexaminedtodeterminethenumberoffields the plotted limits' From these trial where the values consistently fell within countsitwasconcludedthatthemostefficientmethodwouldbethecountingof seventyfieldsfromeachsectionexamined,atthemagnificationof24ox. being undertaken the eyepiece lens of the However, prior to the final counting microscopewaschangedfrom6xtolOx,duetoatechnicalproblem.Asaresultof thischangeamagnificationof240xwasnolongerpossiblesothatthenearest subsequent increase in the number of fields magnification of 200x was used with a persectiontoeighty.lnthiswayatotalof3,36opointswasappliedtoeach section. sampled (Weibel, 1970b) to cover the com- Each section was systematically a distance of approximately r'75 mm' plete section with each field separated by t04. landed on non-parenchyma was Any field in which more than half the points above (weibel et al.' lg8la)' and the excluded, as these were included in level II section moved to the next field' (d) Level IV Fromthemedialaspectofeachblockselectedforlevelllasmallslicewas diced into l-l mml pieees and taken for EM examination. This tissue was processedasdescribedpreviously(Section2.5.2).Ultra-thinsectionsfromone randomlyselectedblockfromeachofthesixregionsofeachlungweremounLed ethanolic uranyl acetate and Reynolds on 200-mesh cu/Rd grids, stained with 50% electron microscope' All micro- lead citrate, and examined in a Phillips EM 100 magnification' graphs were taken at approximately 1'400x most suitable method for subsampling (i) subsampling method: To determine the from each of the six regions of sections a trial was conducted using one section fromanimalAs .Twomethodsofsubsamplingwerecompared,one,aligned and the oLher' a form of systematic systematic quadrat (sa) (weibel et al., 1966) (SAWG) (Cruz-Orive and Weibel' 1981; quadrat samPling with area-weighting Müller et al., 19Bl). Inthealignedsystematicquadratsubsamplingmethodthegridsquereinthe UpperleftsideofthesectionistranslatedtothecentreoftheEMscreen.The Ieft hand corner of that qrid square image was then translated so that the upper position of the image was deter- would be included in the micrograph. The exact and the left of that grid square with the mined by aligning the grid bars at the top topleftandbottomleftortoprightmicrographareamarkerlinesetchedintothe EM screen (see Fig' 6, Weibel et al'' 1966)' Thesectionimagewasthentranslatedtothenextfifteenconsecutivegrid each grid square aligned as described' squares and the upper left hand corner of in a 4 x 4 pattern (see Fig' 6a' Cruz-orive The sixteen grid squares were arranged in the sampled area' no micrograph was and weibel, lg8l). If there was no tissue of sixteen micrographs were taken bu[ this was noted. In this way a maximum r.05. number varyinq, in this trial, from eight possible for each sec[ion with the actual to thirteen. Theunaliqnedsystematicquadratsamplingwithtissue.weightinq(STWa) area-weighted quadrat (sAwa) system was devised to approximate the systematic (I9Bl) and Mütler et al' (l98l)' The sAwG described by cruz-orive and weibel systemwasdesignedtoincludeasmuchofthereferencephase'thealveolar septa,aspossibleineachrandomelectronmicrograph.TheSTWGsystemwas when septum was present in the grid similar to that for sG subsampling; however, squalecornertheimagewasrealignedtoincludeasmuchaspossibleoftheseptal corners of grid squares which contained tissue within the area of the micrograph. notissuewereignored.Afixednumberofeightmicrographswastakenofeach found in the upper Ieft hand corners then section, and if insufficient material was in turn to the Iower right, Iower left, and a similar sampling procedUle waS applied ifnecessaryUpperrighthandcolnersofthegridsquares.Thismodificationtothe as the facilities available did not allow sAWG subsampling system was necessary fortheappropriatealterationstotheEMscfeenasdescribedbyMüIIergt-e!. from one region of animal AS4' (1981). Figures 68 to75 represent an sTWG sample Themicrographsfromthetwosubsamplingmethodswerecontactprintedon (Du Pont) together with the "waffle" strips of 70 mm wide ortho s Litho film (Ladd), which was photographed at the end of calibration grid, of 2,4OO lines/mm microscope. The strips of positive imaqes each recording session on the electron (weibel et al'' 1966) at a final magnification were examined with a projection unit ofapproximatetyg,200x.MountedontheScreenoftheprojectionunitwasa Weibelmulti-purposetestsystemof84lines22mminlengthwhichwasusedto the structures within the alveolar determine the volumes and surface areas of septa. UsingtheresultsfromthesetwoproceduresthevaluesforVc,saandSc two methods: were calculated from the whole lung using r06. parenchymal volume, and the t) Lwo-level sampling using level I values for resultsfromsQsubsamplingofelectronmicrographsasthesecondleve]; the sTWG subsampling at level IU four-level sampling using the results from four. was subsequently used to lr4ethod two resulted in the Ieast standard error and examine all sections. the harmonlc mean (ii) Harmonic mean barrier thickness: For estimation of (thp) an appropriate logarithmic bamier thickness of the tissue (tþt) and plasma micrographs from four sections scale had to be devised. To do this a total of eight ofanimalAs4wereanalysedwiththeprojectionunitmountedwitha2cm-square tissue thickness were lattice grid (weibel, 1970a). The intercepts measured for the horizontal direction' from where the lines of the grid, in either the vertical or to the Iumenal surface of overlaid the airspace surface of the epithelium through was from the the endothelium. For the plasma thickness, the intercept measured (see Fig' 5, Weibel' surface of an erythrocyte to the lumenal endothelial surface 1970a).Inthosecasesofadoublecapillarynetworkintheseptum,onlytheplasma thickness toward the thin part of the septum was measured' wele made of From these micrograpl'ts 2J6 measufements, in millimetrest marked in 0'05 mm the tissue and plasma thicknesses using a pair of dial calipers d iv isions. (1/l) as were the The reciProcals of these measurements were calculated reciprocal values ranged logarithms of the reciprocals (log l/t). The logarithmic smallest' The log Ul values from -2.00 for the greatest thickness, to +1.00 for the indicate the distribution of were plotted on graph paper at 0.02 unit intervals to 0'1 units which gave 2) values. The log l/l values were classed into intervals of to accommodate the few evenly-sized classes and two slightly larger end classes of classes the log 1/l extreme values in either direction. To decrease the number 0'15 units and 0'25 units' values were regrouped into scales with intervals of at either end were forming seventeen and eleven classes respectively' The classes again slightlY larger. 107 the eleven and seventeen class The linear values of the class limits from upper class limits starting from zero and scales were calculated. These formed the finishingatonehundred.Thetwoscaleswerethendrawnat4xenlargement, photographicallyreducedtotheircorrectsizeandprintedontoKodalithsheet film (Kodak). as the original Each scale was then applied to the same micrograPhs measurements.Thezeropointwasappliedtooneendofeachinterceptandits immediately beyond the other Iength classified into the class with its upper limit I979a)' Several of the smallest classes end of the intercept (see lig.5.2O, Weibel, due to problems with in both scales were not included on the drawn scale the calipels were set at the resolution between the lines. Instead, for these classes Upperlimitforaparticularclassandtheinterceptlengthclassifiedinthesame manner as for the drawn scale' were calculated using the results The values for th for tissue and for plasma for the original linear from both the eleven and seventeen class scales, and was calculated from the equation' measurements. The harmonic mean thickness (Weibet and Knight 1964) rh = Zl3 . lh ' intercept length' where Ih is the harmonic mean for each scale the linear mid- To calculate the harmonic mean intercept tength the linear uppel class Iimits' point (tm) of each scale had to be determined from The value for Ih was then derived usinq and the reciprocal value, J./lm, calculated. the equation' (weibel' ]e70a) t/ th , t" . 1/ Im t" = ,[, ]/ ,[, where]/Ihisthereciprocaloftheharmonicmeaninterceptlength, fc is the frequencY of class c' and c' l/lmc is the reciprocal of the linear midpoint of class plasma thicknesses the harmonic For the linear measurements of the tissue and mean was calculated from the equation' (weibel'l97oa) 1/¡¡=l/n î. Llli I =-L 109. equa[ion was calculated by Each of the values on the right hand side of the above substitution of the appropriate values into the following equations: oro, = Kt. -j# oPo2-KP'-T-=ñSc+Se (Weibel' l97oa) Den^ o6r.Vc -/- = where KtandKparethepermeationcoefficientsfor02intissueandplasma resPectivelY' Og, is the rate of O2 uptake by whole blood, and ery- Sa,5c and Se are the surface areaS of the alveoli, capillaries throcytes, resPectivelY, plasma barriers' rht and thp are the harmonic thicknesses for tissue and respectivelY, Vc is the volume of the capillaries' were' For the physical coefficients' Kt, Kp and O2, the values used and Kt = Kp = 4.I .l0-I0cmZ'sec-f'tbar-f, (Gehr l98lb) oo2= 1.87 . l0-2mt 02.rl-1.r""-1.mbar-I "t "1., was estimated (ii) Average capillary diameter: The average capillary diameter capillary is from the ratio Vc/Sc. The model for this estimation assumes that each for the a uniform cylinder (Gehr et al., 1978a). The derivation of the equation capillarydiameterisbasedonthecylindricalmodelwhere' c I inder vol uîe Vc) cyli er surface area (sc and where T is the radius of the cYIinder, L is the length of the cYIinder. which comes the This equation reduces lo the expression Vc/Sc - tl2, from average capillary diameter, d = 4 ' Vc/Sc in microns' of relative (iii) Index of alveolar diameter: The ratio St/Vp is an indicator that the alveolar diameters (Maina et al., IgB2). A hiqh ratio indicates 5a Vy" and Vp V¡ parenchyma is divided into small alveoli. In this ratio st = ' = ' diameter index was VVp, ro that St/Vp is expressed in rn*2/rntl. The alveolar 11"0. for several other estimated for the Australian sea lion, together with those species for comParison. 4.2.7 Animal and Regional Differences the six To test for differences between the five animals, and between with the results for the sampling sites, the results for levels I, II and III, together surface densities from capillary volume density and the alveolar and capillary the HSD ANOVA II Ievel IV were analysed using nested analysis of variance, personal com- package (Madigan and Lawrence, l9B2) (courtesy of L' Richards, m unicatioh). morphometric para- Regional differences in the absolute values of various (sokal and Rohlf, 1969) meters were tested for using a t-test by paired comparison or anterior and posterior' based on the difference between the dorsal and ventral, values for each Parameter' of the absolute For those morphometric parameters in which the calculation body size, only the results for value required the lung volume, which is related to animalsAsLJr24r25and26wereconsideredforinvestigationofregional Sc' Vc' Vc/Sa' the differences (Section 4.4'2(a))' These parameters included Sa' and otor'Those estimated capillary diameLer, the alveolar-diameter index, the five animals parameters whose regional differences were compared between werethearithmeticandharmonicmeanbarrierthicknessesofthetissueandthe the proportion of plasma, and the haematocrit. Also tested were the results for double capillary septa estimated at Ievel III' 4.2.4 IntersPecies ComParison compared to results The results for the Australian sea lion population were between species from other studies on terrestrial mammals' In the comparisons a species mean' only the results from four of the sea lions were used to establish animal AS4 due to This was due to the fact that no weight was measured for Most methods of transport problems encountered during the first field trip' for this reason the results species comparison ate done on a body weight basis and from animal A54 were excluded' tll. to quantify interspectes Allometric regression equations have been used et al' (r98lb) have presented allomelric comparisons of morphometric data. Gehr parameters from a wide range of terrestrial regressions for several morphometric population' The data from the present study mammalian species, their'rextended" wereplottedonlog-toggraphsdrawntoshowtheregressionlinebasedonthe by Gehr g!--g!. (198tb)' This enabled allometric regression equations presented visualappraisalofthedifferencesbetweenthesealionsandtheterrestrial differences between the sea lions and species studied. To test for any significant terrestrialspeciesthepredictedvalueforeachmorphometricparameterwas calculatedbysubstitutingtheweightofeachSealionintotheappropriate (sokal and Rohlf' 1969) was allometric equation. A t-test for paired comparisons between the predicted values and then used to test for any significant difference parameter' the measured values for each morphometric is presented after the results' Discussion of the methods (Section 4'4'I) 4.' RESULTS 4.5.1 SPecies Results (a) Level I volume densities TheresultsforlevellarepresentedinTable].Themeanvalueforthefive lung was non-parenchyma and animals indicated that about z3o/o of the "coarse" 77Vo was I'coarse'r ParenchYma' (b) Levet II volume densities TheresultsforlevelllarealsoshowninTableJ.Forthepopulationabout non-parenchyma with about 93o/o'rfine' 7Vo of rrcoarse" parenchyma was in fact parenchYma. Thetotalvolumedensityoftheparenchymainthelung,Vv(p2,¡)isgivenbY The results for this are shown in the equation, Vv(p2,¡) = Vv(Pl,¡) ' Vv(p2'p1)' Tabte].Thepopulationmeanforthetrueparenchymalvolumedensityinthelung was 7I.5%. IT2. Table 5: Morphometric results - levels I and tr Level I Level II Level I x II True Parenchyma animal Coarse Parenchyma Fine Parenchyma P1 tSE P2 tSE Vv(t2,¡) 4 o.7 69 0.017 o.924 0.008 0.711 23 o.738 0.028 0.940 0.008 o.694 24 o.7 62 0.018 o.934 0.010 o.7I3 25 0.770 0.01_5 o.9I5 0.006 0.705 26 0.8il_ 0.015 o.925 0.002 0.750 popn o.7702 o.9276 o.7146 ISE 0.0117 0.0041 0.0094 (c) Level III volume densities The results are presented in Table 4. The results for level III indicate that ePithel ium' I3.7Vo of the parenchyma was occupied by respiratory /\, . This value gas-exchalge tissue included those portions of" which lined the interacinar and the interlobular septa. Of the interacinar^ septa approximately 3lo/o of their volume was occupied by respiratofypttfittTïhitst for the interlobular septa this was only 1070. Of the alveolar septa about 27Vo conlained a double capillary network, whilst the remainder contained a single capillary network. Table 4: Morphometric results - level trI Level III Total voluo E¡action VoImo fraction of Volwo f¡action of Volue fraction of of respiratory epitheliw respiratory epitheliu ¡espiratory epiEheliu double capillary sopta animal of interacinar septa of intê¡lobular septa +SE V +SE ü +sE +SE 4 o.I32 0.010 o.286 0.001 0.091 0.001 0.270 0.017 23 0.099 0.006 0.326 0.016 0.118 0.015 0.210 0.014 24 0.165 0.007 0.327 0.011 0.115 0.004 o.284 0.011 25 0.148 0.007 0.296 0.01_1 0.094 0.004 o.2t7 o.o24 26 0.1_41 0.009 o.298 0.015 o.097 0.006 o.292 0.011 popn 0.1170 o.3066 0.1010 o.2666 +SE 0.0109 0.0084 0.0056 0.0147 ll l. (d) Level IV volume and surface densities TheresulLsfor[hevolumeandsurfacedensitiesestimatedattheEMlevel the haematocrit values' are presented in Table 5, together with (e) Absolute values 6 presents the absolute values of the (i) MorPhometric lung Parameters: Table presents oxygen diffusing capacity' Table 7 morphometric lung parameters and the theabsolutevaluesoftheharmonicandarithmeticmeanbarrierthicknesses. (ii)Capiltarydiameter:Theestimatedcapillarydiameterforthesealionwas approximatelY 7.5 ¡rm' alveolar diameter in the sea lion was (iii) Alveolar diameter index: The index of calculated as 32 mm2/mmJ' 4.1.2 Animal and Regional Differences (a) Level I Atlevelltherewereanterior-posteriordifferenceswithinthesealionsata 5Tolevelofsignificance.However,therewerenosignificantdifferencesbetween the animals (Table B)' (b) Levet tr between animals' nor At level lI there were no significant differences (Table 9)' between the different sampling sites (c) Levets III and IV between animals' The At both levels III and IV there were differences level for level III (Table L0)' For level iV differences were significant at the l% capillary volume was significant at the 5% the difference between animals for the levelandthedifferenceforcapillarysurfaceareabetweenanimalswassigni- showed no significant difference ficant at the l-% level. The alveolar surface area between the animals (Table 1l)' sea Iions at levels III and IV, the Due to these differences between the individualvaluesforeachanimalwereusedthroughouttocalculateeachpar- ticularmorphometricparameterbeforethepopulationmeanwascalculated. TherewerenosignificantdifferencesbetweenregionsateitherlevellllorlV. Table 5: Morphometric results - Ievel [V Level IV +SE Sy(e,s) Vy(c,s) +SE Hct Sy(c,s) *SE +5E Sy(a,s) +SE -I anim al _'l cm -1 cm cm* o.5tt o.o35 457 o.490 o.o3t t1642 L86 tr95t 4 3r34t 34t+ 0.021 o.576 0.011 5,O87 385 o.549 IB2 51163 25t o.oz5 2t 2,953 o.660 0.028 o.67t L?I 6,566 t89 2,4ro 290 2,737 0.485 0.0r4 ?4 3rz 0.591 o.o24 ?r875 256 4r2O7 25 21559 zTB o.o52 o.474 o.o5J 3,4t8 665 o.476 100 3,598 249 26 2,968 o.5532 o.5482 3,^Ot.o 4,650.? 0.016r popn 2,846.6 0.0lrB T83.7 548.t *SE L65.2 P -ñ Table 6: Morphometric results - absolute values Absolute Values Vc Vc/Sa oto, Sc Se Mb VL 5a -f . tbar-I 2 ml mtlm2 ml O2.sec animal 2 m m2 k9 L m 2..337 r44.3 I7B.B L.47 ].BB r22.O Lt3.O 4 282.1, r.B6 3.416 151.8 162.5 26r.4 ?3 60.6 7.49 6.111 527.r 529.7 2.7 4 6.84 L9J.5 2L9.7 24 65.3 2.3L 4.809 22L.5 324.r 455.O 87.1 7.37 I97.I 3.992 25 22I.I 306.L 1.60 6.10 L90.9 2JL.4 26 95.O J50.34 r.996 4.IJJO L7r.O6 L9J.62 295.60 0.6375 popn 77.00+ 6.336 64.8I 62.91, o.2i5 o.66L 14.75 19.40 tSE 8.33 * Mean calculated from four animals only' thicknesses Table 7: MorPho metric results - barrier îin ten +5E 1 tissue rep rSE rSE tissue th plasma rSE +SE um ìlm th *SE um um animal um um O.t+38 o.o29 o.385 o.022 o.644 o.o47 o.o79 0.0I0 2.932 o.205 4 0.601 o.034 0.110 0.017 o.454 o.o37 O.7 t+5 0.117 0.090 0.005 3.344 o.222 23 0.598 o.o23 0.1r6 o.327 0.020 o.t39 o.360 o.o32 0.675 o.o23 0.054 0.004 4.009 24 0.501 16 o.069 o.389 o.o36 0.210 o.433 o.o29 o.7 0.05r o.r25 0.0r9 3.746 25 0.680 o.706 0.11-l- O.]BB 0.040 o.233 o.55r 0.061 0.0J1 0.087 0.019 J.t37 26 o.575 o.6972 o.3704 3.4336 0.4J66 0.5910 0.0870 0.0r74 o.0232 popn 0.1970 o.o33L o.o2B7 0.0rr4 P tSe P 116 Table B: Level I - analysis of variance of Mean Variance Source of Degrees of Sums squares squares ratios v ar i ance freedom 0.0r7 0.004 I Between animals 4 0.040 0.004 4* APa within animals 10 0.001 AB within AP 15 0.015 Total 29 o.o72 xp(0.05 a: A - anterior slice of lung P - Posterior slice of lung A - aPical surface of lung slice B - basal surface of Iung slice Table 9: Level II - analysis of variance Variance Degrees of Sums of Mean Source of squales ratios variance freedom squares 0.0006 2 Between animals 4 0.0024 0.0014 0.0001 I APa within animals IO 0.0044 0.0001 DV within'AP L5 Total 29 0.0r02 a A - anterior slice of lung P - posterior slice of lunq D - dorsal region of lung slice V - ventral region of lung slice Tabte I0: Level III - Analysis of variance Mean Variance of Degrees of Sums of Source squ ares rat ios varlance freedom squares 0.0016 9*x Between animals 4 0.0141 0.0041 0.0004 L' APa within animals t0 0.0049 0.0001 DV within AP l5 Total 29 o.0235 xx p ( 0.01 a: A - anterior slice of lung P - Posterior slice of lung D - dorsal region of lung slice V - ventral region of lung slice 117. Table Il: Level IV - analysis of variance Mean Variance Source of Degrees of Sums of squares ratios Parameter variance freedo m squares o.r23 0.01r 4.4+ Vv" Between animals 4 0.007 l APa within animals IO 0.074 0.007 DV within AP É 0.105 Total 29 o.toL 0.010 ?.5 ,Vac Between animals 4 0.018 0.004 0.8 AP within animals 10 0.019 0.005 DV within AP I5 0.074 Total 29 0.151 0.0r9 9.5x* ,Vcc Between animals 4 0.074 0.002 0.4 AP within animals t0 0.02r 0.005 DV within AP 15 0.075 Total 29 0.170 *p(0.05, xxP(0.01 a: A - anterior slice of lung P - Posterior slice of lung D - dorsal region of Iung slice V - ventral region of lung slice (d) Absolute values parameters t-tests were For the absolute values of various morphometric between the mean values for dorsal apptied to detect any significant differences posterior regions' The results for and ventral regions and between anterior and these tests are presented in Tables I2 and 1l' animals' only one was For those parameters compared between the five this was the arithmetic mean thick- found to differ significantly between regions; interstitiar thickness was significantly ness of the arveolar septar interstitium. The the 5% level, and was 22o/o different between the dorsal and ventral regions at thickness between the greater in the ventral regions. The difference in interstitial the f o/o Ievel and was 46% greater anterior and posterior regions was significant at four animals only there were posteriorly. of the parameters compared between rl8. Table12:t-testfordifferencesbetweendorsalandventralregions mean difference SE t Parameter nâ 0.014 T.I76 Hct 5 0.040 o.o42 0.1r9 tht 5 0.005 0.008 o.250 rhp 5 0.002 o.226 o.26I I tissue 5 0.059 o.o27 o.o37 tep 5 0.00I o.o33 -4.2r2x lin 5 -0.L39 o.o24 -I.667 Ten 5 -0.040 7o double o.o29 o.o22 l.llB capillary 5 5.7 49 -6.O29+x Sa 4 -34.663 rB.53Z 0.055 Sc 4 r.015 29.006 1.709 Vc 4 49.560 0. lB5 4.276x Vc/sa 4 o.79r o.778 r.337 ¿r.Vc/Sc 4 1.040 r.o52 5.75rx St/Vp 4 -6.050 0.480 0.5r0 oao, 4 o.745 x* difference signif icant, P < 0'0I * difference signif icant, P < 0'05 a: n - number of animals and posterior regions Table IJ: t-test for differences between anterior SE t parameter nâ mean difference 0.014 -o.353 Hct 5 -0.012 o.o52 -0.481 tht 5 -o.o25 0.017 o.165 rhp 5 0.0I1 o.r7 4 -0.816 ? tissue 5 -o.r42 o.o47 o.zr3 1ep 5 0.010 0.046 -5.97gxx îin 5 -o.275 0.018 -o.447 îen 5 -0.017 % double 0.018 -0.105 5 -0.004 capillarY L9.844 -1.r4r Sa, 4 -27.633 22.560 -o.27r Sc 4 -6.rrt 14.287 1.184 Vc 4 L9.770 0.285 2.O98 Vc/5a 4 0.598 r.o34 o.455 4.Vc/Sc 4 0.450 3.643 -0.191 St/Vp 4 -r.425 0.197 o.492 oto, 4 0.097 of animals xx difference signif icant, p ( 0'0J- a: n - number I 19. dorsal and ventral regions' but none several significant differences between surface area was significantly between anterior and posterior. The alveolar greaterintheventral,orupper,regionsatthel%levelbyapproximately20%' was also significantly greater in the ventral The index of alveolar diameter, St/Vp, was again about 20% greater' As St/Vp is a regions, but only at the 5o/o Ieve¡; it this result indicates that the alveoli in reciprocal indicator of alveolar diameter, dorsally' Since St/Vp is only a relative the venLral region wele smaller than those that the ventral alveoli ate 2oo/o smaller index this result cannot be taken to imply thanthosedorsally.Theonlyotherparametertodiffersignificantlywasthe greater dorsally this difference was capillary loading, vc/sa. Despite being 40% siqnif icant only at tl'te 5o/o level' t+-1.3 InhersPecies ComParrson Thegraphsofthesealiondata,togetherwiththeregressionlineforter- restrialmammals,wouldappeartoindicatethattheresultsforthesealionwere 86 and B7)' However' after application significantly different (Figs. 82,83,84, 85, two were found to differ of a t-test for each morphometric parameter, only and the capillary loading (Table I4)' significantly; the total capillary surface area predicted and measured absolute values Table 14: t-test for differences between SE t parameter mean difference 156.OO -3.I34 VL -2,369.OO r6.57 r.372 Sa 22.7t IT.93 -3.223x Sc -38.45 67.99 -2.L62 Vc -146.98 o.253 -3.692x Vc/Sa -o.934 0.016 -2.000 tht -o.o72 -r.289 oro, -o.954 0.740 dif ference signif icant, P ( 0'05 Thetotalcapillarysurfacealeawassignificantlygreateratthe5%olevel grea1er than that predicted for a terrestrial where the mean measured sc was 23o/o speciesofmeanweighL,Tlkg.Thecapillaryloadingwasalsosignificantlygreater Figure 82. Allometric plot of lung volume for I'exlended" population of terrestrial mammals, V¡ = 45.gg4 Mbl'059, Mb is body mass in kg (Gehr et al., l98lb). Dots represent individual sea lions, open square the species mean. Figure Bf. Allometric plot of alveolar surface area for "extended" population of terrestrial mammals, Sa = 3.t42 Mb0'949, Mb is body mass in kg (Gehr et al., 1981b). Dots represent individual sea lions, open square the species mean. 120 100 a a tra t¡J =Ð oJ oz D 1 o'oool o.oo1 100 l,ooo BODY MASS (kg) ô E l¡J G, o¡¡¡ É É, (tÐ É, oJ t¡¡ z o'o1 o.ool 10 100 l,ooo BODY MASS (kg) Figure 84. Allometric plot of capillary surface area for I'extended" population of terrestrial mammals, g¿ = 2.727 MbtO'952, Mb is body mass in kg (Gehr ql al., l98lb). Dots represent individual sea lions, open square the species mean. Figure 85. Allometric plol of capillary volume for rrextended'r populaIion of terrestrial mammals, Vc = 3.198 Mbl'000, Mb is body mass in kg (Gehr eL al., lg8fb). Dots represent individual sea lions, open square the species mean. CAPILLARY SURFACE AREA tMz) (ml) CAPILLARY VOLUME -9 I o I o o o I o o o o@ 0 o@ o =Þ Ø U, =Þ Ø a U, ral- a a a x o ,o- o -o o J to J Figure 86. Allometric plot of tissue harmonic mean thickness and capillary load- ing for "extended" population of terrestrial mammals, rht = 0.416 Mb0'050 and Vc/Sa = 0.957 Mb0'05I, Mb is body mass in kg (Gehr et al., f981b). Do[s represent individual sea lions, open square the species mean. Figure 87. Allometric plot of oxygen diffusing capacity for "extended" population of terrestrial mamm"¡r, DLoz = 0.049 Mb0'991, Mb is body mass in kg (Gehr et al., 198lb). Dots represent individual sea lions, open square the species mean. tml/m2r Tht t¡mt (ml O2'sec-1'mbai1) CAPILLARY LOADING DIFFUSING CAPACITY o Io o oo o o I I o o o o @o @ 0 o o 3 Þ 3 U' Þ Ø a U' O¡ U' a a F o a o G¡J a a €g= o 'o o T\) f\) r23. at the 5% Ievel with a mean measured 7\o/o ÇreaLer than that predicted. Apart from these two, the mean measured value for various other parame[ers were considerably greater than predicted but were not found to be statistically signi- the f icant. This was due, in part, to Lhe sm all sample size involved, and to associated relatively large standard error of the means. These parameters included the lung volume which was approximately 50%o greater than the predicted value, the capillary volume which was nearly 6070 Çreater than predicted, the tissue harmonic mean thickness which was l-4% greater, and the total diffusing capacity which was 26o/o greater. The mean alveolar surface area was the only parameter less than that predictedi it was about -Il% smaller. 4.4 DISCUSSION 4-4-I Discussion of Methods (d Level I I'coarse" The results for the eslimation of the parenchymal volume density from each lung slice of animal 454 using the two grid systems are presented in Table 15. Also included in the table are the estimated ranges for the true volume densityl These estimated ranges show that the error for 22 mm Weibel grid was approximately 60/o and approximately 3Vo for the l- cm grid. Therefore, the approximately five-fold increase in the number of test points generated by the I cm grid reduced the error of the estimate by 50o/o. To gain advantage of this increased precision the I cm grid was used for all animals. (b) Level II In the trial count of one 2 '¡rm section with the three grid systems tested, it was found that the 1 cm-square lattice generated over 1,300 test points. This number of points required an unreasonable and hence costly amount of time to count. It has been demonstrated that in conducting stereological estimations no benefit is gained in the efficiency of the final estimate from very precise exami- nations of small areas. Rather, less precise estimations of more areas oI even r24. more animals, depending on the cost of each, is often more efficient and can result in a more accurate final estimate (Gundersen and Østerby, 198J-; Mathieu et lg8l). Another factor in the accuracy of the final result is that of the "I., bìological variation between animals, as it has been demonstrated that the major source of variation in results often stems from the differences between the indi- vidual animals (Shay, I975:. Gundersen and Østerby, J-98l; Mathieu et al., J-98l; Gupta et al., 19Bl). For these reasons the I cm Iattice was excluded from further tri als. Table 15: Results for grid systemsr trials - level I grid 22 mm Weibel grid J- cm-square lattice grid syste m slice Pp1/P1 Vvt ¿Vv ep1/P1 Vv+ ¡Vv A/AA J7l44 B4.l t 5.5 17II2I4 79.9 ¡ 2.7 A/B 401 48 83.3 * 5.4 164l?r7 15.6 + 2.9 M/A 4rlst 77.4 x 5.7 t751240 72.9 ¡ 2.9 M/B 4r153 77.4 + 5.7 r9gl26e 73.6 ¡ 2.1 PIA t4l 44 77.3 t 6.3 r6tl2r5 75.8 t 2.9 PIB 301J7 B 1.l x 6.4 r49lr76 83.5 + 2.8 a: A - anterior slice of lung M - middle slice of lung P - posterior slice of lung A - apical surface of lung slice B - basal surface of lung slice The two Weibel grid systems gave comparable results for the one seclion counted, so it was decided to include all six sections from animal A54 in the trial. The results of this trial are presented in Table 16. From these results it can be seen that the ll mm grid was marginally, and consistently, more precise and even though it involved counting nearly twice the number of points generated by the 22 mm grid, it was used for the other animals. However, in retrospect, it would appear that the overall efficiency of the final result would not have been greatly impaired by using tl-te 22 mm grid. fhe 22 mm grid would be suitable for lwo reasons: one, the effect of biological variation mentioned above; and two, when r25. the variation at level II is compared to that for the other Ievels of the four-level sampling design (Section 4.4.Ì(e)), it can be seen that it is generally much smaller so that its contribution to the overall variation is minimal. Table 16: Results for qrid systemsr trials - level II grid 22 mm Weibel grid ll mm Weibel grid system section ep2lP¡ Vvr ÂVv PpzIPT Vv+ aVv ADA rB4l204 9O.2 + 2.I 3601381 93.O t I.3 AV 1941224 86.6 + 2.3 4221444 95.0 + l-.0 MD 2rrl249 84.7 + 2.3 4051433 93.5 + I.2 MV 2211242 9l-.1 + l.B 4331 412 9I.7 + I.3 PD 2141242 BB.4 r 2.1 4201 470 89.4 t 1.4 PV r95l?26 86.3 t 2.3 4rrl 44e 9I.5 ¡ I.3 a A - anterior slice of lung M - middle slice of lung P - posterior slice of lung D - dorsal region of lung slice V - ventral region of lung slice (c) Level III The effect of improved resolution with increasing magnification has been examined recently (Gehr et al., Keller et al., 1976; I976i Paumgartner et al., 1981). In these studies it was found that improved resolution has a profound influence on the surface density of a particular feature due to greater detail of its contours, particularly when examined electron microscopically. The volume density was found to be consistent, excepl at very low magnifications (-40x) whére it was significantly greater than at even moderate magnifications (-100- 400x) (Keller et al., I976; Gehr et al'r I976)' However, at very high magnifications the volume density of small organelles was found to increase until the critical magnification (110,000x) was reached (Paumgartner et al., l9Bl). The influence of the improved resolution between 60x and 240x magnifica- tion was most pronounced in the determination of the volume proportion of the thicker septa occupied by respiratory surface (Figs. BB and 89). For the Figure BB. Graph showing progressive means of percenlage of interacinar septal volume occupied by respiratory surface, delermined at three levels of magnifica- lion. 126 50 49 4 ¡ A A A ¡ ^ 42 ^ 41 E 40 A o oct) g o o oL 31 o. 30 ooao 29 -a o o 24 a O O 22 21 o 30 80 90 100 110 Number of fields 6Ox t 5% limits ! 5% limits o^----120x ------24Ox t 5% limits Figure 89. Graph showing progressive means of percentage of inlerlobular septal volume occupied by respiralory surface, delermined at three levels of magnifica- tion. Percentage (%l N J J l\) (,| J J J J¿ ¿ (o J ('' o Ìs 8ts I I .t ¿ I I I o I u| I I I o À) I ¡ o I I I ¡ I (, I o t o I I o I ¡ z I ¡ I o I I I =5 I I I CT I t I I lo E8 I I ¡ 19 ¿ I I ]\¡ o) 9.o I o Þ I o o o +o ¡ I x x x ¡ d! ¡ I ¡ þ t+ l+ l+ ¡ cLo I (rl ('l (,| I I o ¡ ¡ àQ àe ñ I I @ I ì o o I 3 3 3 I ,1ì I =o 30 n, I q t I I ¿ I I o I I I J N) ¿ I ! o¿ I o r28 interacinar septa the respiratory surface was 487o of the total septal volume at 60x magnification, but decreased to l0o/o at 24Ox, and for the interlobular septa the respiratory surface was 22%o of the total septal volume at 60x magnification and only l4o/o at 24Ox. Despite this decrease in the proportion of the respiratory surface of lhese septa with increased magnification, the contribution of this respiratory surface, from both types of septa, Lo the total volume of the respira- tory surface within the parenchyma, remained stable at approximately l0% for each Ievel of magnification (Fiq. 90). This was because there was a similar decline in the volume proportion of the total respiratory surface within the parenchyma which was reduced from 22o/o at 60x magnification to t5o/o aL 24Ox (Fig. 92). To gain advantage of the improved resolution the magnificalion of 240x was selected for examination of all sections. For the volume density of most features estimated aL 24Ox magnification the progressive mean remained within the plotted limits after 60 to 70 fields had been counted (Figs. BB, 89, 9Ir 92 and 93). It was therefore determined that 70 evenly distributed systematically sampled fields would be required from each section. As mentioned in the methods above (Section 4.2.2(c)), the 6x eyepiece was not available for the complete counts so a f0x eyepiece was used. The magnifica- tions then available were l-00xr ZOOx and 400x. When the 400x magnification was tested a significant proportion of fields fell completely within the airspace or the connective tissue core of the thicker septa. Therefore a great increase in the number of fields counted would have been required to achieve reliable results for the more sparsely distributed features of interest. Consequently the final ma9- nification selected was 200x with the number of fields examined increased to 80. However, to achieve similar precision of the results at 200x to those at 240x, the number of fields need not have been increased but rather marginally reduced below 70 due to the increase in the field of view at 200x magnification. Figure 90. Graph showing the contribution of the interacinar and interlobular septa to the total respiratory surface volume, determined at three levels of mag- nif ica tion. Figure 9I. Graph showing proqressive means for combined volume proportion of single and double capillary alveolar septa within the parenchyma, determined at three levels of magnification. 129 I I I 31 r I d I I o 29 I C'ì o I c o I o I o ô. t ,, 25 ,t t t a 2 10 20 30 40 50 60 70 80 90 100 110 Number of fields ¡-6Ox r ----12Ox a ------24Ox 16 14 I T Ë 13 o T o) yo I C A o A I 11 A ob T--- À 10 aoa -:-- õ---o-î 9 o O o__t__- 7 10 20 30 40 50 60 70 80 90 100 110 Number of fields .- 6Ox ! 57" limits a --- 12Ox + 57. limits o------24Ox ! 5% limits Figure 92. Graph showing progressive means of total proportion of respiratory tissue volume within the true parenchyma, detenmined at lhree levels of maqni- f ic ation. 130 24 23 ¡ tl 21 -oo.- o 19 AA o)o ^a +tc o ---a- A (J A oL o. 1 --r- ---t--- 15 aoooo OO 14 o 13 10 20 30 40 50 60 70 80 90 100 110 Number of fields ¡ 60 x ! 5Y" limits A t 5% limits o ----12Ox------24Ox t 5% limits Figure 93. Graph showing progressive means of the percentage of airspace volume wi[hin lhe parenchyma, determined at three levels of magnif ication. 131 87 86 OO O ooo 85 I o o o 83 A A 82 ' A A A 81 A d og) o +.g o 78 o It oL o- 73 10 20 30 40 50 60 70 80 90 100 110 Number of fields ¡ 60 x ! 1% limits A --- -12Ox t 1% limits o ------24Ox !1% limits r32. (d) Levet tV (i) Subsampling method: The results for the two methods of subsampling of eleclron micrographs are presented in Table 17. The table shows the mean density, from the six sampling sites, for each feature under consideration. The results for the two methods are significantly different but this is because each method is applicable to a particular overall sampling design. The SQ subsampling was used in two-level sampling whilst the STWQ subsampling was used in four-level sampling. Table 17: Resutts of comparison between SGì and STWG¡ subsampling - level IV SG subsampling STWA subsampling Sv(a,s) sv(c,s) Vv(c,s) Sv(a,s) Sv(c,s) Vv(c,s) _l _r cm- cm--l cm* cm*-l density r,596.9 r,194.5 0.068 3,223.8 4,t3O.5 o.507 SE 313.2 422.8 0.012 178.2 245.O 0.016 %o SE 23.4 23.6 r1.6 5.5 6.r 3.2 The feature of interest in Table l-7 is the percentage standard error of the mean. The mean value for the estimation of the alveolar and capillary surface densitiés and the capillary volume density all have a large percentage standard error for SG subsampling. For the STWG subsampling the equivalent results are consistently small, being between f our to f ive times sm alle'r than f or the SG subsam pling. (ii) Harmonic mean barrier thickness: Tabte .l-B shows the values for the harmonic mean thickness of the tissue and the plasma calculated directly from the Iinear measurements, and using the eleven and seventeen class log scales. The last column in the table shows the percentage variation between th calculated from the linear measurements and h for the two log scales. For the tissue, direct application of the eleven class scale overestimated the rh by about 2% whilst the seventeen class scale overestimated it by about 3.5Vo. For the plasma, the eleven class scale underestimaled the t h by 27o and the I1J. seventeen class scale underestimated by nearly 6%. Therefore, using the eleven class scale resulted in a th within 2o/o of the measured th which was smaller than the seventeen class scale variation, and for lhis reason the eleven class scale was used. Table I8: Results of comparison between Il and 17 class logharithmic scales for harmonic mean barrier thickness (.h .hc)a L - . 100 r/rh th thL Tissue Harmonic Mean Thickness linear measurement 0.150 4.440 ll- class Iog scale o.r47 4.535 -2.14 17 class log scale 0.r45 4.598 -3.56 Plasma Harmonic Mean Thickness linear measurement o.897 o.l4i l-L class log scale 0.916 o.728 +2.O2 L7 class log scale o.952 0.700 +5.7 9 a: .hL - linear harmonic mean thickness thc - log class determined harmonic mean thickness (e) Comparison of two-level and four-level samping designs The results for each level of the two sampling designs, together with the whole lung densities for the three parameters estimated are presented in Table l-9. The whole lung densities for the parameters under consideration, were calculated by the appropriate serial multiplication for each sampling design. The whole lung density was calculated for each feature from each of the six sampling sites from which the animal mean was determined. The percentage standard errors of the means for the whole lung densities were generally larger than for any of the component levels. The percentage standard errors were approximately 2.5 Limes greater for the three paramelers with the two-level design than with the four-level design. The main reason for the greater vaniability in the resulLs of the two-level design was the presence of the interacinar and interlobular septa. With this design it was only possible, at the EM Ievel, to differentiate between the nespiratory whole lung densities Table 19: comparison between sGl and STWGI subsampling - Four-level SamPling Design Whole Lung Densities Level Densities I II III IV Vy(c,L) Sy(a,s) Sy(c,s) Vy(c,s) Sy(a,L) Sy(c,L) Vy(p1,L) Vy(P2¡P1) V(s,pZ) _l -t cm-_l cm*_l cm- cm 0.048 4,03O.5 0.507 307.3 ]85.8 density o.769 o.924 o.L32 3,223.8 to.5 43.J 0.001 0.010 T78.2 245.O 0.016 SE 0.017 0.008 9.9 rt.2 6.3 5.5 6.r J.2 %SE 2.2 0.9 7.6 Two-level SamPling Design Whole Lung Densities Level Densities IA IIA Sy(c,p) Vy(c,P) Sy(a,L) Sy(c,L) Vy(c,L) Vy(p1,L) Sy(a,p) -l -1 cm -1 cm cm -1 cm r,404.3 o.o56 r,596.9 r,794.5 0.068 r,24B.B density o.769 0.0r2 298.3 ]BB.B 0.009 0.0r7 373.2 422.8 SE 27.7 I6.I 2J.4 23.6 17.6 23.9 %SE 2.2 P Þ r35. portion and the connective tissue core of these septa, and for this reason they had to be included at this level. However, due to the small area of tissue sampled in each block for EM, the number of these septa included was both very small and highly variable. In the trial with this design a number of quadrats from some sections were completely occupied by the core of these septa, whilst in other sections sampled no thicker septa were included. Therefore, to achieve reliable results with the two-level design, the number of samples examined at the EM level would need to be greatly increased. This would then become an inefficient approach because of the increased cost and time involved in producing and analysing an adequate number of micrographs (Shay, I975i Gundersen and Østerby, lgBl). The presence of these irregularly-distributed septa also influenced the variability of the results for the four-level design. Level III, the level at which these septa were considered, had the largest percentage standard error of the four levels. However, this was only marginally larger than the results for level IV. The effect of the presence of the thicker septa was also apparent in the differences between the whole lung density mean values for the two sampling designs. Both the alveolar and capillary surface densities were much larqer for the two-level design. This was due to the limited sampling of thicker septa with the SG subsampling which introduced a bias toward the thinner alveolar septa. This bias would indicate that the alveolar septa have a greater surface density through- out the parenchyma than is actually the case. 4.4-2 Discussion of Results (Ð BoAy weights and reproductive status The weight of only four of the animals was recorded at the time of capture, and it was only from these animals that there was any indication of their repro- ductive status (Table l). From these observations it would appear that the animals AS2l , 24, 25 and 26 were mature breeding females. For animal AS4 its weight could be predicted from the relationship between standard length and body weight for female Australian sea lions from Seal Bay' 136. log L = I.52 + 0.33 log w (Walker and Ling, l98l) where L is standard length in cm, and w is body weight in kg. However, only curvilinear lengths were measured which, if substituted for standard length in the above relationship, will give an inflated weight prediction. By substitution of the curvilinear lengths of animals A525 and 26, those captured at Seal Bay, the measured weights were approximately 25o/o smaller than those predicted, whilst for animals AS2l and 24, from Dangerous Reef, the measured weights were approximately 40%o less than predicted. Substitution of the curvi- linear length of animal A54, also from Dangerous Reef, gave a predicled weight of 48 kq which, if similarly overestimated by 25-4O%, would result in a body weight of 29-36 kg. This result is 40-50% less than the smallest animal weighed in this series, that is, 60 kg which was at the bottom of the range for adult females weighed in the field (Walker and Ling, 1980, lgBl). This predicted weight for A54, together with the lack of evidence in regard to reproductive status, would indicate that this animal was not a mature breeding female. Since no weight was recorded for animal AS4, and the indication that it was not an adult animal, the morpho- metric results for this animal have been excluded from the species comparison. The differences observed between the four other animals may have been a result of age, seasonal variation, or locational differences. No reliable method has been developed for accurate age deLermination of otariid seals, so any differences could not be determined. The two pairs of animals were collected six months apart so that seasonal variations, such as availability of food, may have resulted in weight differences. The two animals from Seal Bay had weigh|"s 25% ìess than predicted, whilst for those from Dangerous Reef this difference was 40%. Thus there may also be a Iocational difference between the two populations. However, insufficient data has been recorded from the Dangerous Reef population to establish such a difference. Another possible factor for the difference between the two groups may have been their reproductive status. Both animals from Seal ril Bay were lactating and suckling young pups, whilst neither animal from Dangerous Reef was lactating although animal AS24 was pregnant with a three month old foetus (J.K. Linq, personal communication). (b) Lung inflation and fixation It has been suggested that instillation fixation of lungs with an inflation pressure between 20 to l0 cm H20 above mid-chest height is sufficient to get complete unfolding of the alveolar surface (Weibel, I97Oa, I973). It has recently been demonstrated that the degree of inflaIion is increased by fixation in situ, and is affected by the type of fixative, the initial flow rate of fixative and the final transpulmonary pressure (Hayatdavoudi et al., 1980). Glutaraldehyde fixes tissues rapidly and for this reason needs high initial flow rates to ensure adequate inf lation. In the present study the fixative was instilled via the trachea to a pressure of l0 cm H20 above mid-chest height. Due to the size of the trachea in the sea Iion and the reinforced airways, the fixative would have been rapidly distributed throughout the lung at a sufficient pressure to result in near complete unfolding of the alveolar surface. That this was in fact the case was evident in low power micrographs where the alveoli were large and open' and at the EM level where the integrity of the tissue com ponents was m aintained and there was only very occasional folding or pleating of the septal wall as observed in terrestrial species (Gil and Weibel, I972; Assimacopoulos et al., I976; Weibel et al', I973; Mazzone et al., l97B). No regions of septal wall folding or pleating were found in those blocks randomly selected for morphomelric evaluation' Increasing inflation pressures and over-inflation of the lung by instillation of fixative or by air in perfusion fixed lungs results in decreased capillary volume (Weibel et al., I973; Fopest, I976; Mazzone, 1980) and deformation of capillary shape (Assimocopoulos et al., I976). Gualitative evidence of overinflation was present in some regions of animal A526 where some of the alveolar capillaries appeared flattened and quantitatively the increase in the capillary volume was much less than for the other animals when compared to the predicted values. I lB. The value for total lung volume of fixed lungs has been shown to represent between 70 to gOVo of the total lung capacity determined by other methods (WeibeÌ et al., 19BJ-a; Gallivan, l9Blb). It has also been suggested that large lungs may lose about 5o/o of their volume during the first few hours after instillation fixatiôn (WeiUel et al., l98la). This decrease has been attributed in part to some residual elastic recoil and to trapped air in the more craniaÌ regions of the apical lobes. If residual elastic recoil does in fact decrease the lung volume, then it may be expected to cause a greater decrease in the sea lion lung due to the greatly increased content of elastic tissue. No work has been conducted to determine if this is in fact the case. (c) Absolute values (i) parenchymal volume: The volume fraction of the parenchyma in the sea lion lung was 7I.5Vo. This value is well below the range from 80-90 % found in terrestrial species (Table 20). The decrease in lung parenchymal volume is a consequence of the increased number of the thickest connective tissue septa. These are the septa which divide the lung into segments and are generally greater than I mm in thickness. In terrestrial mammals the septa which demarcate the division of the parenchyma into bronchopulmonary segments are generally only apparent subpleurally. These septa arise from the pleura and penetrate into the lung for a distance of only 2-3 cm (Krahl, 1964; Reid, 1967). In the sea Iion lung the intersegmental septa also arise from the pleura but maintain a relatively constant size as they continue throughout the parenchyma. There was no apparent regional variation in the distribution of these septa and they appear as a regular feature throughout the parenchyma. ePrthelirrm volume fraction - - . - (ii) Respiratory tissue: The A of respiratory /\ within the paren- chyma was IJ.7o/o for the sea lions sampled. This included the respiratory epitheliunt total lining the interacinar and interlobular septa. Thisiivalue is similar to that observed in the larger terrestrial species examined, but only about 6O-7OVo of that in small mammals (Table 20). The tissue and capillary components of the alveolar septa respiratory epithel iutrr Table 20: Species comparison of parenchyman airspace and 71 volumes expressed as percentages sprra eu-thelium , -. Source Species Pa in lung Airspace in P inP I ISSUe Capillary shrew 90 7B 22 l1 11 Gehr et al., J.980 b rat BZ BO 20 9 1l Burri and Weibel, I97Ia, bat B5 82.9 I7.I 7.7 9.4 Maina et al., I9BZ guinea pig 94 77 23 I2 11 Forrest and Weibel, I975 -9 -9 dog B2 B2 ]B Gehr et al., I98J-a macague 90 90 10 Boatman et al., I979 human 90 86.5 13.5 7.8 5.7 Weibel, 1963; Gehr el al.r I97Ba -8.5 horse B6 82.9 I7 -8.5 Gehr and Erni, l9B0 84 9I 9 4 5 Gallivan, l98lb cow B2 89 1l 5 6 Gallivan, l98lb sea lion 72 86.3 L3.7 6.3 7.4 this study a: P - parenchyma P v.l \o r40. each constitute approximately 50 o/o of the septal volume. This division of the septal volume is similar to that found in other species (Table 20)' tnis was despite the fact that approximately 27o/o of the alveolar septa carried a double capillary network. In the sea lion the volu me proportions of the intra-alveolar septal components were determined at level IV which excluded the interacinar and interlobular septa. If these septa had been included at level IV then the tissue fraction within all respiratory septa would have been markedly increased and that for capillary volume decreased. At level III the distinction between lhe respiratory epitheJ"ium and the connective tissue of these septa was possible. From this ìt was found that approximately 3\o/o of the interacinar septal volume was occupied by respiratory tissue, and for the interlobular septa this value was l-0%. The epithel- iun respiratory'- of the interacinar and interlobular septa contributed enithel ium approximately ^--""one-third of the total respiratory' volume determined at ^ Ievel III. (d) Regional differences (i) parenchymal distribution: At level I there was a significant anterior-posterior difference in the volume density of "coarse" parenchyma at the 570 level. The volume proportion of "coarse" parenchyma was approximalely 7o/o greater pos- teriorly than anteriorly. This regional difference was not apparent for the "fine" parenchyma determined at level II, nor was it apparent for the total volume fraction of parenchyma in the lung. The significant anterior-posterior difference detected at level I is primarily a reflection of the inhomogeneous distribution of the coarse non-patenchymal structure, that is, the major intrapulmonary airways, the pulmonary blood vessels, and the prominent intersegmental septa. Many of these structures enter the lung at the hilum and divide repeatedly into smaller and smaller branches which supply all the parenchyma with air and blood. The bronchi, and the pulmonary arteries which closely follow the bronchi in the more proximal regions, divide by an irregular dichotomy (Weibel, 1963; Thurlbeck and Wang,I974), so that at any t4l. given disLance from the hilum any two airways of the same generalion will nol necessarily be of bhe same diameber. This irregularity is most marked between branches of the axial pathways (Reid, L967). Each axial pathway runs directly from the hilum to the distal pleural surface of the segment it supplies, so bhat its length and bhe number of its qenerations depend on the size and shape of that segment. In this way the intrapulmonary bronchi, and their associated blood vessels, are inhomogeneously distributed throughout the lunq parenchyma' There- fore, the position along the anterior-posterior axis of the lung from which each of the three I cm lung slices was collecled, would determine, to a degree, the proportion of the "coarseil non-parenchymal structules present. This is borne out by the fact that there were no significant regional differences in the distribution of the "fineu parenchyma. No work has yet been done lo determine the distribution of major airways and blood vessels in the sea lion lung which would most effectively be achieved by resin casting. (iD Body mass independent parameters: Within the five animals only one parameter was found to differ significanlly between the regions. This was the arithmetic mean thickness of the septal interstitium which was significantly greater ventrally than dorsally, and significantly greater posleriorly than anteriorly. This result is surprising when one considers the distribution of stresses in the lung. West (1978) has shown that the stresses within the human lung are greatest at the apex in both the vertical and lateral directions. This means that the mechanical forces which act on the alveolar walls are greaLest in the apical region, so that in the sea lion lhese forces would be greatest in lhe dorsal region. This redistribution of str.esses with a change in orientalion of the lung has been demonstrated in dogs by Glazier g!_gl. 0967). They found that the stresses in the apical region of upright doqs behaved similarly to lhose in the dorsal region of standing dogs. Therefore if increased stresses due to lhe lung were counteracted by septal reinforcement one would expecl this to occur in [he dorsal region in lhe lungs of quadrupeds and the sea lion, and in the apical region in human lungs. r42. Reinforcement such as this has not been demonstrated in those species examined. The ventro-posterior thickening of alveolar septa in [he sea lion may well be a response to the type of respiralion of these animals and to diving. The Australian sea lion, in common with many marine mammals, possesses an obliquely oriented diaphragm and a ftexibte rib cage (Scholander, 1964; Green, L972; Harrison and Kooyman, 19BJ-). These structures, together with airway reinforcement' have been demonstraled to enable very high expiratory flow rales in anolher otariid, the california sea lion (Kerem et al., 1975; Matthews, L977). It has also been demonslrated in excised california sea Iion lungs, that as a consequence of airway reinforcement the alveoli are able to empty of air almost compleLely when under pressure (Denison eI al., ]-971). This alveolar collapse has been predicted to be similar to that which occurs during diving due to the increased pressure at depth (scholander, 1940). If the reinforcement of alveolar septa, together with the enlarged inleracinar and interlobular sepba, are structural modifications to assist respiration and Iung collapse, lhen it is possible that the forces acting during these activities are greater ventro-posteriorly which has resulted in an increased thickening of the interalveolar septa in this region' (iii) Body mass dependent parameters: For those features compared between the four animals only, there were seVeral regional differences' The alveolar surface area was approximately 2To/o gteater in the ventral (upper) region which was significant at the 1% level. The index of alveolar diameter (St/Vp) was also sig- nificangy greater in the venlral region at the 5o/o level, again being about 20% larger ventrally. The alveolar diameter index is a reciprocal indicator of alveolar diameter, so that the larger the index value then the smaller the alveoli. There- dif ference in f ore the venlral alveoli were sm aller than those dorsally. This alveolar diameter is reflected in the associated increase in Sa, since more of the smaller alveoli will be contained in a unit volume. This decrease in the size of the uppermost alveoli was the reverse to [hat found in dogs (Glazier et al'' 1967; siegwarl et al., I97L; Gehr and weibel, I97Ð. Glazier-9!-g!. (1967) found that the L4J. extenl of the difference in size between upper and dependent alveoli was influenced by the vertical distance between the upper and lower regions and the transpulmonary pressure at the time of lung fixation. They found bhat the size difference in alveoli was greater with increased vertical distance, and with low inflation pressure. In the sea lions the verlical dis[ance was only l] to 20 cm, whilst the inflation pressure was approximalely 30 cm H20. These conditions would therefore tend to minimise the dorso-venLral differences but not accounL for the reversal of the findings in dogs. Associated wi[h the increase in alveolar size in the upper region of dog lungs, Gehr and Weibel (I974) also demonstrated a significant, although small, increase in alveolar volume density in the ventral (upper) region and a I07o smaller tissue volume density ventrally. This difference also indicates the presence of larger alveoli in the upper region. Gehr and Weibel (Ig74) also presented various factors that may contribute to regional differences' of which the shape of the lhorax and the weigh[ of abdominal contents are the most likely to affect the sea lion. In lhe sea lion, lhe thorax, as with the whole body, is elongated in shape. Associated with this shape is a very obliquely oriented diaphragm which extends anteriorly from lhe xiphoid process to the dorsal body wall at the level of the second lumbar vertebra (Green, l'97Ð. As a result of these struclures and a very flexible lhoracic cage, the lungs tend to undergo dorso- ventral compression when in the normal prone posilion (Scholander' 1964)' It has been demonstrated thaL in Lhe horse in the supine position the posterior regions of the lungs are compressed by the abdominal contents (McDonnell, I974; McDonnell et al., I979). Therefore it is possible that in lhe sea lion fixation at a pressure of l0 cm H20 diminished any expected difference in alveolar size due to the vertical gradient, and Lhat the weight of the flexible thorax, together with that of the abdominal contents, caused a reduction in alveolar volume in the ventral (upper) regron. The other fealure which showed a significant difference belween lhe dorsal and ventral regions, relates to the capillary arrangemenl in the alveolar walls. r44. the 570 This was the capillary loading, Vc/Sa, which was significantly different al level; it was found lo be approximately 4070 grealer in the dorsal (dependent) region. concomitant wilh lhis was the finding that lhe capillary volume was alveolar approximately l)70 greater in the dorsal region, Lhat the proportion of the septa with a double capillary arrangement was 1270 greater dorsally, and thal capillary diameter was 14% greater dorsally. Howevet, none of these differences were found to be sta[istically significant. The combined effect of these small differences, with the siqnificanlly smaller Sa in the dorsal region, was sufficient to result in a significantly greater capillary loading dorsally' This increase in capillary loading and the smaller increase in capillary volume are in aqreemenl with the results of Glazier 9!-9!. (1969) and Gehr and Weibel Q97Ð' Both groups lhe found that, in the dog, the dependent region showed a significant increase in capillary volume. This increase also agrees with the finding in humans thal blood (West' flow also shows regional differences, being greater in the dependent regions 1978). (e) Interspecies comParison (i) Absolute and specific values: Table 21 presenls the absolute values of various morphometric parameters from the sea lion, together with those from four ter- reslrial species. The terrestrial species included are those whose body weighb was most like that of the sea lion. Also included in the table are several weight- specif ic results. comparison of the absolute resulbs would appear to indicate that some of the morphomelric paramelers are increased in the sea lion' Howevelt examinaLion of the specific results indicates a different interpretation. The only specific result for the sea lion that is consistently greater than for the other species is the lung volume. DespiÈe this increase in lung volume, the OtO, per kg body weight is similar lo lhat for lhe other species, except humans. It has been shown previously (Weibel, L972) that the neduced OrO, for hum,ans is more like that for inactive or captive species than of active, free-living species' r45. Tabte 2J.: Species comparison of absolute and specific m orPho metric Parqmeters Sea lion Humana Dogb wildebeestb waterbuckb Mb(kq) 77 74 46 LO2 It0 7.84 vL(L) 6.95 4.34 2.89 7;68 383 sa(m2) t8l r43 177 39L 3tB 5c(m2) 209 L26 L32 28L 584 Vc(mt) 393 2r1 2t4 473 tht ( um) 3.56 2.22. rht (um) o.59 o.62 o.53 0.37 0.46 thp (pm) 0.09 0.15 0.22 0.r7 oto, 4.582 2.470 2.84I 6-305 1.24r (ml O2.sec -1.rb".-1) vL/Mb(L.ke-r) 0.090 0.059 o.o63 0.075 0.071 2.k9-r) sa/vu(m 2.38 I.9J 3.85 J.B3 J.4B z.kg-l) sc/wb(m 2.7r 1.70 2.87 2.75 J.O1 vc/ wb(mt.kq-r) 5.10 2.BB 5.O9 4.64 5.3r DLo./Mb 0.060 o.o33 o.062 o.o62 0.066 {r tó2..""-r.mbar-r.t q-t) a: from Gehr et al., L97Ba, excePt D Loz from b b: from Gehr et al., 1981b- That the sea Iion possesses large alveoli is reflected in the specific alveolar (Table surface area, which is closer to that for humans than for the other species ZI). Another indicator of alveolar diameter is the ratio St/Vp. The alveolar diameter indices for several species are presented in Table 22' The index for the sea lion is the same as that for man, which is the smallest index indicating the largest alveoli. Alveolar diameter was not delermined in this study, but has been measured in humans (Dunnill, L962; Weibel, 196Ð and in the california sea lion and ofher otariid seals (Denison et al., L97I; Denison and Kooyman, I973)- In adult humans the alveolar diameter has been estimated to be 25O-280 urn, whilst in otariids it was estimated to be 200-250 ym. These results indicate that lhe alveoli 146. of humans and otariids are of similar dimensions which is also indicated by the equivalence in the alveolar diameter indices of humans and the Australian sea lion. From this it can be concluded Lhat the diameter of the alveoli in the Australia sea lion is in the region of 250 um' The estimated capillary diameter of the sea lion,7.5 Pm' was marginally greater than that for most of the larqer terrestrial species, and was more than twice the diameter of the capillaries in lhe smaller animals. Only one terrestrial mammal had a larger estimated capillary diameter and that, perhaps surprisinqly, was the guinea pig (Forrest and Weibel, 1975). Comparison of the values for the capillary haematocrit do not reflect this variation in capillary diameter (Table ZZ). The smallest recorded haematocrit was from the horse (Gehr and Erni, I9B0) which had an estimated capillary diameter similar lo many terrestrial species, whilst the guinea pig, although possessing the widest capillaries, only had a typically terrestrial haematocriL. The converse situation occurred in the bat (Maina et al., I9B2) which had the smallest capillary diameter together with the highesl haematocrit. For lhe sea lion the capillary diameter was larger than for most species, and the haematocrit value was only less than that for the bat. (ii) Atlometric equations - I'extended" terrestrial mammal population: Interspecies comparisons of predicted values to those measured in the sea lion, using the (198lb)' that terrestrial mammal allometric regressions of Gehr -91-4' indicate only two parameters are significantly differenb, the tolal capillary surface area and the capillary loading. The increase in total Sc would appear to be a direcl resull of the double capillary network, as the amount by which the measured Sc exceeds thal predicLedr 23To, is in close agreement with the proportion of double capillary alveolar septa presenl, LhaL is, 27o/o' In studies of other, non-mammalian species, which possess a complete or partial double capillary nelwork, lhe alveolar and capillary surfaces were divided info respirat,ory and non-respiralory regions (Hughes and weibel, I976; Perry, f91B; J.981; l98l). A similar approach was also adopted by Wangensteen and Table 22: Species comparison of alveolar diameter index, capillary diameter and haematocrit Alveolar Diameter Index Capillary Diameter Haematocrit Source (mmZlmml) ( r,m) from Gehr et al., 1980 shrew t0B 4.r4 o.466 from Main a et al., 1982 bat L2I 3.r5 o.6tr from Forrest and Weibel, 1975 guinea pig 57 8.10 o.499 from Gehr et al., I9BIa, b dog 6I 7.O9 o.46t from Gehr and Erni, 1980 horse 63 6.73 o.334 from Gehr et al., I978a human t2 6.77 o.46 this study sea lion 32 7.53 0.548 ts {N 148. Weibel (f982) in their morphometric study of gas exchange in the avian egg. The respiratory regions were generally def ined as those parts of t'he alveolar and capillary surfaces which formed an air-blood barrier and contributed to gas exchange. The non-respiratory regions were the remaining portions of each surface which do not assist in gas exchange. In most morphometric studies of mammalian lungs all of [he alveolar and capillary surfaces were considered lo be respiratory. This assumption was also made al the outset of this study because the proportion of the double capillary septa and the extent (volume proportion) of the thicker septa were unknown. However, the effective capillary surface area in the sea lion can be approximated by excluding those portions of the capillary surf ace which abutt the cenLral connective lissue core of the double capillary alveolar septa and the interacinar and interlobular septa as these do not form part of the air-blood barrier. Since ZJo/o of the alveolar septa contained a double capillary network and almost31o/o of the respiratory tissue volume consisted of interacinar and interlobular septa, then approximately half of the surface area of these capillaries is ineffective in gas exchange; the effective Sc is then only about 7\o/o of the total. The correcled values for the effective Sc of the four animals used in the species comparisons are presented in Table 21. When compared to the predicted values, the effective Sc is not significantly different (mean difference 23.87, SE I2.2J, t I.952), and lhe mean sea lion value is in fact 14% less than that predicted. When the values for the effective Sc are substituted inlo lhe calculation of DLO2,Lhe total diffusing capacity is reduced by 5-B% (Table 23), and the corrected measured value for OrO, is about lB% greater than predicted. No such correction is possible for the alveolar surface area, so the mosb appropriate method to accurately estimate thist and the Sc, would be to adopt an approach similar to thal of Huqhes and Weibel (I916) and Perry (1978; J-981; 19Bl) where the surfaces are divided into respiratory and non-respiratory components. No alteration was made to the capillary volume since the binding of oxygen lo haemoglobin is considered lo involve the entire r49. erythrocyte volume contained within the capillaries (Wangensteen and Weibel, r9B2). Table 23= Elrective capillary surface area and corrected oxygen diffusing capacity for the Australian sea lion effective Sc corrected OtO, animal 2 m ml O2.sec-1.mb"r-l 2t I2T.4 3.249 24 151.6 5.693 25 r52.4 4.440 26 l_60.4 3.7 64 mean 146.45 4.2865 rSE B.5B o.5285 Associated with this increased Sc was a marked, but nob statistically sig- nificant, increase in the capillary volume which was nearly 610/o greater than predicted. This increase was not significant because of Lhe small sample size and the large standard ertot of the mean of Lhe measured values. The Vc ranged from 2BZ ml up to 510 ml. This variability in the Vc reflects, in part, the variation of the capillary haematocrit values which ranged from 0.474 LoO.673, with a mean of 0.548. A higher haematocrit indicates a grealer number of erythrocytes present per unit volume which will tend to make lhe capillaries bulge more and hence increase their volume. The extent of capillary bulging is also influenced by the conditions prevailing during lung fixation (Section 4.4.2(b)). The mean and range of haematocrit values measured morphomelrically are similar to the results recorded from peripheral blood samples taken from a larger sample of adult female Australian sea lions (Needham et al., L9B0). The mean from that study was 0.557 with a range from 0.481 to O.642. The resulls for the haemalocrit and the capillary volume in the present study, therefore, appear to be realistic estimates for this species. The capillary surface area to alveolar surface area ratio for lhe Lotal Sc had a mean of t.Ll5. This indicates bhat the tolal capillary sunface area was 150. consistenEly greater than that of the alveoli, which is the reverse to thal found in most terrestrial mammals. In terrestrial species Sc/Sa ranges from 0.7 in the horse (Gehr and Erni, 1980; Gallivan, 19B1b) up to about 0.9 in the human (Gehr et al., l978a) and a range of African mammals (Gehr et al., lg8fb). However, sub- stitution of the effective Sc into this ratio results in a mean of 0.705' so thab the effective capillary surface area is less than the total alveolar surface area. These results for Sc/5a again reflect an increase in the lotal capillary surface area but not in the effective surface area. The alveolar surface area was 1l-70 less than that predicted but was not statistically significant. Tenney and Remmers (I96J) reported that marine mammals have a reduced alveolar surface area when compared to terrestrial mammals. They concluded that this was largely due to [he large size of their alveoli. The results of the present study are in agreement with those of Tenney and Remmers in regard to bolh the reduced alveolar surface area and the large alveoli (Section a.a.2(eXii)). The reduction in alveolar surface area, together with the increase in capillary volume, are also reflected in the capillary loading which was 78% qreater than predicted. This is an index of the suitability of the parenchymal lissue for gas exchange (Hughes and Weibel, I976) where a low ratio, Vc/Sa, implies favourable conditions for gas exchange because of the high degree of exposureofthecapillaryblood(Perry,]978).ThehighVc/Sainthesealion, Z.IJ ml/m2 from animals A323-26 only, indicates thal the parenchyma is slightly less specialized for gas exchange than that of terrestrial mammals where the values are generally in the ranqe I.L to l.l m Ilm2. The only species from the ilextended" population of Gehr gl_4. (l981b) which were found to have a capillary loading similar to thal of the sea lion, were the domesticated bovine species in which Vc/Sa ranged from l.B2 to 2.25 mtl^2. From the allometric regression equation for the capillary loading of the "extended" population (Gehr et al., lgBlb), the increased result for the sea lion was found lo be significanLly different at the 5o/o level. t5 1. Despite this increased capillary loading, the specific capillary volume of 5.10 ml/kg for the sea lion is similar to lhat for terrestrial mammals of similar body mass (Table 2l). Therefore lhe double capillary nelwork may represent a compensation for the decreased proportion of parenchyma in the lung so that the total capillary volume present is similar to lhat in terrestrial species, and thus maintains a comparable specific diffusing capacity' More recently, in studying the lungs of various reptiles, Perry (1981) has used the reciprocal of the capillary loading to evaluate the degree of specializa- tion. This ratio, Sqq/Vc, where 54p is the respiratory portion of the lung surface, implies that a large surface-to-volume ratio in the capillaries is beneficial to gas exchange. In comparing his results with those for various species, Perry found the rat, as the representative mammal, to have the highest value, 8.2 cmzlcnl' Th" species with the Iowest ratio were those which have a complete double capillary network in lheir lungs. These were the turtle and the lungfish with values of aboul 0.8 cmZl"^3. The reptiles, with a mixed type of lung, had inlermediate values of 3-4 cmZlcrJ. In the sea lion, also with an intermediate lung type, S4p/Vc was 4.-l cmilcm2. Ho*"u"r, in calculating the value for lhe rat, Perry assumed 100% of the alveolar surface to be respiratoryr âs was done for the sea lion. This assumption results in a slighl overestimation of the respiratory surface area in the sea lion, so that the true SO*/Vc value would be even closer to thal for the reptiles. The measured lung volume, although not significanlly different, was found to be approximately 50% greaber than predicted. This increase in the lung volume has been observed in other marine mammals (Slijper, L962; Tenney and Remmers, 1963 Kooyman, L97t). Slijper (196Ð found a similar increase in the lung volume of the smallen toothed whales, whilst Kooyman (I973) found marine mammal lungs generally to be slightly larger than those of terrestrial mammals. Kooyman (L973) presenled an allometric prediclion equation for marine mammal lung volumes; when the mean weight of the Australian sea lion was substituted into this 152. equation, the predicted lung volume was very similar to the mean value recorded in this study. The net result of these differences in the morphometric parameters of lhe sea lion was an increase in the oxygen diffusing capacity. The mean D¡gr, using the effective capillary surface alear for the four sea lions was 4.287 mI O2/sec.mbar which was 18% greater than predicted but this was not found to be statisticalty significant. When the diffusing capaciby is broken down into the values for each of lhe three componenbs from the model, then the con- tribution of each can be appreciated (Table 24). Also included in the table are the values for the membrane diffusing capacity, D¡yr which represents the combined contribution of the lissue and plasma layers. Table 242 Oxygen conductance values for individual components of the mormphometric model for estimating the oxygen diffusion capacity Dt Dp D M De oao, anim al m I O2.sec -l .mbar-l 23 9.366 87.t93 8.458 5.275 3.249 24 14.LZL 257.655 13.381 9.905 5.69J 25 ro.5J6 78.146 9.284 8.509 4.440 26 t2.525 89.894 I0.99J 5.724 t.7 64 mean IL.6J7O L28.2220 t0.5105 7.35J) 4.2865 rSE I.0539 43.2174 1.0887 1. r_111 o.5285 From Tabte 24 it can be seen that 02 conductance is most rapid within the plasma layer and slowest within the erythrocytes. The high conductance of the plasma layer is primarily a result of the very small plasma harmonic mean thick- ness, which for the four animals considered here had an avelage value of 0.089 um. This thp is similar to lhe smallest measured in terrestrial mammals, from several species of shrew (Gehr et al', l9B0)' The small thp in lhe sea lion resulls from the large capillary volume and the associated high haemaLocrit. The high haematocrit means that a large proportion of the erythrocytes will lie close lo the capiltary surface and so reduce the thickness of the plasma layer. r53. The tissue barrier conductance is markedly less than that of lhe plasma, largely due to the tissue harmonic mean thickness. The average tht for these four animals was 0.589 um, an order of magnilude greater than that for plasma, similar [o the difference between the average tissue and plasma conductances (Table 24). The conductance of the erythrocytes, as the slowest, is the rate limiting step in the overall diffusing capacity. The rate of O2 uptake by whole blood, which is required for calculation of De, has only been reliably determined in humans (Holland et al., 1977). As the size of erythrocytes in mosL terrestrial species is similar and shows no correlalion with body mass, the human value has generally been used in morphometric studies (Gehr et al., lgBlb). Whether this uptake rate is appropriate for marine mammals is not known; therefore the human value has been used here since it at least allows for comparisons between the sea lions and other species. In regard to the overall diffusing capaciby of the sea lion lung, the increase above that predicted for a terrestrial mammal of similar body mass results from an increase in the capitlary volume and a high haematocrit. The effect of these two factors is important to the erythrocyte conductance for 02 but both also contributes to the reduced plasma harmonic mean thickness and consequently to the high plasma conductance. 4.5 SUMMARY The mean total lung volume of lhe five animals studied was 6.34 L of which lI.5Yo was true parenchyma. Of this true parenchyma I3.7 7o wâs respiralory tissue. The alveolar surface area was 171.I t2, th" total capillary surface area I9J.6 r2, th" capillary volume )50.3 ml, bhe tissue harmonic mean thickness 0.59 ¡rm, the plasma harmonic thickness 0.09 ym, the tissue arithmetic mean thickness 3.43 ym, and the oxygen diffusing capacity 4.Il ml O2.sec-l.mbar-1. There were significant regional differences in the interstiLial thickness and lhe alveolar surface area. The alveolar septal interslitium was 22o/o gtea|er venlnally 154. than dorsally, and 460/o greaLer posberiorly than anteriorly. The alveolar surface area was about 2oo/o qtealer ventrally. For the four animals whose results were compared to the predicted values for the terrestrial species of sim ilar body weightr lwo parameters wete significantly different. These were the capillary loading which was nearly B0% greater than predicted, and the total capillary surface area whicl't was 23o/o greater. However, the effective capillary surface area involved in gas exchange was l-470 less lhan that predicted. Other parameters which were qrealer than predicted included the lung volume, which was 50% greater, the capillary volume, 60% greater, and the tissue harmonic mean thickness, 147o greater. The alveolar surface area was Ll% less than predicted. The total oxygen diffusing capacity was 260/o greater than predicted, but consideration of the effective capillary surface area resulted in a diffusing capacity 18% greater than predicted. 155. Chapter 5 PULMONARY PATHOLOGY 5.I INTRODUCTION The occurfence of pulmonary disease in free-living and captive pinnipeds is a well recognized and common phenomenon (Daitey, L97O; Sweeney, I97Ð' Of the two types of parasites which inhabit Lhe respiratory tract of pinnipeds, lhe pul- monary nematodes often lead to death of the host (Johnston and Ridgway, 1969; Dailey, I97O; Fleischman and Squire, I97O; Migaki et al., I97Ii Morales and Helmboldt, I97I; Sweeney and Gilmartin, I974; Stroud and Dailey, I97B:' Stroud and Roffe, IgTg), whilst the nasal mites, although commonly pnesent, are only of minor significance clinically (Keyes, 1965; Kurochkin and Sobolewsky, I975; Dunlap et al., 1976; Kim et al., 1980; Fay and Furman, I9B2). These parasiles occur in both the Phocidae and lhe Otariidae, often being of t.he same genus bul generally different sPecies. The two groups of parasites which have been described from the respiratory system of otariids are the lungworms of the genus Parafilaroides (Nematoda: Metastrongytoidea), and the nasal and lung mites of the g enus Orthohalarachne (Acarina: Halarchnidae). The recorded incidence and otariid host species for these parasites are listed in Table 25. Of lhese species recorded, only O. altenuaLa has previously been rep orted from N. cinerea (Domrow, I974)- The distribution of lhese parasites in the Australian sea Iion is shown in Fig. 94' 5.2 RESULTS 5.2.1 Macroscopic (a) vites Two species of orthohalarachnid mites have been collected from all animals examined, and identified as O. altenuata (Banks) and O. diminuala (Doetsch) (R. Domrow, personal com munication) Of the O. attenuata examined only adult 156 females and larvae were identified, whilst of lhe O. diminuata examined one adult male was identif ied Logelher with adult females and larvae. Table 25: Recorded incidence of pulmonary parasites in otariid species P arasite Source Host Or bhoh alarach nae O. attenuata O. diminuata Northern fur seal + + Keyes, 1965; Furman and Smith' I973; Kurochkin and SobolewskY, I975i Dunlap et al., I976; Kim et aI., J-980. Australian fur seal + + Domrow, I974. Steller's sea lion + + Margolis, I956; Slroud and Dailey, I97B; lay and Furman, T982. California sea lion + + Dailey, L97O; Furman and Smith' ),973; Sweeney and Gilmartin, I974; Ogata el al., 1979; Kim et al., I980. Australian sea lion + Domrow, I974. P araf ilaro ides Stellerrs se a Iion P. decorus Stroud and Dailey, I97B; Stroud and Roffe, 1,919. P. nanus Dougherty and Herman, 1947; Delyamure, 1955. P. prolificus Dougherty and Herman, 1947; Margolis, I956. California sea lion P. decorus Dougherty and Herman, 1947; Delyamure, 1955; Dailey, I97O; Fleischm an and Squire, I97O; Migaki et al., I97L; Morales and Helmbotdt, L97I1' Sweeney, I974; Sweeney and GilmarIin, I974; Stroud and Dailey, 1978. The larqer species, O. attenuata was recovered throughout the nasal tur- binales; the hexapod larvae predominantly on the anterior, or maxillary turbinates (Fig.95); the large octopod females deeply embedded in the ethmoid turbinates (Fiq. 96). A few larvae and adults were also found in lhe nasopharynx. The larvae r57. of the smaller species, O. diminuata were found in the nasal lurbinates (Fiq. 95), nasopharynx and trachea (Fiq. 97), whilst the adulls were only recovered from the f ixaLive collected from the lungs. The live larvae collecled from the nasal turbinates of animal AS25 were incubated in petri dishes containing a shallow layer of physiological saìine at room temperature for variable periods of time up to )B days. Of the total of 15 live larvae recovered, two were O. diminuata and ll were O. attenuata. Both O. diminuata specimens died within 48 hours of incubation during which time one had moulted to produce an octopod adult female. Only three of the O. attenuala larvae moulted to become adult females. These started to moult between 3 and 6 days post incubation. No further evidence of moulting took place after B days' incubation even though none of these adults had completely shed their cuticles. Over the next two days these three adult females died. Four O. attenuala larvae survived incubation for as long as 38 days but showed no signs of moulting. At no stage were nymphal forms observed. (b) Worms On gross examination the lungs of all animals were apparently normal. However, on close inspection a small number of creamish-white subpleural nodules were found in all Iungs and section of the lung revealed a number of similar lesions (Fiq. 9B). Some of these nodular lesions were found to contain adult worms (Fiqs. 99 and 100). Specimens of adult females, an adult male and larvae which were recoveled from bhe lung fixative have been preliminarily identified as a new species of Paraf ilaroides (M.D. Dailey, personal communication). The basis for this tentative identificalion is lhe very small size of mature wotms, parlicularly the male, when compared to P. decorus which is lhe most common lungworm in the California sea Iion (Dougherty and Herman, L947; Fleischman and Squire, I97O; Sweeney, 1972; Sweeney and Gilmartin, 1974) and is also common in the Steller's sea lion (Stroud and Dailey, I97B). l5B 5.2.2 Microscopic (a) Nasal turbinates In the turbinates the larval mites were found on the mucosal surface, often within a thick layer of mucus (Fiq. LOl), or deeply embedded in the mucosa (Fig' 102). The adull female O. aL,tenuata were generally found with their heads deeply embedded in the mucosa with only their elongated, gravid opisthosoma protruding into the lumen (Fiq. 96). Where the mites had become deeply embedded in the mucosa there was marked erosion and flattening of the epithelium, wilh minimal associated haemor- rhage (Fiq. l-02). There was an inflammatory infiltrate into the lamina propria often involving the periosteum of the bony trabeculae. This infiltrale was prin- cipally mononuclear in cell type, with macrophages, lymphocytes and plasma cells. The bony trabeculae showed areas of osleolysis adjacent to the mite wilh new bone formation on the opposite surface (Fiq. f03). The cellular exudate surround- ing the m ites was m ixed in type, consistinq of both m ononuclear cells and neutrophils (Fig. 104). Intact areas of turbinate mucosa were covered, in part, by a purulenl exudate characterised by polymorphonuclear cells and macrophages (Fig. 101). (b) Nasopharynx and trachea Larvae of both species of mile were found sitting on and embedded in the epithelium of the nasopharynx and trachea (Fig. I05). The mites tended to flatten the underlying epibhelium with some erosion and ulceration at the points of attachment (Fiq. 107) and the feeding site (Fiq. 105). As a result of this ulceration the mites were surrounded and covered by an exudate of mucust some red blood cells, f ibrin and cellular debris (Figs. 106 and 107). The lam ina propria immediately subjacent to the mites was thickened due to a heavy infiltrate of mononuclear cells, particularly macrophages, some lymphocytes and plasma cells, and many eosinophils (Figs. 107 and l0B). I59. (c) Lung Intact adult worms, males and larviparous females, and larvae were occasionally observed Iying free in the lung parenchyma (Figs. 109' 110 and l-ll). The presence of these worms elicited either no obvious inflammatory response or a minimal cellular reaction. This consisted of hypertrophy of the epithelial cells of the alveolar septa, an exudate of plasma cells, small lymphocytest some macro- phages and an occasional multinucleate gianl cell in the affecbed alveoli (Fig. IL2). There was minimal involvement of the adjacenl alveolar septa (Figs. 109 and l-l-2). Associated with this reaction was a variable amount of hyperaemia of nearby blood vessels (Figs. l-09 and l-I0) and a predominantly lymphocytic inLerstitial infiltrate of adjacent bronchi (Figs. lIl and lLl). In the terminal and lobular bronchi the larvae caused an acute bronchitis someLimes with an associated bronchopneumonia. The acute bronchitis was characterized by an increase in mucus secretion, a fibrinous exudate with many neutrophils and some macrophages (Figs. 114 and ll-5), and acute congestion of the bronchial and surrounding alveolar vessels with a moderale interstitial infiltrate of lymphocytes. The bronchopneumonia consisted of bronchi filled with inflam- matory cells (Figs. 116 and 1l-7) as were many of the immediately neighbouring alveolar structures (Fiq. I16). The bronchial interstitium was often heavily infiltrated by tymphocytes (Fiq. 117). The pneumonic inflammatory response was often centred around individual larvae or intact adult wotms (Figs. J.LB, ll9 and L21). The cells involved were predominantly mononuclear, particularly Iympho- cytes and macrophages with some plasma cells and multinucleate giant cells (Figs. lL9 and I2l) with an occasional polymorphonuclear cell. The intense pneumonic reaction often involved complete and adjacent acini (Fig. I2l) or even compleL,e lobules (Fiq. 120) being contained by the interlobular or inLeracinar septa. In some regions the reaction was so intense as to obliterate the alveolar septa (Figs. 1lB, 119 and 121) or to even disrupt their integrity (Figs. Il9 and I23). In other areas the inflammatory response had become purulent due l_60. to the degeneration and necrosis of the inflammatory cells (Fiq. I22). An occasional larva was observed migrating through the perivascular tissue (Fiq. 124). (d) Other pathology In some ateas there was a marked acute hyperaemia of the vessels of whole acini. This hyperaemia involved the vessels in the walls of the terminal airways and all the alveolar walls; however, there was no fluid or cellular exudate into the lumen nor was there any cellular infiltrate into the alveolar walls. Some bronchi associated with these lesions were heavily infiltrated with lymphocytes, whilst others showed apparent marked hyperplasia and hypertrophy of bronchial muscle and had variable numbers of lymphocytes in their walls. A number of these lesions were examined by exhaustive serial sectioning but no aetiological agent could be identified. Occasionally an atelectic acinus was found in which the alveolar walls appeared thickened and fibrotic, possibly the result of more proximal bronchial obstruction. Figure 94- Diagram showing the distribution of pulmonary p arasites in Neophoca c tnerea. Figure 95. Macroscopic view of rostral por[ion of maxillary turbinates. Note lhe presence of larval mites on the mucosal surface; most larvae are Lhe larger O. attenuata (A), with small numbers of O. diminuata (D). 1.4x. Figure 96- Macroscopic view of ethmoid turbinates, showing embedded adult f emale O. atlenuata with protruding opist,hosoma ( t ). 1.2x. 161 ADt,lLT ORTHOHALARACHNE ATTENUATA ADULT ORTHOHALARACHNE DIMINUATA ,À LARVAL ORTHOHALARCHNE ATTENUATA ORTHOHALARCHNE DIMINUATA ADULT AND LARVAL PARAFILAROIDES SP 94 Figure 97. Scanning electron m icrograph of trachea showing larval O. diminuata. f[x. Scale = 500 Um. Figure 98. Maeroscopic view of lung parenchyma with pathologic subpleural nodule(t).fOx. Figure 99. Macroscopic view of lung parenchyma showing adult worms ( t ). 10x. Figure 100. Scanning eleclron micrograph of adult worms. Note minimal evidence of assoeiated pathology. 50x. Scale - 500 ym. 162 I )sa Figure LOI. Macroscopic view of larval mites on maxillary turbinate. Nole mucous layer covering larvae. llx. Figure J.02. Light micrograph of larval mite embedded in nasal turbinate. Note erosion and flattening of epithelium. Glycolmelhacrylale. 70x. Scale = 250 um. Figure I03. Light micrograph of embedded larva. Note inflammatory infiltrate into lamina propria ( t ), osteolysis (ol) and new bone formation (n b) of turbinate bone; and purulenl rhinitis (x) of inlact epithelium. Glycolmethacrylate. 140x. Scale = l-50 Um. Figure 104. High power light micrograph of embedded larval mite (M) showing cellular exudate of mononuclear cells and neutrophils. Glycolmelhacylate. 220x. Scale = L00 Um. 163 "lt -ã ¿' 103 \, Figure 105. Scanning electron micrograph of larval mite on tracheal surface. Nole ulceration at feeding siLe ( t ) with some associated haemorrhage and exudate. 140x. Scale = 100 um. Figure 106. Scanning electron micrograph of larval mibe on tracheal surface covered by exudate of red blood cells, fibrin, inflammatory cells and cellular debris. l_40x. Scale = 100 um. Figure 107. Light micrograph of larval mite embedded in epithelium of naso- pharynx. Note exudate of ned blood cells, inflammatory cells and mucus ( t ); protrusion of "fool" (f) into basement membrane at point of altachment, and cellular infiltrale into lamina propria (l p). H and E. 190x. Scale - 100 um. Figure 108. High power light micrograph of nasopharyngeal epithelium showing flattening of epithelium (ep), cellular infiltrale in lamina propia (t p), and inflam- matory cells miqraling through the basement membrane ( t l. H and E. 42Ox. Scale - 50 pm. 164 ( a T t I a' ì I i I (' ) I lp t a r;, \ *'' ( i \ .'_ep \ a \ t- ',{ ,ì \ '\r f i I "l ¡ Figure J.09. Light micrograph of larviparous female Parafilaroides I yrng ln an alveolar ducl. Note minimal inf lammatory response wilh engorgement of adjacenl blood vessels ( t ). Gtycolmethacrytate. 70x. Scale = 200 um. Figure J-I0. Light micrograph of larval Paraf ilaroides Iying free within an alveolar sac. Note absence of inflammatory reaction. H and E. 140x. Scale = l-00 um. Figure IlI. Liqht micrograph of adult male Parafilaroides lying free within an alveolar duct (P). Note absence of inflammatony response immediately adjacent to worm, but lymphocytic infiltrale inlo interstitium of nearby bronchus ( t ). H an¿ E. 50x. Scale = 500 um. Figure LIz. H iqh power light micrograph of adult Parafilaroides (P) showin g inflammatory exudate of plasma cells, lymphocytes, macrophages and multi- nucleale giant cells. Note also the thickening of the alveolar septal epithelium ( t ) and the Iimiied inlerslitial inflammatory infillration. H and E. 23Ox. Scale = 100 um. 165 t. \l' ¡t\L /4 l. ì J t , ¡ t .$, t g Ð , , I ¡ 109 11o! ît å.- t -r t t I \ 3g I û I I Or Figure tll. Higher power light micrograph of lobular bronchus showing lympho- cytic interstitial infiltrate. H and E. 250x. Scale = 50 um. Figure 114. Light micrograph of tertiary bronchus showing fibrinous exudale within Lhe lumen. H and E. 23Ox. Scale - 50 ym. Figure J-I5. High power light micrograph of tertiary bronchus showing cellular nature of bronchitic exudate. Note larval Parafilaroides (P) present within airway lumen. H and E. 140x. Scale = 50 um. Figure Ll6. Light micrograph frorn bronchopneumonic region. Note massive exudate into bronchial lumen (Ð and alveoli (alv). H and E.50x. Scale = 250 um. (o (o r È-\._._ ¡r' , a ') a/ / ?t ,l I I Figure ll7. High power light micrograph fnom bronchopneumonic region showing the fibrinous exudate with mononuclear cells in the bronchial lumen (L), and the lymphocytic infiltrate in the inlerstitium ({c). H and E.240x. Scale = 50 pm. Figure ll8. Ligh micrograph of pneumonic reaction sumounding two larval Para- filaroides (P). H and E. 140x. Scale = 100 pm. Figure 119. Light microqraph of pneumonic lesion. NoLe larval Parafilaroides (P); and the chronic inflammatory exudate with disruplion of alveolar structures (alv). H and E. 210x. Scale - 50 ¡rm. Figure 120. Low power light micrograph of subpleural parenchyma showing patho- logic involvement of whole acini and lobules. H and E. lOx. 167 \a \I I 119 - Figure l2ì-. Liqht micrograph of subpleural lesion. Note inflammation surrounding larviparous females, and involvement of adjacent acini (åç). H and E. JOx. Scale = 500 um. Figure 122. Liqht micrograph of pneumonic region showing adult Parafilaroides (P), and degeneralion and necrosis of inflammatory cells to form purulent exudate (x). Note the interstitial mononuclear cellular infiltrate ( t ). H and E. B0x. Scale - 200 ym. Figure I25. High power light micrograph of pneumonic exudate of lymphocytes. plasma cells, macrophages and multinucleate giant cells (#). Note disruption of alveolar septa ( t ). U and E. 200x. Scale - 100 um. Figure L24. High power liqht micrograph of larval Parafilaroides m igrating through perivascular tissue ( t ). H and E. 100x. Scale = 50 um. 168 I :ò' t I ) v I Jí"'' l¡ â. r1 ,f\ ¡ I J I a ìa"' t¡ I t n a ì: (¡ -- r69 5.3 DISCUSSION 5.f-l In vitro Development The in vitro development of larvae of bolh species occurted more rapidly than described by Furman and Smith (L971). This may possibly have been due to the extended time from capture of the host animal to the examination of the skutl. Neither probonymphal nor deutonymphal octopod instars were noled during the rapid developmenl of either species. 5.3.2 Pathology (a) Mites The lesions in the nasal cavities and nasopharynx associated with the mites were essentially the same as those described previously in other otariids (Kurochkin and Sobolewsky, I975; Dunlap et al., )'976; Kim et al., 1980; Fay and Furm an, I9B2). The distribution of larval mites was similar to that described in other otariids (Dunlap et al., L976; Kim et al., 1980; Fay and Furman, 1982). Most adult females of O. attenuata were found embedded deep within the eLhmoid turbinates, whilst only very few were found in the nasopharynx, the preferred habitat in olher host species (Dunlap et al., I976; Kim et al., 1980). Further detailed inspection of the mites to delermine the numbers present in each Iocalion was not undertaken. No mites were found associaled with any parenchymal lesions and the most distally situated mites observed were in the terminal bronchi. However, some of the lesions in the parenchyma, with no apparent aetiology, were similar to those described for lunq mite infestation in OId World monkeys (Kim, 1977; 1980). Despite this, until an adult O. diminuata is demonstrated within a parenchymal Iesion it can only be speculated that those recovered from lhe lung washings were responsible for these, ot anyr lesions. (b) Worms Fatal verm inous pneu monia as a consquence of inf estation with Para- filaroides spp in young otariid seals has been well documenled (Fleischman and 170. Squire, L97O; Dailey, L97O; Migaki et al., L97L; Morales and Helmboldt, l97J-; Sweeney and Gilmartin, 1974; Stroud and Dailey, I97B; Slroud and Roffe, 1979). It has been suggesled that the mosl severe lesions of the verminous pneumonia are the result of a hypersensitivity reaction in response to dead or degeneraling adull worms. This was indicated by the observation that the most severe reaction often surrounds degeneraLing adults and that these lesions contain many eosinophils (Fleischman and squire, L97o; Migaki et al', 1971)' A pronounced inflammatory reaction has also been observed in response to the presence of intacL larvae in the airways and parenchyma (Fleischman and Squire, L97O; Morales and Helmboldt, 1971). In terrestrial species an intense mixed cellular inflammatory reaction occurs in the lung in response to the migratory phase of larval nematodes regard- less of whether or not lhe adult predeliction si[e is in the lungs (Jubb and Kennedy, I97O; Smith et al., I972; Leid and Williams, I979).In the present study at intense reaction was observed bolh in relation to the larvae and to the intact adult worms. However, those adults surrounded by an inflammalory response wele generally larviporous females so that the reaction may have been to lhe presence of newly released larvae rather lhan to the adults. If this were in fact the case then the observed response would be more in line with lhat reporled for olher otariids (Fleischman and Squire, I97O; Migaki eL al., I97I; Morales and Helmboldt, L97l). In other olariid species the strong immunological reaction to the worms would appear to be importanl in limiting the infection rate in older animals (Morales and Helmbotdt, 197I; Stroud and Dailey, 1978) and to even resull in voiding the host of these parasites by about four years of age (Sweeney, I974; Sweeney and Gilmartin, L91Ð. This is supported by the results of a survey of 5l stranded California sea Iions in which lungworms were most prevalenL in lhe l5 to 20 month-old age group (Sweeney and Gilmartin, l97lr). However, in contrasb to this, a smaller survey of adull California sea lions found Iungworms [o be present in eighl of eleven animals examined (Fleischman and Squire, 1970). Four of the 17 l. sp ecimens of Neoph oca exam ined in the present study were adult breeding females, whilsl the fifth animal was an immature female. All five of these animals were found lo be infected with lungworms, so despite an immunological reaction to the presence of lhe worms, it would appear that in those animals examined it was insufficient to rid them of these parasites but at the same time it may well have limited the degree lo which the hosts became infested. Within the infecLed animals, the distribution of the disease may also have been limited by the presence of the interacinar and interlobular septa. These septa would tend to contain the parasites within individual acini or lobules as well as limit the spread of the inflammatory cells and exudate as the presence of these septa precludes any channels for collateral ventilation between adjacent parenchymal units. The specimens of Neophoca examined in this study were all apparently healthy ab the time of capture. Some animals in the local populations did, however, show several of the symptoms observed in infected California sea lions (Fleischman and Squire, L97O; Sweeney, I974; Ogata et al., I919). These included sneezing, "coughing", and the discharge of tenacious phlegm, but none of the more severe symptoms found in moribund, infected animals. Thus it would appear lhat the presence of orthohalarachnid mites and Parafilaroides worms are of minor consequence generally, but may be important in younger animals or during times of stress such as diving when the full respiratory capacity may be required. The presence of mites, together with the associated inf lammalory response, in the nasal passages may, if they reached sufficient numbers, inhibit the high expiratory flow rate which has been demonstrated for other olariids (Kerem et al., L975; Matthews, 1977). A potential obstruction due to the mites was not apparent in lhe animals examined. By involving whole acini and even whole lobules the worms present in lhe lung would reduce the amount of the alveolar surface available for potenLial gas exchange. The proportion of the lung parenchyma affected by these parasites has nol been delermined nor has lhe distribution of Iesions, although from preliminary examinations of the maberial available, mosl 172. lesions occurred subpleurally. If infeslation with lungworms were to be a pro- gressive condition, ralher than a self-limiling one as proposed for other otariids (Morales and Helmboldt, l97l; Sweeney and Gilmartin, ),974; Stroud and Dailey, I97B), then it could result in the situation where the lung's function as a gas exchange organ is severely impaired. 5.4 SUMMARY Two species of orthohalarachnid m ites were present in lhe respiratory system of the Australian sea lion. Most larvae of both species, Orthohalarachne attenuata and O. diminuata were found lying on the nasal turbinates with small numbers in the nasopharynx and trachea. Adult female O. altenuala were deeply embedded in the epithelium of the ethmoid turbinates, whilst adult O. diminuata were only recovered from the fixative from the lungs. The mites caused localized erosion and ulceralion of the epithelium. Specimens of a sp ecies of Parafilaroides were found in nodules and I ying free in the parenchyma; some were also recovered from the lung fixative. The most common pathology associated with the larval and adult worms was eilher an acute bronchitis or bronchopneumonia. Occasionally worms were present in the paren- chyma with no surrounding inflammatory response. With gross infestation by the nasal mites the associated response may well cause an obstruction to air flow in the nasal passages so thal the generally high expiratory flow rale would be reduced. For this to occur the mites would need to be present in very large numbers. The presence of worms in lhe lung parenchyma together with the associated inflammatory reaction will lead to a reduction in the available gas exchange surface. In large numbers these parasites may severely impair the normal respiralory function of the lunqs. r73. Chapter 6 DISCUSSION 6.I LUNG STRUCTURE The lungs of the Australian sea Iion possess several structural characteri- stics which distinguish them from the Iungs of lerrestrial species. Foremost amongst these, being obvious both grossly and m icroscopically, is the great increase in the number of thick septa throughout the parenchyma and the amount of connective tissue within them. These septa divide the lung into discrete, inde- pendent units of smaller and smaller size. The largest of these septa divide the parenchyma into large bronchopulmonary seqments which are furlher divided into many pulmonary lobules. Each lobule contains a variable number of acini also separaled from each olher by thickened septa. Within each acinus are two or more alveolar ducts, a number of alveolar sacs and many alveoli whose septal walls are thickened by an increased content of connective tissue. The alveolar septal thickening is so exlreme in parts that approximately one-quarter of the alveolar septa contain a double capillary network, a feature which in adult mammals is unique to various aquatic species. Another prominent feature of the lungs of Neophoca, also observed in other oLariids (Denison and Kooyman, L973), is the reinforcemenl of Lhe lerminal air- ways, here termed lobular bronchi, with hyaline cartilaqe plales and a thickened muscularis. The epithelial tining of these airways is, for the most part, similar to that of the bronchioles in terrestrial mammals. One feature of this epithelium in these animals, rarely observed in Lerreslrial species and only occasionally observed in oLher marine species, is Lhe presence of significanl numbers of alveolar type II-like cells in the most distal porlions of these airways. Although no histochemical studies were conducted on lhese cells, ultrastruclurally they are almosl identical to lhe type II cells of the alveolar septa. 17 4. 6.2 LUNG MORPHOMETRY Morphomelrically there are a number of parameters which show marked differences from those predicted for a terrestrial species of similar body weight. The [olal lung voìume is 500/o greater than predicted, a well recognised feature of marine mammals lungs (Goudappel and Slijper, J.958; Kooyman, 1973), The alveolar surface area is about ll% less than predicted, whilst bhe tolal capillary surface area is 23% greaLer than predicted, resulling in a Sc/Sa ratio of l.l-l which is the reverse of lhal found in most lerrestrial species. However, when the capillary surface of the double capillary septa and the interlobular and interacinar sepla which is nol involved in gas exchanqe is considered, the effective Sc is 14% less than predicted, and the Sc/Sa ratio is 0.7I, similar to that in many terrestrial species (Gehr et al., 1978a; Gehr and Erni, 1980; Gallivan, 1981b; Gehr et al', IgBIb). The totat capillary volume is 60% grealer than predicted, in part due to the double capillary septa. The tissue harmonic mean barrier thickness is l47o greater than predicted, whilst the plasma harmonic mean thickness,0.09 um, is similar Lo the smallest yet measured in mammals; several species of shrew (Gehr et al., f9B0). The net result is a maximal oxygen diffusing capacity 26¿/o gteater than predicted, whilst consideration of the approximalions for lhe effective Sc results in a diffusing capacity 18% grealer lhan predicted. Therefore the true maximal oxygen diffusing capacity for adults of this species lies in the range 4.281 Lo 4.582 ml o2.sec-l.tb"r-1. In marine mammals the blubber accounts for 20-1070 of the total body weight (Kooyman, I973) whilst in humans the fat component is only 15% of total body weight (Schmidt-Nielsen, I97Ð. However, the proporLion of body fal is directty correlated with body size, and caged wild species exhibil increased fat accumulation (Pitts and Bullard, I96B). Thus some of the larger wild African species in the study by Gehr g!_el. (l98lb) may have had a body fat content approaching that of smaller marine mammals. Therefore, had lhe allometric equalions been based on tean body weight, Lo consider only the metabolically r75. aclive tissues, the morphomelric results for the sea lion could be expected to have been further increased when compared lo the terrestrial mammals. 6.3 DIVING PHYSIOLOGY Despile these differences in structure and morphometry, the role of the lung in enhancing the diving duration of these animals would appear to be minimal. As outlined in [he Introduclion (Chapter l) there are a number of non-pulmonary aspects which characterize diving duration in marine mammals. Of particular importance is the obsenvalion that the vast m ajority of dives, in pinnipeds at least, are well within the maximum duralion possible for each particular species. Not only are most dives much shorter Lhan the maximum possible, m'ost are also within the aerobic dive limit for that species. (Kooyman et al., t9B0; Kooyman, regz). 6-l.I Aerobic Dive Limit The prolonged dive limit displayed by pinnipeds is a result of two features in particular, the large total body oxygen store available during a dive, and the initiation of the diving response upon submersion. (a) Body oxygen stores The major conLribution to the total body oxygen store is that of lhe blood due to an increased oxygen storage capacity. This large blood oxygen slore results from increases in the blood volume, haematocrit and haemoglobin concentration (Lenfant, 1969; Clausen and Ersland, 1969; Lenfant et al., I97O; Ridgway et al., I97O; Simpson et al., 1970). The total body oxygen store is augmented by an increased muscle oxygen storage capacily due to an increased myoglobin con- centration (Man'kovskaya, I975; Blix, I976; Castellini, I98I; Castellini and Somero, LgBI). The degree to which the lunq oxygen conLributes to the lotal body oxygen store during diving is largely species dependent. Phocid seals tend to exhale eilher prior to diving or during the inilial descent phase so that the diving lung volume is only 20-60% of the total lung capacity (Scholander, 1940; Kooyman r1 6. et al., I970; I972). Otariids and cetaceans are generally considered to dive on full inspiration (Kooyman and Andersen, J-969; Ridgway et al., 1969; Kooyman, I973). However, the diving lung volume of California sea lions, *ni"n have been observed to exhale small volumes during descent, has been eslimated to be 5O-9OVo of their vital capacity (Kerem et al., I975; Malthews, I977). It would therefore appear that otariids and cetaceans have a greater reliance on lung oxygen during the early stages of a dive than do phocids. Regardless of the diving lung volume, all marine mammals studied undergo progressive compression collapse of the lung due to the increased hydrostatic pressure at depth (Kooyman, I973). Therefore the effectiveness of the lung as an oxygen store will diminish with increasing depth and will cease beyond the depth at which complete collapse occurs, estimated to be about 70 m (Kooyman et al., I97\ Ridgway and Howard, I979). (b) Diving response The diving response, which has been shown to be common to many verte- brale classes, is most readily observed as a bradycardia. The intensity of this bradycardia is not the result of a simple reflex response. In the Weddell seal it has been demonstrated that a more profound decrease in heart rate occurs with an increase in the anticipated dive duration up to about l5 min. (Kooyman and Campbell, 1972) which is reasonably similar to the aerobic dive limit of 2O to 26 min. for this species (Kooyman et al., l-980). Associated with this bradycardia is a decrease in cardiac output and a peripheral vasoconstriction so that arterial blood pressure is maintained, or even increased slightly (Irving et al., 1942; Elsner et al., 1964; Elsner et al., I96Q Murdaugh et al., 1968; Sinnett et al., I97B; Zapol et al., I979; Blix et al., 198l). As a consequence of the peripheral vaso- constriction, perfusion through the peripheral tissues and organs is reduced slightly during aerobic dives but is markedly reduced, or even completely stopped, during maximum breath hold dives (Bron et al., 1966; Zapol et al., 1979; Blix et al., 1983). r71. 6.J-2 Biochemistry Biochemical studies have indicated that marine mammals have no majon adaptations to enhance anaerobiosis but, rather, their biochemisLry is geared to minimizing the effects and clearance time of anaerobic end-products (Hochachka et al., 1977; Murphy et al., J.980; Castellini eb al., l9BÌ; Hochachka, 1981). The skeletal musculature, where Lhe majority of the primany anaerobic end-product, lactate, is formed, has an increased bufferinq capacity (Castellini and Somero, l9Bl). The buffering capacily of the bìood is also increased to minimize pH changes during the laclate washout from the muscle when the peripheral perfusion is re-established at lhe end of the dive (Scholander, J-940; Lenfanl, 1969; Lenfant et al., I97O; Lutz, I9B2). There are also biochemical modif icalions which allow the central organs to utilize laclate during and after a dive but particularly during recovery, when blood lactate levels are highest, so that the relurn to pre-dive levels is hastened (Hochachka et al., I977; Murphy et al., L9B0). 6-l.f Pulmonary Function The struclural modificalions of the pulmonary sys[em appear Lo be of major importance in enabling the Australian sea lion to rapidly load lhe body oxyqen stores prior lo a dive and to limit the absorption of nitrogen when diving Lo depth. Otariids and cetaceans have a large lidal volume which approaches B0 to 9O% of total lung capacity, and as such is very similar lo their vital capacity (Irving et al., I94I; Scholander and Irving, I94I; Olsen eL al., 1969). Normal res- piration generally involves a single respiratory cycle of exhalalion followed by inhalation during a brief period at lhe water's surface. This type of respiration results in high flow rates, particularly during expiration, and near complete alveolar emptying al end-expiration (Kerem et al., I975; Malthews, 1917; Kooyman and Sinnelt, 1979; Kooyman and Cornell, l-98L). It has previously been suggested (Denison el al., I97I; Denison and Kooyman, 1973; Kooyman, 1973) that the attainment of such high flow rates is made possible by the reinforcement of the terminal airways; this is assisted by the elastic recoil of the lung, a flexible r-78. thorax and the obliquely oriented diaphragm (Slijper, 1962). In lerreslrial mammals peak expiratory flow rates are achieved at about 80%o tolal lung capacity and lhen rapidly decrease until flow is obstructed by collapse of the lhin- walled bronchioles at low lung volumes (Hyatt et al., 1958; Pride et al., 1961; Macklem and Mead, l96B). In otariids the peak expiratory flow rale is mainlained at a plateau even down to lung volumes as low as 2070 of total lunq capacity (Kerem et al., 1975; Matthews, 1977). Expiration by marine mammals is generally considered to be passive, as with humans (Olsen et al., 1969). However, Kerem g!-e!. Q915) have suggested that in the California sea lion lhe variable expiralory flow rates for a constant expired volume are dependent on the muscular effort that can be developed from a par- ticular chest wall position, whereas in terrestrial mammals the maximum expiratory flow rates are independent of lhe expiratory effort and depend in part on the elastic recoil of the lung (Leith, I916). Despile an increased interstilial connective lissue component in marine mammal lungs, the lung elastic recoil and compliance are similar to those of terrestrial species (Olsen et al., 1969; Gillespie, l98l). However, the thoracic wall of marine mammals has a relatively hiqh com- pliance (Gillespie, IgBi). This resulls in a very low lung relaxation volume so lhal the functional residual capacity differs very little from lhe residual volume and minimum alveolar volume (Leith, I976). The lung relaxation volume in phocids, and probably most aquatic species, only constitutes aboub 3.5o/o of the lotal lung capacity as compared to about 20 to 6O% in many terrestrial species (Gillespie, re93). Such a low functional residual capacity means that inflation, which starts al a low lung volume, needs to overcome the tendency for the alveoli to collapse further. This is achieved by the surfactant which reduces Lhe alveolar surface lension, and by lhe distribution of inflationary forces throughout the parenchyma along the connective Lissue networks. The surfactant of marine mammals has rarely been examined, bul in two reports fnom pinnipeds the amounl of sunfactant r19. present, Lhe phosopholipid components and surfactant activity, were found to be similar Lo terrestrial mammals (CIements et al., I97O; Denison et al., f97L). However, there is one report, as a personal communication, that the amount of surfacIant present in the lungs of the Weddell seal is considerably less than that in terrestrial mammals (Boyd, I975). More recently the effect of intermittent increased tidal volumes on surfactanl release has been examined in terrestrial mam mals. Lavage of isolated perf used rat lungs af ler a single deep brealh indicated that surfactant release had been increased, and it was also observed that this large brealh resulted in the opening of ateleclic alveoli and an increase in the oxygen diffusing capacity (Nicholas el al., I9BZ). Monphological examination of rat alveolar type II cells after periodic deep breaths of four times the normal tidal volume, indicated that an increased tidal volume slimulated surfaclant release (Massaro and Massaro, 1983). Whelher or not a normally larqe tidal volume, such as lhat of marine mammals, has a similar effect on surfactant release has yel to be determined. If this is in fact the case, it may help compensate for the apparent reduclion in the volume of alveolar type II cells in the lung of lhe Australian sea lion. As a consequence of lhe reduction in surface lension by the sunfactanl, lhe inflationary pressure required lo reinflate or reopen Lhe alveoli is kept lo a minimum. The inflationary pressure which is crealed by active inspiralion involves the flexible thorax which is capable of moving relaIively large distances during respiration (Leith, I976) and the obliquely oriented diaphragm which is common lo many marine mammals (Green, I972). The large inspiratory volume, moving at a reduced flow rate as compared to expiration (Kerem et al., 1975; Matthews, 1977; Kooyman and Cornell, 19BI) tends to fill the lung lo near total capacity so that alveolar recruitment will be al or near its maximum during normal respiralion. Associated with the large tidal volume is a respiralory arrhythmia which leads to an increased heart rate during inspiration with an increased cardiac output and hence a grealer pulmonary blood flow (lrving et al., 1963; Kooyman and IBO. Campbell, I972; Lin et al., L972; Casson and Ronald, I975; Tanjii et al., I975; Dormer et al., I977;Pasche and Krog, 1980). This increased blood tfà1r¿ wiff lead to a greater pulmonary capillary filling and/or capillary recruitment during inspira- tion when alveolar oxygen tensions are maximal. Therefore the increased alveolar recruilment, resulling from the large lidal volume, together with the increase capillary filling and/or recruitment should result in less ventilalion-perfusion mismatching. This in turn will enable better utilization of the available diffusing capacity than occurs in terrestrial mammals at rest. Reduction in ventilalion-perfusion inhomogeneity has been suggested as a factor in the increase of the oxygen diffusing capacity in heavily exercising humans (Veicsteinas et al., I976).In exercising humans the respiratory and cardio- vascular events partially parallel those of marine mammals at rest; there are increases in oxygen consumplion, ventilalion and cardiac output so thal the physiologically determined pulmonary diffusing capacity during exercise is increased nearly f.5 Limes above that at rest (Veicsteinas et al., I976). Although the physiologically determ ined oxygen dif f using capacity is increased during exercise, this value is still only 5O-65Vo of that measured morphometrically (Siegwart et al., I97I; Weibel, I972; I973; I979b; 19Bl). This is in part due to the fact that instillalion fixation of the Iung exposes the tolal alveolar surface so that the diffusing capacity is 20 to 50% greater than that for air-filled perfusion-fixed lungs (Weibel et al., I973). Whether lhe morphometric diffusing capacity of marine mammals is similarly I.5-2 times greater lhan that determined physiolo- gically during heavy exercise in humans (Weibel, I98l) has yel lo be established. However, it has been demonstrated that the oxyqen utilization by the lung of marine mammals is almost twice that of terrestrial species (Scholander and Irving, 1941; Andersen, 1,966). This implies a greater utilization of the available diffusing capacity, which may in part be a reflection of the inspiralory apnoea common to many marine mammals (Scholander, J.940; Scholander and Irving, I94I; Olsen el al., L969). lB1. During normal respiration, particularly in otariids and cetaceans, lhe ventilation and perfusion of the lung are likely to be well matched since alveolar recruitment approaches the maximum as does lhe capillary filling and/or recruit- ment. However, there is evidence in terrestrial mammals that on a microscopic level the ventilation and perfusion of the Iungs are not ideally matched due to the stratification of oxygen tensions within the alveolar duct system (Adaro and Piiper, I976; Piiper, I979; Nixon and Pack, 1980; Weibel, 19Bl). This occurs because ventilation of the alveoli, or more accurately ventilation of lhe alveolar surface of the blood-gas barrier, involves two mechanisms: one, the bulk flow of inspired air to the level of the respiraLory bronchioles, the other, diffusion of oxygen from the respiratory bronchioles to the blood-gas barrier along an oxygen concentration gradient (Weibel, I9Bl). It has been suggested that the depth to 'which the bulk flow of air reaches in the alveolar duct system is dependent on the tidal volume, with deeper penetralion occurring with larger tidal volumes (Weibel, 1981). In marine mammals bulk flow can therefore be expected to extend well within the alveolar duct system due to the large tidal volume. It has furlher been suggested lhal the degree of oxygen stratificational inhomogeneity depends on the lenglh of lhe alveolar duct system (Weibel et al., lg8]b). These aulhors report that preliminary evidence suggests a body size dependence for this duct system, so lhat in larger animals it is longer which results in a reduced oxygen tension at the blood-g as barrier. In Neophoca, although the alveolar ducl length has not been delermined, the structural arrangement of the acinus whereby the first generalion alveolar ducls arise directly from Ihe reinforced lobular bronchi, will assisl the inspiralory bulk flow and hence this dislance for oxygen diffusion and the stratifi- calion of oxygen tensions. Weibel (i-98l) has proposed a model of "parallel perfusion and series ventila- tion" in which lhe capillaries of all alveoli within an acinus are equally perfused whilsl the ventilation of alveoli is dependent on the stratified oxygen diffusion gradienl within lhe acinus. In this m odel lhere is an axial oxygen dif fusion 182. gradient along the alveolar ducts and a radial oxygen diffusion gradient within lhe indiv idual alveoli. Weibel has suggesled this resulls in a higher oxygen con- centration within bhe alveoli closest to the alveolar duct enlrance and a lower concentration in those further distal, resulting in mismatching of venlilation and perfusion. These stralificational gradient,s are Iikely to be reduced in the otariid Iung due to the large tidal volume and the simple acinar geomelry. If this is the case then grealer advantage can be taken of the maximal diffusinq capacity as measured morphometrically. Thus it is likely that if the oxygen diffusing capacily were also Lo be determ ined physiologically lhere would be less discrepancy belween the two than observed in terrestrial species. 6.f-4 Increased Hydrostatic Pressure and Lung Collapse The degree lo which this increased pulmonary diffusing capacity will be affected during the early phase of a dive will depend on lhe rate of descent, the rate at whìch the aìveoli collapse and the way in which the alveolar septa fold during collapse. (a) Rate of lung collapse The averaqe rate of depth change for the Norlhern fur seal, is 70 m/min. (Kooyman el al., I916) whilst the mean rate of descent for the Weddell seal is abou[ 35 mlmin., with a maximum recorded of about 100 m/min. (Kooyman, L967). However, the descent rate for prolonged dives greater lhan 20 min. by Weddell seals, which are to a mean depth of ll0 m, is generally less than that for shorter deeper dives (Kooyman et al., I97l-). With such descent rates the depth at which complete alveolar collapse has been predicted to occur, 70 m (Kooyman el al., I97),; Ridgway and Howard, L979), will be quickly achieved. The descent rale, together with the accompanying increase in hydrostatic ptessure, will determine the rate of alveolar collapse. From a lung model, the ability fon alveolar collapse was predicted to depend on a small alveolar volume to dead-space volume ratio, and terminal airways which are at least five times as rigid as the alveoli (Denison and Kooyman, 1973). In phocids a small alveolar to dead-space volumes ratio is IBJ. achieved by exhaling to only 20-60%o of lolal lung capacity either before a dive or in early descent (Scholander, 1940; Kooyman el al., I970; I97Ð. Olariids and ceIaceans tend lo dive on full inspiralion (Kooyman and Andersen, 1969; Ridgway et al., 1969; Kooym an, I973), allhough the California sea lion has been observed lo exhale small quantities of gas on descent which resuìts in a diving lung volume of 5}-g}o/o of the vilal capacity (Kerem el al., 1975; Matthews, 1977)- The large diving lung volume will result in a slightly larger alveolar to dead-space volumes ratio, but this will be partially offset by the more extensive rigidly reinforced terminal airways which marginally increase the dead-space volume. Togelher with the fact thal the airways are considerably more rigid than the alveolar walls, this will ensure bhat alveolar collapse precedes thal of the airways. Evidence from phocid sea-[s in a pressure chamber indicates that the alveoli are in fact the first respiratory slructures to respond to increased hydrostatic pressure (Kooyman et al., 1970). Radiographs, taken of these seals when pressurized, demonstraled substantial but not complete collapse of the lrachea whilst Lhe medium-sized bronchi and "bronchioles" mainlained their pre-dive diamelers even at ptessures equivalent to a depth of 3O6 m. The authors concluded that there had been con- siderable collapse of the alveoli at pressures equivalent to Iess than 30 m depth. (b) Nature of lung collapse Sludies of lhe excised lungs from the California sea Iion have demonstrated that otariid lungs are capable of near complete alveolar collapse (Denison et al., L97l). These authors only examined lhis lung material light microscopically and so did not observe the way in which the alveoli collapsed. The only studies which indicate the nature of septal folding have been those conducted on ral lungs perfusion fixed al different positive inflation pressures (Gil and Weibel, I972; Weibel et al., 1973). The mechanies of these two situations will most certainly have been differenl, as lhe sea lion lungs were exposed to variations in exlernally applied pleural pressure. However, the fealure of interest is the behaviour of the alveolar septa wilh decreased inflation and lhe influence of this on the diffusing I 84. capacity. The major morphological feature at low inflation was that with foldinq of the septum the larger components, such as the epithelial and interstitial cell bodies and connective tissue fibres, were displaced into the depth of Lhe septum. In this way the m ajority of the ef f ective blood-gas barrier consisted of the thinnest portions, those of the fused epithelium and endothelium (Gil and weibel, Ig72). In the present study evidence of similar alveolar folding due to incomplete inflation was observed in occasional areas; however the degree of submaximal inflation was unknown. As the structural amangement of the alveolar septal components in Neophoca is similar to that of terrestrial species, it is likely that the folding of the septa, due to compression or at end-expiration, would be similar to that observed in the rat. In the rat a smooth alveolar surface was maintained at sub- maximal inflation pressures by the displacement of the surfactant which was thinnest over the thin exposed portions of the gas-exchange barrier and formed pools in the folds and plications of the thicker reqions (weibel et al', l97l)' This surfactant is important for alveolar stability and reinflation by minimizing the surface tensions as the alveolar surface area is reduced. In the present study, due to the method of fixation, the majority of this alveolar surface lining layer was removed leaving only occasional pockets of surfactant associated with septal f olds. (c) Effect of lung collapse on gas-exchange The morphometric results of the rat lung showed that with submaximal inflation the available pulmonary diffusing capacity is reduced by as much as 50 to 75% when compared to completely unfolded instillation fixed lungs (Weibel et al., Ig73).5imilar changes in the oxygen diffusing capacity of marine mammal lungs can be expected to occur with compression collapse which will reduce not only oxygen diffusion but also nitrogen absorption. Therefore the effect of alveolar collapse on limiting nitrogen absorption is dependent not only on total lung collapse, estimated to occur by about 70 m (Kooyman et al., I97I; Ridgway and rB5. Howard, I979), but also on the degree to which the alveolar collapse, prior to this point, reduces the capacity for nitrogen absorption. However, this decrease in the effective alveolar surface area and subsequent decrease in diffusion capacity will be partly offset by the increase in the partial pressure of nitrogen in the reduced lung volume. This leads to an increase in the nitrogen diffusion gradient, from alveolar air to capillary blood, resulting in enhanced nitrogen absorption across the exposed regions of the blood-gas barrier. That this is the case has been demonstrated in dolphins diving to less than 100 m (Ridgway and Howard, 1979). The results from that study indicated significant levels of nitrogen absorption at depths of less Lhan 70 m; therefore, despite partiaì alveolar collapse and diminished nitrogen absorption, marine mammals appear lo be potentially capable of absorbing sufficient nitrogen to suffer from nitrogen narcosis and decom- pression sickness. During simulaLed dives to depths between l0 and 212 m phocid seals displayed increased blood nitrogen tensions which were lower than the level theoretically possible at such depths. Even though the arterial nitrogen levels, particularly those in the harbor seals, were above the level at which nitrogen narcosis occurs in humans and bubble formation occurs in cats, these animals displayed no symptoms (Kooyman et al., 1912). 6.3-5 .Nitrogen Absorption and Decompression Sickness Repetitive breath hold dives by humans to depths of 15 to 20 m have resulted in decompression sickness (Paulev, 1965); therefore it would appear that marìne mammals are potentially capable of suffering from increased nitrogen absorption when diving, particularly to depths less than 70 m and when making repetitive dives. However, the mechanism by which they avoid such consequences remains to be elucidated. Kooyman (1973) outlined five possible mechanisms by which marine mammals may avoid the problems of excessive nitrogen absorption: i) qreater tolerance to nitrogen through greater solubility in tissues and blood; ii) distribution of nitrogen throughout the tissues; iii) restriction of the duration and depth of dives; t86. iv) the presence of tissues to specifically absorb nitrogen; v) prevention of nitrogen absorplion. There is no evidence available to fully support any of these mechanisms. Nitrogen solubility in seal blood was not found to differ significantly from that in human blood (Kooyman et al., \972). Although lhere is extensive peripheral vaso- constriction during dives to depth and during restrained or simulated dives (lrving et al., 1942; Van Citters et al., 1965; Elsner et al., L966; Stevens, L97I; Sinnett el al., I97B;Zapol et al., 1979), the results of Ridgway and Howard (1979) indicate an effective perfusion of skeletal muscle whilst diving to depths of at least 70 m. After divinq, lhe nilrogen tensions in the dorsal epaxial muscles of the dolphins were two to three times greater than lhat at rest at the surface, rèsulting in levels similar to arterial nitrogen tensions which have caused problems in humans (Kooyman et al., I972), Therefore this distribution of nitrogen would not appear lo be an effective prevenlalive mechanism as lhe levels present in the muscle lissue are sufficiently high to potentially cause problems with decompression during rapid ascenb. The diving behaviour of the few pinniped species studied in detail indicate that the duration, depth and frequency of dives are potenlially capable of result- ing in decompression sickness (Kooyman -et al., I976; t9B0; Kooyman, I9B2; Gentry ef al., lg8l), as has been shown to occur in humans (Paulevr 1965) and other ler- restrial mammals (enitp, I97Ð. No special nitrogen-absorbing tissues have yet been identified although nitrogen is approximately five times more soluble in blubber than other tissues, and Kooyman (1913) speculated that the blubber may act as a nitrogen sink during ascent when the peripheral vasoconstriclion relaxes. However, the nitrogen tensions in the blubber of repetitively diving dolphins was less than that in the dorsal epaxial muscles (Ridqway and Howard, 1979), so this also appears lo be an unlikely explanation. The prevention of nitrogen absorption by alveolar collapse has been discussed above and was also shown to be inadequate to prevent decompnession sickness. lB7. Another mechanism which may prevent decompression sickness in marine mammals relates to intravascular events which result from decompression. In terresL,rial species these include red blood cell clumping, rouleaux formation, the sludging of blood in small vessels, a decrease in the number of platelets, and haemoconcentration (Pnitp, I974:, Strauss, 1979) which are believed to resull from the gas-blood interface of nitrogen bubbles. Associated with these changes is the electron microscopic observation that bubble growth in decompression sickness involves the coating of the bubbles by a thin Iayer of fibrinogen, and the aggrega- tion of lipid micelles and platelets (Philp, I97Ð. There are no reports in the lilerature of similar changes in the blood of diving marine mammals. However, the in vitro clotting time of cetacean blood has been shown to be prolonged as compared to terreslrial species, whilst the blood of several celacean species is deficient in blood clotting factor 12 (Hageman factor) (Robinson et al., 1969; Ridgway el al., I91O; Ridgway, 1972). Whether or not these affect intravascular coagulation has not been established. The clotting time of blood from the Soulhern elephant seal, Mirounqa leonina, however, did not differ from thal of terrestrial mammals (Bryden and Lim, 1969). Platelet counts in marine mammal blood have rarely been undertaken with one report indicating that in the killer whale, Orcinus orca, the number of platelets were within the normal human range whilst those of the bottle nose dolphin were somewhat low (Robinson et al., L969). It therefore appears that differences in intravascular properties of marine mammals may be an importanl factor by which these animals are able to avoid the problems associated with excessive nitrogen absorption. 6.4 CONCLUSION The role of the lung in marine mammals is to rapidly supply sufficient oxygen for the body oxygen slores which will enable full utilizalion of their aerobic diving capabilily. The m ajor oxygen store is the blood, whose storage capacity is increased above that of terrestrial mammals. This results from ]- BB increases in lhe whole blood volume, haematocrit, and haemoglobin concenlration. Thus, in',rapid" breathers such as Neophoca cinerea, the role of the lung is to rapidly oxygenate this large blood pool as it circulates through the pulmonary capillaries during the animals'brief sojourn al the air-water interface. To achieve this a large readily available oxygen source needs to be delivered to the gas- exchange surface, which then needs to rapidly pass from the alveoli lo the erythroc ytes. The necessary supply of òxygen to the blood-gas barrier is made possible by several slructural, and associated physiological, modifications of the general terrestrial pallern. Foremost amongst these is the large tidal volume of the "rapid" breathers. The ability to move such volumes of air is achieved'in pant by the presence of cartilage-reinforced lower airways. These lobular bronchi remain patent at low Iung volumes and during periods of high air flow. The high expiralory flow rate and large tidal volume lead to a very small residual volume. Thus at end- inspiration the inspired oxygen concentration will rem ain relatively hiqh' as compared to the atmospheric concentration, due to a small physiologic dead-space volume of gas present to dilute the inspired air. The relatively high inspired oxygen concentration is then more fully utilized through the structural arrangement of the terminal respiratory unit, or acinus. In Neophoca each lobular bronchus supplies a number of acini whose alveolar ducts arise directly from the bronchial terminal cartilage rings. This acinar geometry, logether wibh the reinforced airways, enhances the bulk flow distribution of inspired air, and hence diminishes lhe distance necessary for inlra-ductal and intra-alveolar diffusion by oxygen. This reduction in stratificational inhomo- geneity within the acinus is com plem enled by an inspiratory tachycardia, and hence increased pulmonary blood f low, so that any ventilation-perfusion m is- matching will be minimized. The effect of Lhis increased oxygen concentration at the blood-gas barrier and the diminished ventilation-perfusion mismatching will be enhanced by the 189. increase in the effective morphometrically determined oxygen diffusion capacity of the lung. The increase in oxygen diffusion capacity, when compared to ter- restrial species, is achieved primarily through increases in the oxygen conduclance of the plasma componenl of [he morphometric model and the pulmonary capillary volume. The increase in the plasma oxygen concltlctance results from a markedly increased haematocrit which increases the volume of the pulmonary capillaries and consequently decreases the plasma harmonic mean thickness. If interspecies comparison of the pulmonary oxygen diffusing capacity were to consider lean body weight rather than total body weight, then the sea Iion oxygen diffusion capacity would be further increased when compared to terrestrial species. This is due to the large proportion, up to )Oo/o, of the body weighl of marine mammals which is relalively mebabolically inactive adipose tissue. The increased bulk flow of an increased inspired oxygen concenLralion which reaches the bÌood-gas barrier, together with an increased pulmonary diffusion capacity, lead to the high oxygen utilization which has been observed in marine mammals. The normal respiratory cycle of these animals contributes to the oxygen uLilization as there is an extended inspiratory pause even when resling on Iand. In this way the high oxygen requirement for the predominantly aerobic dives of marine mammals is made available by these structural and physiologic modifi- cations [o the respiratory system. The structural modifications of the respiratory system are also importanl for the ability of marine mammals to dive to depth. it has been well established that when mammals dive lo depth there is collpase of gas-containing organs and struct.ures due to the increase in hydrostatic pressure. In marine mammals, the nu.mber of gas-containing structures has been minimized, in particular the bony sinuses of lhe skull which are enlirely absent. The major gas-containing organ is the lung which in marine mammals has been shown to undergo near complete collapse at deplh. The only slructures wibhin the lung which mainlain their patency are the cartilage-reinforced intrapulmonary airways and the trachea, 1 90. whilst the alveoli collapse completely. This capacity of the alveoli to collapse prior to [he small airways is important to prevent rupture of [he alveolar slructures and to minimize nitrogen absorplion. Alveolar collapse is achieved by the increased airway rigidity of these species, and a small alveolar volume to dead-space volum e rabio. Sim ilar collapse of the alveoli occurs in "rapid" breathers at the end of expiration due to Lhe large tidal volume and the resullant small residual volume. Reinflation of the alveoli is made possible by two features, one physico- chemical and one struclural. The surfactant secreted by the type II cells of lhe alveolar epithelium, and possibly also the type II-like cells of the lobular bronchi, acts to reduce the surface tension of the alveoli as their volume, and hence surface area, is diminished. The effect of the surfactant is to minimize the force required to reinflate the alveoli durinq inspiralion and ascent, and to maintain alveolar stability at low lung volumes. The structural modification which assisls reinflation is lhe increase in the connective tissue septa throughout lhe paren- chyma. These septa form a continuum which passes from the pleura of the lung surface to lhe adventitia of the airways and larger blood vessels. In this way the negative inflationary pressure crealed within lhe thoracic cavity is readily trans- mitted throughout the parenchyma. During inspiralion the negative intrathoracic pressure is created by the lateral movement of the thoracic wall, and lhe caudal and ventral movement of the diaphragm. The thoracic wall of marine mammals is highly compliant which enables it to move relaIi.vely large distances during res- piration and so create the necessary negative intrathoracic pressure for inspiration. The action of the thoracic wall inspiratory musculature is aided by the obliquely orienled diaphragm which will act in a similar fashion. Also imporlanl for reinflation of lhe alveoli during inspiration is lhe more peripherally distributed bulk flow of inspired air, due to the airway reinforcement and acinar geometry. The increased alveolar seplal rigidity, as determined by an increased inler- slitial component, was found to be significantly greater in the venlro-posterior r91. region of lhe lung. This may ref lect an increased stress on this region durinq normal respiration and when divinq to depth. When at depth the increased hydro- stalic pressure will act on the whole body and tend to move the abdominal conten[s toward the collapsing thorax, particularly during the descent phase. Similar shifts of abdominal organs will also tend to occur, bul to a lesser degree, during expiration, partly as a result of lhe position of the very obliquely orienled diaphragm, Therefore, the increased reinforcement of the delicate alveolar septa in lhe ventro-posterior regions may have developed in response lo this mechanical stress. Although bhe lungs of marine mammals undergo compression collapse when diving to depth, this reduction in the available gas-exchange surface is insufficient to completely inhibit nitrogen absorption. It has been demonstrated that the absorption of potentially hazardous nitrogen occurs to a depth of at least 70 m. This resulbs, in part, from the likely paltern of alveolar septal collapse, in which exposure of the thinnest portions of the blood-gas barrier is m aintained until complete collapse occurs. The other factor which will influence nitrogen absorp- lion is the increase in the par[ial pressure of nitrogen when al depth so thal the diffusion qradient from alveolar air to capillary blood is increased. Therefore there are likely to be other mechanisms in marine mammals which operate to minimize or eliminate the complications of excessive nilrogen absorption. 6.5 FUTURE RESEARCH There are a number of avenues of research arisinq from this study which would likely conLribute to further understanding of lhe role of Lhe respiralory system in lhe diving capability of marine mammals. These include: i) the estimation of the morphometric pulmonary dif fusing capacity of another marine mammal species for which various physiologic estimates of pul- monary function have been determined. This would perhaps be best achieved 192. by examination of a phocid species, in particular lhe Weddell seal, since this species has been identified as the master diver amongst pinnipeds, and whose pulmonary function has been extensively studied; ii) the determination of the rate of oxygen uptake by the erythrocytes of one or several aquatic species, as this has only been assessed with any accuracy for human red blood cells. The rate of oxygen uptake by erythrocytes is an imporlant faclor for the erythrocyte component of the morphometric lung model and any differences between this rate for humans and marine mammals may further affect the observed increase in the pulmonary diffus- ing capacity of the Australian sea lion; iii) establishing the relalionship between the pulmonary diffusing capacity determined morphometrically and that determined physiologically for a marine mammal. In humans it has been demonslrated that the morphometric diffusing capacity of the lung is I.5-2 times greater than thal determined physiologically durinq heavy exercise. From the available evidence in regard lo oxygen utilization and the efficiency of lhe lung as a gas-exchange organ' it is likely that the discrepancy between the results of these two melhods would be diminished in marine mammals; iv) the exam ination of the intravascular events, in an aquatic species, associaled with the compression and decompression when diving to depth. In terreslrial species the development of a hypercoagulable state and associated haematological changes have been demonstrated to be important consequences of rapid decompression and the formation of intravascular bubbles. It appears lhat marine mammals may possess certain haematoligic and biochemical adaptalions which prevenl, or limit, such intravascular events. 193 BIBLIOGRAPHY Adaro, F. and Piiper, J. I916. Limiting role of stratification in alveolar exchange of oxygen, R iration Ph siolo 26, 195-206. Allan, E.M. 1978. The ultraslructure of the brush cell in bovine lung . Research in Veterinar Sc ience.25 , 3r4-3r1. Andersen, H.T. 1966. Physiological adaptations in diving verlebrates. Phys ioloqical Reviews, 46, 2I2-24J. Andersen, J.B. and Jespersen, W. 1980. Demonstration of intersegmental respiratory bronchioles in normal human lunqs. European Journal o! Respiratory Diseases, 6I, 331 -34I. André-Bouqaran, J., Pariente, R., Legrand, M. and Cayrol, E. I915. Normal ultrastructure of small bronchi and bronchioli in man. Patholoqie Bioloqie,23, 629-638. Assimacopoulos, 4., Guggenheim, R. and Kapanci, Y.1,916. Changes in alveolar capillary configuration at differenl levels of Iung inflation in the rat. An ultrastructural and morphometric study. LaboraLory Investiqation, 34, IO-22 Banks, W.J. l98l. Aop Iied Veterinarv Histoloqv. Baltimore: Williams and Wilkins Baskerviìle, A. I97O. Ultrastructural studies of the normal pulmonary lissue of the prg R esearch in Velerinarv Science. ll-, 150-r55 Becci, P.J., McDowell, E.M. and Trump, B.F. l-978. The respiratory epithelium. II. Hamster trachea, bronchus, and bronchioles. Journal of the Nalional Cancer Institute, 6I,55I-56L Bélanqer, L.F. 1940. A study of the histological struclure of the respiraLory portion of the lungs of aquatic mammals. American Journal of Anatomy,67, 437-469. Bennetl, H.S., Wyrick, A.D., Lee, S.W. and McNeil, J.H. Jr. L976. Science and art in preparing tissues embedded in plaslic for light m icroscopy, with special refenence to glycol meLhacrylate, glass knives and simple stains. Stain Technoloqy, 5I' lI-97. 194. morphometric Berendsen, P.B. and De Fouw, D.O. I1BZ- An ultrastructural comparisonoftherightandleftsidesandlobesofthedoqlung . Anatomical Record, 2O4r 39-44- scular ResPonses to Blix, A.S. I975. The Elicitatio n and Requlation o f the Cardiova Sect' Physiol Divino. Transo, Norway: University of Transo' Inst' Med' Biol'' pp 5-2O. in mammals and Blix, A.s. Ig76. Metabolic consequences of submersion asphyxia 169-178. birds. B iochemical Societ UK) m ostum 4r, and its distribution BIix, 4.s., Elsner, R. and Kjekshus, J.K. l-981. cardiac output ioloqica through capillaries and A-V shunts in diving seals AcLa Phys Sc andinav ia. llB, l_09-116. Philadelphia: w'B' Bloom, w. and Fawcett, D.W. 1968. A Textbook of Histoloqy' Saunders Co. and Martin' C'J' 1919' Boatman, E.5., Arce, P', Luchtel, D't Pump' K'K' (ed')' pulmonary function, morphology and morphometrics, in Bowden, D'M' AqinqinNo n-humanPrimates.NewYork:VanNostrandReinhold,pp.292-'13. der Lungen bei der respiratorischen Bohr, c. 1909. ueber die spezifische Tätigkeit ioloov. 27, 22I-28O. Cited bY Gasauf nahme. Scand inav ian Archives of Phys pulmonary diffusion capacity' We.ibel, E.R. 1970a. Morphometric estimation of . I. Modet and method, Respiration Physiology, LI,54-75. in Ridgway, s'H' and Bonner, w.N. 198I. Southern Fur Seals - Arctocephalus, V ot. l. The Walrus, Sea Harrison, 3.¡. (eds): Handbook of M arine Mammals, pp. 16l-208' Lions, Fur seals and sea otter. London: Academic Press, Boshier' D.P.andHilI'P.McN.IgT4.Structuralaspectsofventilationand weO )' in Hamison' R'J' (ed')' d iffu sion in the Weddell seal tf-eptonycnotes 2. London: Academic Presst pp Func tional Anato myofM arine Mammals, Vol. I97 -229. Boyd, R.B IITJ.Acomparativeanatomicalstudyoftherespiratorysystemsof the Antarctic Weddell and Crabeater seals' D issertation Abstracts Internati on al 34,5292-8. I95. Boyd, R.B. l-975. A gross and micnoscopic study of the respiratory analomy of the Anfarctic Weddell seal, Leptonychotes weddelli, Journal of Morpholoqy, I47, 309-336. Boyden, E.A. l97l-. The slructure of the pulmonary acinus in a child of six years and eight months, American Journal of Anatom r32,275-3OO. Breeze, R.G. and Wheeldon, E.B. 1977. The cells of the pulmonary airways, American Review of R i ra Ior D isease 116, 705-777. Breeze, R.G., Wheeldon, E.B. and Pirie, H.M. L976. Cell structure and funclion in the mammalian lung: ïhe trachea, bronchi and bronchioles. Veterinary Bullef in, 46, 3I9-337. Bron, K.M., Murdaugh, H.V. Jr, Millen, J.E., Lenthall, R., Raskin P. and Robin, E.D. 1966. Arterial constrictor response in a diving mammal. Science, 152, 540-543. Bryden, M.M. and Lim, G.H.K. 1969. Blood parame[ers of the Southern elephant seal (Milsglgq _!ge!il_Ð Linn.) in relalion to diving . Comparalive Biochemislry and Physio loqy, 28, I39 -I48. Burns, J.J. I9Bt. Ribbon seal - Phoca fasciata in Ridgway, S.H. and Harrison, R.J. (eds), Handbook of Marine Mammals, Vol. 2.994q. London: Academic Press, pp. B9-I09. Burri, P.H. I974. The postnatal growth of lhe rat lung. III. Morphology, Anatom icaì Record IBO,77-98. Burri, P.H. and Weibel, E.R. I97la. Morphometric evaluation of changes in lung sLructure due to high altitude, in Porter, R. and Knight, J. (eds), Hiqh Altitude Physioloqy: Cardiac and Rqsplratory 4SgeqÞ. Edinburgh: Churchill Livingstone, pp. 15-10. Burri, P.H. and Weibel, E.R. L971b. Morphomelric estimation of pulmonary diffusion capacity. II. Effect of POZ on the growing [ung, Resp iralion Physioloqy, lI , 247 -264. 196. Burri, P.H., Dbaly, J. and Weibel, E.R. 1974. The poslnatal qrowLh of the rat lung. I. Morphometry, Anatomical Record r7B,1II-730. Burri, P.H. and Sehovic, S. L979. The adaptive response of Lhe rat lung after bilobectom y, American Review of R esoiratorv Disease. Ll9, 769-771. Buller, P.J. 1982. Respiratory and cardiovascular control during diving in birds and mammals, JournalofE erimental Biolo LOO, L95-22I. Butler, P.J. and Jones, D.R. 1982. The comparative physiology of diving in v ertebrates, Advances in Comp arative Physioloqy and Biochemistry, B, r79- 364. Casson, D.M. and Ronald, K . 1.915. The Harp seal, Pagophilus qloelþfldjçu! (Erxleben, 1977) - XIV. Cardiac arrhythmias, Comparative Biochemistry and Physiolooy, 50A, JOl -3I4. Castellini, M.A. 1981. Biochemical adaptations for diving in marine mammals. DissertaLion Abstracts International 41, 4401_-8. Castellini, M.A. and Somero, G.N. 1981. Buffering capacity of vertebrate muscle: Correlations with potentials for anaerobic function, Journal of Comparative Physioloqy, I43, l9I-I98. Castellini, M.4., Somero, G.N. and Kooyman, G.L. L98l-. Glycolytic enzyme activilies in tissues of marine and terresLrial mammals, Physioloqical Zooloqy, 54,242-252. Castleman, W.L., Dungworth, D.L. and Tyler, W.S. 1975. Inlrapulmonafy airway morphology in three species of monkeys: A correlated scanning and transm ission electron m icroscopic study, American Journal of Anatomy, I42 ro7-r22. Clausen, G. and Ersland, A. 1968. The respiralory properties of the blood of lwo diving rodenls, lhe beaver and lhe water vole, Respiration Physioloqy, 5r 350- 359. Clausen, G. and Ersland, A. 1969. The respiralory properties of the blood of the bladdernose seal (Cystophora cristata), Respi¡ation Physioloqy, 7 , l-6. r97. Clements, J.4., Nellenbogen, J. and Trahan, H.J. 1970. Pulmonary surfactanI and evolution of the lungs, Science, 169, 603-604. Craig, A.B. Jr and Pasche, A. J-980. Respiratory physiology of freely diving harbor seals (Phoca vitulina), Physioloqical Zooloqy, 53, 4I9-432. Crapo, J.D., Marsh-Salin, J., Ingram, P. and Pratt, P.C. l978a. Tolerance and cross-tolerance using NOZ and O2. II. Pulmonary morphology and m orphome try, Journal of Applied Ph vs io loo v. 44, J7O-379. Crapo, J.D., Peters-Golden, M., Marsh-Salin, J. and Shelburne, J.S. l97Bb. Pathologic changes in the lungs of oxygen-adapted rats: A morphometric analysis. Laboratory Investiqalion, 39, 640-653. Cruz-Orive, L.M. and Weibel, E.R. l-98L. Sampling designs for stereology, Journal of Microscopy, I22, 235-257. Cutz, E. and Conen, P.E. L971. Ultrastruclure and cytochemistry of Clara cells, American Journal of Patholo ov. 62 , r21-r34. Dailey, M.D. I91O. The transmission of Parafilaroides decorus (Nematoda: Metaslrongyloidea) in the Calif ornia sea lion (Zat"pnus catitorn . Proceedinqs of the Helm!1llploqical Society of Washinqton, 37, 215-222. Daly, M. de B. lg8l-. The diving response, in Pethes, G. and Frenyo, V.L. (eds), Advances in Animal and Comparalive Physioloqy. Buda pest: Akademiai Kiado, pp 289-296 Delyamure, S.L. 1955. Helminthofauna of marine mammals (ecoloqy and phyloqeny) (1968), Academy of Sciences USSR, I-52I. Translated from Russian. Israel Program for Scientif ic Translations, Jerusalem. Denison, D.M., Warrell, D.A. and Wesl, J.B. 1971. Airway structure and alveolar emptying in lhe lungs of sea lions and dogs. Respiration Physioloqy, 1,3r 253- 260. Denison, D.M. and Kooyman, G.L. L973. The struclure and function of Lhe small airways in pinniped and sea oller lungs, Re iralion Ph siolo 17,1-10. l9B. Dhindsa, D.S., Metcalfe, J., Hoversland, A.S. and Hartman, R.A. I974. Comparalive studies of the respiratory functions of mammalian blood. X. Killer whale (Orcinus orca, Linnaeus) and beluga whale (Delphinaplerus leucas), Resp iration Phvsioloov. 20 ,93-ro3. Domrow, R. 1974. Notes on halarachnine larval morphology and a new species of Pneumonvssus Banks (Acari: Dermany ssidae). Journal of the Australian Entomolo ical Sociel 13, I7-26. Dormer, K.J., Denn, M.J. and Stone, H.L. L917. Cerebral blood flow in the sea lion (Zalophus californianus) during volunlary dives, Comparalive Biochemistry and Physioloqy, 5BA, 11-lB. Dougherty, E.C.and Herman, C.M' 1941. New species of the g enus Parafilaroides Dougherty, 1946 (Nematoda: Metastrongylidae), from sea-lions, wilh a list of the lungworms of the PinniP edia. Proceedinq s of the Helmintholoqical Society of Washington, 14, 77-87. Dunlap, J.S., Piper, R.C. and Keyes, M.C. I976. Lesions associated with Orthohalarachne attenuata (Halarachnidae) in the Northern fur seal (Callorhinus ursinus). Journal of Wildlife Diseases I2, 4?-45. Dunnill, M.S. 1962. Postnatal growth of the lung, Thorax, Ll,329-333. Dunnill, M.S. 1964. Evaluation of a simple method of sampling the lung for quantitative hislolog ical analysis, Thorax 19, 443-448. Dykes, R. W. 1974. Factors related Lo the dive ref lex in harbor seals: Sensory contributions from the lrigem inal region, C anadian Journal of Phvsioloov and Pharm acoloqy, 52, 259 -265. EIias, H. and Hyde, D.M. 1980. An elementary introduclion to stereology (quantitative microscopy), American Journal of Anatomy, I59, 4II-446. EIsner, R. l-965. HearI rate response in forced versus trained experimental dives in p innipeds, Hvalradets Skrifter 48,24-29. Elsner, R.W., Franklin, D.L. and Van Citters, R.L. 1964. Cardiac oulput during diving in an unrestrained sea lion, N a ture 202, Bo9-BLo. r99. EIsner, R.W., Franklin, D.L., Van Citters, R'L' and KenneY, D.W. 1966- Cardiovascular defense against asphyxia' Science r53,94r-949. Exner, H.E. Ig16. Derivations of the working rela[ionship of stereology. A. Volume fnaction analysis, in underwoocl, E.8., de wit, R. and Moore, G.A. (eds), Proceedinqs Fourth International Conqr ess for Stereoloqy. National Bureau of Sbandards, Special Publication 411. Washingbon: Government Printing Office' pp 446-448. Fanning, J.C. Ig7l. The submicroscopic structure of the dolphin lung. Ph.D. Thesis, UniversitY of Adelaide. Fanning, J.C. and Harrison, R.J. 1974. The structure of the trachea and lungs of the South Australian bottle-nosed dolphin, in Harrison, R.J. (ed.), Functional Anatomy o f Marine Mammals, Vol. 2. London: Academic Press, pp 23I-252' Fay, F.H. and Furman, D.P. L982. Nasal miles (Acari: Halarachnidae) in the spotled seal, Phoca laroha Pallas, and other pinnipeds of Alaskan waters. Jou rnal of Wildlife Diseases rB, 63-68. Ferren, H. and Elsner, R. I978. Diving physioloqy of the ringed seal: Adaptations and implications, in Alaska Fisheries: 200 Years and 200 Miles of Chanqe. proceedings 24th Alaska Science Conference, August I918. Sea Grant Report 79-6, August L979. College, Alaska: Alaska Division, American AssociaLion for the Advancement of Science. Fiebiqer, J. 1916. Uber eigentümlichkeiten im Aufbau der Delphinlunge und ihre physiologische Bedeutungt Anabom ischer Anzeiqer, lt8 , 540-565. Cited by Belanger, L.F. 1940. A study of the histological structure of the respiratory portion of the lungs of aquatic mammals, American Journal of Anatomy,67, 437 -469. Fleischman, R.W. and Squire, R.A. I97O. Verminous pneumonia in lhe California sea lion (Zalophus californianus). Patholoqia Veterinaria. 7, 89-101. Forresf, J.B. Ig76. Lung lissue plasticity: Morphometric analysis of anisotropic strain in liquid filled lungs, Respiration Physioloqy. 21,223-239. 200. Forresl, J.B. 1979. Structural aspects of gas exchange, Federation Proceedin 38,209-214. Forrest, J.B. and Weibel, E.R. 1975. Morphometric estimation of pulmonary diffusion capacity. VII. The normal guinea pig lung' Re iration Ph lo 24, L9L-202. Frost, K.J. and Lowry, L.F. l98L. Ringed, Baikal and Caspian seals - Phoca hispida, Phoca sibirica and Phoca caspica, in RidgwaY, S.H. and Harrison, R.J. (eds), Handbook of Marine Mammals Vol. 2, lgglg. London: Academic Press, pp. 29-51. Fung, Y.C. and Sobin, S.S. 1969. Theory of sheet flow in lung alevoli. Journal of App Iied Phvsioloqy, 26 , 47?-488. Funman, D.P. and Smith' A.W. 1973. In vitro development of two species of Orthohalarachne (Acarina: H alarac hnidae) and ad aptations of lhe life cycle for endoparasitism in mammals, Journal of Medical Entomoloqy, I0, 415-416- Gail, D.B. and Lenfant, C.J.M. 1983. Cells of the lung: Biology and clinical implications, American Review of Respiralory Disease, 1,27, 366-J87. Gallivan, G.J. l-980. Hypoxia and hypercapnia in the respiratory control of lhe Amazonian manatee (Trichechus inunquis) , Physiological Zool!W, 5J, 254-26I Gallivan, G.J. l-98la. VenLilation and gas exchange in unrestrained harp seals hoca roenlandic . Comparative Biochemist rYa nd Phvsioloov. 694 ,809-811. Gallivan, G.J. l98l-b. Comparative pulmonary slructure and function in two larqe mammals, the horse and the cow. Ph.D. thesis, McMaster University. Gallivan, G.J. and Best, R.C. I980. Metabolism and respiration of the Amazonian m anatee richechus inu u is), Physioloqical Zooloqy, 5J, 245-253. Gamsu, G., Thurlbeck, W.M., Macklem, P.T. and Fraser, R.G. I97I. Peripheral bronchographic morphology in the nonmal human lung, Investiqative Radioloqy, 6, l6l_-170. 201. Gaskin, D.E. 1982. The Ecoloqy of Whales and Dolphins. London: Heinemann Educational Books. Geelhaar, A. and Weibel, E.R. I97I. Morphometric estimation of pulmonary diffusion capacity. III. The effecl of increased oxygen consumption in Japanese waltzing mice, Respiration Physiolooy, II, J54'366. Gehr, P. and Weibel, E.R. 1974. Morphometric esLimabion of regional differences in the dog Iung, Journal of Applied Physioloqy. 37, 648-653. Gehr, P., Bachofen, M. and Weibel, E.R. l978a. The normal human lung: Ultrastructure and morphometric estimation of diffusion capacity . Respiration Physioloov, J2, I,ZL-I4O. Gehr, P., Hugonnaud, C., Burri, P.H., Bachofen, H. and Weibel, E.R. 1978b. Adaptation of the growing Iung to increased Vgr: III. The effecl of exposure to cold environment in rats, Respiration Physioloqy, 32, 345-J5J. Gehr, P. and Erni, H. 1980. Morphometric eslimalion of pulmonary diffusion capacif y in two horse lungs, Respiration Physioloqy, 4I' I99-2IO. Gehr, P.,Sehovic, S., Burri, P.H., Claassen, H. and Weibel, E.R. 1980. The lung of shrews: Morp hometric estimalion of diffusion capacity, Respiralion Physioloqy, 40 ,3t-47. Gehr, P., Siegwart, B. and Weibel, E.R, l9BLa. Allometric analysis of the morphometric pulmonary diffusing capcity in dogs, Journal of Morpholoqy, 168, 5-r5. Gehr, P., Mwangi, D.K., Ammann,4., Maloiy, G.M.O., Taylor, C.R. and Weibel, E.R. 1981b. Design of [he mammalian respiratory system. V. Scaling morphometric pulmonary diffusing capacity to body mass: wild and domestic mammals, Respiration Physiolooy. 44, 6I-86. Genlry, R.L., Kooyman, G.L. and Goebel, M.E. 1981. Feeding and diving behavior of Northern fur seals, Callorhinus ursinus. Unpublished m anuscript. 202. Geraci, J.R. and Smith, T.G. L975. Funclional hematology of ringed seals (Phoca hispida) in the Canadian Arctic, Journal of Fisheries Research Board of Canada, 32, 2559 -256t+- Gil, J. and Weibel, E.R. 1969. Improvements in demonstration of lining layer of lung alveoli by electron microscopy. Respiralion Physioloqy. B, L3-36. Gil, J. and Weibel, E.R. L97I. Extracellular lining of bronchioles after perfusion- fixation of rat lungs for eleclron microscopy' Anatom ical Record t69, r85- 200. Gil, J. and Weibel, E.R. I912. Morphological study of pressure-volume hysleresis in rat lungs fixed by vascular perfusion, Respiralion Physioloqy, 15, I9O-2I3. Gillespie, J.R. I98l. Mechanisms that determine functional residual capacity in different mammalian species. American R eview of Resoiratorv Disease. f 2 B, 7 4-17 . (Suppl.) Gillespie, J.R. and Tyler, W.S. 1967. Quantitative electron m icroscopy of lhe interalveolar septa of the horse lung, American Review of Respiratory D isease 95,477-483. Glauert, A.M. 1975. Fixalion, dehydralion and embedding of biological specimens, in Glauerl, A.M. (ed.), Prac lical Methods in Eleclron Microscoov. Amsterdam: North Holland, l(1), pp l'-2O7. Glazier, J.8., Hughes, J.M.B., Maloney, J.E. and West, J.B. l-967. Vertical gradient of alveolar size in lungs of dogs frozen intact, Journal of Applied Physioloqy. 23,694-705. Glazier, J.8., Hughes, J.M.B., Maloney, J.E. and Wesl, J.B. 1969. Measurement of capillary dimensions and blood volume in rapidly frozen lunqs, Journal of Applied Physiology, 26, 65-76. Glazova, T.N. 1977. Physiological and biochemical characteristics of the blood of the Greenland seal, P h ilus enland icus Journal of Evolulionary Bioche m istry and Phvsioloqy. 1l ,220-224. 203. Goudappel, J.R. and Slijper, E.J. I958. Microscopic slructure of the lungs of lhe botflenose whale, Nalure, IB2, 479- Green, R.F.1972. Observations on the anatomy of some cetaceans and pinnipeds, in Ridgway, S.H., Mammals of the Sea: Bioloqy and Medicine. Springfield: C.C. Thomas, pp 247-297. Gundersen, H.J.G. and Østerby, R. 1981. Optimizing sampling efficiency of stereological studies in biology: Or'Do more less well!', Journal of Microscopy, r2r, 65-73. Gupta, M., Mayhew, T.M., Bedi, K.s., sharma, A.K. and White, F.H. Inter-animal variation and its influence on the overall precision of morphometric estimates based on nested sampling design s. Journal of Microsco nr, r47-r54. Haies, D.M., Gil, J. and Weibel, E.R. I9BI. Morphometric study of rat lung cells. I. Numerical and dimensional characteristics of parenchymal cell population, American Review of Res p iratorv Disease, l-2J ,5J3-54L. Haldiman, J.T., Henry, R.W., Henk, W.G., Abdelbaki, Y.2., Al-Bagdadi, F.K. and Duffield, D.w. l-980. Respiratory anatomy of the bowhead whale (_Egþ9rlg myslicetus) of Alaska, American Zoologisl, 20,759 (abstract). Hance, A.J. and Cryslal, R.G. 1975. The connective tissue of lung, American Review of Resp iratory Disease, lJ- 2, 651-7 rL. Hansen, J.E., Ampaya, E.P., Bryant, G.H. and Navin, J.J. L975. Branching pattern of airways and air spaces of a single human terminal bronchiole, Journal of Applied Physioloqy, 38' 9Bt-989. H arrison, R.J. and Tomlinson, J.D.W. 1963. Anatomical and physiological adaplions in diving mammals, View inls in Biolo 2, rr5-L62. Harrison, R.J., Ridgway, S.H. and Joyce, P.L. 1972. Telemelry of heart rale in diving seals, Nature. 237r ZBO. Harrison, R.J. and Kooyman, G.L. 1981. Diving in marine mammals, in Head, J.J. (ed.), Carolin a Bioloov Read ers. 6. Carolina Bio logical Supply Company. 204. Haslefon, P.S. I972. The internal surface area of the adult human Iung, Journal of A natomv. lI2 , l9l_-400. Hayatdavoudi, G., Crapo, J.D., Miller, F.J. and O'Neil, J.J. l9B0' Factors determ ining degree of inf Iat ion in intracheally f ixed rat lungs. Journal of Aoolied Ph vsioloov - ResÞirator v Environmental and E xercise Phvsioloqy. 4 B, 389-193. Haynes, F. and Laurie, A.H. I9J1. On the histological struclure of cetacean lungs, D iscov ery Reoorts. l-7 , I-6' Heezen, B.C. 1957. Whales entangled in deep sea cables, Deep-Sea Research,4, 105-l_r5. Hepburn, D. I896. Halichoerus qrypus: the qrey seal. Observations on its external appearances and visceral anatomy. IV. Examinalion of the viscera, Journal of Anatom y and Physioloqy, 30, 4BB-501. Hepburn, D. IgI2. Scottish National Antarctic Expedilion: Observalions on the ana[omy of the W eddell seal (Leptonychotes weddelli). III. The respiralory system, and Lhe mechanism of respiration, Transactions of the Ro I Societ Ed inburoh. 48 ,32r-332. Heusner, A.A. 1983. Body size, energy metabolism, and the lungst Journal of Applied Ph ysioloqy - ResPirator y Environme ntal and Exercise PhysioloqY, 54 867-873. Hijiya, K., Okada, Y. and Tankawa, H. 1977. Ultrastructural study of the alveolar brush cell, Journal of Electron Microsc 26,32r-329. Hochachka, p.W. 1981. Brain, lung and heart functions during diving and recovery. Science, 2I2,5O9-5I4. Hochachka, P.w., Liggins, G.c., Gvist, J., Schneider, R., snider, M.Y., Wonders, T.R. and Zapol, W.M. I977. Pulmonary metabolism during diving: Conditioning blood for the brain, Science. I9B' Bll-814. Holland, R.A.B., Van Hezewijk' W. and Zubzanda, J. 1977. Velocity of oxygen uptake by partlY saturated adult and fetal human red cells, Respiralion Physioloov.29 ,2O3-Jr4. 205. Hughes, G.M. l-978. A morphological and ultrastructural comparison of some vertebrate lungs, in, Klika, E. (ed.), XIXth Morpholooical Conoress Symposia. Prague: Charles University' pp 393-4O5. Hughes, G.M. and Weibel, E.R. 1976. Morphometry of fish lungs, in Hughes, G.M. (ed.), Respiration of Amphibious Vertebrates. London: Academic Press, pp 2I3- 232. Hugonnaud, C., Geht, P., Weibel, E.R. and Burri, P.H. 1,977. Adaptation of lhe growing lung to increased oxygen consumplion. II. Morphometric analysis, Resp iration Phvsioloqv. 29, l-10. Hui, C.A. 1975. Thoracic collapse as affected by the retia thoracica in the dolphin, R iration Ph to 25, 63-70. Hyatt, R.E., Schilder, D.P. and Fry, D.L. L958. Relationship between maximum expiratory flow and degree of lung inflaLion. Journal of Applied Physioloqy, 13, 33r-336. Hyde, D., Orthoefer, J., Dungworth, D., Tyler, W., Carter, R. and Lum, H. )-918- Morphometric and morphologic evaluation of pulmonary lesions in Beagle dogs chronically exposed to high ambient levels of air pollutants, Laboratory Investioation. JB ,455-469. Hyde, D.M., Samuelson, D.4., Blakeney, W.H. and Kosch, P.C. I979. A correlative light microscopy, transmission and scanning electron microscopy study of the ferret lung, Scanninq Electron Microscoov. III, 891-898. Inselman, L.S. and Mellins, R.B. l98l. Growth and Development of the lung, Journal of Paediatrics. 98 , l-l_5. Irving, L. 1939. Respiration in diving mammals, Physioloqical Reviews. 19, Il'2- 134. Irvinq, L., Solandt, O.M., Solandl, D.Y. and Fisher, K.C. 1935. The respiratory metabolism of the seal and its adjustmenl to diving, Journal of Cellular and Comparative Physioloqy, 7, r31-r5I. 206. Irvinq, L., Scholander, P.F. and Grinnell, S.W. I94I. The respiration of the porpo ise, Tursiop s truncalus, Journal of Cellular and Comparalive Physioloqy, I7, l-lt5-168. Irving, L., Scholander, P.F. and Grinnell, S.W. 1942. The regulation of arterial blood pressure in the seal during diving, American Journal of Physiology, 135, 557 -566. Irving, L., Peyton, L.J., Bahn, C.H. and Peterson, R.S. L961. Aclion of the hearL and breathing during the development of fur seals (Callorhinus ursinus). Physioloq ical Zooloqy, J6 , r-2o. Jeffery, P.K. and Reid, L. 1915. New observations of rat airway epithelium: A quantilative and electron microscopic study, Journal of Anatomy, I2O, 295- 320. Jeffery, P.K. and Reid, L.M. l-977. The respiratory mucous membrane, in Brain, J.D., Proctor, D.F. and Reid, ¡.¡¡4. (eds), Resoiratory Defence Mechanisms: Parf l. New York: Marcel Dekker, pp. I93-245- Johnston, D.G. and Ridgway, S.H. l-969. Parasitism in some marine mammals. Journal of the American Veterinary Medical Association, I55, 1064-101 2. Jones, D.R., Fisher, H.D., McTaggart, S. and West, N.H. 1971. Heart rate during brealh-holding and diving in the unrestrained harbour seal (Phoca vitulina richard i), Canadian Jo urnal of Zooloov. 5 r,67t-680. Jubb, K.V.K. and Kennedy, P.C. I97O. Patholoqy of Domestic Animals. Vol. 1. 2nd Ed. London: Academic Press. Kapanci, Y., Assimacopoulos,4., Irle, c., Zwahlen, A. and Gabbiani, G. L974, "Contrac tile interstiLial cells" in pulm onary alveolar sep ta: A possible regulator of ventilalion/perf usion ralio? UltrastrucLural, immunofluorescence' and in vitro sludies, Journal of Cell Bioloqy. 60 , J75-392. Karnovsky, M.J. 1965. A formaldehyde-glutaraldehyde f ixative of high osmolality for use in electron microscopy. Journal of Cell Bioloqy, 27' I37A-I3BA. 201. Keller, H.J., Friedli, H.P., Gehr, P., Bachofen, M. and weibel, E.R. 1976. The effects of optical resolution on the estimation of stereological parameters, in Underwood, E.E., de Wit' R. and Moore, G.A. (eds), Proceedinqs Fourth International Conqress for Stereoloqv. National Bureau of Standards , Special Publication 43I. Washington, DC: US Government Printing Office, pp 409-4J-0. Kennedy, 4.R., Desrosiers, 4., Terzaghi, M. and Little, J.B. I918. Morphometric and histological analysis of lhe lungs of Syrian golden hamsters, Journal of Anatomy, I25,521-55J. Kerem, D.H., Kylslra, J.A. and Saltzman, H.A. I975. Respiratory flow rates in the sea lion, Undersea Biomed ical Research. 2, 20-27. Keyes, M.C. l-965. PathologY o f the Northern fur seal. Journal of the American Veterinarv Medical A ssociation. l-47 , l_090-1095 Kim, J.C.S. I911. Pulmonary acariasis in Old World monkeys. Veterinary Bulletin, 47, 249-255. Kim, J.C.S. 1980. Pulmonary acariasis in OId World monkeys: A review, in Montali, R.J. and Migaki, G. (eds), The Comparalive PaLholoqy of Zoo Animals. Washington, DC: Smithsonian Institution Press, pp. 383-394. Kim, K.C., Haas, V.L. and Keyes, M.C. 1980. Populations, microhabitat preference and effecls of infestation of two sp ecies of Orthohalarachne (Halarachnidae: Acarina) in the Northern fur seal. Jou rnal of Wildlife Diseases. I 6, 45-5r King, J.E. and Harrison, R.J. 196I. Some notes on the Hawaiian monk seal, Pacific Science. l'5, 282-293. Kooyman, G.L. 1966. Maximum diving capacilies of lhe Weddell seal, Leptonychotes weddelli, Science. I5I, I553-L554. Kooyman, G.L. 1967. An analysis of some behavioral and physiological characterislics relaled to diving in the Weddell seals, in Llano, G.A. and Schmitt, W.L. (eds), Bioloqy of lhe Antarctic Seas, III. Antarctic Research Series, Vol. 11. Washington: American Geophysical Union, pp 227-26I 208. Kooyman, G.L. 1973. Respiratory adaptalions in marine mammals, American Zoo ist. ll ,457-468. Kooyman, G.L. 1982. How marine mammals dive, in Taylor, C.R., Johansen, K. and Bolis, L. (eds), A Comp anion to Animal Phvsioloov. Cambrid ge: Cam bridge Universily Press, pp 15l-160. Kooyman, G.L. and Andersen, H.T. L969. Deep diving, in Andersen, H.T. (ed.), The Bioloqy of Marine Mammals. New York: Academic Press, pp 65-94. Kooyman, G.L., Hammond, D.D. and Schroeder, J.P. 1970. Bronchograms and tracheograms of seals under pressure, Science, 169, 82-84. Kooyman, G.L., Kerem, D.H., Campbell, W.B. and Wright, J.J. l97l-.. Pulmonary function in freely diving Weddell seals, Leptonychotes weddelli, Respiration Physioloqy, I2, 27 L-282. Kooyman, G.L. and Campbell, W.B. 1972. Heart rates in freely diving Weddell seals, Leotonvcho le s weddelli. Comparative Biochemistry and Physioloqy, 4JA, 3I-36. Kooyman, G.L., Schroeder, J.P., Denison, D.M., Hammond, D.D., Wright, J.J. and Bergman, W.P. I972. Blood nitrogen tensions of seals during simulated deep d iv es, American Journal of Phvsioloqy, 22J , 101_6-1020. Kooyman, G.L., Kerem, D.H., Campbell, W.B. and Wrighl' J.J. 1973. Pulmonary gas exchange in freely diving Weddell seals, Respiration Physioloqy, I7,283- 290. Kooyman, G.L., Norris, K.S. and Gentry, R.L. 1975. Spout of the gray whale: Its physical characterisbics, Science. l90r 908-910. Kooyman, G.L., Gentry, R.L. and Urquhart, D,L. I976. Norlhern fur seal divinq behavior: A new approach to its study, Science, I93r 4IL'4L2. Kooyman, G.L. and Sinnett, E.E. 1979. Mechanical properties of lhe harbour porpoise lung, Phocoe na ohocoena. Respiration Physioloqy. 36, 287 -3O0. 209. Kooyman, G.L., Wahrenbrock, E.4., Castellini, M.A., Davis, R.W. and Sinnett, E.E. 1980. Aerobic and anaerobic melabolism during voluntary diving in Weddell seals: Evidence of preferred pathways from blood chemistry and behaviour, Journal of Compar ative Phvsioloov 8.. Il 8,3t5-J46. Kooyman, G.L. and Cornell, L.H. l9BL. Flow properties of expiration and inspiration in a trained bottle-nosed porpoise, Ph o ical Zool 54, 55-6r. Krahl, V.E. 1964. Analomy of the mammalian lung, in Handbook of Physioloqy, Section R ira ti Volume l-. Washington, DC: American Physiological Society, pp 213-284. Kuhn, C., Callaway, L.A. and Askin, F.B. L974. The formation of granules in the bronchiolar Clara cells of the rat. I. Electron m icroscopy, Journal of Ultrastructure Research 49, tB7-4O0. Kurochkin, Y.V. and SobolewskY, E.I. L975. Nasal mites Orthohalarachne attenuata of Norther f ur seals. Raooorts et Proces-Verbaux des Reunions. Conseil International oour ItExoloration de la Mer. 169, 362. Lambert, M.W. 1955. Accessory bronchiole-alveolar communications, Journal of Patholoqy and Bacterioloqv. 70 , 3rL-3I4. Lapennas, G.N. and Reeves, R.B. 1982. Respiratory properties of blood of lhe gray seal, Halichoerus qrypus, Journal of Comparalive Physioloqy, I49' 49-56- Laurie, A.H. 1933. Some aspects of respiration in blue and fin whales, Discovery Reports, 7, 363-406. Lechner, A.J. 1978. The scaling of maximal oxygen consumption and pulmonary dimensions in small mammals, Respiration Physioloqy, 34,29-44. Leith, D.E. 1976. Comparative mammalian respiratory mechanics, Physioloqist, t9, 485-510. Lenfant, C. 1,969. Physiological properLies of blood of marine mammals, in Andersen, ¡.7. (ed.)¡ The Bioloqy of Marine Mammals. New York: Academic Press, pp 95-1J.6. 2IO. Lenfant, C., Elsner, R., Kooyman, G.L. and Drabek, C.M. 1969. Respiratory function of blood of the adult and fetus Weddell seal, Leptonychotes weddelli, American Journal of Ph iolo 216, 1595-1597. Lenfant, C., Johansen, K. and Torrance, J.D. l-970. Gas transport and oxygen slorage capacity in some pinnipeds and the sea otter, Respiralion.-P¡y-s-t-qþ-911-' 9,277-286. Lewis, P.R. and Knight, D.P. 1977. Staining meLhods for sectioned material, in Glauert, A.M. (ed.), Praclical Methods in Electron Microscopy. Amslerdam: North Holland, :(I)' pp 1-111. Lin, Y.-C., Malsuura, D.T. and Whittow, G.C. I972. Respiratory variation of hearL rate in the California sea lion, Americ an Journal of Phvsioloov.222 ,260-264. Ling, J.K. and Walker, G.E. I977. Seal studies in South Australia: Progress report for the period January I976 to March 1977. South Australian Naturalist, 52, LB-27. Leid, R.W.Jr and Williams, J.F. 1919. Helminth parasites and the host inflammatory system. Chemical Zo II,229-27r. Lutz, P. 1982. Oxygen transport in vertebrale blood: Challenges, in Taylor, C.R., Johansen, K. and Bolis, L. (eds), A Comp anion to Animal Physioloqv Cambridge: Cambridge University Press, pp 65-72. McDonnnell, W.N. I974. The effect of anaesthesia on pulmonary gas exchange and arterial oxygenation in the horse. PhD thesis, Cambridge University. Cited by Gallivan, G.J. 198lb. Comparative pulmonary structure and function in two large mammals, the horse and the cow. PhD thesis, McMas[er University. McDonnell, W.N., Hall, L.W. and Jeffcott, L.B. I979. Radiographic evidence of impaired pulmonary function in laterally recumbent anaeslhelised horses. Equine Veterinary Joqrnaf II, 24-32. McDowell, E.M., Barrett, L.4., Glavin, F., Harris, C.C. and Trump, B.F. 1978. The respiratory epithelium. I. Human bronchus, Journal of the National Cancer Inst i lu te 6I,539-549. 2II. Macklem, P.T. and Mead, J. 1968. FacLors determining maximum expiralory flow in dogs, Journal of Applied Physioloqy. 25, I59-169. Macklem, P.T, Proctor, D.F. and Hogg, J.C. 1970. The stability of peripheral airways, Respiration Physioloqy. B, I9I-2O3. McLaughlin, R.F., Tyler, W.S. and Canada, R.O. l-961-. A study of the subgross pulmonary anatomy in various mammals, American Journal of Anatomy, l0B, r49-165. Madigan, S. and Lawrence, V. 1982. A General Analysis of V ariance Prooram for the Apple II,H SD Anova II Packaqe. Norlhri dge: Human Systems DYnamics. Maina, J.N., King, A.S.and King, D.Z. L982. A morphometric analysis of the lung of a species of bat, Res iration Ph iolo 50, l-11. Man'kovskaya, I.N. I975. The conlent and distribulion of myoglobin in muscle tissue of Black Sea dolPhins, Journal of EvolutionarY Biochem istry and Physioloqv, II, 22I-225. Margolis, L. L956. Parasilic helminths and arthropods from pinnipedia of the Canadian Pacific coast. Journal of Fish eries Researc h Board of Canada. lJ 489-505. Mariassy, A.T., Plopper, C.G. and Dungworth, D.L. 1975. Characteristics of bovine Iung as observed by scanning electron microscopy,Anatomical Recor4 1"83, 13- 26. Marlow, B.J. and King, J.E. L974. Sea lions and fur seals of Auslralia and New Zealand, J ournal of the Australian Mammal Society, l- , rL7 -rJ5. Marfin, H.B. 1966. R.espiratory bronchioles as the pathway for collateral ventilation, Journal of lied Ph iolo 21., 1443-1,447. Massaro, G.D. and Massaro, D. I9Bl. Morphologic evidence lhat large inflaLions of the Iung stimulate secretion of surfactant. American Review of Re ira lor D isease 1,27, 235-236. 2I2. Mathieu, O., Cruz-Orive, L.M., Hoppeler, H. and Weibel, E.R. -ì.98L. Measuring error and sampling variation in slereology: Comparison of the eff iciency of various methods for planar image analysis, Journal of Microsco Lzr,75-BB. Matthews, R.C. 1971. Pulmonary mechanics of California sea lions, Zalophus californianus. Master's Thesis, San Diego State University. Mazzone, R.W. 1980. Influence of vascular and transpulmonary pressures on the functional morphology of the pulmonary microcirculation, M icrovascular Research, 20,295-306. Mazzone, R.W, Durand, C.W. and West, J.B. 1978. Electron microscopy of lung rapidly frozen under controlled physiological conditions, Journal of Applied Physiolocy - R esoiralory Environmental and Exercise P hvsioloov. 45 ,325-33J. Meyrick, B. and Reid, L. 1968. The alveolar brush cell in rat lung a third pneum onocyte, Journal of Ultrastructure Research 2t,7r-BO. Migaki, G., van Dyke, D. and Hubbard, R.C. 1971. Some histopathological lesions caused by helminths in marine mammals. Journal of Wildlife Diseases 7,2Br- 289. Miller, M.E., Christensen, G.C. and Evans, H.E. 1,964. Anatomy of the Dog. Philadelphia: W.B. Saunders Co. Miller, P.J. 1971. An elastin stain, Medical Laboratory Technoloqy, 28, I4B-I49. Morales, G.A. and Helmbotdt, C.F. I97),. Verminous pneumonia in the California sea lion (Zalophus californianuÐ Journal of Wildlife Diseases 7, 22-27. Müller,4.E., Cruz-Orive, L.M., Gehr, P. and Weibel, E.R. 1981. Comparison of lwo subsampling methods for electron microscopic morphometry, Journal of M icroscopy, I23, 35-49. Murdaugh, H.V.Jr, Cross, C.E., Millen, J.E., Gee, J.B.L. and Robin, E.D. 1968. Dissociation of bradycardia and arlerial constriction during diving in lhe seal P hoca vi[ulina. Science. 1,62, 364-365. of the Metab olic activities P'W' 1980. Journal d Hochachka' Weddell seal' W'M' an erY in the B'' ZaPoI, and recov SIO MurPhY, dur ing drving Exercise Ph braïn e ntal and Iung , and Env ronm heart, Re t1 aIor dPh iol o fA Iie 48, 196-605' specimens render biological techniques [o J.A. r97B' Non-coating fi, L15-194' iúurPhY' cttonM icro .5c ann Ele tof conductive andT reaImen forD nosis The B ASIS or maI t916' The N Co' MurraYt JÍ. W .8. Saunders ase. PhiladelPhia: lm on a1 D lse Pu rheLu BaILimore: isIo of [om and H cIion aI An a 197 Z. I un Nagaishi, C' Park Press' U niversity of the Auslralian D' l9B0' H aematologY C'F' and Sheriff, 16, lol-ro7' D'J'' Cargill, fW iIdIife D rse ASES Needham, a Jour nalo Ne cac of sea lion, consequences The pulmonary H.A. I9B2' J.H.T. 3¡i Barr¡ Power' 49, trïlz4 Nichoias, T'E'' ïoIo .R ira Lio nPh deeP breaLh a in pulmonary phosphotipid alveolar ceII and Iarge 19 67. Bronchiolar Niden, A'H' Lt24. nce t5B,1373' I 5 cle and m etabolism structure o n the fine So me observaLions (ed'), EIecLron E. 1966' yeda' R. and Yamada' cells' in U Nident A.H. bronc hiolar [he non-c iliated ction of d., PP' fun uzen Co' LL . TokYo: Mar '99-600' M on alveolar gas gas dif fusivity Effec I of altered .R lr ator packr n 1980' lied P olo W' Journal ofA Nixon, "n¿ ebical studY' a iheor 48, 141-L'3' exchange S iolo I tse Ph m ent al and mite Env of PulmonarY f. L919. A case and Suzuki' na C olle 5irouzur H. zabU Veterl Itagaki' H'' lle lin oft heA Ogata' M., sea lion' Bu of a Calïfornian ïnfestation in the pilot of ventilation 4, zo1-2o9 ' R' 196g.Mechanics F .C. and Elsner' C'R'' HaIe, 1-L49. Olsen, ioÌ 1 L' ira Eion Ph whale' Re 214. Pack' R'J'' Al-ugaily, L'H' and Morris, G. l-981. The ceils of the tracheobronchial epithelium of the mouse: A quanlitative light and electron micnoscope study, Journal of Anatomy, I3Z, 7I_84. Pasche, A' r976' The effect of hypercapnia on respiraLory characteristics divinqbehaviouroffreelydivingseals,@26,IBJ-Ig4. and Pasche, A. and Krog, J. 1980. Heart rate in resting seals on land and in water. Com rative B ioche m is lr and Ph siolo 67A,77-83. Paulev, P. 1965. Decompression sickness following repeated breath_hold dives, Journal of Aoolied Physiolo qv. 20, IO2B-IO3I. Paumgartner, D., Losa, G. and Weibel, E.R. I98l. Resolution effect on the stereological esli mati on of surface and volume and its interprelation in of fractal terms dimensions, Journal of M icrosco I2I, 5I-63 Perry, S.F. I978. euanti tative anatom y of the lungs of the red_eared turile, Psuedem S SCII ta ele ans Respiration ph yst oloqy, 35,245-262. Pemy, s.F. r98l. Morphometric anarysis of purmonary structure: Methods for evaruation and comparison of unicamerar rungs, Mikroskopie, 38, 278_293. Perry, S.F. l_98l Reptilian lungs: Functionar anatomy and evorutiòn, Advances Anatom in Emb and Cell Bi olo 79, I-BI. Petrik, P. and Collet, A.J. L 974. Quantitative electron mìcroscopic autoradiography of in vivo incorpo ration of lH-choline, JH-leucine, lH_acetate lH-galactose and in non-ciliated bronchiolar (Clara) cells of mice, American Journal of Anatomy, Ij9, 5Ig_534. Philp, R.B. 1974. A review of blood changes associated with compression_ decompression: Rela tionship to decompression sickness. Undersea Biomedical Research l_, tt7-150. Piiper, J. I979. Series ventilation, diffusion in airways, and stratified inhom ogeneity, Federati on Proceedin 38, 17-2I. 2L5. Pinkerton, K.8., Barry, 8.E., O'Neil, J.J., Raub, J.4., Pratt, P.C. and Crapo, J.D. 1982. Morphologic changes in the lung during the lifespan of Fischer 344 rats, American J ournal of Anatomv. 164 , r55-r74. Pitts, G.C. and Bullard, T.R. 1968. Some interspecific aspects of body composition in mammals, in Body Composilion in Animals and Man. Proceed ings of a Symposium, May L961. Washinglon, DC: National Academy of Sciences, pp. 45- 70. Pizey, N.C.D. 1954. The struclure of lhe pinniped lung, Journal of Anatomy, BB, 552 (abstract). Plopper, C.G., Mariassy, A.T. and H ill, L.H. l-980a. Ultrast,ructure of the noncilialed bronchiolar epithelial (Clara) cell of mammalian lung. I. A comparison of rabbit, guinea pig, rat, hamster and mouse. Experimental Lunq Research, I, I39-I54. Plopper, C.G., Mariassy, A.T. and Hill, L.H. l-980b. Ultrastructure of the nonciliated bronchiolar epitheliaÌ (Clara) cell of mam malian lung. II. A comparison of horse, steer, sheep, dog and cat. Experimental Lunq Research, \ t55-169. Plopper, C.G., Hill, L.H. and Mariassy, A.T. l9B0c. Ultrastructure of the nonciliated bronchiolar epilhelial (Clara) cell of mammalian lung. III. A study of m an with comparison of 15 m am m alian sp ecies. Experimental Lunq Research 1, It7-180. Pride, N.8., Permutt, S., Riley, R.L. and Bromberger-Barnea, B. 1967. Determinants of maximal expinatory flow from t,he lungs, Journal of Applied Physioloqy, 2l, 646-662. Pump, K.K. L964. The morphology of the finer branches of the bronchial tree of the human lung, Diseases of lhe Chest, 46, Jl9-399. Pump, K.K. 1969. Morphology of the acinus of the human lung, Diseases of the Chesf, 56, 126-134. Reid, L. 1967. The Patho of Em h ma. London: Lloyd-Luke. 216. Repenning, C.A. and Tedford, R.H. I971. Otarioid seals of the Neogene, United Slales Geoloq ical Survev Professional Paper, 992 , I-93. Repenning, C.4., Ray, C.E. and Grigorescu, D. L979. Pinniped biogeography, in Gray, J. and Boucot, A.J. (eds), Historical Biqqeqqlep¡)4 elatp_fqçlq!þlrqlq the Chanqinq Environmenl. Oregon State University Press, pp 357-369. Rhodin, J.A.G. 1966. Ultrastructure and funclion of the human lracheal mucosa, American R eview of Resoiralorv Disease. 9J , 1-r5. Rhodin, J.A.G. I974. Histoloqy: A TexL and Atlas. New York: Oxford University Press. Rhodin, J.A.G. L978. Microscopic anatomy of Lhe pulmonary vascular bed in the cat lung, M icrovascular Research, I5 , L69-r93 Ridgway, S.H. L912. Homeoslasis in the aquatic environment, in Ridgway, S.H. (ed.), Mammals of the Sea: Bioloqy and Medicine. Springf ield: C.C. Thomas, pp 590-7 47 . Ridgway, S.H. 1976. Diving mammals and biomedical research, Oceanus, )-9,49- 55. Ridgway, S.H. and Johnston, D.G. L966. BIood oxygen and ecology of porpoises of three qenera, Science r5r, 456-458. Ridgway, S.H., Scronce, B.L. and Kanwisher, J. \969. Respiration and deep diving in the bottlenose porpoise, Science. 166, 165I-1654. Ridqway, S.H., Simpson, J.G., Patton, G.S. and Gilmartin, W.G. 1970. Hematologic f ind ings in certain small cetaceans, Journal of lhe Amenican Veterinary Medical Associalion r57, 566-575. Ridgway, S.H., Carder, D.A. and Clark, W. L975. Conditioned bradycardia in the sea lion Zalophus californianus, Nature, 256, J7-38. Ridgway, S.H. and Howard, R. 1919. Dolphin lung collapse and inlramuscular circulation during free divinq: Evidence from nitrogen washoul, Science,206, 1r.8 2-r.l_Bl. 2rl. Robin, E.D., Murdaugh, H.V.Jr, Pyron, W., Weiss, E. and Soteres, P. 1963- Adaptations to diving in the harbour seal - gas exchange and ventilatory response to CO2r American Journal of Physioloqy, 2O5, II75-II77- Robinson, 4.J., Kropatkin, M. and Aggeler, P.M. 1969. Hageman factor (factor XII) deficiency in marine mammals. Science 166, 1420-1422. Robinson, N.E. 1982. Some functional consequences of species differences in lung anatom y, Advances in Veterinary Science and Comparative Medicine, 2 6, I-32 Ronald, K. and Healey, P.J. I9BJ.. Harp seal - Phoca qroenlandica, in Rid 9waYt S.H. and Harrison, R.J. (eds), Handbook of Marine Ma mmals. Vol. 2, Seals. London: Academic Press, pp. 55-87. Rosenquist, T.H., Bernick, S.. Sobin, S.5. and Fung, Y.C. 1973. The structure of the pulmonary interalveolar microvascular sheet, Microvascular Research 5) r99-2r2. Ryan, S.F. 1969. The structure of the interalveolar septum of the mammalian lung, Anatomical Record 165,467-484. Ryan, S.F. L973. The structure of the primary lung lobule, Annals of Clinical and Laborator Science 3, r47-r55. Schalm, O.W., Jain, N.C. and Carroll, E.J. I975. Velerinary Hematoloqy. lrd Ed Philadelphia: Lea and Febiger Scheffer, V.B. L976. A Natural History of Marine Mammals. New York: Charles Scribner's Sons. Schmidt-Nielsen, K. I979. Animal Phvsioloqv: Adaptation and Env ironment. Second Edition. Cambridge: Cambridge University Press. Scholander, P.F. l-940. Experimental investigations on the respiratory function in diving mammals and birds, Hvalradets Skrifter 22, I-I3r. Scholander, P.F. 1964. Animals in aquatic environments: Diving mammals and b irds, in Handbook of Physioloqy - Adaptation to the Environment. Washi ngton, DC: American Physiotogical Society, pp 729-739. 2LB. Scholander, P.F. and Irving, L. 1941-. Experim enlal invesligations on the respiralion and diving of the Florida m anatee. Journal of Cellular and Comparative Physioloqy, 17 , r69-L9I. Schreider, J.P. and Raabe, O.G. 198l-. Struclure of the human respiratory acinus, American Journal of Anatom L62,22r-232. Semba, R. 1979. Contributions to semithin sectioning on a conventional rotary microtome, Stain Technolo 54,25r-255. Shapunov, V.M. 197I. Evaluation of the economy and effectiveness of external respiration in the dolphin Phoceana phoceana, Journal of Evolutionar Bioche m istry and Physioloqy, 7 ,33I-335. Shay, J., L975. Economy of effort in electron microscope morphometry, American Journa I of Patholoov. Bl,5O3-5L2. Siegwart,8., Gehr, P., Gil, J. and Weibel, E.R. I971. Morphometric estimation of pulmonary diffusion capacity. IV. The normal dog lung, Respiration Physioloqy, I3, r4I-I59. Simon, L.M., Robin, E.D., Elsner, R., Van Kessel, A.L.G.J. and Theodore, J. I914. A biochemical basis for differences in maximal diving time in aquatic m am m als, Comp arative Biochemistry and Physioloqy, 478 ,2O9-2L5. Simpson, J.G., Gilmartin, W.G. and Ridqway, S.H. L970. Blood volume and olher haematologic values in young elephant seals, American Journal of Veterinary Research, )r, L449-1452. Simpson, J.G. and Gardner, M.B. 1912. Comparative microscopic anatomy of selected marine mammals, in Ridgway, S.H. (ed.), Mammals of the Sea: Bioloqy and Medicine. Springfield: C.C. Thomas, pp 298-418. Sims, B. I974. A simple method of preparing I-2 pm sections of large tissue blocks using glycol methacrylate, Jou rnal of Microscoov. lOL , 223-227. Sinnett, E.E., Kooyman, G.L. and Wahrenbrock, E.A. 1978. Pulmonary circulalion of the harbour seal, Journal of Applied Physioloqy - Resoiratory Environmenlal and Exencise Physioloqy, 45, 1 IB-721. 2r9. Slijper, E.J. 1962. Whales, Lranslated by Pomerans' A.J. London: HuLchinson. Smith, H.4., Jones, T.C. and Hunt' R D. 1972. Veterinary Patholoqy. 4th Ed. Philadelphia: Lea and Febiger. Smith, M.N., Greerrberg, S.D. and Spjut, H.J. I979. The Clara cell: A comparalive ultrastructural study in mammals, American Journal of Anatomy, I55, I5-3O. 5mith, P., Healh, D. and Moosavi, H. I974. The Clara cell, Thonax 29, 1.47-163. Sobin, S.S., Tremer, H.M. and Fung, Y.C. I970. Morphometric basis of the sheet- flow concept of the pulmonary alveolar m icrocirculation in the cal, Circulation Research, 26 , 397 -4r4. Sokal, R.S. and Rohlf, F.J. 1969. Biometry. San Francisco: Freeman. Sokolov,4.S., Kosygin, G.M. and Shustov, A.P. I97I. Slructure of lungs and trachea of Berinq Sea pinnipeds, in Arsenrev, V.A. and Panin, K.I. (eds), Pinnipeds of the North Pacìfic. Translated from Russian. Jerusalem: Israel Program for Scientific Translations, pp 25O-262. Spencer, M.P., Gornall, T.A. and Poulter, T.C. 1967. Respiralory and cardiac activity of killer whales, American Journal of Physioloqy. 22r 974-98I. spicer, s.s., Mochizuki, I., selser, M.E. and Martinez, J.R. 1980. Complex carbohydrales of rat tracheobronchial surface epithelium visualized u ltras lruc tural ly. Ame rican Journal of Anatomv. 15 8,93-LO9 stahl, w.R. 1967. Scalinq of respiratory variables in mammals, Journal of Physioloqy, 22 , 453-460. Stevens, C.M. l-971. The cardiovascular responses to breath-hold diving in the free-swimming California sea lion, Zalop hus californianus. Disserlation Abslrac Ls Internalional.32 , l_lBt-B - 1182-B Stirling, I. I972. Observations on the Australian sea lion, Neophoca cinerea (Peron). A ustralian Journal of Zooloov. 20 , 27r-279. 220. Storey, K.B. and Hochachka, P.W. L974. Glycolytic enzymes in muscle of lhe Pacific dolphin: Role of pyruvate kinase in aerobic-anaerobic transilion durinq div ing, Comparative Biochem is lry and P hvsioloov.49B ,1l_9-128. Strauss, R.H. 1979. Diving medicine, American Review of Respiralory Disease, ll9,1001-ro23. Stroud, R.K. and Dailey, M.D. t978. Parasites and associated pathology observed in pinnipeds stranded along the Oregon coast. Journal of Wildlife Diseases, J-4, 292-298. Stroud, R.K. and Roffe, f .J. I919. Causes of death in marine mammals stranded along the Oregon coasl, Journal of Wildlife Diseases t5, 9r-97. Sweeney, J.C. I914. Common diseases of pinnipeds. Journal of the American Ve terin ary Medical AssociaLion, 16 5, 805-810. Sweeney, J.C. and Gilmartin, W.G. I914. Survey of diseases in free-living California sea lions. Jour nal of Wildlife Diseases. l0, J7O-376 Tanji, D.G., Weste, J. and Dykes, R.W. 1975. Interactions of respiration and the bradycardia of submersion in harbor seals, - anadian Journal of Phvsioloov and Pharm acoloqy, 53, 555-559. Tarasoff, F.J. and Kooyman, G.L. I913a. Observalions on the anatomy of the respiratory system of lhe river otter, sea otter, and harp seal. I. The lopography, weighb, and measurements of the lungs, Canadian Journal of Zooloqy, 51,, 163-170. Tarasoff, F.J. and Kooyman, G.L. I97ib. Observations on lhe anatomy of the respiratory system of the river otler, sea otter, and harp seal. II. The trachea and bronhial tree, Canadian Journal of Zoolo 5L, L7l.-r77. Tenney, S.M. and Remmers, J.E. L963. Comparative quantitalive morphology of the mammalian lung: Diffusing area, Nalure, I97, 54-56. Thurlbeck, W.M. 1967. The internal surface area of nonemphysemalous lungs, American Rev iew of Resoiratorv Disease. 95 ,165-773 22r. Thurlbeck, W.M.and Wang, N.S. 1974. The structure of the lungs, in Widdicombe, J.G.W (ed.), Res irator Ph siolo MTP International Review of Science Physioloqy Series l, Volume 2. London: Bullerworths, pp 1-30. Tyler, W.S., Gillespie, J.R. and Nowell, J.A. 197I. Symposium on pulmonary and carcliac function. I. Modern funclional morphology of lhe equine lung, Equine Veteninary Journal, t, 84-94. Van Citters, R.L., Franklin, D.L., Smith, O.A.Jr, Walson, N.W. and Elsner, R.W. 1965. Cardiovascular adaptations lo diving in the northern elephant seal M iroun a ust iros tr i Comp arative Biochemistrv and Phvsioloqv. l6, 267- 27 6. Veicsteinas, A., Magnussen, H., Meyer, M. and Cerretelli, P. I976. P.ulmonary 02 diffusing capacity at exercise by a modified rebreathing method. European Journal of App lied Physioloqy, 35 ,79-BB. von Hayek, H . L960. The Human Lung (translated by Krahl, V.E.). New York: Hafner Publishing Co. Wahrenbrock, E.A., Maruschak, G.F., Elsner, R. and Kenney, D.W. I974. Respiration and metabolism in two baleen whale calves, M arine Fisheries Review (US Nat.Mar.Fish.Service), 36, J-9. Walker, G.E. and Ling, J.K. 1980. Body weights and dimensions of adult female Australian sea lions, Neophoca cinerea, Journal of Mammolo 6r, 164-165. Walker, G.E. and Ling, J.K. I981. Australian sea lion - Neophoca cinerea, in Ridgway, S.H. and Harrison, R.J. (eds), Handbook of Marine Mammals Vol. I The Walrus. Sea Lions, Fur Seals and Sea Otter. London: Academic Press, pp 99-118. W angens teen, D. and Weibe I, E.R. 1982. M orpho m etric ev alualion of chorioallanLoic oxygen transport in the chick embryo. Respiration Physioloqy, 47, r-2O. Weber, R.E., Johansen, K. and Abe, A.S. l9Bl-. Myoglobin from the burrowing rep I ile Amphisbaena alba. ConcentraLions and functional characleristics, ComDarative Biochemislrv and Phvsioloov. 68A , r59-165. 222. Weibel, E.R. 1961. Morphomelry of lhe Human Lunq. Berlin-Gottingen-Heidelberg: Spri nger-Verlag. Weibel, E.R. 1970a. Morphometric estimation of pulmonary diffusion capacity. I. Model and method, ResÞ iration Phvsiolooy. 1l ,54-75. Weibel, E.R. 1970b. An aulomatic sampling stage m icroscope f or stereology, Journal of Microsco 9, l-lB. Weibel, E.R. 1972. Morphometric estimation of pulmonary diffusion capacity. V. Comparative morphometry of alveolar lungs, R iration Ph iolo r4, 26- 43. Weibel, E.R. L973. Morphological basis of alveolar-capillary gqs exchange, Physioloc ical Reviews. 5 3, 419-495. Weibel, E.R. I979a. Stereoloq ical Methods. Volume 1. Practical Methods for Bioloq ical Morphometry. London: Academic Press. Weibel, E.R. 1979b. Oxyqen demand and lhe size of respiratory slructures in mammals, in Wood, S.C. and Lenfant, C. (eds), Evolution of Respiralory Processes: A C om oarative oroa ch. pp 289-346. New York: Marcel Dekker Weibel, E.R. 1981. Morphological basis for üO/O distribution, in Hutas, I and Debreczeni, L.A. (eds), Advances in Physioloqical Science. Volume 10. Respiration. Budapest: Akademiai Kiado, pp J.79-J-89. Weibel, E.R. l98l. How does lung structure affec[ gas exchange? Chesl, 83,657- 665. Weibel, E.R. and Gomez, D.M. l-962. Architecture of the human lung, Science, t37, 517-585. Weibel, E.R. and Knighl, B.W. 1964. A morphometric study on the thickness of lhe pulm onary air-blood barrier, Journal of Cell Bio 2r, J67 -384. Weibel, E.R., Kistler, G.S. and Scherle, W.F. 1966. Practical stereological melhods for m orphometric cytology, Journal of Cell Bio 3O,23-JB. 22i Weibel, E.R., Untersee, P., Gil, J. and ZuIauf, M. L971. Morphometric estimalion of pulmonary diffusion capacity. VI. Effect of varying positive pressure inflation of air spaces, Respiratory Physioloqy. IB, 285-108. Weibel, E.R., Bachofen, M. and Gehr, P. I975. Estimation of pulmonary diffusing capacity in human lungs by morphometry, Proqress in Resp iratorv R esearch. B, 110-140. Weibel, E.R. and Gil, J. 1.917. Struclure-function relationships at the alveolar level, in West, J.B. (ed.), Bioenqineerinq Aspects of the Lung. New York: Marcel Dekker, pp. l-BJ-. Weibel, E.R., Claassen, H., Gehr, P., Sehovic, S. and Bumi, P.H. 1980. The respiratory system of the smallest mammal, in Schmidt-Nielsen,. K., Bolis, L. and Taylor, C.R. (eds), Comparative Physioloqy: Primitive Mammals. Cambridqe: Cambridge University Press, pp lBJ--19J.. Weibel, E.R., Gehr, P., Cruz-Orive, L.M., MÜller, 4.E., Mwangi, D.K. and Haussener, V. l98l-a. Design of the mammalian respiratory system. IV. Morphometric estimation of pulmonary diffusing capacity; critical evaluation of a new sampling method, Respiratory Physioloqy, 44, 39-59. Weibel, E.R., Taylor, C.R., Gehr, P., Hoppeler, H., Mathieu, O. and Maloiy, G.M.O. I98Ib. Design of the mammalian respiratory system. IX. Funclional and structural limils for oxyg en flow. Respiralory Physioloqy, 44, 15I-164. Weibel, E.R., Taylor, C.R., O'Neil, J.J., Leith, D.E., Gehr, P., Hoppeler, H., Langman, V. and Baudinette, R.V. 1983. Maximal oxygen consumption and pulmonary diffusing capacity. A direct comparison of physiologic and m orp hometric measurements in canids. Respiralion Physioloqy, 5 4, L7J-rBB. Welsch, U. and Drescher, H.E. L9B?. Liqht- and electron m icroscopical observations on lhe terminal airways and the alveolar parenchyma of the Antarctic seals Lobodon carcinophaqus and Leptonychotes weddelli, Polar Bioloqy, l, 105-l-14. West, J.B. 1978. Regional differences in lhe lung, Chest, 4, 426-437 224 Widdicombe, J.G. and Pack, R.J. 1982. The Clana cell, European Journal of Respiratory Diseases, 63, 2O2-22O. Wislocki, G.B. 1,929. On the structure of the lungs of the porpoise t_Iq!þpg truncatus) American Journal of Anatomy, 44, 47 -77 . Wislocki, G.B. 1915. The lungs of the manalee Trichechus latirostris) compared with those of other aquatic mammals, Bioloqical Bulletin, 6 8,385-396. Wislocki, G.B. L942. The lungs of the cetacea, with special reference to the harbour porpoise (Phoca"na phoc""na , Analom ical Records, 84, lt7- LzI. Wislocki; G.B. and Belanger, L.F. 1940. The lungs of the larger cetacea compared to those of smaller species, Bioloqical Bulletin, 7B , 289-297. Young, 1., Vazzone, R.W. and Wagner, P.D. 1980. Identification of functional lung unil in the dog by graded vascular embolization, Journal of Applied Physioloqy - Resoirat,orv Environmental and Exercise Ph ysioloqy. 49, rJz-r4r. Zapol, W.M., Liggins, G.C., Schneider, R.C., Qvisl, J., Snider, M.T., Creasy, R.K. and Hochachka, P.W.I979. Regional blood flow during simulated diving in the conscious Weddell seal, Journal of Aoolied Ph ysioloqy - Respinalory Environmental and Exercise Physioloqy, 47 , 968-97).