Aust. J. Zool., 1984, 32, 721-42
Changes in the Male Reproductive Organs of the
Dugong, Dugong dugon (Sirenia :Dugondidae)
with Age and Reproductive Activity
H. Marsh,* G. E. Heinsohn* and T. D. GloverB
A
Zoology Department, James Cook University, Townsville, Qld. 481 1.
* Department of Veterinary Anatomy, University of Queensland, St Lucia, Qld 4067.
Abstract
The anatomy and histology of the male reproductive tract of the dugong (Dugong dugon) is described.
Each testis and its adjacent epididymis lie immediately caudal to the corresponding kidney. The seminal
vesicles are large but there is no discrete prostate gland and the bulbo-urethral glands are also diffuse.
Both qualitative and quantitative examination of the testes and epididymides of 59 males whose ages
have been estimated from tusk dentinal growth layer counts indicate that the male dugong does not
produce spermatozoa continuously, despite the absence of a distinct breeding season. Individual
dugongs were observed with testes at all stages between complete quiescence and full spermatogenesis,
and only 10 of the 40 mature males had fully spermatogenic testes and epididymides packed with
spermatozoa. Androgenic and spermatogenic activity of the testes appeared to be in phase, but the
testicular histology of some old males suggested that they may have been sterile for long periods.
Introduction
The dugong, Dugong dugon (Miiller), one of the four extant species of Sirenia or seacows,
is of particular interest as the only existing herbivorous mammal which is strictly marine.
Although the dugong is listed as vulnerable to extinction (Thornback and Jenkins 1982),
aerial surveys over the last six years have confirmed that sizeable populations still occur in
the shallow seas around northern Australia (Heinsohn et al. 1976, 1978; Ligon 1976;Marsh
et al. 1981).
Since 1969, specimens which ideally included skulls and reproductive tracts have been
collected from 160 north Queensland dugongs, in order to obtain information about the
animal's life history that is basic to the development of effective conservation and
management procedures. Marsh (1980) counted tusk dentinal growth layers to determine the
ages of these animals and presented strong evidence for annual layer deposition, indicating
that dugongs have a life span of 50-60 years. Concomitant preliminary examination of the
reproductive tracts suggested a minimum pre-reproductive period of 9-10 years for both
sexes, and a cycle of male sexual activity in which not all animals are in rut simultaneously
(Marsh 1980).
Subsequent qualitative and quantitative histological studies of the testes and
epididymides of 59 dugongs of a wide range of ages have been performed with the aim of
defining precisely the terminal reproductive state of each male. Results of a parallel study of
49 female reproductive tracts, and on breeding cycle, life history and population dynamics
are published separately (Marsh et al. 1984a, 1984b).
Although Home (1820), Riha (1911), Hill (1945), Gohar (1957), Petit (1925), and
Harrison (1969) have described various aspects of the gross reproductive anatomy of the
male dugong, even the exact location of the testes (Harrison 1969) seems unclear and no
previous histological records are known to us. Accordingly, this paper also includes details of
the gross anatomy and histology of the male reproductive organs.
H. Marsh, G. E. Heinsohn and T. D. Glover
Materials and Methods
Material
Source. Male reproductive tracts were obtained between May 1969 and October 1981 from the
following dugongs in northern Queensland: 3 1 animals that were accidentally drowned in nets (usually
shark nets set for swimmer protection) near Townsville (19" SS., 146"15'E.) had material collected from
them within 2 h-2.5 days of death (Heinsohn 1972;Heinsohn and Spain 1974);25 that were speared and
drowned for food by Aborigines near Mornington I. (16"301S.,13Y30'E.) had material collected from
them within 0.5-12 h of death (Marsh et al. 1981); and three that were similarly killed for food by
Islanders in Torres Strait (10"30fS.,142'3'E.) were sampled within 0.5-12 h of death.
Description. Body-length frequencies of the dugongs from Townsville and Mornington I. are
illustrated (Fig. 1)arranged in 0.2-m body-length classes. The ages were estimated from dentinal growth
layers in the tusks (Marsh 1980),which erupt and wear in adult males, so that minimum age estimates
only are available for these animals. The age-frequency distribution and tusk-eruption status of males
from Townsville and Mornington I. are summarized in Fig. 2. Data on all males including the three from
Torres Strait are shown in Tables 1, 3 and 4.
Fig. 1
12 r
Fig. 2
Body length im)
Age
(YI
Fig. 1. Body-length frequency diagrams (by 0.2-m classes) of male dugongs collected in 1969-81 from:
(a) Mornington I.; (b) near Townsville. Open bars, tusks not erupted; stippled bars, tusks erupting; solid
bars, tusks erupted and worn.
Fig. 2. Frequency diagrams of 5-year age classes of male dugongs collected in 1969-81 from:
(a) Mornington I.; (b) near Townsville. Open bars, tusks not erupted; stippled bars, tusks erupting; solid
bars, tusks erupted and worn, age estimates a minimum only. Animals have been assigned to the age
class which corresponds to the number of growth layer groups in their tusks, irrespective ofwhether the
tusks are worn.
Field collection procedures. Each carcass was measured and dissected, attention being focussed on
the collection of the stomach contents, skull and reproductive tract. One-centimetre cubes were
removed from the centre ofthe right testis and from the right epididymis opposite the caudal pole of the
testis. These samples were fixed in Cleland's solution (Rowley and Heller 1966) for 24 h. After incision
of each testis, the entire reproductive tract was preserved by injection with, and immersion in, 10%
seawater formalin.
The reproductive tracts of six freshly killed adults were examined in more detail by excising samples
from the efferent ductules, epididymis and ductus deferens at several levels. Other accessory organs
were also sampled. The samples were fixed in Bouin's fluid for 24 h. Before the epididymis was fixed,
spermatozoa were collected from an arbitrarily chosen level of the mesonephric duct (opposite the
caudal pole ofthe testis) by smearing the cut surface across several slides and allowing them to dry in the
air. One testis and the corresponding testicular artery from an adult were removed and frozen.
Dugong Male Reproductive Organs
Measurement, Histological Preparation and Examination of Reproductive Specimens
Each fixed testis was trimmed of epididymis and fat and weighed to 0.1 g. The specimen of testis that
was fixed in Cleland's solution was included in the weighing. Testis length, width at midlength, and
dorso-ventral thickness at midlength were measured with vernier calipers. In order to check that all
regions ofboth testes from each dugong were at the same stage of reproductive activity, three extra tissue
blocks were taken from each testis along the longitudinal midline and two from along the transverse
midline. These specimens were post-fixed in Bouin's fixative (Lillie and Fullmer 1976, p. 61) for 24 h
before subsequent processing.
Tissues for histology were dehydrated in ethanol, cleared in xylene, embedded in Paraplast and
sectioned at 5 ym or 6 ym. Sections were stained with Mayer's haemalum and Young's eosin-erythrosin
(Hand E) (Marsh et al. 1977), Masson's trichrome, or the PAS procedure and Mayer's haemalum (Culling
1974). Some sperm smears were stained with Mayer's haemalum; others were coated with
gold-palladium and examined with the scanning electron microscope.
After being thawed, the testicular artery of the frozen testis was injected with latex. The testis was
subsequently cleared with salicylate for examination of the arterial architecture. The blood vessels were
then examined.
Region of efferent ductules
Testis
Middle level of epididymis
Lower level of epididymis
(sperm store)
Ductus deferens
Seminal vesicle
Disseminated bulbo-urethral glands
Pen~s(cut o f f )
Fig. 3. Schematic diagram of the male reproductive tract of the dugong.
Each slide was scanned qualitatively at xlOO and x400 in order to assess overall spermatogenic
activity. The diameters of 100 round seminiferous tubules, randomly selected from all samples from
each testis, were measured with a calibrated ocular micrometer at x400 magnification. In retrospect, this
sample was unnecessarily big, because the standard deviation per testis was not very large (see Tables 1,
3,4). Subsequent quantitative histological studies were limited to well preserved material. The cell types
present were identified under oil immersion at x1000. The sections that were used for cell identification
included those stained with PAS, trichrome, and H and E. At least 10 cells of each type were measured by
an ocular micrometer at x1000.
Quantitative information on testicular activity was obtained by use of phase analysis. This analysis
was based on a series of phases, each of which shows a certain cellular association of the seminiferous
epithelium. A series of 12 phases was recognized with this procedure, which differed slightly from
standard methods developed for mammals which undergo continuous spermatogenesis (Courot et a!.
1970).
Since no qualitative differences were observed in the histological appearance of the seminiferous
tubules in different regions of one testis or between contralateral testes, the phase analysis was
performed on one H-and-E-stained section from the Cleland's fixed sample from each dugong. (The
fixation of these samples was superior to the others.) The phase in each of 200 transverse sections of
tubules was identified at a magnification of x400. This sample size was comparable to those used for
H. Marsh, G. E. Heinsohn and T. D. Glover
Fig. 4. (A) Ventral view of the left kidney and testis of an adult dugong in situ. The testis is 9 cm long.
t, Testis; k, kidney. (B) Ventral view of a dissection of part of the left side of the reproductive tract of a
10-year-old dugong with fully spermatogenic testes. The testis is 11 cm long. The lower epididymis has
been displaced to the right. e, epididymis; s, seminal vesicle. (C) Photomicrograph of seminal vesicle
from a dugong aged at more than 28 years, with fully spermatogenic testes. The pseudostratified
epithelium consists of numerous cuboidal principal cells and a few round basal cells. H and E. (D)
Photomicrograph of cross-section of bulbo-urethral gland of the same dugong as Fig. 4C. Note the
columnar cells with basal nuclei. PAS and Mayer's haemalum. (E) Scanning electron micrograph of
spermatozoa from the lower level of the epididymis of a dugong aged at least 36 years.
Dugong Male Reproductive Organs
similar analyses of the testes of the African elephant (Johnson and Buss 1967) and the albino rat
(Roosen-Runge and Giesel 1950).
Within the lobules of each testis, the volume fractions of the tubular and intertubular components
(Leydig cells, other cells, space) were estimated by use of a Weibel graticule (Wild Heerbrugg Ltd) and
the principles of stereology (Schaefer 1970). This technique involved recording the elements appearing
at 42 points (the ends of each of 21 lines of equal length arranged in 7 equally spaced rows) in each of a
series of random fields at x400. Slides from three dugongs (one immature and two mature animals) were
sampled at 10, 15 and 20 fields at x400 in order to determine an appropriate sample size. The testis of
one of the two mature dugongs was producing spermatozoa but the seminiferous tubules of the other
contained spermatogonia and Sertoli cells only. On the basis of results calculated at each of the above
levels, a sample of 10 fields from one section of Cleland's fixed tissue per testis was selected. Counts
beyond this level produced only slight variations.
Observations
General Anatomy
Fig. 3 summarizes diagrammatically the principal anatomical features of the male
reproductive tract.
,Testis
Large epididymal arteries
The testes are dorso-ventrally flattened and ovoid. They are abdominal, each lying caudal
and slightly lateral to the corresponding kidney (Fig. 4A). Inguinal canals and inguinal
ligaments are absent. Each testis is encased in a cream-coloured fibrous tunica albuginea
from which connective tissue septa converge on the mediastinum, dividing the testis
internally into distinct lobules which are easily recognizable macroscopically. The dugong
has a straight testicular artery (Fig. 5) which divides into three or four branches before
entering the substance of the testis. The absence of a pampiniform plexus, as we11 as the
lack of a helical testicular artery, characterize the testis as that of a typical testicond
mammal.
At least 14 ductuli efferentes emerge from the cranial pole of each testis to join the
proximal region of the mesonephric (Wolffian) duct. Adjacent to the dorsolateral side of
each testis, the mesonephric duct is highly convoluted and forms a distinct epididymis (Fig.
4B) which extends several centimetres caudally and medially beyond the testis before
gradually straightening out into the ductus deferens.
Large, thick-walled, ovoid, seminal vesicles (Figs 3,4B, 4C) lie dorsal to the bladder. The
tissue surrounding the prostatic urethra consists mainly of bundles of smooth muscle and
erectile tissue with limited, poorly developed, urethral mucous glands. We have found no
discrete structure corresponding to a pars compactum of a prostate gland, so prostatic tissue
in dugongs is likely to be disseminated along the wall of the pelvic urethra. Bulbo-urethral
glands are reported to be absent in sirenians (Riha 1911; Harrison 1969) and we have
not been able to identify distinct glands macroscopically. Histologically, however,
bulbo-urethral type glandular tissue (Fig. 40) was identified dispersed among the muscle
H. Marsh, G. E. Heinsohn and T. D. Glover
surrounding the urethra close to the root of the penis, in a dugong whose testes were
producing spermatozoa. Because of the limited amount of adequately fixed tissue available,
we were not able to determine how this bulbo-urethral type tissue changed with the level of
testicular activity.
The penis (Fig. 6) is not visible externally unless erected, when it is extruded through the
preputial orifice about halfway between the anus and umbilicus. On morphological grounds
the penis should be classified as vascular, meaning that erection is probably achieved
entirely by vascular engorgement. The penile blood vessels are extensive and the corpus
cavernosum penis substantial. At the root of the penis the corpus cavernosum is undivided
but in most of the shaft it is divided by a septum. Near to the apex of the penis, the corpus
cavernosum penis extends upward and laterally to fill two lateral and ventral bulges that
form a glans penis or corona glandis. Beyond this expansion, the urethra, surrounded at first
by both corpus cavernosum erectile tissue and corpus spongiosum, extends beyond this
corona to form an erectile urethral process. Similar bulbs of erectile tissue occur in other
eutherian species and have been seen in the penis of the nine-banded armadillo Dasypus
novemcinctus (personal observation). The anatomy of the penis and its muscles has been
described in detail by Riha (1911) and Petit (1925). Harrison (1969) includes photographs
of the glans.
Fig. 6. Ventral view of the 20-cm-long penis of the same dugong as in Fig. 4B.
Immature Dugongs
(i) The seminiferous tubules
The seminiferous tubules of immature dugongs are narrow and range in mean diameter
from about 36 to 66 pm (Table 1; Fig. 7). The basement membrane of each tubule is lined
with numerous spermatogonia and Sertoli cell nuclei (Fig. 8A). Variable numbers ofprimary
spermatocytes (Table 1) are sometimes present in the centre of the tubules but spermatids
and spermatozoa are absent. Lumina are only clearly apparent in the seminiferous tubules of
some adult animals.
The Sertoli cell nuclei (Fig. 8A) are oval, up to 10.5 pm long, and usually aligned with
their long axes perpendicular to the basement membrane. Each nucleus has a typical and
conspicuous nucleolus, evenly distributed chromatin, and a distinct nuclear membrane of
irregular outline. The Sertoli cell cytoplasm is characteristically indistiinct.
Two types of spermatogonia are present. Type A spermatogonia (sensu Bloom and
Fawcett 1975) (Fig. 8A) each have a round to oval nucleus up to 10.5 pm in diameter with
fine, evenly distributed chromatin granules, one or two distinct nucleoli and a distinct
membrane. The cytoplasm stains faintly so that the cell borders may be difficult to see. Type
B spermatogonia tend to be slightly smaller than type A, the nucleus of these cells is about
7.5 pm in diameter and has coarser and more abundant chromatin than those of type A.
Dugong Male Reproductive Organs
Table 1. Age, body length, mean formalin-fixed testis weight and details of the
seminiferous tubules of dugongs with immature testes
Values for tubule diameters are means t standard deviations. NA, not available
Accession
number
Body
length (m)
Age
(y)
Mean testis
weight (g)
Tubule
diameter (pm)
Tubules with
spermatocytes (Yo)
Townsville
2
5.1
NA
6.5
12.5
14.4
NA
NA
11.6
NA
15.5
t l l
Mornington I
7.8
15.8
16.5
KA
16.2
Torres Strait
9.1
Mean tubule diameter (gm)
Fig. 7. Relationship between the seminiferous tubule diameter, testis weight and stage of testicular
development or activity. .Immature. approaching puberty. @Mature: resting (phase - 3
predominant). A Mature: intermediate (phase -2 predominant). o Mature: recrudescent (phases - 1,0,
or 1-5 predominant, few or no spermatozoa in epididymis). o Mature: fully spermatogenic (phases 1-5,
many spermatozoa in epididymis). rn Mature: spermiogenesis finishing (phase 6 predominant).
H. Marsh, G. E. Heinsohn and T. D. Glover
Fig. 8. (A) Photomicrograph of immature testicular tissue from a 5.5-year-old dugong showing narrow
closed seminiferous tubules, the peripheries of which are lined with spermatogonia (arrowed) and
numerous Sertoli cell nuclei (arrowhead). Sertoli cell nuclei, spermatogonia, 'and occasional
spermatocytes may also occur in the centre of the tubules. Leydig cells (L) are prominent and
vacuolated. (B)-(F) The five cellular associations observed in the seminiferous tubules of mature
dugongs with testes in a state of arrested or incomplete spermatogenesis. Phase -4 is the least active
phase; phase 0 the most. All material obtained within 0.5 h of death from dugongs with erupted tusks
and estimated to be older than 20 years. (B)Phase -4. Sertoli cell nuclei (arrowhead) and types A and B
spermatogonia (arrowed) line the periphery of solid tubules which have connective tissue fibres;
occcasional Sertoli cell nuclei and vacuolated Sertoli cell cytoplasm in the central area. The lamina
propria surrounding the basement membrane is about 5 pm thick. Note the interstitial fibrosis and the
vacuolated Leydig cells (L). (Q Phase - 3. Like phase - 4 but with low numbers of spermatocytes and
more Sertoli cell nuclei in the central area of the tubules. Note the intratubular pigment deposits
(arrowed) and type A spermatogonium (arrowhead). This phase was also observed in the testes of No.
112, which was considered to be approaching puberty. (D) Phase - 2. Like phases -4 and - 3 but with
numerous primary spermatocytes (arrowed) in tubules which may be open or closed. This phase was
also observed in the testes of No. 112,which was considered to be approaching puberty. (E)Phase - 1.
Primary (arrowed) and secondary (arrowheads) spermatocytes present. Cellular associations organised.
Lumen usually present. Leydig cells (L) hypertrophied in this animal. (F)Phase 0. Like phase - 1 but
with a few spermatids (arrowed) or spermatozoa present. All sections stained with H and E.
Dugong Male Reproductive Organs
The primary spermatocytes each have a round nucleus about 8.5-10.5 pm in diameter, in
which the chromatin configurations typical of pachytene spermatocytes are clearly visible.
The nuclear membrane is poorly defined, but the cytoplasm is conspicuous.
(ii) The intertubular tissue
Intertubular tissue (Fig. 8A) is prominent and abundant. The Leydig cells are polyhedral
and each has an eccentric round or oval nucleus with one or two centrally placed nucleoli,
granular chromatin, and a prominent nuclear membrane. The cell boundary is often
indistinct. Most cells contain a large cytoplasmic vacuole (Fig. 8A); however, delayed
fixation must be taken into account in interpreting this feature.
Leydig cells, about 5 pm in nuclear diameter and with cytoplasmic vacuoles about
6.5 pm in diameter, were visible in sections from the youngest dugong from which well fixed
testicular tissue was available (No. 150, estimated age 1.5 years), even though much of its
other intertubular tissue was undifferentiated. The Leydig cells of the other two immature
animals examined in detail (No. 103, 5.5 years; No. 117, 15.5 years) were slightly larger,
with nuclei averaging 6.8 pm and 6.3 pm in diameter and cytoplasmic vacuoles averaging
9.5 pm and 7 Fm in diameter, respectively. Of the lobular tissue, 27% in No. 103 and about
19% in No. 117 consisted of Leydig cells (Table 2).
Table 2. Details of the relative volumes of intertubular and tubular tissue in the interlobular area of the
testes of dugongs
Dugongs are ranked in ascendingorder of reproductive maturity and activity as determined by testicular histology. The
phase of spermiogenesis follows the reproducrive category; for interpretation, see text
Reproductive
activity
ranking
Acc.
No.
Reproductive category and
predominant phase
Tubule
volume
(yo)
Inter-tubule volume (Yo)
Space
Leydig
Other
cells
Immature
Immature
Approaching puberty (-2, Regressed (- 4)
Resting (-3)
Resting (- 3)
Intermediate (- 2)
Intermediate (-2)
Intermediate (-2)
Recrudescent (- 1)
Recrudescent (- 1, 0, 1)
Recrudescent (1)
Recrudescent (1-4)
Fully spermatogenic (1-5)
Fully spermatogenic (1-5)
Spermiogenesis finishing (6)
(iii) Testis size
If males from Townsville and Mornington I. are considered separately, our results
indicate that the size of the testes of immature dugongs tends to increase with increasing age
(Fig. 9) and body length (Table 1). Single testis weights of the 18 immature dugongs
examined ranged from 2 g in an animal of body length 1.28 m, to 16.5 g in one of unknown
body length; testis lengths ranged from 2.5 to 4.8 cm in the same animals. The body lengths
of these immature dugongs ranged from 1.28 to 2.49 m and their ages from 0 to 6 years
(Townsville population, from which specimens aged between 6.5 and 8.5 years inclusive
were unavailable) and up to 15.5 years (Mornington I. population) (Table 1).
(iv) Tusk eruption
None of the immature dugongs examined had erupted tusks.
H. Marsh, G. E. Heinsohn and T. D. Glover
(v) Criteria for classifying a male dugong as immature
A dugong was classified as immature if it had a single testis weight of less than 20 g and its
testicular histology was similar to that shown in Fig. 8A.
Adult Dugongs and those Approaching Puberty
Forty-one dugongs from 9 to more than 36 years old had testes which were larger than,
and histologically distinct from, those of the immature animals, as can be seen by comparing
Figs 8B-8F and Fig. 10 with Fig. 8A. Spermiogenesis was occurring in the testes of only 16
dugongs. These animals spanned the entire adult age range and included four whose tusks
had not yet erupted (Nos 4, 38, 146, 156; Table 3). This suggests that the tusks erupt after
puberty (first spermiogenesis). It was not possible to use testicular histology alone to
distinguish males approaching puberty from those whose testes were recrudescing after a
period of quiescience. This distinction was made more difficult by the variation in the age of
sexual maturity and the absence of a sharply defined breeding season (see Marsh et al.
1984b).
(i) Cell types present in the seminiferous tubules
Spermatogonia. Types A and B spermatogonia, similar to those in immature testes, lie in
typical fashion adjacent to the basement membrane (Figs 8B, 8C).
Fig. 9. Relationship between testis
weight and age for immature
dugongs from Townsville (0) and
Mornington I . (m).
Spermatocytes. Mature primary spermatocytes (Fig. 8 0 )are similar to those in immature
testes, pachytene forms being especially conspicuous. Secondary spermatocytes (Fig. 8E) are
relatively difficult to find, as is usual in mammalian testes where equational divisions are
completed rapidly. The occasional secondary spermatocyte that is seen has a round nucleus
about 7
in diameter (intermediate in size between a primary spermatocyte and a round
spermatid) with a distinct membrane. The cytoplasm is conspicuous.
Spermatids. Early spermatids (Fig. 10A) each have a spherical nucleus about 5 ym in
diameter surrounded by an easily discernible membrane and distinct cytoplasm, but as
spermateliosis commences the nucleus elongates and its affinityfor haemotoxylin increases
(Fig. 10B). The cytoplasm is progressively reduced (Fig. 10C) and a distinct tail becomes
visible (Fig. 10E). Remnants of spermatid cytoplasm are to be seen as residual bodies
(Fig. 1OD).
Spermatozoa. The spermatozoon of the dugong (Fig. 4-5) has a flattened head about 5 ym
long and approximately 3 ym wide, which stains lightly with haematoxylin. The tail is about
60 pm long. Testicular spermatozoa retain the cytoplasmic droplet proximally in the region
Dugong Male Reproductive Organs
Body
length
(m)
Age
(y)
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Just
Yes
Yes
Yes
Yes
Tusks
erupted
AConsidered to be approaching puberty
Acc.
No.
Dale of
death
Mcan
testis
wt (9)
(w)
Tubulc
diameter
3
-2
90
3.5
10
95
Torres Strait
1.5
Mornington Island
-4
State of tubules (percentage at each phase)
-1
0
1
2
3
4
For explanation, see Table 3. Accession numbers in italics are of animals examined within 1 h of death
5
Table 4. Summary of the age and reproductive status of Mornington I. and Torres Strait dugongs with adult testes
6
Density of
sperm in
epididymis
Dugong Male Reproductive Organs
of the neck of the cell but, as yet, the level of the epididymis at which the droplet migrates to
the end of the midpiece has not been determined.
Fig. 10. Photomicrographs of cross-sections of several seminiferous tubules from a dugong (minimum
age 36 years) with fully spermatogenic testes. The diameter ofindividual tubules is larger and the tubule
lamina propria thinner (c. 2 l m thick) than in Fig. 7. The intertubular connective tissue is reduced; the
Leydig cells are large and vacuolated. (A) Phase 1. Sertoli cell nuclei and spermatogonia line basement
membrane. Primary and secondary spermatocytes organised into layers. Lumen surrounded by round
spermatids. (B)Phase 2. Spermatids elongating and becoming associated with Sertoli cells. (C)Phase
3. Spermatid bundles well organized. Maturation divisions occurring among primary and secondary
spermatocytes. This stage does not usually occupy whole tubule cross-section. (D) Phase 4. Two
generations of spermatids present. Elongate spermatids have tails and are more central to lumen.
( E )Phase 5. Sperm line tubule lumen. Round spermatids also present. (F)Phase 6. Spent tubule at the
end of period of sexual activity. Lumen of tubule very wide and surrounded by an incomplete
assemblage of cell layers which may contain a subset of any of the cell types. All sections stained with H
and E. Material fixed within 0.5 h of death.
Sertoli cells. Sertoli cells (Figs 8B, 8C) are morphologically similar to those in immature
dugongs but they occur less frequently.
Conspicuous deposits of light brown pigment (possibly lipofuscin) (Fig. 8C) occurred in
the seminiferous tubules of all dugongs more than 20 years old. Small amounts were
similarly present in No. 154 which was 15 years old.
H. Marsh, G. E. Heinsohn and T. D. Glover
(ii) The phase analysis
Eleven phases of activity were recognized in the seminiferous epithelia of adult dugongs.
These phases are illustrated in Figs 8B-8F and 10A-1 OF, and are described in the legends to
the plates. Phases 1-5 (Fig. 10) are characteristic of active spermiogenesisand represent the
orderly sequence ofevents that occurs during the transformation of round spermatids (phase
1)into mature spermatozoa (phase 5). They correspond to phases 9 and 5-8 used by Johnson
and Buss (1967) for African elephant testes, and represent a simplified version of the more
elaborate phase series developed for animals which undergo continuous spermatogenesis
(Courot et al. 1970).
Some dugongs with testes in which spermatogenesis was active had epididymides which
were relatively devoid of spermatozoa. These animals, i.e. Nos 116, 140 and 2 11 (Table 4),
had a low proportion of tubules at phases 4 and 5 combined, in contrast to dugongs at the
height of reproductive activity with large numbers of spermatozoa in their epididymides,
e.g. Nos 38 and 146 (Table 3), and Nos 109, 210, 213 and 214 (Table 4). This observation
indicates that spermiogenesis was just starting in Nos 116, 140 and 21 1, a conclusion
supported by the observation that a majority of the tubules ofNo. 116 were still at phases - 3
to 0 inclusive (Table 4). These phases are characteristic of dugongs with testes in which
spermatogenesis is arrested or incomplete (see below and Table 2). However, even though
No. 104 (Table 4) had few spermatozoa in its epididymis, it had a high proportion (45%) of
tubules at phases 4 and 5.
We consider tubules at phase 6 (Fig. 10F) which were only observed in Nos 114 and 214
(Table 4) to be characteristic of spent testes towards the end of a period of production of
spermatozoa. No. 2 14, which still had epididymides packed with spermatozoa, had only 4%
of tubules at phase 6. In contrast, No. 114, which had 68.5% of tubules at phase 6, had few
tubules at any of the other phases and only moderate numbers of spermatozoa in its
epididymides.
Adult dugongs in which spermatogenesis is arrested or incomplete have tubules in the
phase range from -4 to 0. Phase -4 (Fig. 8B) represents extreme inactivity and only three
animals (No. 68, minimum age 31.5 years; No. 80, minimum age 24.5 years; No. 127,
minimum age 33 years) had at least 50% of tubules at this stage. The shapes of the pulp
cavities of the tusks of these animals suggest that they were all considerably older than the
minimum estimated age. They may have been aspermatogenic for long periods or even
permanently. The testes of 13 dugongs had 50% or more of their tubules at phase - 3 (Fig.
8C), which suggests that this is the usual resting phase. Phase -2 (Fig. 80) is an intermediate
phase and was the predominant (>50°/o of tubules) phase in the testes of eight dugongs. We
consider that phases - 1 and 0 (Figs 8E and 8 F respectively) are part of the sequence of
increased activity that precedes spermiogenesis, and that animals with a large proportion
(>50%) of tubules at one or more of these phases, e.g. Nos 39,83, 179 (Table 3) and No. 116
(Table 4), were recrudescing, i.e. in the process of becoming fully spermatogenic.
No. 112, aged 14.5 years from Mornington I., had half its seminiferous tubules at phase
-3, the remainder at phase -2, and a mean single testis weight of 24.6 g, about halfway
between that of the heaviest immature testis (16.5 g) and the lightest of the other
post-immature testes which were not producing spermatids or spermatozoa (the weights of
which ranged from 35 to 96 g; see below). We considered that No. 112 was probably
approaching puberty.
(iii) The intertubular tissue
As in other mammals, dugong testes showing active spermatogenesis contain more fluid
than resting, intermediate or recrudescent testes and are thus more difficult to fix. The
relative volume of testicular intertubular tissue also tends to decrease with increasing
spermatogenic activity, as a result of the concomitant increase in tubule diameter (Table 2).
In comparison with that in dugongs with resting, intermediate or recrudescent testes (Fig. 8),
Dugong Male Reproductive Organs
the intertubular connective tissue of fully spermatogenic dugongs appears conspicuously
reduced (Fig. 10). However, when the relative volume of Leydig cells in testes at various
stages of activity (Table 2) was multiplied by the corresponding testis weight to give an
estimate of total Leydig cell weight per dugong, it showed that the total Leydig cell weight
actually tends to increase with increasing testicular activity (Fig. 11). This has also been
regularly observed in the testes of other eutherians.
This result (Fig. 11) suggests that there is an increase in androgenic activity in animals
approaching puberty, and that androgenic activity in adult dugongs is synchronous with the
activity of the seminiferous epithelium. The relationship between the spermatogenic cycle in
the testes and blood androgen levels clearly needs to be analysed.
(iv) Testis size
The mean testis weight ofa dugong which is approaching puberty or is mature, reflects the
mean testis tubule diameter (Fig. 7) and depends on the animal's age and reproductive stage;
testis weight usually increasing with increased activity. We observed little size difference
between contralateral testes. In general, the size of the testes of dugongs older than 20 years
was greater than that of younger dugongs at the same phase (Table 5). Among dugongs less
than 20 years old, the testes of animals with unerupted tusks were usually lighter than those,
at the same phase, from animals with erupted tusks (Table 5). The single testis weights of
dugongs with testes which were producing spermatozoa ranged from 82.7 to 332 g; testis
--
.... .
30 r
8
m
X m -
9
E
I
10
-
.
.
a
a
.
a
0
2
,
4
6
8
Fig. 11. Relationship between reproductive
activity ranking of male dugongs (see Table 2)
and an estimate of the weight of Leydig cells in
their testes calculated from the following
formula: volumetric fraction of Leydig cells
(O/o) X mean testis weight X 2.
10
Reproductive activity ranking
lengths from 8.7 to 12.5 cm. Three young animals-Nos 36, 38 and 156 (Table 3)-all of
which had epididymides packed with spermatozoa, had single testis weights ranging from
82.7 to 105 g, appreciably less than those ofthe six other dugongs at the same phase, whose
testis weights ranged from 153to 332 g. The tusks ofNo. 38 (9 years) and 156 (10.5 years) had
not erupted; those of No. 36 (12 years) were erupting. In contrast, the testis of another young
dugong with unerupted tusks (No. 146, 10.5 years) weighed 225 g at the same phase of
activity, a weight comparable with that of some of the older dugongs (Nos 2, 109,213)at the
same phase.
(v) Epididymis and ductus deferens
We did not attempt a detailed anatomical study of these organs. However, examination
of the sections of tissue taken at several standardized positions from three mature animals,
each of which were sampled within '/2 h of death, indicates, not unexpectedly, that the
histology of the epididymis changes with the level of testicular activity and generally seems
to conform to descriptions of that in other species (see Nicander 1957a, 19573; Glover and
Nicander 1971; Nicander and Glover 1973).
Nos 2 10 and 2 13 (Figs 12A,128,120), both of which were at the height of spermatogenic
activity, had similar epididymal histology. The density and distribution of spermatozoa
were also similar. Near the cranial pole of the testis, the efferent ductules (Fig. 12A) and the
initial segment of the epididymis contained a low concentration of spermatozoa. In this
H. Marsh, G. E. Heinsohn and T. D. Glover
region, the lumen of the epididymis was lined with a pseudostratified epithelium composed
of principal and basal cells. The principal cells were columnar and 25-65 ym high, with
round basal nuclei about 6 ym in diameter, and basophilic cytoplasm. These cells bore
stereocilia up to 16 ym long. The distribution and occurrence of basal cells was typically
irregular and occasional; the nuclei of these cells were about 8 ym long. The epithelium and
adjacent highly vascular lamina propria were extensively folded into finger-like projections
up to 0.5 mm long, causing the outline of the lumen to be irregular.
These projections became more numerous and elaborate in more distal parts of the
epididymis, reaching their maximum development in the region just above the caudal pole
of the testis (Fig. 12B), where the height of the principal cells ranged up to 80 ym and their
oval nuclei were up to 16 ym long.
Table 5. Summary of details of testes weights of adult dugongs according to stage of testicular
activity
Values of testes weights are means, with ranges in italics; sample sizes in square brackets. No. 68, which had 50%
of tubules each at phase -3 and -4, is classed as -4; No. 112, with 50% each at phase -2 and -3, is classed
as -3
Reproductive
category
Predominant
phase in
testes
Regressed
-4
Resting
-3
Intermediate
-2
Recrudescent
Formalin-fixed testis weight (g)
Dugongs t 2 0 y old
Dugongs >20 y,
Unerupted tusks
Erupting or
erupted tusks
erupted tusks
-I
-1, 0, 1
Fully spermatogenic
Spermiogenesis
finishing
1-5c
6
AClassified as approaching puberty.
BFew or no sperm in epididymis.
CMany sperm in epididymis.
The epididymis gradually became less convoluted near the caudal pole of the testis. In
this region the epididymal lumen was about 1.6 mm in diameter. This appeared to be the
main sperm store (Fig. 120). We believe that this level of the duct is both homologous and
analogous to the terminal segment of the epididymis (Glover and Nicander 1971). The
ductus deferens, which was also fairly well filled with spermatozoa, was entirely straight. The
mucosal folding persisted in the terminal segment and ductus deferens but was less
elaborate. The principal cells were lower, only 16-25 pm high, and, they had shorter nuclei
(6-12 pm high). Basal cells in the ductus deferens were still abundant.
The surrounding muscle layers exhibited a proximodistal increase in thickness
throughout the epididymis. Initially, the contractile cells were very slender and the bundles
they formed were mostly oriented circumferentially. Near the sperm store the layer of
circular fibres was surrounded by a layer of longitudinal and oblique fibres. The ductus
deferens had very thick muscular walls.
The epididymides of No. 145 (Fig. 12C), a mature dugong with resting testes (majority of
tubules at phase - 3), were devoid of spermatozoa and differed histologically from those of
the others in the following aspects: (I) the principal cells were much shorter and lacked most
Dugong Male Reproductive Organs
of their supranuclear cytoplasm, and were also devoid of stereocilia, which is not typical for
eutherians; (2) the epididymal lumen was narrower; (3) the surrounding muscle layer(s) were
thicker.
Fig. 12. (A) Photomicrograph of a transverse section of one of the efferent ductules near the cranial pole
of the testis of a dugong (minimum age 28 years) with fully spermatogenic testes. H and E. (B) An
epithelial fold extending into the lumen of the epididymis of the same animal as Fig. 12A just above the
caudal pole of the testis. Note the stratified epithelium with elongate ciliated principal cells, the round
basal cell nuclei and the lumen packed with spermatozoa. H and E. (C) Cross-section of the epididymis
just above the caudal pole of the testis of an adult dugong (minimum age 27.5 years) with resting testes
(most of the seminiferous tubules at phase - 3). Compare with B and note the complete absence of
spermatozoa and the reduced supranuclear cytoplasm of the principal cells. H and E. (D) Part of the
main sperm store near the caudal pole of the testis. This specimen is from the same animal as Figs 12A
and 12B.
No spermatozoa were observed in the epididymides of any of the 24 adult dugongs whose
testes were not producing spermatids or spermatozoa. Few or no spermatozoa were
observed in the epididymides of five dugongs with active testes, at least three of which
showed only the early stages of spermateliosis, i.e. spermatozoa were not observed but
various stages of spermatid development were evident. Moderate amounts of spermatozoa
were observed in the epididymides of one dugong whose testes were at the end of a period of
spermiogenesis, and large quantities of spermatozoa were observed in the epididymides of
only 10 dugongs, all of which were at the height of spermiogenesis.
(vi) Criteria for classifying a male dugong as approaching puberty
One dugong was classified as approaching puberty. Its single testis weight was less than
30 g, its testis was similar histologically to Figs 8C and 8 0 , and spermatozoa were absent
from its epididymides.
H. Marsh, G. E. Heinsohn and T. D. Glover
(vii) Criteria for classifying a male dugong as mature
A dugong was classified as mature if its single testis weight was 30 g or more and its testis
was similar histologically to Figs 8B-8Fandlor Fig. 10,irrespective ofwhether spermatozoa
were present in its epididymides.
Discussion
Activity of the Testis and Epididymis
Gross and histological studies of the testes and epididymides of 59 dugongs from north
Queensland indicate that adult male dugongs are not always fertile. However, there is no
definite correlation between testicular activity and season (Marsh et al. 19843) and both
spermatogenically active and inactive animals may be found simultaneously in the same
population.
The testes of a large proportion (approximately 60%) of the 41 dugongs that were either
approaching puberty or fully adult showed arrested or incomplete spermatogenesis. These
animals spanned the entire range of the ages and sizes of adult dugongs and included 18
individuals with erupted and worn tusks. (The tusks of male dugongs begin to erupt after
puberty (Table 3; Marsh 1980; Marsh et al. 1984b).We therefore consider it highly unlikely
that all these dugongs were merely approaching puberty, although No. 112 was classifiedas
being at that stage on the basis of its testis weight. At least some mature male dugongs are
thus not in continuous breeding condition in the wild. Some old males have unusually
regressed testes (most tubules at phase -4). These animals are probably sterile for long
periods or even permanently.
The reproductive condition of many of the pubertal and adult dugongs is intermediate
between complete testicular quiescence (phases -4 and -3) with no epididymal
spermatozoa, and full spermatogenesis (phases 1-5) with large numbers of epididymal
spermatozoa (see Tables 3,4; Figs 8, 10).The androgenic and spermatogenic activities of the
testis appear to be in phase (Fig. 11) and only animals with fully spermatogenic testes have
large numbers of spermatozoa in the epididymis.
We are unable to consider the relationship between male reproductive activity and social
status, as no pertinent information on dugong social behaviour is available except for reports
by Australian Aborigines of the presence of 'whistler' dugongs (e.g. Roughsey 1972) which
are considered to be dominant males. If male reproductive activity is dependent on social
factors, the presence of a significant proportion (about 20%) of mature males with testes
entering or leaving the fully spermatogenic state suggests that social status is very fluid.
The breeding pattern of the male dugong in relation to that of the female is discussed in
Marsh et al. (1984b).
Comparison of the Breeding Pattern of the Male Dugong with that of Other
Paenungulates
Eye-lens protein analyses have confirmed the monophyletic origin of the paenungulate
orders Hyracoidea, Sirenia and Proboscidea, and place the paenungulates as the second
offshoot from the main eutherian line after the endentates (de Jong and Zweers 1980). Male
African elephants show evidence of a seasonal cycle in the mean diameter of their
seminiferous tubules (Laws 1969) but, in contrast to the dugong, adults with arrested or
incomplete spermatogenesis have not been reported. However, the male breeding patterns
of the different genera of hyrax (e.g. Procavia and Heterohyrax) have many similarities to
that of the dugong, especially in the histological changes in the testis and epididymis (Glover
and Sale 1968; Millar and Glover 1970). Although away from the equator hyrax
reproduction is markedly seasonal, in the tropics many species do not have a well defined
breeding season, and observations on behaviour of rock species indicate that, as in the north
Queensland dugong (Marsh et al. 1984b),the periods of activity of individual males are not
Dugong Male Reproductive Organs
coordinated (Glover and Sale 1968). Sometimes there are seasons of widespread sexual
activity in a population; conversely,there may be periods of very little sexual activity. There
is also evidence that some male rock hyraxes remain reproductively quiescent for
indefinitely long periods or even permanently (Glover and Sale 1968).
The Dugong as a Testicond Mammal
The dugong's testes are permanently situated in the abdominal cavity immediately
caudal to the kidneys. In this respect they are similar to the testes ofthe other paenungulates,
the elephant (Short et al. 1967) and the hyrax (Glover and Sale 1968). Also like the other
testicond mammals the dugong has a straight testicular artery. However, the structure of the
dugong epididymis suggests the beginnings of some caudal migration. The terminal segment
of the epididymis or conventional sperm store, appears, for example, to be situated just
caudal to the distal pole of the testis and is thus unlike the corresponding structures in the
hyrax and the elephant which (although they exhibit important anatomical differences) are
both remotely situated from the testes in, or near to, the pelvic cavity.
The presumed advantages of testicular migration, which occurs in most eutherian
mammals, have been the subject of much discussion (e.g. Bedford 1977;Carrickand Setchell
1977) but they still remain obscure. Although the detrimental effects of abdominal
temperatures on spermatogenesis are well known, and the susceptibility of spermatozoa in
the epididymis of scrotal mammals to increased temperature has been demonstrated
(Glover 1960; Cummins and Glover 1970), both spermatogenesis and sperm storage
apparently take place unimpeded at abdominal temperature in testicondids.
Could this be, perhaps, because abdominal temperatures in testicond mammals are lower
than those of scrotal mammals? Available evidence does not suggest this although results for
sirenians suggest some thermolability. Gallivan et al. (1983) concluded that Amazonian
manatees Trichechus inunguis have a limited capacity for thermiogenesis. The core
temperature of an animal weighing 210 kg (about the weight of a 2.25-m dugong) ranged
from 34 to 35.6'C at water temperatures between 20 and 34°C. Our unpublished
observations on a dugong showed an abdominal temperature of 37°C after we opened its
body cavity within 25 min of its death by drowning. In contrast, low rectal temperatures
have been recorded for sirenians (Dekeyser 1952; unpublished observation). However, other
large testicondids such as the African elephant (Short et al. 1967), the Indian elephant
Elephas elephas (Waites and Setchell1969, quoted by Carrick and Setchell1977) and various
cetaceans (for references see Irving 1969; Ridgway 1972) have body temperatures which are
as high as those of many scrotal mammals.
The anatomical barriers to testicular descent in cetaceans, and the advantages of their
having abdominal testes, have been discussed by de Smet (1977). His arguments apply
equally to sirenians. Weber (1893) claimed (without firm evidence) that the abdominal
position of the testes in cetaceans was a secondary condition and that their terrestrial
forebears may have had scrotal testes. As the testes remain just caudal to the kidneys in all
paenungulates, their position in sirenians is obviously not a secondary adaptation to the
marine environment.
The gross anatomy of the testes and excurrent ducts of the dugong resembles that of
cetaceans (personal observation) much more closely than it does that of the African elephant
(Short et al. 1967) or the rock hyrax (Glover and Sale 1968). The similarities (relatively
elongate testes each with an adjacent distinct epididymis) reflect the similarities in the
general morphology of the abdominal cavity in these sirenians and cetaceans. This is
dictated by the morphological adaptations ofboth groups to their aquatic environments, e.g.
absence of hind limbs, and associated reduction of the pelvic girdle and generally
streamlined shape.
The microscopic anatomy of the dugong epididymis (Fig. 12) is unusual, with a number
of features similar to that in the other large testicondids, including the African elephant
H. Marsh, G. E. Heinsohn and T. D. Glover
(Short et al. 1967) and various odontocete cetaceans (Simpson and Gardner 1972; Kasuya
1978, personal communication 1979; Kasuya and Brownell 1979). For instance, the
epididymal epithelium and adjacent highly vascular lamina propria are elaborately folded
(Fig. 12), resulting in a very large epithelial surface area with an excellent blood supply. This
arrangement may be functionally significant in relation to the maturation and storage of
spermatozoa at relatively high temperatures, since in most eutherian mammals these events
occur at lower temperatures in a scrotum. Yet, maturation of spermatozoa can occur at body
temperature where no such specialized epididymal structure exists (Glover 1973; Bedford
1977). However, the elaborately folded epithelium could be associated with sperm storage
since, in the dugong, the sperm store is more deeply situated within the abdomen than it is in,
for example, the rock hyrax. Alternatively, the folding may be necessitated by the diameter
of the epididymis in large testicondids, which is at least twice that in large scrota1
mammmals (personal observation).
Our examinations of the epididymis of the dugong have so far been relatively superficial,
largely because of problems of fixation, and the maturation and storage of spermatozoa in
this species warrant closer scrutiny.
Concluding Remarks
In the absence of a sharply defined breeding season, the variation in testicular and
epididymal activity between individual dugongs documented in this study raises many
interesting questions. For example, does the dugong's apparent lack of a pineal gland (Dexler
1912) render it unable to show seasonal changes in reproductive activity induced by
photoperiod? Although some of these questions could be answered by a carcass study on a
larger sample, field and laboratory studies of known individuals over several years will be
necessary to understand the reproductive biology of the male dugong. Monitoring the
hormonal status of animals whose reproductive behaviour is being studied would be
particularly valuable.
Acknowledgments
Dugong research at James Cook University has been financed by the Australian National
Parks and Wildlife Service, the Australian Research Grants Commission and the Marine
Sciences and Technologies Grants Scheme. We acknowledge the assistance of the many
people, particularly Dr A. V. Spain and Messrs B. R. Gardner and P. W. Channells, who
have helped in collecting dugong specimen material since 1969. We also thank Messrs L.
Winsor and P. Osmond for technical assistance with histology, Mr R. Hall, Ms P. Moreton,
Ms S. Francis, Mr Z. Florian and the Photography Department at James Cook University for
assistance with the figures, and Professors J. M. Taylor, P. K. Anderson and R. P. Kenny,
Drs D. Domning and G. Rathbun, and two anonymous referees for their critical comments
on the manuscript, which was largely completed while the senior author was a visiting
investigator at the University of British Columbia.
References
Bedford, J. M. (1977). Evolution of the scrotum: the epididymis as prime mover. In 'Reproduction and
Evolution: Proceedings of the Fourth International Symposium on Comparative Biology of
Reproduction'. (Eds J. H. Calaby and C. H. Tyndale-Biscoe.) pp. 171-82. (Australian Academy of
Science: Canberra.)
Bloom, W., and Fawcett, D. W. (1975). 'A Textbook of Histology.' 10th Ed. (W. B. Saunders Co.:
Philadelphia, London, Toronto.)
Carrick, F. N., and Setchell, B. P. (1977). The evolution of the scrotum. In 'Reproduction and
Evolution: Proceedings of the Fourth International Symposium on Comparative Biology of
Reproduction'. (Eds J. H. Calaby and C. H. Tyndale-Biscoe.) pp. 165-70. (Australian Academy of
Science: Canberra.)
Dugong Male Reproductive Organs
Courot, M., Hochereau-de Reviers, M.-T., and Ortavant, R. (1970).Spermatogenesis. In 'The Testis'.
(Eds A. D. Johnson, W . R. Gomes and N . L. Vandemark.) Vol. 1 , pp. 339-432. (Academic Press:
New York and London.)
Culling, C. F. A. (1974). 'Handbook o f Histopathological Techniques.' 3rd Ed. (Buttenvorths:
London.)
Cummins, J. M., and Glover, T . D. (1970). Artificial cryptorchidism and fertility in the rabbit.
J. Reprod. Fertil. 23, 423-33.
Dekeyser, P. L. (1952).Note sommaire sur la temperature rectal du Lamantin (Trichechussenegalensis
Link). Bull. Mus. Natl Hist. Nut. 24, 243-6.
de Jong,W . W., and Zweers, A. (1980).Study o f eye lens protein confirms relationship o f manatee with
hyraxes and elephants. Acta Amazonica 10, 897-902. [In Portuguese.]
de Smet, W . M. A. (1977).The position o f the testes in cetaceans. In 'Functional Anatomy o f Marine
Mammals'. (Ed. R. J. Harrison.) Vol. 3, pp. 361-86. (Academic Press: London.)
Dexler, H. (1912).Das Him von Halicore dugong Erxl. Morphol. Jahrb. 45, 97-189.
Gallivan, G. J., Best, R. C., and Kanwisher, J . W . (1983).Temperature regulation in the Amazonian
manatee (Trichechus inunguis): A preliminary study. Physiol. 2001.56, 255-62.
Glover, T . D. (1960).Spermatozoa from the isolated cauda epididymis o f rabbits and some effectso f
artificial cryptorchidism. J. Reprod. Fertil. 1, 121-9.
Glover, T . D. ( 1 973).Aspects o f sperm production in some East African mammals. J. Reprod. Fertil. 35,
45-53.
Glover, T . D., and Nicander, L. (1971).Some aspects o f the structure and function o f the mammalian
epididymis. J. Reprod. Fertil. Suppl. 13, 39-50.
Glover, T . D., and Sale, J . B. (1968). The reproductive system o f male rock hyrax (Procavia and
Heterohyrax). J. 2001.(Lond.) 156, 351-62.
Gohar, H. A. F. (1957).The Red Sea dugong. Publ. Mar. Biol. Stn A1 Ghardaqa 9, 1-49.
Hamson, R. J . (1969).Reproduction and reproductive organs. IV. Sirenia. In 'The Biology o f Marine
Mammals'. (Ed. H. T . Andersen.) pp. 336-42. (Academic Press: New York.)
Heinsohn, G. E. (1972).A study o f dugongs (Dugongdugong)in northern Queensland, Australia. Biol.
Conserv. 4, 205-13.
Heinsohn, G. E., Lear, R. J.,Bryden, M. M., Marsh, H., and Gardner, B. R. (1978).Discovery o f a large
resident population o f dugongs o f fBrisbane, Australia. Environ. Conserv. 5, 91-2.
Heinsohn, G. E., and Spain, A. V . (1974).Effectso f a tropical cyclone on littoral and sub-littoral biotic
communities and on a population o f dugongs (Dugong dugon (Miiller)). Biol. Conserv. 6,
143-52.
Heinsohn, G. E., Spain, A. V . , and Anderson, P. K. (1976). Populations o f dugongs
(Mammalia:Sirenia): Aerial survey over the inshore waters o f tropical Australia. Biol. Conserv.
9, 21-3.
Hill, W . C. 0. (1945).Notes on the dissection o f two dugongs. J. Mammal. 26, 153-75.
Home, E. (1820).Particulars respecting the anatomy o f the dugong. Philos. Trans. R. Soc. Lond. 2,
3 15-23.
Irving, L. (1969).Temperature regulation in marine mammals. In 'The Biology o f Marine Mammals'.
(Ed. H. T . Andersen.) pp. 147-75. (Academic Press: New York.)
Johnson, 0 . W., and Buss, I. 0. (1967). The testis o f the African elephant (Loxodonta africana).
I . Histological features. J. Reprod. Fertil. 13, 11-21.
Kasuya, T . (1978).The life history o f Dall's porpoise with special reference to the stock o f the Pacific
coast o f Japan. Sci. Rep. Whales Res. Inst. Tokyo 30, 1-64.
Kasuya, T., and Brownell, R. L. Jr (1979). Age determination, reproduction and growth o f the
franciscana dolphin, Pontoporia blainvillei. Sci. Rep. Whales Res. Inst. Tokyo 31, 45-67.
Laws, R. M. (1969). Aspects o f reproduction in the African elephant (Loxodonta africana).
J. Reprod. Fertil. Suppl. 6, 193-21 7.
Ligon, S. H. (1976). A survey o f dugongs (Dugong dugon) in Queensland. J. Mammal. 57, 580-2.
Lillie, R. D., and Fullmer, H . M. (1976). 'Histopathologic Technique and Practical Histochemistry.'
(McGraw-Hill:New York.)
Marsh, H . (1980). Age determination o f dugong (Dugong dugon (Miiller)) and its biological
implications. In 'Age Determination o f Toothed Cetaceans and Sirenians'. [Int.Whaling Comm.
Rep. (Spec. Issue 3).] pp. 181-201.
Marsh, H., Gardner,B. R., and Heinsohn, G. E. ( 1 98 1). Present-day hunting and distribution of dugongs
in the Wellesley Islands (Queensland): Implications for conservation. Biol. Conserv. 19, 255-67.
H. Marsh, G. E. Heinsohn and T. D. Glover
Marsh, H., Heinsohn, G.E., and Channells, P.W. (1984~).Changes in the ovaries and uterus of the
dugong, Dugong dztgon (Sirenia: Dugongidae) with age and reproductive activity. Aust. J. Zool. 32,
734-66
Marsh, H., Heinsohn, G. E., and Marsh, L. M. (1984b). Breeding cycle, life history and population
dynamics of the dugong, Dugong dugon (Sirenia: Dugongidae). Aust. J. Zool. 32, 767-87.
Marsh, H., Heinsohn, G. E., and Spain, A. V. (1977). The stomach and duodenal diverticula of the
dugong (Dugongdugon). In 'Functional Anatomy of Marine Mammals'. (Ed. R. J. Harrison.) Vol. 3.
pp. 271-95. (Academic Press: London.)
Millar, R. P., and Glover, T. D. (1970). Regulation of seasonal sexual activity in an ascrotal mammal,
the rock hyrax, Procavia capensis. J. Reprod. Fertil. Suppl. 19, 203-20.
Nicander, L. (1957~).On the regional histology and cytochemistry of the ductus epididymis in rabbits.
Acta Morphol. Neerl. -Stand. 1, 99-1 18.
Nicander, L. (1957b). Studies on the regional histology and cytochemistry of the ductus epididymis in
stallions, rams and bulls. Acta Morphol. Need-Scand. 1, 337-362.
Nicander, L., and Glover, T. D. (1973). Regional histology and fine structure of the epididymal duct in
the golden hamster (Mesocricetus auratus). J. Anat. 114, 347-64.
Petit, G. (1925). Recherches anatomiques sur l'appareil genito-urinaire male des sirenians. Arch.
Morphol. Gen. Exp. 23, 1-326.
Ridgway, S. H. (1972). Homeostasis in the aquatic environment. In 'Mammals of the Sea. Biology and
Medicine'. (Ed. S. H. Ridgway.) pp. 590-747. (Charles C. Thomas: Springfield, Illinois.)
Riha, A. (1911). Das mannliche urogenital System von Halicore dugong Erxl. Z. Morphol.
Anthropol. 13, 395-422.
Roosen-Runge, E. C., and Giesel, L. 0. (1950). Quantitative studies on spermatogenesis in the albino
rat. Am. J. Anat. 87, 1-30.
Roughsey, D. (1972). 'Moon and Rainbow.' (A. H. and A. W. Reed: Sydney.)
Rowley, M. J., and Heller, C. C. (1966). The testicular biopsy: surgical procedure, fixation and staining
technics. Fertil. Steril. 17, 177-86.
Schaefer, A. (1970). The mathematical basis of stereology. Microskopion 18+ 19, 3-16.
Short, R. V., Mann, T., and Hay, M. F. (1967). Male reproductive organs of the African elephant,
Loxodonta africana. J. Reprod. Fertil. 13, 5 17-36.
Simpson, J. G., and Gardner, M. B. (1972). Comparative microscopic anatomy of selected marine
mammals. In 'Mammals of the Sea. Biology and Medicine'. (Ed. S. H. Ridgway.) pp. 298-418.
(Charles C. Thomas: Springfield, Illinois.)
Thornback, J., and Jenkins, M. (1982). The IUCN Mammal Red Data Book, Part 1. (IUCN: Gland,
Switzerland.)
Weber, M. (1893). 'Studien uber Saugetiere.' (Fischer: Jena.)
Manuscript received 6 May 1983; accepted 26 April 1984